VISVESVARAYA TECHNOLOGICAL UNIVERSITY

"Jnana Sangama", Belgavi-590 018, Karnataka, India

A
PROJECT REPORT
On
“Novel Approached Flood Forecasting System Using
Machine Learning and Big Data”
Submitted in Partial Fulfillment of the requirements for the award ofthe degree of

BACHELOR OF ENGINEERING
IN
COMPUTER SCIENCE AND ENGINEERING

Submitted By

SWAPNA K 1SJ18CS109
YARRAMSETTY SYONA SREE 1SJ18CS122
NAMITHA B C 1SJ18CS125
SHIVADARSHINI R 1SJ19CS412

Carried out at
B G S R&D Centre,
Dept of CSE,
SJCIT

Under the guidance of

Prof. Ajay N B. E, M. Tech (Ph.D.)
ASSISTANT PROFESSOR
Dept. of CSE, SJCIT

S J C INSTITUTE OF TECHNOLOGY
DEPARTMENT OF COMPUTER SCIENCE AND ENGINEERING
CHIKKABALLAPUR-562 101
2021-2022
||Jai Sri Gurudev||
Sri Adichunchanagiri Shikshana Trust®

S.J.C INSTITUTE OF TECHNOLOGY, Chickballapur - 562101

Department of Computer Science and Engineering

CERTIFICATE

This is to certify that the project work entitled “Novel Approached Flood Forecasting
System Using ML and BD” is a bonafied work carried out by Swapna K (1SJ18CS109),
Yarramsetty Syona Sree(1SJ18CS122), Namitha B C(1SJ18CS125), Shivadarshini R
(1SJ19CS412) in partial fulfillment for the award of Bachelor of Engineering in
Computer Science and Engineering of Visvesvaraya Technological University,
Belagavi during the year 2021-2022. It is certified that all corrections/suggestions
indicated for internal assessment have been incorporated in the report. The project report
has been approved as it satisfies the academic requirements with respect to project work
prescribed for the Bachelor of Engineering degree.

................................. ................................... ..................................

Signature of Guide Signature of HOD Signature of Principal
Prof. Ajay N Dr. Manjunatha Kumar B H Dr. G T Raju
Assistant Professor, Professor & HOD, Principal, SJCIT,
Dept. of CSE, SJCIT Dept. of CSE,SJCIT Chickballapur

External Examiners:
Name of the Examiners Signature with Date

1………………………………… …………………………………

2…………………………………. …………………………………
.
DECLARATION

We Swapna K (1SJ18CS109), Yarramsetty Syona Sree (1SJ18CS122), Namitha B C

(1SJ18CS125), Shivadarshini R (1SJ19CS412) Student of VIII semester B.E in Computer
Science and Engineering at S J C Institute of Technology, Chickballapur, hereby declare that
this dissertation work entitled “Novel Approached Flood Forecasting System Using Machine
Learning and Big Data” has been carried out at B.G.S R&D Centre, Dept. of CSE, SJCIT under
the guidance of guide Prof. Ajay N, Assistant Professor, Dept. of CSE, S J C Institute of
Technology, Chickballapur and submitted in the partial fulfilment for the award of degree
Bachelor of Engineering in Computer Science and Engineering of Visvesvaraya Technological
University, Belagavi during the academic year 2021-2022. We further declare that the report
had not been submitted to another university for the award of any other degree.

Place : Chikkaballapur Swapna K (1SJ18CS109)

Date: Yarramsetty Syona Sree (1SJ18CS122)
Namitha B C (1SJ18CS125)
Shivadarshini R (1SJ19CS412)

i
ABSTRACT

Flood is one of the most disruptive natural hazards, responsible for loss of lives and damage to
properties. A number of cities are subject to monsoons influences and hence face the disaster
almost every year. Early notification of flood incident could benefit the authorities and public to
devise both short and long terms preventive measures, to prepare evacuation and rescue mission,
and to relieve the flood victims. Geographical locations of affected areas and respectiveseverities,
for instances, are among the key determinants in most flood administration. Thus far,an effective
means of anticipating flood in advance remains lacking. Existing tools were typically based on
manually input and prepared data. The processes were tedious and thus prohibitive for real-time
and early forecasts. Furthermore, these tools did not fully exploit more comprehensive
information available in current big data platforms. Therefore, this paperproposes a novel flood
forecasting system based on fusing meteorological, hydrological,geospatial, and crowdsource
big data in an adaptive machine learning framework. Data intelligence was driven by state of-
the-art learning strategies. Subjective and objective evaluations indicated that the developed
system was able to forecast flood incidents, happeningin specific areas and time frames. It was
also later revealed by benchmarking experiments that the system configured with an MLP ANN
gave the most effective prediction, with correct percentage, Kappa, MAE and RMSE of 97.93,
0.89, 0.01 and0.10, respectively.

ii
ACKNOWLEDGEMENT

With reverential pranam, we express my sincere gratitude and salutations to the feet
of his holiness Paramapoojya Jagadguru Byravaikya Padmabhushana Sri Sri Sri Dr.
Balagangadharanatha Maha Swamiji, his holiness Paramapoojya Jagadguru Sri Sri Sri
Dr. Nirmalanandanatha Maha Swamiji, and Sri Sri Mangalnath Swamiji , Sri
Adichunchanagiri Mutt for their unlimited blessings.
First and foremost We wish to express my deep sincere feelings of gratitude to our
institution, Sri Jagadguru Chandrashekaranatha Swamiji Institute of Technology, for
providing us an opportunity for completing the Project Work successfully.
We extend deep sense of sincere gratitude to Dr. G T Raju, Principal, S J C Institute
of Technology, Chickballapur, for providing an opportunity to complete the Project Work .
We extend special in-depth, heartfelt, and sincere gratitude to HOD Dr. Manjunatha
Kumar B H, Head of the Department, Computer Science and Engineering, S J C Institute
of Technology, Chickballapur, for his constant support and valuable guidance ofthe Project
Work .
We convey our sincere thanks to Project Guide Prof Ajay N , AssistantProfessor,
Department of Computer Science and Engineering, S J C Institute ofTechnology, for
his constant support, valuable guidance and suggestions of the Project Work.
We also feel immense pleasure to express deep and profound gratitude to Project Co
–ordinators Prof. Pradeep Kumar G M and Prof. Shrihari M R, Assistant Professors,
Department of Computer Science and Engineering, S J C Institute of Technology, for their
guidance and suggestions of the Project Work.
Finally, We would like to thank all faculty members of Department of Computer
Science and Engineering, S J C Institute of Technology, Chickaballapur for their support.
We also thank all those who extended their support and co-operation while bringing out
this Project Work.
Swapna K (1SJ18CS109)
Yarramsetty Syona Sree (1SJ18CS122)
Namitha B C (1SJ18CS125)
Shivadarshini R (1SJ19CS412)

iii

TTE
CONTENTS
Declaration i
Abstract ii
Acknowledgement iii
Contents iv
List of Figures v
List of Tables vi

Chapter No Chapter Title Page No

1 INTRODUCTION 1-4
1.1 Overview 1
1.2 Problem Statement 2
1.3 Significance and Relevance of Work 3
1.4 Objectives 3
1.5 Methodology 4

2 LITERATURE SURVEY 5-11

2.1 Literature Survey 5

3 SYSTEM REQUIREMENTS AND SPECIFICATION 12-15

3.1 System overview 12
3.2 Functional Requirements 12
3.3 Non-Functional Requirements 13
3.4 Performance Requirement 14

4 SYSTEM ANALYSIS 16-17

4.1 Existing System 16

4.2 Proposed System 17

5 SYSTEM DESIGN 18-22

5.1 Data Flow Diagram 18
5.2 Activity Diagram 20

5.3 Use Case Diagram 21

6 IMPLEMENTATION 23-26
6.1 Algorithm/Pseudo code module wise 23

7 TESTING 27-30
7.1 Methods of Testing 27

iv
7.1.1 Unit Testing 27
7.1.2 IntegrationTesting 27
7.1.3 Functional Testing 28
7.1.4 ValidationTesting 28
7.2 Test cases 29

8 PERFORMANCE ANALYSIS 31-33

9 CONCLUSION AND FUTURE ENHANCEMENT 34

BIBLIOGRAPHY 35-39

APPENDIX 40-45
Appendix A: Screen Shorts 40
Appendix B: Abbreviations 45
PAPER PUBLICATION DETAILS

v
LIST OF FIGURES

Figure No Name of the Figure Page No.

Figure 1.1 Conceptual diagram of the proposed scheme 04

Figure 5.1 Data Flow Diagram Level 0 18

Figure 5.2 Data Flow Diagram Level 1 19

Figure 5.3 Data Flow Diagram Level 2 19

Figure 5.4 Activity Diagram 20

Figure 5.5 Use case Diagram 1 21

Figure 5.6 Use Case Diagram 2 22

Figure 6.1 ML powered prediction of flood 24

Figure 6.2 Current precipitation 25

Figure 6.3 Damage analysis 25

Figure 6.4 ML powered Prediction 26

Figure 6.5 Flood prediction of a city 26

Figure A.1 Home page 37
Figure A.2 Interactive Plots for Flood Prediction 37
Figure A.3 Interactive plot for damage analysis 38
Figure A.4 Interactive plot for precipitation 38
Figure A.5 Interactive heatmaps for damage analysis 39
Figure A.6 Interactive heatmaps for precipitation 39
Figure A.7 Satellite image analysis for Bangalore in May 40
Figure A.8 Satellite image analysis for Bangalore in July 40
Figure A.9 Prediction analysis of Bangalore 41
Figure A.10 Prediction analysis of Chennai 41
Figure A.11 Prediction analysis of Mumbai 42

vi
LIST OF TABLES

Table No Name of the Tables Page No.

Table 7.1 Unit Testing for Live Data set 29

Table 7.2 Integration Test Preprocessing of Dataset 29

Table 7.3 Validation testing for Heatmaps 30

Table 7.4 Functional testing for Prediction of flood for a particular city 30

vii
CHAPTER 1
INTRODUCTION
1.1 Overview

Natural flood is one of the most recurrent disasters. Unlike stagnant water discharge,
occasionally experienced in poorly planned cities, major flood incidents always cause
considerable damages to properties and, more often than not, loss of lives. Several Asian
countries, particularly Thailand, are subject to both southwest and northeast monsoons and
accordingly facing seasonal deluge almost every year and in most parts of the countries.

Among notable causes, sudden and enduring heavy rain is the most pertinent one in Thailand.
Furthermore, overflow from main rivers along shore sides to surrounding basins can greatly
spread the damages. Although being located further away from a river, an area with
inappropriate land uses are unable to efficiently discharge accumulated precipitation, and hence
are inevitably prone to even more frequent floods. Regardless of causes, however, a flood is
generally sudden and thus almost formidable for the general public and relevant organization
to be adequately prepared for the incident. This is mainly due to the lack of aneffective means
of anticipating the disaster well in advance.

Despite the recent extensive development of computerized flood forecasting systems, they
remained based primarily on present precipitation, monitored by rain stations or rain gauges.
These facilities are normally owned by a meteorology department or similar organizations.
Besides, they are scantly located in a few areas due to costly installation and maintenance.
Hence, it is difficult to determine precipitation or predict flood accurately, especially in areas
with no such facility. To remedy this issue, precipitation in these areas were typically estimated
either by inter- or extrapolation from those with rain stations present
.
Due to a limited number of these stations and those readings in one area may not be a good
representative to others. Therefore, estimated precipitation was insufficiently accurate tomake
a realistic forecast. Conventional meteorological readings, e.g., precipitation, temperature, and
humidity, etc., took really long time to measure, process, record, andtransfer to relevant
organizations . Analyses based on past precipitation were known to be associated with several
short comings.

1
Novel Approached Flood Forecasting System using ML and BD Introduction

For instance, they contribute to inaccurate and often outdated flood prediction. Limited sample
size, inadequate computing capability, and inefficient prediction methods were all undermining
the real potential of this scheme. Nonetheless, with the recent advances in distributed
computing and especially modern machine learning (ML), resembling human intelligence,
computerized flood forecasting, based on thematic factors has widely been investigated. In
addition, as the number of both open and proprietarydata providers escalates, Big Data has now
become a central source of information in such pursuits

Thus far, according to recent surveys, most flood forecasting systems relied primarily on either
monitored precipitation data or those obtained from a single source. Beside the mentioned
limitations, existing systems remained lacking in other various aspects. For example, in a case
where monitoring facilities or communication network of ones became malfunctioned, there
would be no precipitation data available for imperative analyses. To the best of our knowledge,
there was also no tool (software) that can accommodate area specific forecasting well in
advance. Furthermore, existing tools were highly dependent on demanding data preparation
and compilation from various sources, including Big Data. As a consequent, automated and
spontaneous notification of flood incidents to the public and authorities, or realistic anticipation
of ones has remained a grand challenge.

In addition, there have been recent developments in flood forecasting systems based on ML.
These systems embedded both attributes and crowd sourcing data into their ML frameworks.
However, most existing systems operate by analyzing these data offline on premise before
presenting their prediction results on various platforms. A typical practice was proposed in,
where an ML was trained with real-time rainfalls, streamflow, and other data. It was unclear,
nonetheless, how a prediction result was verified against an actual event, which was obtained
from crowd sourcing.

1.2 Problem Statement

• Flood is one of the most disruptive natural hazards, responsible for loss of lives and
damage to properties. A number of cities are subject to monsoons influences and hence
face the disaster almost every year.
• Early notification of flood incident could benefit the authorities and public to devise

Dept. Of CSE SJCIT 2 2021-22

Novel Approached Flood Forecasting System using ML and BD Introduction

• Short and long terms preventive measures, to prepare evacuation and rescue mission,
and to relieve the flood victims.
• Existing tools were typically based on manually input and prepared data The processes
were tedious and thus prohibitive for real-time and early forecasts.
• Existing tools did not fully exploit more comprehensive information available incurrent
big data platforms.
• This project proposes a novel flood forecasting system based on fusing meteorological,
hydrological, geospatial, and crowdsource big data in an adaptive machine learning
framework
• Data intelligence was driven by state of-the-art learning strategies. Subjective and
objective evaluations indicated that the developed system was able to forecast flood
incidents, happening in specific areas and time frames.
• System will be configured with an MLP ANN give the most effective prediction.

1.3 Significance and Relevance of Work

• To elucidate the merits of the proposed scheme, the experiments were carried out on
Chennai coastal places. We will consider the geographical information for the
preparation of dataset.
• These areas are under influences of both southeast and northwest monsoons. As a
consequence, they both are prone to heavy floods.
• In this approach would show that no restriction on data nor fundamental processes was
imposed regarding these specific provinces. Therefore, the proposed scheme could be
generalized and applied equally well to other areas

1.4 Objectives
To collect the flood forecasting data such as
1. Geospatial,
2. Meteorological and hydrological, obtained from GLOFAS,
3. Crowdsource (or volunteer) data.

• Normalization of data and make a common coordinate frame, a pre-processing need to

be done. Data integration pre-processed data.

• Prediction and evaluating the results using MLP ANN, RF, SVM.

Dept. Of CSE SJCIT 3 2021-22

Novel Approached Flood Forecasting System using ML and BD Introduction

1.5 Methodology
• This project proposed a novel distributed flood forecasting system, based on integrating
meteorological, hydrological, geospatial, and crowd source data.
• Big data made available by prominent agencies were acquired by means of various cross
platform APIs. Forecasting was performed based on these data learned bymodern ML
strategies.
• They were decision tree, RF, Naïve Bayes, MLP and RBF ANN, SVM, and fuzzy logics
• It was elucidated empirically that the developed system could be used to alert the public
and authorities alike of not only a current flood but also future ones.
• This system also enhanced user experience via responsive graphical interfaces,
interoperable on different computing devices including mobiles
• This advantage effectively encouraged greater contribution of crowdsource data from
the public, enriching data aggregation and hence increasing system accuracy and
reliability.

Figure 1.1: Conceptual diagram of the proposed scheme

Dept. Of CSE SJCIT 4 2021-22

CHAPTER – 2
LITERATURE SURVEY

2.1 Flood Forecasting System Based on Integrated Big and Crowdsource Data
by Using Machine Learning Techniques.[48]
Authors : S. Puttinaovarat and P. Horkaew

Flood is one of the most disruptive natural hazards, responsible for loss of lives and damage to
properties. A number of cities are subject to monsoons influences and hence face the disaster almost
every year. Early notification of flood incident could benefit the authorities and public to devise both
short and long terms preventive measures, to prepare evacuation and rescue mission, and to relieve
the flood victims. Geographical locations of affected areas and respective severities, for instances,
are among the key determinants in most flood administration. Thus far, an effective means of
anticipating flood in advance remains lacking. Existing tools were typically based on manually input
and prepared data. The processes were tedious and thus prohibitive for real-time and early forecasts.
Furthermore, these tools did not fully exploit more comprehensive information available in current
big data platforms. Therefore, this paper proposes a novel flood forecasting system based on fusing
meteorological, hydrological, geospatial, and crowdsource big data in an adaptive machine learning
framework. Data intelligence was driven by state of-the-art learning strategies. Subjective and
objective evaluations indicated that the developed system was able to forecast flood incidents,
happening in specific areas and time frames.
Disadvantages:-
➢ Data considered in this study, such as GLOFAS, were not of intrinsically high
spatial resolution

2.2 Flood Prediction Using Machine Learning Models: Literature Review

Authors : Amir Mosavi , Pinar Ozturk and Kwok-wing Chau[49]
The research on the advancement of flood prediction models contributed to risk reduction, policy
suggestion, minimization of the loss of human life, and reduction of the property damage associated
with floods. To mimic the complex mathematical expressions of physical processes of floods, during
the past two decades, machine learning (ML) methods contributed highly in the advancement of
prediction systems providing better performance and cost- effective solutions. Dueto the vast
benefits and potential of ML, its popularity dramatically increased among hydrologists.

5
Novel Approached Flood Forecasting System Using ML & BD Literature Survey

Researchers through introducing novel ML methods and hybridizing of the existing ones aim at
discovering more accurate and efficient prediction models. The main contribution of this paper is to
demonstrate the state of the art of ML models in flood prediction and to give insight into the most
suitable models. In this paper, the literature where ML models were benchmarked through a
qualitative analysis of robustness, accuracy, effectiveness, and speed are particularly investigated to
provide an extensive overview on the various ML algorithms used in the field. As a result, this paper
introduces the most promising prediction methods for both long-term and short-term floods.
Furthermore, the major trends in improving the quality of the flood prediction models are
investigated.
Disadvantages :-
➢ This paper suggests that the drawbacks to major ML methods in terms of accuracy,
uncertainty, performance, and robustness
➢ Long-term prediction

2.3 Building an Intelligent Hydroinformatics Integration Platform for Regional

Flood Inundation Warning Systems.[50]
Authors:- Li-Chiu Chang and Fi-John Chang
This paper first summarizes the ML methods proposed in this special issue for flood forecasts and
their significant advantages. Then, it develops an intelligent hydro informatics integration platform
(IHIP) to derive a user-friendly web interface system through the state-of-the-art machine learning,
visualization and system developing techniques for improving online forecast capability and flood
risk management. The holistic framework of the IHIP includes five layers (data access, data
integration, servicer, functional subsystem, and end-user application) and one database for
effectively dealing with flood disasters. The IHIP provides real-time flood-related data, such as
rainfall and multi-step-ahead regional flood inundation maps. The interface of Google Maps fused
into the IHIP significantly removes the obstacles for users to access this system, helps communities
in making better-informed decisions about the occurrence of floods, and alerts communities in
advance. The IHIP has been implemented in the Tainan City of Taiwan as the study case. The
modular design and adaptive structure of the IHIP could be applied with similar efforts to other cities
of interest for assisting the authorities in flood risk management.

Disadvantages:-
➢ Web server deals merely with sending requests and responses between clients and the
IHIP.

Dept. of CSE, SJCIT 6 2021-22

Novel Approached Flood Forecasting System Using ML & BD Literature Survey

2.4 Flood Hydrograph Prediction Using Machine Learning Methods[51]

Authors :- GokmenTayfur , Vijay P. Singh, TommasoMoramarco and SilviaBarbetta
This study discusses the application of the artificial neural network (ANN), the genetic algorithm
(GA), the ant colony optimization (ACO), and the particle swarm optimization (PSO) methods for
flood hydrograph predictions. Flow field data recorded on an equipped reach of Tiber River, central
Italy, are used for training the ANN and to find the optimal values of the parameters of the rating
curve method (RCM) by the GA, ACO, and PSO methods. Real hydrographs are satisfactorily
predicted by the methods with an error in peak discharge and time to peak not exceeding, on average,
4% and 1%, respectively. In addition, the parameters of the Nonlinear Muskingum Model (NMM)
are optimized by the same methods for flood routing in an artificial channel. Flood hydrographs
generated by the NMM are compared against those obtained by the numerical solutions of the St.
Venant equations. Results reveal that the machine learning models (ANN, GA, ACO, and PSO) are
powerful tools and can be gainfully employed for flood hydrograph prediction.They use less and
easily measurable data and have no significant parameter estimation problem.
Disadvantages :-
➢ Delay should be improved

2.5 Development of Heavy Rain Damage Prediction Model Using Machine

Learning Based on Big Data.[42]
Authors :- Changhyun Choi, Jeonghwan Kim
We developed a model for the prediction of heavy rain damage based on the big data provided by
the Korea Meteorological Administration and machine learning that can maximize the prediction
performance of the model, and the model could be used in implementing a proactive disaster
management system. However, this study has some limitations on the number of damage data and
the use of hydrometeorological data. We used heavy rain damage data of 22 years from 1994 to 2015,
which were provided in Annual Natural Disaster Report published by the Korean government, as the
dependent variables of the model. Actually, there are more damage data but the reliable data were
used in this study. Therefore, if we have more data, we can get better prediction performance of the
model. Also this study just used hydrometeorological data such as temperature, precipitation,
humidity, and so on as the independent variables of the model. Therefore, we may need more data
which are related to disaster prevention projects, disaster recovery budget, and socioeconomic factors
such as increasing ratio of impervious area, per capita income, and ratio of vulnerable populations
of local governments. Taking into account such damage- related factors in

Dept. of CSE, SJCIT 7 2021-22

Novel Approached Flood Forecasting System Using ML & BD Literature Survey

addition to hydrometeorological factors will make it possible to develop a more reliable prediction
model for heavy rain damage.
Disadvantages :-
➢ This study has some limitations on the number of damage data and the use of
hydrometeorological data
➢ System is unstable

2.6 Designing Deep-Based Learning Flood Forecast Model With ConvLSTM

Hybrid Algorithm[31]
Author :- Mohammed Moishin , Ravinesh C. Deo, Ramendra Prasad , Nawin Raj , and Shahab
Abdulla
Efficient, robust, and accurate early flood warning is a pivotal decision support tool that can help
save lives and protect the infrastructure in natural disasters. This research builds a hybrid deep
learning (ConvLSTM) algorithm integrating the predictive merits of Convolutional Neural Network
(CNN) and Long Short-Term Memory (LSTM) Network to design and evaluate a flood forecasting
model to forecast the future occurrence of flood events. Derived from precipitation dataset, the work
adopts a Flood Index (IF ), in form of a mathematical representation, to capture the gradual depletion
of water resources over time, employed in a flood monitoring system to determine the duration,
severity, and intensity of any flood situation. The newly designed predictive model utilizes
statistically significant lagged IF , improved by antecedent and real-time rainfall data to forecast the
next daily IF value. The performance of the proposed ConvLSTM model is validated against 9
different rainfall datasets in flood prone regions in Fiji which faces flood-driven devastations almost
annually. The results illustrate the superiority of ConvLSTM- based flood model over the benchmark
methods, all of which were tested at the 1-day, 3-day, 7- day, and the 14-day forecast horizon. For
the next best model, the RMSE values were 0.105, 0.154,
0.213 and 0.282 in that same order for the four forecast horizons. In terms of the difference in model
performance for individual stations, the Legate-McCabe Efficiency Index (LME) were 0.939, 0.898,
0.832 and 0.726 for the four forecast horizons, respectively.
Disadvantage :-
➢ Potential use in disaster management and risk mitigation in the current phase of extreme
weather events
➢ Drawbacks to major methods in terms of accuracy, uncertainty, performance, and robustness

Dept. of CSE, SJCIT 8 2021-22

Novel Approached Flood Forecasting System Using ML & BD Literature Survey

2.7 Prediction of Flood by Rainfall using MLP Classifier of Neural Network

Model[12]
Author :- P. Vijayalakshmi , M. Vandhana, K. Vasanthapriya, P.B. Thushikka
Floods are among the most destructive natural disasters, which are highly complex to model. The
research on the advancement of flood prediction models contributed to risk reduction, policy
suggestion, minimization of the loss of human life, and reduction of property damage which are all
associated with floods. To mimic the complex mathematical expressions of physical processes of
floods, during the past two decades, neural network methods contributed highly in the advancement
of prediction systems providing better performance and cost-effective solutions. To prevent this
problem and to predict flood using rainfall dataset, the neural network based techniques are used.
The analysis process started from variable identification, missing value treatments, data validation
and data cleaning/preparing, exploratory analysis and evaluation. Various modules are built
signifying the use of each of the analytical process made. Machine learning algorithms are made
used for prediction. The Algorithms like, Logistic Regression, Support Vector Machines, K-Nearest
Neighbor, Multi-Layer Perceptron Classifier. Among them, the one which provides the highest
accuracy is made to predict the Flood. The performance of Flood prediction takes place by
calculating accuracy along with the evaluation classification report,identifying the confusion matrix
and the final result is shown as GUI based application by providing corresponding Input Parameters.
Disadvantages :-
➢ It can’t thereby better determine the regularity of rainfall data and achieve more accurate
prediction results of flash flood.
➢ Model for an urban area in which short-term forecasting of a flood and it is complex scenario
including multiple cells is imagined where a possible real-world scenario is undertaken.

2.8 Satellite Flood Inundation Assessment and Forecast Using SMAP and
Landsat[34]
Author :- Jinyang Du , John S. Kimball, Justin Sheffield, Ming Pan , Colby K. Fisher, Hylke
E. Beck, and Eric F. Wood
The capability and synergistic use of multisource satellite observations for flood monitoring and
forecasts is crucial for improving disaster preparedness and mitigation. Here, surface fractional water
cover (FW) retrievals derived from Soil Moisture Active Passive (SMAP) L-band)

Dept. of CSE, SJCIT 9 2021-22

Novel Approached Flood Forecasting System Using ML & BD Literature Survey

brightness temperatures were used for flood assessment over southeast Africa during the Cyclone
event. We then focused on five sub catchments of the Pungwe basin and developed a machine
learning based approach with the support of Google Earth Engine for daily forecasting of FW and
30-m inundation downscaling and mapping. The Classification and Regression Trees model was
selected and trained using retrievals derived from SMAP and Landsat coupled with rainfall forecasts
from the NOAA Global Forecast System. Independent validation showed that FW predictions over
randomly selected dates are highly correlated with the Landsat observations. The forecast results
captured the flood temporal dynamics from the event; and the associated 30-m downscaling results
showed inundation spatial patterns consistent with independent satellite synthetic aperture radar
observations. The data-driven approach provides new capacity for flood monitoring and forecasts
leveraging synergistic satellite observations and big data analysis, whichis particularly valuable for
data sparse regions
Disadvantages :-
➢ Delay should be improved

2.9 River Flooding Forecasting and Anomaly Detection Based on Deep

Learning[26]
Author :- Scott Miau and Wei-Hsi Hung
Pluvial floods are rare and dangerous disasters that have a small duration but a destructive impact
in most countries. In recent years, the deep learning model has played a significant role inoperational
flood management areas such as flood forecasting and flood warnings. This paper employed a deep
learning based model to predict the water level flood phenomenon of a river in Taiwan. We combine
the advantages of the CNN model and the GRU model and connect theoutput of the CNN model
to the input of the GRU model, called Conv-GRU neural network, and our experiments showed that
the Conv-GRU neural network could extract complex features of the river water level. We compared
the predictions of several neural network architectures commonly used today. The experimental
results indicated that the Conv-GRU model outperformed the other state-of-the-art approaches. We
used the Conv-GRU model for anomaly/fault detection in a time series using open data. The efficacy
of this approach was demonstrated on 27 river water level station datasets. Data from Typhoon
Soudelor were investigated by our model using the anomaly detection method. The experimental
results showed our proposed method could detect abnormal water levels effectively.

Dept. of CSE, SJCIT 10 2021-22

Novel Approached Flood Forecasting System Using ML & BD Literature Survey

2.10 Rainfall–runoff modelling using Long Short-Term Memory (LSTM)

networks[19]
Author :- Frederik Kratzert, Daniel Klotz, Claire Brenner, Karsten Schulz, and Mathew Herrnegger

Rainfall–runoff modelling is one of the key challenges in the field of hydrology. Various approaches
exist, ranging from physically based over conceptual to fully data-driven models. In this paper, we
propose a novel data-driven approach, using the Long Short-Term Memory (LSTM) network, a
special type of recurrent neural network. The advantage of the LSTM is its ability to learn long-term
dependencies between the provided input and output of the network, which are essential for
modelling storage effects in e.g. catchments with snow influence. We use241 catchments
of the freely available CAMELS data set to test our approach and also compare theresults to the well-
known Sacramento Soil Moisture Accounting Model (SAC-SMA) coupled with the Snow-17 snow
routine. We also show the potential of the LSTM as a regional hydrological model in which one
model predicts the discharge for a variety of catchments. In our last experiment, we show the
possibility to transfer process understanding, learned at regional scale, to individual catchments and
thereby increasing model performance when compared to a LSTM trained only on the data of single
catchments. Using this approach, we were able to achieve better model performance as the SAC-
SMA + Snow-17, which underlines the potential of the LSTM for hydrological modelling
applications.

Dept. of CSE, SJCIT 11 2021-22

Novel Approached Flood Forecasting System Using ML & BD Literature

Dept. of CSE, SJCIT 12

2021-22
CHAPTER – 3
SYSTEM REQUIREMENTS SPECIFICATION
System Requirement Specification is used for the programming contexture that are basically
being for the functionality of the system can do, and also for the work behind the organization
for describing and to understand the client’s needs. The purpose of SRS givesthe requirement
to be master planed of a system or sub-system. It generally provides for the capable clients
needful conditions at a particular instance of time before the work is finalize.

3.1 System Overview

A software requirements specification is a complete that describes the behavior of system to be
developed. Use case techniques can be used to find the product of functional requirements or
supplementary requirements. A nonfunctional requirement performs the engineering
requirements, quality of standards.

3.1.1 Purpose:

The purpose of SRS document is to list the user requirements in the organized manner. It
defines all the constraints and software requirements needed to understand the application and
documents. The user should be able to understand the purposed system after going through the
SRS documents and should be in position to incorporate some changes required.

3.2 Functional Requirement Specification

The functional requirements are designed to carry out to the clients. The requirements used
by the clients should be very well defined for the operation of the system. The clients
understand what the services to be provided, objectives to be defined and how the system will
react with particular input. The development of functional requirement leads to the specific
operations, major requirements to develop project, the types of software to be tested and what
does the system need to be specified. The input and output data should be taken to the use with
characteristics of clients data. The specified task of functional requirement of system should
understand the brief description of clients needs. The resources those are very essential to
estimate the operations, costs, analysis and the information to be carried.

3.2.1 Functional Requirements

• The analysis and design of the proposed system focused on its main functional

12
Novel Approached Flood Forecasting System using ML and BD SRS

requirements. They included gathering not only key geospatial factors from institutional
data sources but also those crowdsourced from individuals.
• Users are divided into two groups. Firstly, the one labeled as ‘‘User,’’ are the general
public, community leaders, or government officers who are anticipating the event and
need access to forecasts so as to prepare appropriate measure in accordance.
• The users are able to browse map data as well as relevant information geographically,
such as observed and predicted precipitations. The prediction includes accumulated
amount (in mm.) and probability (in percent) of precipitation, etc.
• Based on relevant factors within an area, they may query prediction of flood event,
along with its likelihood and possible severity.
• The outcomes can then be verified against the flooding data, crowdsourced and reported
via the Flood Mitigation System (thaiflood.org). Secondly, the other group labeled as
‘‘Admin/ Officer’’ are authorized agents who are operating the services.
• In addition to generic functionalities, their administrative tasks include membership
management as well as updating relevant data and forecasted results.

3.3 Non Functional Requirements

Non-Functional requirements are indirectly specified for the structure of project. The
requirements with the specified function may carry different constraints to perform the
system. The quality of the system are presides to measure the constraints of the system
capabilities. The measure of Non-Functional requirements allows the clients to indirectly
concentrate on the system analysis.
Qualities of Non-Functional Requirements are
• Usability: Design very simple no pre-requisite technical knowledge is required
• Maintainability: Maintenance is very less.
• Scalability: it is scalable for multiple devices which does not degrade the
performance.
• Portability: it can be used in anywhere on independent of platform.

▪ Data used in flood forecasting could be categorized into four main groups. They were
1) Geospatial,
2) Meteorological and Hydrological, obtained from GLOFAS,
3) Hourly rainfalls prediction from TMD Big Data platform, and
4) Crowdsource (or volunteer) data.

Dept. Of CSE, SJCIT 13 2021-22

Novel Approached Flood Forecasting System using ML and BD SRS

• They were stored in geo-database and then processed by one of modern ML strategies.
• Since all data involved in this study were acquired from various sources data needs to
preprocess in terms of spatial resolution, interpolation was made based on their
geographic coordinates.
• The key elements were data acquisition and interchange between the system and
respective sources and their intelligence via MLs.

• Meteorological and hydrological data were acquired from the Global Flood server,
called GLOFAS. Meteorological data consisted of accumulated precipitation, and the
probability of precipitation at different levels, predicted daily.
• The prediction is based on ECMWF (European Center for Medium-Range Weather
Forecasts) model.

3.4 REQUIREMENTS

3.4.1 HARDWARE NECESSITIES

The most extensively watched approach of fundamentals delineated in some running system
applications is a physical PC resources , everything considered known contraption , hardware
entities once over is anyway significant part of the time as could be normal joined by an
apparatus resemblance list , particularly if there ought to their event of working structures. A
HCL notes endeavored , flawless and once an incongruent gear devices for a certain running
structure. The going with set packs separates the various bits of mechanical assemble stray
pieces.

All personal computers running systems are made methodologies for a particular PC structure.
The hugeness of leading getting ready unit is a focal structure essential for anything. Highest
programming executing on x86 dealing with delineate master minding ability as copy and time
speed of CPU.

3.4.2 HARDWARE SYSTEM CONFIGURATION :-

❖ Processor : Intel dual core

❖ Speed : 1.1 Ghz

Dept. Of CSE, SJCIT 14 2021-22

Novel Approached Flood Forecasting System using ML and BD SRS

❖ RAM : 4GB (min)

❖ Hard Disk : 50 GB
❖ Key Board : Standard Window Keyboard
❖ Mouse : Two or Three Button Mouse

3.4.3 SOFTWARE SYSTEM CONFIGURATION:-

❖ Operating System : Windows 10 , 11

❖ Front End : HTML, CSS

❖ Back End : Python

❖ Database Connectivity : "https://weather.visualcrossing.com”

Dept. Of CSE, SJCIT 15 2021-22

CHAPTER - 4

SYSTEM ANALYSIS

The software development is generally carried out with the System Analysis and Design. The
system analysis provides various detailed information of existing performance that may lead
to the configuration of the new system. The problem in the existing system will be the
drawbacks and to overcome this problem, the proposed system may raise with the improvement
with the solutions for existing system. The process of studying the system analysis in order to
identify the goals and objectives of system

4.1 Existing System

All the Existing system statistically analyzed floods based on the historical data of water levels,
and seven points of water inlets and outlets. The resultant forecast was presented as point-wise
flood levels, rendered on a two-dimensional map. Its shortcoming was that the analysis did not
consider any geospatial data. In addition, forecasts could only be made at specific outlets, but
not at those arbitrarily queried by users. Although it was developed as a web application, it did
not support responsive web technology, and as such was unable to equally well satisfy user
experience on all other devices apart from personal computers (PC). According to recent
surveys, most flood forecasting systems relied primarily on either monitored precipitation data
or those obtained from a single source. Beside the mentioned limitations, existing systems
remained lacking various aspects.
Existing tools were typically based on manually input and prepared data. The processes were
tedious and thus prohibitive for real-time and early forecasts. Furthermore, these tools did not
fully exploit more comprehensive information available in current big data platforms.

4.1.1 Disadvantages
• Monitoring facilities or communication network of one’s became malfunctioned,
there would be no precipitation data available for imperative analyses.
• There was also no tool (software) that can accommodate area specific forecasting
well in advance.
• Existing tools were highly dependent on demanding data preparation and
compilation from various sources, including Big Data

16
Novel Approached Flood Forecasting System using ML and BD System Analysis

4.2 PROPOSED SYSTEM

To elevate the limitations stated above, the proposed system thus analyzed and designed a
flood forecasting system that improved over the current ones. The aspects considered herein
were supports of responsive web technology, automation of key processes ,and availability
and usability of the system. To this end, the proposed system was developed by using both
meteorological and hydrological models in forecasting accumulated precipitation from data
obtained from TMD big data and GLOFAS, and ML models in forecasting flood situations in
given areas. The analyses were made based on meteorological, hydrological, geospatial and
crowd sourcing data.
Our novel flood forecasting system based on fusing meteorological, hydrological,
geospatial, as well as crowd sourcing data, and integrating them into an ML framework.
These data were compiled from various big data platforms, by using online application
programming interfaces (API). The forecasting mechanism was driven by a machine
learning strategy. To determine the most suitable one for the task, several state-of-the-art
MLs, i.e., decision tree, random forest, naïve Bayes, artificial neural networks, support
vector machine
4.2.1 ADVANTAGES
• Forecasting was performed based on these data learned by modern ML strategies.
They were decision tree, RF, Naïve Bayes, MLP and RBF ANN, SVM
• The proposed system also enhanced user experience via responsive graphical
interfaces, interoperable on different computing devices including mobiles.
• This advantage effectively encouraged greater contribution of crowd source data
from the public, enriching data aggregation and hence increasing system accuracy
and reliability

Dept. Of CSE, SJCIT 17 2021-22

CHAPTER – 5
SYSTEM DESIGN
This document gives the design of the overall project. Software development is the phase which
is very important for the supernova of the software, which is called as design phase. The design
phase should satisfy the functional and non-functional requirements for theeffectiveness for
satisfying all the constraints and objectives of the project. It mainly concentrates on the
modules that needed for system. The design phase depends mainly on the specification of
feasibility survey.

5.1 Data Flow Diagram

The information stream outline demonstrates the graphical portrayal, similar to game plans it
is utilized to speak to the information through the sources of info, different sorts of information
examination will be completed and the coveted yield will be produced. These parts will be
utilized to demonstrate the framework and it will be displayed by to contemplatequickly
regarding the info. In the framework outline the DFD will demonstrate the stream of whole
parts. The stream of data will in arrangement of change utilizing this framework.

Figure 5.1: Data Flow Diagram Level 0

In the figure 5.1 , user can search the particular city which is been needed, it shows the result
of the particular city through the efficient prediction by using the RF.

18
Novel Approached Flood Forecasting System using ML and BD System Design

Figure 5.2: Data Flow Diagram Level 1

In the figure 5.2, it shows that the user types the city name and through the parameters like cloud,
temperature and wind it reads the data and predicts the flood.

Figure 5.3: Dataflow Diagram Level 2

In the figure 5.3, the user can view map related data and processes the data as input, it collects the
data from the rainfall API which is been updated the flood situations . The user collects the updated
data and flood forecast data and predicts the flood.

Dept. Of CSE, SJCIT 19 2021-22

Novel Approached Flood Forecasting System using ML and BD System Design

5.2 Activity Diagram

Figure 5.4: Activity Diagram

The above figure represents the activity model of the novel flood forecasting project, it collects
the data by processing of recording and transmitting the readings of an instruments and it is
automated communication process from multiple data sources. The Telemetry works by using
sensors at a remote source, the sensors measure physical data and electrical data which gets
converted to electrical voltages and c0mbined with timing data, they form a data system which
is transmitted over wireless, wired or hybrid medium. The weather forecasting is the
application of scene and technology to predict the conditions of the atmosphere for a given
location and time, it is a scientific estimate of future weather condition. Weather condition is
the state of atmosphere at a given time expressed in terms of the most significant weather
variables. The significant weather variables bring forecast differ from the place to place.
Weather radar which is also called weather surveillance radar and Doppler weather radar is a
type of radar used to locate precipitation, calculate its motion, and estimate its, Modern weather
radars are mostly pulse-Doppler radars, capable of detecting the motion of rain droplets in
addition to the intensity of the precipitation. The images are collected from satellite. These all
data is been collected and modules is been transmitted.

Dept. Of CSE, SJCIT 20 2021-22

Novel Approached Flood Forecasting System using ML and BD System Design

5.3 Use Case Diagram

A utilization case in programming designing and frameworks building is a portrayal of a
framework's conduct as it reacts to a demand that starts from outside of that framework. As it
were, an utilization case depicts "who" can do "what" with the framework being referred to.
The utilization case method is utilized to catch a framework's behavioral necessities by
specifying situation driven strings through the useful prerequisites.
Utilize cases portray the connection between at least one performing artists (an on- screen
character that is the initiator of the communication might be alluded to as the 'essential
performer') and the framework itself, spoken to as a succession of basic strides. Performing
artists are something or somebody which exists outside the framework ('black box') under
review, and that partake in a grouping of exercises in a discourse with the framework to
accomplish some objective. On-screen characters might be end clients, different frameworks,
or equipment gadgets. Each utilization case is a total arrangement of occasions, depicted from
the perspective of the on-screen character. Flood forecasting was rendered based on learning
of thematic data by an ML method. The resultant forecast was displayed on a web browser.
With this platform, users may access this information from various devices. The design phase
should satisfy the functional and non-functional requirements for the effectiveness for
satisfying all the constraints and objectives of the project. It mainly concentrates on the
modules that needed for system. The design phase depends mainly on the specification of
feasibility survey.

Figure 5.5: Use Case Diagram

Dept. Of CSE, SJCIT 21 2021-22

Novel Approached Flood Forecasting System using ML and BD System Design

Figure 5.6: Use Case Diagram

As shown in the above diagram, the user can view the map and data which is related ,he can
view the weather forecast and floods occurring situation. The data is been related by
considering the parameters like rainfall, GLOFAS and weather forecast, through considering
these parameters flood is been predicted and it updates the relevant data and flood
forecasting.

Dept. Of CSE, SJCIT 22 2021-22

CHAPTER 6

IMPLEMENTATION

6.1 Algorithm
Prediction Module
Step 1: Start prediction model
Step 2: Users should select a state required to enter the state name
Step 3: It checks the all-weather and parameters
Step 4: It display safe/unsafe for the flood and 2 also displays temperature, humidity, cloud,
parameters
Step 5: Stop

User Module
Step 1: Start user model
Step 2: view map related data
Step 3: If city == entered city name
If true
View flood statues
Step 4: flood forecasting information
Step 5: end if condition
Step 6: Stop

Data Integration
Input :{City name}
Output :{View map related data}
Begin
Step 1: Add city name
Step 2: connected with map data
Step 3: checks the weather data
Step 4: Shows the weather related data
End

Data Processing
Input: {City name}
Output: flood result
Begin
Step 1: Read city name
Step 2: Preprocess query
Step 3: Checks the city name in map
Step 4: Apply RF
Step 5: Find flood related data
Step 6: returns result
End

23
Novel Approached Flood Forecasting System Using ML and BD Implementation

Floods are one of the most dangerous and frequent natural disasters in the world. Every year,
thousands of people die from the floods and millions of people lose their homes and livelihoods.
In many states around India and China's Yellow River Valley, some of the world's worst floods
have killed thousands, if not millions of people.

MODULES
There are three modules in the Novel Approached Flood Forecasting System Using ML And BD.
They are:-
• PLOTS
• INTERACTIVE HEATMAPS
• FLOOD PRIDICTION

PLOTS

The visualizations show flood predictions, damage predictions, and heavy rainfall predictions
across India, taking in factors such as precipitation, wind speed, humidity, temperature, cloud
cover, as well as previous data history. We can view plots showing flood prediction, precipitation
analysis, and damage analysis.

Flood prediction: The plot below shows our ML powered prediction of where a flood is going to
occur, marked by red dots

Figure 6.1: ML powered prediction flood

Dept. Of CSE,SJCIT 24 2021-22

Novel approached flood forecasting System using ML & BD Implementation

Precipitation: The plot below shows the current precipitation data across the nation. The larger
bubbles indicate more precipitation.

Figure 6.2: current precipitation

Interactive Heatmaps
The heatmaps show flood predictions, damage predictions, and heavy rainfall predictions across
India, taking in factors such as precipitation, wind speed, humidity, temperature, cloud cover, as
well as previous data history.

Damage Analysis
The plot below shows cost and damage analysis, based on the flood risk prediction. The
colorscale of the heatmap indicates the extent of predicted monetary damage, measured in INR.

Figure 6.3: damageanalysis

Dept. of CSE, SJCIT 25 2021-22

Novel approached flood forecasting System using ML & BD Implementation

Flood Prediction

The plot below shows our ML powered prediction of where a flood is most likely to occur given
the current environmental factors, marked primarily by the darker red spots.

Figure 6.4: ML powered prediction

Figure 6.5 : Flood prediction of a city

Dept. of CSE, SJCIT 25 2021-22

CHAPTER 7

TESTING

The primary objective of testing is to correct the bugs, issues or blunders. To recognize mistakes
the test engineers needs to test every individual segments of the venture arrange module. Every
module is tried for the better execution and by checking the modules the mistakes are recognized.
It confirms that the frameworks achieved its prerequisites are definitely not. By examining the
every modules and mistakes are pester out of framework to get particular yield

Dissimilar to every module gives the required yield, the assurance of test designer prompts the
rightness of framework projects. The last module stage is intended to maintain a strategicdistance
from the disappointments and to expel deficiencies. So it's ideal to give the testing stage to the
advancement of the venture.

7.1 Testing methods

The testing methods is one of the essential stages in the framework testing. It serves to the general
population who are working outside the testing field. The correct arranging makes the item to build
up to customer. The test arrange gives the documentation of the exercises performed for
programming testing and its serves to approve the yield work.

7.1.1 Unit Testing

Unit testing as the name portrays that the testing procedure is completed with the testing where
every individual models are tried in a steady progression. The operation to perform unit testing is
to figure out where every module testing is approve or not. The investigation of testing gives the
fruitful result and to perform correct report determination. The capacity of unit test additionally
upgrades the level of testing before the reconciliation procedure. By testing every module the
blunders are recognized in before stages and this may prompt the yearning yield of the projects.
Unit testing isolates the every individual piece of modules and redresses whether the module is
executed or not. The essential execution is to give an end-clients to enhancing the application
programming, business handle and the level of framework setup.

27
Novel Approached Flood Forecasting System Using ML & BD Testing

7.1.2 Integration Test

It is test where every one of the exhibitions are planned with the product testing procedure and
individual set programming's are coordinated to perform in a gathering to run the one program.
The fulfillment of this testing leads just when exhibitions of every necessities, programming
modules and programming design. The yield execution makes when all reconciliation test modules
are determined to play out the testing procedure with craving input. At long last mix testing
furnishes end-client with accuracy of the yield with determined programming testing

7.1.3 System Test

Framework Testing is one of the testing procedure where the fruition of testing stage is for the
most part relies on upon System. Framework testing gives the spine support to all the testing stage
on the grounds that once the consummation of all the testing procedure the framework testing plays
out the Hardware and Software Requirement Specifications and the Softwaresituated examination
of framework. This depends on the desire of end-client, where it ought to fulfilled to get crave
result.

7.1.4 Functional testing

This is a type of black box testing that is based on the specifications of the software that is to be
tested the application is tested by providing input and then the results are examined that need to
conform to the functionality it was intended. Functional testing of a software is conducted on a
complete, integrated system to evaluate the system’s compliance with its specified requirements.

7.1.5 Validation testing

Validation testing determining the system complies with the requirements and performs functions
for which it is intended and the organization’s goals and user needs. Validation is done at the end
of the development process and takes place after verification are completed. Performed after a
work product is produced against established criteria ensuring that the product integrates correctly
into the environment. Determination of correctness of the final software product by a development
project with respect to the user needs and requirements.

Dept. of CSE, SJCIT 28 2021-22

Novel Approached Flood Forecasting System Using ML & BD Testing

7.2 Testing Principles

The testing goal of the framework programming is to make a point to substantiate the framework
condition by dissolving the flaws and blunders.

Some of the basic principles of testing are:

1. “A successful test to be carried out which holds the capability of finding errors.”

2. “Planning of testing phase may required long time.”

3. “The tests are performed according to the clients need.”

4. “Effective test should require for detecting all errors.”

5. “Exhaustive testing is not possible”

Table 7.1: Unit Testing for Live Data set

Unit Testing Data set upload
Feature bring Tested Live Data set
Description Through API live data of weather will be retrieved
Sample input API
Expected output Dataset of weather
Actual Output Web page is designed to display the weather for different cities
Remarks Success

Table 7.2: Integration Test Preprocessing of Dataset

Integration Test Dataset Normalization

Feature bring Tested Preprocessing of Dataset

Description Read live weather data from API, perform the normalization any null
values will be removed
Sample input Dataset of weather
Expected output Successful preprocessing of dataset

Actual Output Preprocessing will be done and weather data will be displayed

Remarks Success

Dept. of CSE, SJCIT 29 2021-22

Novel Approached Flood Forecasting System Using ML & BD Testing

Table 7.3: Validation testing for Heatmaps

Validation testing Heatmaps
Feature bring Tested flood predictions, damage predictions, and heavy rainfall
predictions
Description The heatmaps show flood predictions, damage predictions, and
heavy rainfall predictions across India
Sample input Input from trained model
Expected output flood predictions, damage predictions, and heavy rainfall
predictions
Actual Output Maps will be displayed for flood prediction, and damage
analysis,
Remarks Success

Table 7.4: Functional testing for Prediction of flood for a particular city
Functional testing flood prediction
Feature bring Tested Prediction of flood for a particular city
Description predictions about future floods
Sample input City Name
Expected output Successful prediction
Actual Output Prediction says safe or unsafe from the flood.
Also display weather data such as humidity, wind speed,
temperature .
Remarks Success

Dept. of CSE, SJCIT 30 2021-22

CHAPTER 8

PERFORMANCE ANALYSIS

To ensure maximum versatility and most extensive coverage of crowdsourcing, the proposed
methodology adopted responsive web design paradigm in developing the website. The web
application once deployed was able to support various devices, ranging from personal and
portable computers to mobile phones and tablet computers, with varying screen sizes and
resolutions.

By using the Bootstrap framework, the rendered CSS and JavaScript automatically aligned and
adjusted the layout of components and controls to maintain uniform appearance andhence
satisfying user experience (UX). Furthermore, the development cost was minimized as focuses
were placed on core functionalities and customizable contents, instead of variationsof
frontend interfaces and layouts across platforms.

Meteorological and hydrological data that were acquired from GLOFAS are illustrated in
Figure 13. In this figure, accumulated precipitation in each area was color coded. Specifically,
blue, light green and dark red, represents low (10 – 25 mm), moderate (25 – 100 mm), and high
(> 100 mm), respectively. Areas with neutral one indicated those with no precipitation.
Likewise, the data would also be used in forecasting. Another crucial information was the
probability of precipitation. In this study, the respective probability in each area was color
coded in three levels, i.e., 50 mm, 150 mm, and 300 mm. Examples of these levels are
illustrated in Figures 14 to 16. In each figure, pixel intensities indicate the likelihood of
precipitation (i.e., 0 to 1) at that location being of the respective levels. In Figure 14, for
instance, light and dark green indicate low and high probability of 50 mm precipitation in a
given location. Similarly, Figures 15 and 16 display the probability of precipitation of 150 mm
and 300 in blue and red, respectively.

Yellow, orange, and red pins represented flood levels of less than 20 cm, 20 – 50 cm, and
greater than 50 cm, respectively. Without crowdsourcing data, there would be no better
means of acquiring rainfall duration and its intensity, as well as drainage problem, which
were all crucial

Flood forecasting was rendered based on learning of thematic data by an ML method. The
resultant forecast was displayed on a web browser. With this platform, users may access this

31
Novel Approached Flood Forecasting System using ML & BD Performance Analysis

information from various devices. An example of forecasted flood is shown snapshots. The
strength of its impacts was determined based its level, and accordingly coded in different
colors. Green pin indicate that the area was not at all affected by flood. On the contrary, yellow,
orange, and red pin represents flooded areas, whose levels were less than 20 cm, 20 – 49 cm,
and greater than 50 cm, respectively. Figure 19 shows an example when no flood incident was
anticipated.

The performance of these MLs is listed in Tables 5 and 6. It is evident that MLP ANN, SVM,
and RF were placed in top three ranks, in terms of classification accuracies (i.e., 97.83%,
96.67%, and 96.67%) and Kappa coefficients (i.e., 0.89, 0.84, and 0.84). Statistically, Kappa
values of greater than 0.8 indicate highly accurate forecast, while those between 0.4 – 0.8 were
moderate performers. Naïve Bayes, and fuzzy logic fell in the latter category. These trends
were similarly exhibited in MAE and RMSE, the closer to 0, the more accurate the forecast.
Taken into account the results of, not only flooded regions but also those unaffected by flood,
the balance between relevant and irrelevant predicted samples had to be considered.

Dept. of CSE SJCIT 32 2021-22

Novel Approached Flood Forecasting System Using ML & BD Conclusion

Comparison of previous algorithm and updated algorithm

Some data considered in this study, such as GLOFAS, were not of intrinsically high spatial
resolution, However, they were accurate. The accumulated precipitations and their
probability at different levels, for instance, corresponded to the actual event. Their API
system was also reliable. These characteristics were updated by the proposed system. Their
resolutionshortcomings were remedied by incorporating other more detailed layers as well as
crowdsource factors into the ML framework. Possible improvements for this issue include
involving the Internet of Things (IoT) in measuring actual meteorological data with preferred
coverage

Dept.Of,CSE,SJCIT 33 2021-22
CHAPTER 9

CONCLUSION& FUTURE ENHANCEMENT

The proposed methodology is a novel distributed flood forecasting system, based on integrating
meteorological, hydrological, geospatial, and crowdsource data. Big data made available by
prominent agencies were acquired by means of various crossplatform APIs. Forecasting was
performed based on these data learned by modern ML strategies. They were decision tree, RF,
Naïve Bayes, MLP and RBF ANN, SVM, and fuzzy logics. Evaluation results on studied areas
indicated that the system could forecasted flood events highly accurately.

This system also enhanced user experience via responsive graphical interfaces, interoperable on
different computing devices including mobiles. This advantage effectively encouraged greater

In prospects, the system can be readily employed in existing floods management schemes, e.g.,
those led by government agencies or non-profit organizations. Moreover, thanks to distributed
architecture, the system can reach wider public, and therefore serves as an effective means of
communicating with them (and especially the flood victims), regarding current status and
development of the disaster.

In prospects, the system can be readily employed in existing floods management schemes, e.g.,
those led by government agencies or non-profit organizations. Moreover, thanks to distributed
architecture, the system can reach wider public, and therefore serves as an effective means of
communicating with them (and especially the flood victims), regarding current status and
development of the disaster.

Future Enhancement

Future improvements of the system include initial flood representation and its extent being adapted
to the current location of the device, so that they can be instantly made aware of by its user.
Moreover, flooded location pined by an icon may be augmented with color-coded regions, so that
the conditions (e.g., levels and extents) of affected areas may be better comprehended.

34
BIBLIOGRAPHY
[1] Azhoni, A., Holman, I., & Jude, S. (2017a). Contextual and interdependent causes of

climate change adaptation barriers: Insights from water management institutions in

Himachal Pradesh, India. Science of the Total Environment, 576, 817– 828.
[2] Baan, P., & Klijn, F. (2004). Flood risk perception and implications for flood risk
management in the Netherlands. International Journal of River Basin
Management, 2, 113– 122.

[3] Alfieri, L., Salamon, P., Bianchi, A., Neal, J., Bates, P. D., and Feyen, L.: Advances in
pan-European flood hazard mapping, Hydrol. Process., 28, 4928– 4937,https://doi. org/
10.1002/hyp.9947, 2014b.
[4] E. Mayoraz and E. Alpaydin, ‘‘Support vector machines for multiclass classification,’’ in
Proc. Int. Work-Conf. Artif. Neural Netw. Berlin, Germany: Springer, 1999, pp. 833–842
[5] S. Ghosh, S. Karmakar, A. Saha, M. P. Mohanty, S. Ali, S. K. Raju, and P. L. N. Murty,
‘‘Development of India’s first integrated expert urban flood forecasting system for
Chennai,’’Current Sci., vol. 117, no. 5, pp. 741–745, 2019.
[6] Q. Ran, W. Fu, Y. Liu, T. Li, K. Shi, and B. Sivakumar, ‘‘Evaluation of quantitative
precipitation predictions by ECMWF, CMA, and UKMO for flood forecasting:
Application to two Basins in China,’’ Natural Hazards Rev., vol. 19, no. 2, May 2018, Art.
no. 05018003.
[7] S. Ghosh, S. Karmakar, A. Saha, M. P. Mohanty, S. Ali, S. K. Raju, and P. L. N. Murty,
‘‘Development of India’s first integrated expert urban flood forecasting system for
Chennai,’’Current Sci., vol. 117, no. 5, pp. 741–745, 2019.
[8] A. Mosavi, P. Ozturk, and K.-W. Chau, ‘‘Flood prediction using machine learning models:
Literature review,’’ Water, vol. 10, no. 11, p. 1536, Oct. 2018.
[9] L.-C. Chang, F.-J. Chang, S.-N. Yang, I.-F. Kao, Y.-Y. Ku, C.-L. Kuo, and I. Amin,
‘‘Building an intelligent hydroinformatics integration platform for regional flood inundation
warning systems,’’ Water, vol. 11, no. 1, p. 9, Dec. 2018.N. A. Sulaiman, N. F. A. Aziz,
N. M. Tarmizi, A. M. Samad, and W. Z. W. Jaafar, ‘‘Integration of geographic
information system (GIS) and hydraulic modelling to simulate floodplain inundation
level for Bandar Segamat,’’ in Proc. IEEE 5th Control Syst. Graduate Res. Colloq.,
Aug. 2014, pp. 114–119.
[10] Sarawut Jamrussri,Yuji Toda. Simulating past servere flood events ti evaluate the
effectiveness of nonstructural flood countermeasures in the upper chao phraya river
basin, Thailand. Journal of Hydrology Regional Studies,2017,Vol.10,p.82.

35
Novel Approached Flood Foracsting system using ML and BD Bibliography

[11] Scott Miau and Wei-Hsi Hung River Flooding Forecasting and Anomaly
Detection Based on DeepLearning30 Oct 2020, 198384-198402, Vol 8
[12] P. Vijayalakshmi , M. Vandhana, K. Vasanthapriya, P.B. Thushikka
Prediction of Flood by Rainfall using MLP Classifier of Neural Network
Model 2021 ICCES51350.2021.9489161

[13] Thorndahl, S.; Einfalt, T.; Willems, P.; Ellerbæk Nielsen, J.; Ten Veldhuis, M.C.;
Arnbjerg-Nielsen, K.; Rasmussen, M.R.; Molnar, P. Weather radar rainfall data in urban
hydrology.Hydrol. Earth Syst. Sci. 2017, 21, 1359–1380
[14] Cecinati, F.; de Niet, A.; Sawicka, K.; Rico-Ramirez, M.A. Optimal Temporal Resolutionof
Rainfall for Urban Applications and Uncertainty Propagation. Water 2017, 9, 762.
[15] Flood inundation modelling: A review of methods, recent advances and uncertainty
analysis,J. Teng, A. Jakeman, J. Vaze, B. Croke, D. Dutta,S.Kim,BusinessEnviron.
Model.Softw.2017
[16] Weili Duan, Bin He, Daniel Nover,et al. floods and associated socio economic damages
in China over the latest century.Natural Hazards. 2016, Vol.82,No.1,p.401.
[17] S.Zhang, J.Zhang, Y. Liu, and Y. Liu, “A mathematical spatial interpolation method for
The estimation of convective rainfall distribution over small watersheds,” Environ. Eng.
Res.,vol. 21, no. 3,pp.226-232, Sep.2016.
[18] J. P. De Albuquerque, M. Eckle, B.Herfort, and A. Zipf, “Crowdsourcing geographic
Information for disaster management and improving urban resilience: An overview of
recent developments and lessons learned,’’ in European Handbook of Crowdsourced
Geographic Information. London, U.K.: Ubiquity Press, Aug. 2016, pp. 309–321.
[19] Ford, J. D., Tilleard, S. E., Berrang-Ford, L., Araos, M., Biesbroek, R., Lesnikowski, A.
C., Bizikova, L. (2016). Opinion: Big data has big potential for applications to
climatechange adaptation. Proceedings of the National Academy of Sciences, 113,
10729– 10732.
[20] Al-Zahrani M, Al-Areeq A, Sharif HO. 2016. Estimating urban flooding potential near
the outlet of an arid catchment in Saudi Arabia. Geomat Nat Hazards Risk. 1:1-17.
[21] Technical review of large-scale hydrological models for implementation in operational
Floodforecasting schemes on continental level, Ann Kauffeldt , F.Wetterhall,

Dept .Of CSE, SJCIT 36 2021-22

Novel Approached Flood Foracsting system using ML and BD Bibliography

F.Pappenberger, P.Salamon, J.ThielenEnvironmental Science,Environ.Model.

Softw.2016
[22] Xu, H.; Xu, C.Y.; Saelthun, N.R.; Xu, Y.; Zhou, B.; Chen, H. Entropy theory based multi
criteria resampling of rain gauge networks for hydrological modelling–a case study of
humid area in southern China. J. Hydrol.2015,525,138 – 151. 2015, 525, 138–151.
[23] G. Schimak, D. Havlik, and J. Pielorz, ‘‘Crowdsourcing in crisis and disaster
management Challenges and considerations,’’ in Environmental Software Systems.
Infrastructures, Services and Applications. Cham, Switzerland: Springer, 2015, pp. 5670.
[24] T. Simon, A. Goldberg, and B. Adini, ‘‘Socializing in emergencies— A review of the use
of social media in emergency situations,’’ Int. J. Inf. Manage., vol. 35, no. 5, pp. 609–619,
Oct.2015
[25] L. Alfieri, J. Thielen, A European precipitation index for extreme rain-storm and flash
flood early warning. Meteorol. Appl. 22(1), 3–13 (2015).
[26] Development of a spatial decision support system for flood risk management in Brazil that
combines volunteered geographic information with wireless sensor networks April 2015
,Computers & Geoscience 80:84-94

[27] Rico-Ramirez, M.A.; Liguori, S.; Schellart, A.N.A. Quantifying radar-rainfall

uncertainties in urban drainage flow modelling. J. Hydrol. 2015, 528, 17–28
[28] On the Use of Global Flood Forecasts and Satellite-Derived Inundation Maps for Flood
Monitoring in Data-Sparse Regions,B. Romero, F. Hirpa, +4 authors T.Groeve,
EnvironmentalScience Remote. Sens,2015
[29] N. A. Sulaiman, N. F. A. Aziz, N. M. Tarmizi, A. M. Samad, and W. Z. W. Jaafar,
‘‘Integration of geographic information system (GIS) and hydraulic modelling to
simulate floodplain inundation level for Bandar Segamat,’’ in Proc. IEEE 5th Control
Syst. Graduate Res. Colloq., Aug. 2014, pp. 114–119.

[30] S. Puttinaovarat, P. Horkaew, and K. Khaimook, ‘‘Conguring ANN for inundation areas
identication based on relevant thematic layers,’’ ECTI Trans. Comput. Inf. Technol.(ECT
I- CIT), vol. 8, no. 1, pp. 56–66, May 2014.
[31] L.S. Campbell, W.J. Steenburgh, Finescale orographic precipitation variability and gap-
filling radar potential in little Cottonwood Canyon, Utah.Weather Forecast. 29,2014

Dept .Of CSE, SJCIT 37 2021-22

Novel Approached Flood Foracsting system using ML and BD Bibliography

[32] Al-Nammari F., Alzaghal M. Towards local disaster risk reduction in developing
countries:Challenges from Jordan. Int. J.Disaster Risk Reduct. 2015;12:34–41. doi:
10.1016/j.ijdrr.2014.11.005.
[33] J.García-Pintado, J.C. Neal, D.C. Mason, S.L. Dance, P.D. Bates Scheduling satellite-
based sar acquisition for sequential assimilation of water level observations into flood
modelling J. Hydrol., 495 (2013), pp. 252-266
[34] Siriluk Mongkonkerd, Siyanee Hirunsalee, Hidehiko Kanegae, et al. Comparison of
Direct Monetary Flood Damages in 2011 to Pillar House and Non-pillar House in
Ayutthaya, Thailand. Procedia Environmental Sciences. 2013, Vol.17, p.327.
[35] Shunji Kotsuki, Kenji Tanaka. Impacts of Mid-Rainy Season Rainfall on Runoff into
Chao Phraya River, Thailand. Journal of Disaster Research. 2013, Vol.8, No.3, p.397.
[36] Abaza, M., et al., 2014. A comparison of the Canadian global and regional meteorological
ensemble prediction systems for short-term hydrological forecasting (vol 141, pg 3462,
2013). Monthly Weather Review, 142, 2561–2562. doi:10.1175/mwr-d-14-00018.1.
[37] Ly, S.; Charles, C.; Degré, A. Different methods for spatial interpolation of rainfall datafor
operational hydrology and hydrological modeling at watershed scale: A review.
Biotechnol.Agron. Soc. Environ.2013,17,392-406.20,35
[38] Xu, H.; Xu, C.Y.; Chen, H.; Zhang, Z.; Li, L. Assessing the influence of rain gauge density
distribution on hydrological model performance in a humid region of China. J. Hydrol.
2013, 505, 1–12.
[39] J.García-Pintado, J.C. Neal, D.C. Mason, S.L. Dance, P.D. Bates Scheduling satellite-
basar acquisition for sequential assimilation of water level observations into
floodmodelling J. Hydrol., 495 (2013), pp. 252-26.
[40] Rogstadius, J., Vukovic, M., Teixeira, C. A., Kostakos, V., Karapanos, E., and Laredo,

J. A.: CrisisTracker: crowdsourced social media curation for disaster awareness,IBM J.

Res. Dev., 57, 4:1– 4:13, doi:10.1147/JRD.2013.2260692, 2013.

[41] Pappenberger, F., et al., 2013. Visualizing probabilistic flood forecast information:

expert preferences and perceptions of best practice in uncertainty communication.

Hydrological Processes, 27 (1), 132– 146. doi:10.1002/hyp.9253.

[42] Peleg, N.; Ben-Asher, M.; Morin, E. Radar subpixel-scale rainfall variability and

Dept .Of CSE, SJCIT 38 2021-22

Novel Approached Flood Foracsting system using ML and BD Bibliography

uncertainty: Lessons learned from observations of a dense rain-gauge network. Hydrol

Earth Syst. Sci. 2013, 17, 2195–2208
[43] Thorndahl, S.; Einfalt, T.; Willems, P.; Ellerbæk Nielsen, J.; Ten Veldhuis, M.C.; Arnbjerg-
Nielsen, K.; Rasmussen, M.R.; Molnar, P. Weather radar rainfall data in urbanhydrology.
Hydrol. Earth Syst. Sci. 2017, 21, 1359–1380.
[44] Adam, N., Shafiq, B., and Staffin, R. (2012). Spatial computing and social media in the
context of disaster management. Intelligent Systems, IEEE, 27(6), 90 –96.

[45] C. Corral, D. Velasco, D. Forcadell, D. Sempere-Torres, Advances in radar-based flood

Warning systems. The EHIMI system and the experience in the Besòs flash-flood pilot
basin , in Flood Risk Management: Research and Practice, ed. by P. Samuels, S.
Huntington, W.Allsop, J. Harrop (Taylor & Francis, London, 2009), pp. 1295–1303
[46] Dawson R.J., Speight L., Hall J.W., Djordjevic S., Savic D., Leandro J. Attribution of

flood risk in urban areas. J. Hydroinf. 2008;10:275–288. doi: 10.2166/hydro.2008.054

[47] Chaplot,V.; Saleh, A.; Jaynes, D.Effect of the accuracy of spatial rainfall information onthe
modeling of water, sediment, and NO3–N loads at the watershed level. J. Hydrol.

2005, 312, 223–234

[48] Supattra Puttinaovarat, Paramate Horkaew Flood Forecasting System Based on Integrated

Big and Crowdsource Databy Using Machine Learning Techniques Jan 2020, 10.1109
[49] Amir Mosavi , Pinar Ozturk and Kwok-wing Chau Flood Prediction Using Machine Learning
Models: Literature Review (2021) CSDE5383.2021.9718497

[50] Li-Chiu Chang and Fi-John Chang Building an Intelligent Hydro informatics Integration

Platform for Regional Flood Inundation Warning Systems 21 Dec 2018,9, 10.3390

[51] GokmenTayfur , Vijay P. Singh, TommasoMoramarco and SilviaBarbetta Flood

Hydrograph Prediction Using Machine Learning Methods 24 July 2018,10(8),968

Dept .Of CSE, SJCIT 39 2021-22

APPENDIX

APPENDIX A: SCREENSHOTS

Figure A.1: Home page

Figure A.2: Interactive Plots for Flood Prediction

40
Novel Approached Flood Forecasting System using ML and BD Appendix

Figure A.3: Interactive plots for Damage

analysis

Figure A.4: Interactive plots for Precipitaion

Dept. Of CSE, SJCIT 41 2021-22

Novel Approached Flood Forecasting System using ML and BD Appendix

Figure A.5: Interactive heat maps for Damage analysis

Figure A.6: Interactive heatmaps for Precipitation

Dept.Of CSE,SJCIT 42 2021-22

Novel Approached Flood Forecasting System using ML and BD Appendix

Figure A.7: Satellite image analysis for Bangalore in May

Figure A.8: Satellite image analysis for Bangalore in July

Dept.Of CSE,SJCIT 43 2021-22

Novel Approached Flood Forecasting System using ML and BD Appendix

Figure A.9: Prediction analysis of Bangalore

Figure A.10: Prediction analysis of Chennai

Dept.Of CSE,SJCIT 44 2021-22

Novel Approached Flood Forecasting System using ML and BD Appendix

Figure A.11: Prediction analysis of Mumbai

Dept.Of CSE,SJCIT 45 2021-22

Novel Approached Flood Forecasting System using ML and BD Appendix

APPENDIX B: ABBREVATIONS

• ML :- Machine Learning

• MLP :- Multilayer Perceptron

• ANN :- Artificial Neural Network

• GLOFAS :- Global Flood Awareness System

• RF :- Random Forest

• SVM :- Support Vector Machine

• APIs :- Application Programming Interface

• GUI :- Graphical User Interface

Dept.Of CSE,SJCIT 45 2021-22

PAPER PUBLICATION DETAILS

Paper Name: Novel Approached Flood Forecasting System Using Machine Learning and
Big Data

Journal: International Journal of Innovative Research in Computer and Communication

Engineering (IJIRCCE)

Volume: 10

Issue: 06

Published on: June – 2022

47
48
49
International Journal of Innovative Research in Computer and Communication Engineering
| e-ISSN: 2320-9801, p-ISSN: 2320-9798| www.ijircce.com | |Impact Factor: 8.165 |

|| Volume 10, Issue 6, June 2022 ||

| DOI: 10.15680/IJIRCCE.2022.1006139|

Novel Approached Flood Forecasting System

Using Machine Learning and Big Data
Prof. Ajay N, Swapna K, Yarramsetty Syona Sree, Namitha B C, Shivadarshini R
Assistant Professor, Dept. of CSE, S J C Institute of Technology, Chikkaballapur, Bangalore, India
B.E Student, Dept. of CSE, S J C Institute of Technology, Chikkaballapur, Bangalore, India
B.E Student, Dept. of CSE, S J C Institute of Technology, Chikkaballapur, Bangalore, India
B.E Student, Dept. of CSE, S J C Institute of Technology, Chikkaballapur, Bangalore, India
B.E Student, Dept. of CSE, S J C Institute of Technology, Chikkaballapur, Bangalore, India

ABSTRACT: Flood is one of the most disruptive natural hazards, responsible for loss of lives and damage to
properties. A number of cities are subject to monsoons influences and hence face the disaster almost every year. Early
notification of flood incident could benefit the authorities and public to devise both short and long terms preventive
measures, to prepare evacuation and rescue mission, and to relieve the flood victims. Geographical locations of affected
areas and respective severities, for instances, are among the key determinants in most flood administration. this paper
proposes a novel flood forecasting system based on fusing meteorological, hydrological, geospatial, and crowdsource
big data in an adaptive machine learning framework. Data intelligence was driven by state of-the-art learning strategies.
Subjective and objective evaluations indicated that the developed system was able to forecast flood incidents,
happening in specific areas and time frames. It was also later revealed by benchmarking experiments that the system
configured with an MLP ANN gave the most effective prediction.

KEYWORDS: Flood Forecasting System, Geographical, Meteorological, Hydrological, Geospatial, Crowdsource,

MLP ANN, Big Data, Machine Learning.

I.INTRODUCTION

Natural flood is one of the most recurrent disasters. Unlike stagnant water discharge, occasionally experienced in poorly
planned cities, major flood incidents always cause considerable damages to properties and, more often than not, loss of
lives. Several Asiancountries, particularly Thailand, are subject to both southwest and northeast monsoons and
accordingly facing seasonal deluge almost every year and in most parts of the countries.

Among notable causes, sudden and enduring heavy rain is the most pertinent one in Thailand. Furthermore, overflow
from main rivers along shore sides to surrounding basins can greatly spread the damages. Although being located further
away from a river, an area with inappropriate land uses are unable to efficiently discharge accumulated precipitation, and
hence are inevitably prone to even more frequent floods. Regardless of causes, however, a flood is generally sudden and
thus almost formidable for the general public and relevant organization to be adequately prepared for the incident. This
is mainly due to the lack of an effective means of anticipating the disaster well in advance.

Despite the recent extensive development of computerized flood forecasting systems, they remainedbased primarily on
present precipitation, monitored by rain stations or rain gauges. These facilities are normally owned by a meteorology
department or similar organizations. Besides, they are scantly located in a few areas due to costly installation and
maintenance. Hence, it is difficult to determine precipitation or predict flood accurately, especially in areas with no such
facility. To remedy this issue, precipitation in these areas were typically estimated either by inter- or extrapolation from
those with rain stations present. Due to a limited number of these stations and those readings in one area may not be a
good representative to others. Therefore, estimated pre imitation was insufficiently accurate to make a realistic forecast.
Conventional meteorological readings, e.g., precipitation, temperature, and humidity, etc., took really long
IJIRCCE©2022 | An ISO 9001:2008 Certified Journal | 5742
International Journal of Innovative Research in Computer and Communication Engineering
| e-ISSN: 2320-9801, p-ISSN: 2320-9798| www.ijircce.com | |Impact Factor: 8.165 |

|| Volume 10, Issue 6, June 2022 ||

| DOI: 10.15680/IJIRCCE.2022.1006139|

time to measure, process, record, and transfer to relevant organizations. Analyses based on past precipitation were known
to be associated with several shortcomings.

II. LITERATURE REVIEW

Flood Forecasting System Based on Integrated Big and Crowdsource Data by Using Machine
LearningTechniques.This paper proposes a novel flood forecasting system based on fusing meteorological,
hydrological, geospatial, and crowdsource big data in an adaptive machine learning framework. Data intelligence was
driven by state of-the-art learning strategies. Subjective and objective evaluations indicated that the developed system
was able to forecast flood incidents, happening in specific areas and time frames.

Disadvantages:- Data considered in this study, such as GLOFAS, were not of intrinsically high spatial resolution.

Flood Prediction Using Machine Learning Models

The holistic framework of the IHIP includes five layers (data access, data integration, servicer, functional subsystem, and
end-user application) and one database for effectively dealing with flood disasters. The IHIP provides real-time flood-
related data, such as rainfall and multi-step-ahead regional flood inundation maps. Disadvantages:-Web server deals
merely with sending requests and responses between clients and the IHIP.

Flood Hydrograph Prediction Using Machine Learning Methods

This study discusses the application of the artificial neural network (ANN), the genetic algorithm (GA), the ant colony
optimization (ACO), and the particle swarm optimization (PSO) methods for flood hydrograph predictions.

Disadvantages:-Delay should be improved

Development of Heavy Rain Damage Prediction Model Using Machine Learning Based on Big Data.

Authors developed a model for the prediction of heavy rain damage based on the big data provided by the Korea
Meteorological Administration and machine learning that can maximize the prediction performance of the model, and
the model could be used in implementing a proactive disaster management system. However, this study has
somelimitations on the number ofdamage data and the use of hydro-meteorological data

III. METHODOLOGY

This project proposed a novel distributed flood forecasting system, based on integrating meteorological, hydrological,
geospatial, and crowd source data. Big data made available by prominent agencies were acquired by means of various
cross platform APIs. Forecasting was performed based on these data learned by modern ML strategies. They were
decision tree, RF, Naïve Bayes, MLP and RBF ANN, SVM, and fuzzy logics, It was elucidated empirically that the
developed system could be used to alert the public and authorities alike of not only a current flood but also future ones.
This system also enhanced user experience via responsive graphical interfaces, interoperable on different computing
devices including mobiles. This advantage effectively encouraged greater contribution of crowdsource data from the
public, enriching data aggregation and hence increasing system accuracy and reliability.

Objectives

• To collect the flood forecasting data such as 1. Geospatial, 2. Meteorological and hydrological, obtained from
GLOFAS, 3. Crowdsource (or volunteer) data.

• Normalization of data and make a common coordinate frame, a pre-processing need to be done. Data integration
preprocessed data.

IJIRCCE©2022 | An ISO 9001:2008 Certified Journal | 5743

International Journal of Innovative Research in Computer and Communication Engineering
| e-ISSN: 2320-9801, p-ISSN: 2320-9798| www.ijircce.com | |Impact Factor: 8.165 |

|| Volume 10, Issue 6, June 2022 ||

| DOI: 10.15680/IJIRCCE.2022.1006139|

• Prediction and evaluating the results using MLP ANN, RF, SVM.

IV. PROPOSED SYSTEM

To elevate the limitations stated above, the proposed system thus analyzed and designed a flood forecasting system that
improved over the current ones. The aspects considered herein were supports of responsive web technology, automation
of key processes,and availability and usability of the system. To this end, the proposed system was developed by using
both meteorological and hydrological models in forecasting accumulated precipitation from data obtained from TMD big
data and GLOFAS, and ML models in forecasting flood situations in given areas. The analyseswere made based on
meteorological, hydrological, geospatial and crowd sourcing data. Our novel flood forecasting system based on fusing
meteorological, hydrological, geospatial, as well as crowd sourcing data, and integrating them into an ML framework.
These data were compiled from various big data platforms, by using online application programming interfaces (API).

The forecasting mechanism was driven by a machine learning strategy. To determine the most suitable one for the task,
several state-of-the-art MLs, i.e., decision tree, random forest, naïve Bayes, artificial neural networks, support vector
machine

ADVANTAGES

• Forecasting was performed based on these data learned by modern ML strategies. They were decision tree, RF, Naïve
Bayes, MLP and RBF ANN, SVM

• The proposed system also enhanced user experience via responsive graphical interfaces, interoperable on different
computing devices including mobiles.

• This advantage effectively encouraged greater contribution of crowd source data from the public, enriching data
aggregation and hence increasing system accuracy and reliability

V. IMPLEMENTATION

Data used in flood forecasting could be categorized into four main groups. They were geospatial,meteorological and
hydrological, obtained from GLOFAS , hourly rainfalls prediction from TMD Big Data platform , crowd source (or
volunteer) data. They were stored in geo-database and then processed by modern ML strategies.

The key elements in this module were data acquisition and interchange between the system and respective sources and
their intelligence via MLs. Thematic data acquisitions were divided into four groups, i.e., meteorological and
hydrological data, hourly precipitation data, area specific geospatial data, and crowdsource data.

Unlike a physical based approach, data driven ML does not focus on insights into functional models, but intrinsic
relationships between flood relevant factors and corresponding outcomes, learned from the past events. The design of
ML models adopted in this framework and their characteristics are described as follow: Firstly, MLP is a configuration
of ANN with multiple layers and is suitable for complicate learning tasks.

IJIRCCE©2022 | An ISO 9001:2008 Certified Journal | 5744

International Journal of Innovative Research in Computer and Communication Engineering
| e-ISSN: 2320-9801, p-ISSN: 2320-9798| www.ijircce.com | |Impact Factor: 8.165 |

|| Volume 10, Issue 6, June 2022 ||

| DOI: 10.15680/IJIRCCE.2022.1006139|

Figure5.1: Interactive plots

The above figure shows the Flood Prediction , the plots shows ML powered prediction of where a flood is going to
occurred, which are marked by red dots.

Figure 5.2: Damage Analysis

The above figure shows the damage analysis which is based on the flood risk prediction. The color scale of the heatmap
indicates the extent of predicted monetary damage, measured in INR.

IJIRCCE©2022 | An ISO 9001:2008 Certified Journal | 5745

International Journal of Innovative Research in Computer and Communication Engineering
| e-ISSN: 2320-9801, p-ISSN: 2320-9798| www.ijircce.com | |Impact Factor: 8.165 |

|| Volume 10, Issue 6, June 2022 ||

| DOI: 10.15680/IJIRCCE.2022.1006139|

Figure5.3: Flood Prediction for a city

The above figure shows the prediction of the particular place in India about the Flood Prediction whether the particular
place is safe or not from the floods. It consists of parameters like Temperature, Wind speed, Cloud cover, Precipitation
and Humidity.

Figure 5.4: Damage Analysis Graph

The above figure is the prediction of the damage analysis graph of few cities which will affect from floods.

VII.CONCLUSION

The main applications of this proposed approach is employing crowd source data, not only in training but also in
verification. The system obtained data from https://india.mongabay.com/, to which public users could send notifications
of flood situations. It is worth noted that, without dedicate equipment, accurate quantification of relevant data is
prohibitive in practice, especially when provided by the public. Instead of requesting an explicit number, the proposed
system relied on GUI, by which a participant could provide the experienced factors. Particularly, they could choose, for
instance, one out of four different rainfall levels, i.e., none, low, moderate, or heavy, as they had actually

IJIRCCE©2022 | An ISO 9001:2008 Certified Journal | 5746

International Journal of Innovative Research in Computer and Communication Engineering
| e-ISSN: 2320-9801, p-ISSN: 2320-9798| www.ijircce.com | |Impact Factor: 8.165 |

|| Volume 10, Issue 6, June 2022 ||

| DOI: 10.15680/IJIRCCE.2022.1006139|

experienced. The Crowd source data were represented by pin icons. Each pin is coded in three different colors with
respect to reported flood levels. Flood forecasting was rendered based on learning of thematic data by an ML method.
The resultant forecast was displayed on a web browser. With this platform, users may access this information from
various devices.

REFERENCES

[1] Natural Disaster Data Book, ADRC, Hyogo, Japan, 2012. ¯

[2] Natural Disaster Data Book, ADRC, Hyogo, Japan, 2015. ¯
[3] D. Komori, S. Nakamura, M. Kiguchi, A. Nishijima, D. Yamazaki, S. Suzuki, A. Kawasaki, K. Oki, and T. Oki,
‘‘Characteristics of the 2011 Chao Phraya River flood in central Thailand,’’ Hydrolog. Res. Lett., vol. 6, no. 6, pp. 41–
46, 2012.
[4] P. Promchote, S.-Y. S. Wang, and P. G. Johnson, ‘‘The 2011 great flood in Thailand: Climate diagnostics and
implications from climate change,’’ J. Climate, vol. 29, no. 1, pp. 367–379, Jan. 2016.
[5] S. Puttinaovarat, P. Horkaew, K. Khaimook, and W. Polnigongit, ‘‘Adaptive hydrological flow field modeling based
on water body extraction and surface information,’’ J. Appl. Remote Sens, vol. 9, no. 1, Dec. 2015, Art. no. 095041.
[6] S. K. Jain, P. Mani, S. K. Jain, P. Prakash, V. P. Singh, D. Tullos, S. Kumar, S. P. Agarwal, and A. P. Dimri, ‘‘A
Brief review of flood forecasting techniques and their applications,’’ Int. J. River Basin Manage., vol. 16, no. 3, pp. 329–
344, Jul. 2018.
[7] N. Belabid, F. Zhao, L. Brocca, Y. Huang, and Y. Tan, ‘‘Near-real-time flood forecasting based on satellite
precipitation products,’’ Remote Sens., vol. 11, no. 3, p. 252, Jan. 2019.
[8] A. K. Kar, A. Lohani, N. Goel, and G. Roy, ‘‘Rain gauge network design for flood forecasting using multi-criteria
decision analysis and clustering techniques in lower Mahanadi river basin, India,’’ J. Hydrol., Regional Stud., vol. 4,
pp. 313–332, Sep. 2015.
[9] M. Dembélé and S. J. Zwart, ‘‘Evaluation and comparison of satellitebased rainfall products in Burkina Faso, West
Africa,’’ Int. J. Remote Sens., vol. 37, no. 17, pp. 3995–4014, Sep. 2016.
[10] A. Foehn, J. G. Hernández, B. Schaefli, and G. De Cesare, ‘‘Spatial interpolation of precipitation from multiple rain
gauge networks and weather radar data for operational applications in Alpine catchments,’’ J. Hydrol., vol. 563, pp.
1092–1110, Aug. 2018.
[11] F. Cecinati, A. Moreno-Ródenas, M. Rico-Ramirez, M.-C. Ten Veldhuis, and J. Langeveld, ‘‘Considering rain
gauge uncertainty using kriging for uncertain data,’’ Atmosphere, vol. 9, no. 11, p. 446, Nov. 2018.
[12] S. Stisen and M. Tumbo, ‘‘Interpolation of daily raingauge data for hydrological modelling in data sparse regions
using pattern information from satellite data,’’ Hydrolog. Sci. J., vol. 60, no. 11, pp. 1911–1926, Sep. 2015.
[13] Q. Hu, Z. Li, L. Wang, Y. Huang, Y. Wang, and L. Li, ‘‘Rainfall spatial estimations: A review from spatial
interpolation to multi–source data merging,’’ Water, vol. 11, no. 3, p. 579, Mar. 2019.

IJIRCCE©2022 | An ISO 9001:2008 Certified Journal | 5747