Department of Emerging Technology (SB)

III B.Tech –I Semester

Machine Learning
Subject/Course Code:22PCOAM16
Academic Year:2024-25
Introduction to Machine Learning

PRESENATION BY:

K.Ruchitha
22wj8a6630
Aiml-1
What Is Machine Learning
Machine learning (ML) allows computers to learn and make decisions without being explicitly programmed. It
involves feeding data into algorithms to identify patterns and make predictions on new data. Machine learning is
used in various applications, including image and speech recognition, natural language processing, and
recommender systems.

Why do we need Machine Learning?

Machine Learning algorithm learns from data, train on patterns, and solve or predict complex problems beyond the scope of
traditional programming. It drives better decision-making and tackles intricate challenges efficiently.
Here’s why ML is indispensable across industries:
1. Solving Complex Business Problems
Traditional programming struggles with tasks like image recognition, natural language processing (NLP), and medical diagnosis.
ML, however, thrives by learning from examples and making predictions without relying on predefined rules.
Example Applications:
•Image and speech recognition in healthcare.

•Language translation and sentiment analysis.

2. Handling Large Volumes of Data

With the internet’s growth, the data generated daily is immense. ML effectively processes and analyzes this data, extracting
valuable insights and enabling real-time predictions.
Use Cases:
•Fraud detection in financial transactions.

•Social media platforms like Facebook and Instagram predicting personalized feed recommendations from billions of
interactions.
3. Automate Repetitive Tasks

ML automates time-intensive and repetitive tasks with precision, reducing manual effort and error-prone systems.
Examples:
•Email Filtering: Gmail uses ML to keep your inbox spam-free.

•Chatbots: ML-powered chatbots resolve common issues like order tracking and password resets.

•Data Processing: Automating large-scale invoice analysis for key insights.

4. Personalized User Experience

ML enhances user experience by tailoring recommendations to individual preferences. Its algorithms analyze user behavior to
deliver highly relevant content.
Real-World Applications:
•Netflix: Suggests movies and TV shows based on viewing history.

•E-Commerce: Recommends products you’re likely to purchase.

5. Self Improvement in Performance
ML models evolve and improve with more data, making them smarter over time. They adapt to user behavior and refine their
performance.
Examples:
Voice Assistants (e.g., Siri, Alexa): Learn user preferences, improve voice recognition, and handle diverse accents.
Search Engines: Refine ranking algorithms based on user interactions.
Self-Driving Cars: Enhance decision-making using millions of miles of data from simulations and real-world driving.

What Makes a Machine “Learn”?

A machine “learns” by recognizing patterns and improving its performance on a task based on data, without being explicitly
programmed.
The process involves:
Data Input: Machines require data (e.g., text, images, numbers) to analyze.
Algorithms: Algorithms process the data, finding patterns or relationships.
Model Training: Machines learn by adjusting their parameters based on the input data using mathematical models.
Feedback Loop: The machine compares predictions to actual outcomes and corrects errors (via optimization methods like
gradient descent).
Experience and Iteration: Repeating this process with more data improves the machine’s accuracy over time.
Evaluation and Generalization: The model is tested on unseen data to ensure it performs well on real-world tasks.
In essence, machines “learn” by continuously refining their understanding through data-driven iterations, much like humans
learn from experience
Importance of Data in Machine Learning
• Data is the foundation of machine learning (ML). Without quality data, ML models cannot learn, perform, or make accurate
predictions.
• Data provides the examples from which models learn patterns and relationships.
• High-quality and diverse data improves model accuracy and generalization.
• Data ensures models understand real-world scenarios and adapt to practical applications.
• Features derived from data are critical for training models.
• Separate datasets for validation and testing assess how well the model performs on unseen data.
• Data fuels iterative improvements in ML models through feedback loops.

Types of Machine Learning

1. Supervised learning
Supervised learning is a type of machine learning where a model is trained on labeled data—meaning each input is paired with
the correct output. The model learns by comparing its predictions with the actual answers provided in the training data.
Both classification and regression problems are supervised learning problems.
Example: Consider the following data regarding patients entering a clinic. The data consists of the gender and age of the
patients and each patient is labeled as “healthy” or “sick”.
Gender Age Label
M 48 sick
M 67 sick
F 53 healthy
M 49 sick
F 32 healthy
M 34 healthy
M 21 healthy
In this example, supervised learning is to use this labeled data to train a model that can predict the label (“healthy” or “sick”) for
new patients based on their gender and age. For instance, if a new patient (e.g., Male, 50 years old) visits the clinic, the model can
classify whether the patient is “healthy” or “sick” based on the patterns it learned during training.
2. Unsupervised learning:
Unsupervised learning algorithms draw inferences from datasets consisting of input data without labeled responses. In
unsupervised learning algorithms, classification or categorization is not included in the observations.
Example: Consider the following data regarding patients entering a clinic. The dataset includes unlabeled data, where only the
gender and age of the patients are available, with no health status labels.

Here, unsupervised learning technique will be used to find patterns or

groupings in the data such as clustering patients by age or gender. For
example, the algorithm might group patients into clusters, such as
“younger healthy patients” or “older patients,” without prior knowledge
of their health status.
3.Semi-Supervised Learning: Supervised + Unsupervised Learning
Semi-Supervised learningis a machine learning algorithm that works between the supervised and unsupervised learning so it
uses both labelled and unlabelled data. It’s particularly useful when obtaining labeled data is costly, time-consuming, or
resource-intensive. This approach is useful when the dataset is expensive and time-consuming. Semi-supervised learning is
chosen when labeled data requires skills and relevant resources in order to train or learn from it.
We use these techniques when we are dealing with data that is a little bit labeled and the rest large portion of it is unlabeled. We
can use the unsupervised techniques to predict labels and then feed these labels to supervised techniques. This technique is
mostly applicable in the case of image data sets where usually all images are not labeled.

Let’s understand it with the help of an example.

Example: Consider that we are building a language translation

model, having labeled translations for every sentence pair can be
resources intensive. It allows the models to learn from labeled and
unlabeled sentence pairs, making them more accurate. This
technique has led to significant improvements in the quality of
machine translation services.
4. Reinforcement Learning
Reinforcement Learning (RL) trains an agent to act in an environment by maximizing rewards through trial and error. Unlike
other machine learning types, RL doesn’t provide explicit instructions.
Instead, the agent learns by:
Exploring Actions: Trying different actions.
Receiving Feedback: Rewards for correct actions, punishments for incorrect ones.
Improving Performance: Refining strategies over time.
Example: Identifying a Fruit
The system receives an input (e.g., an apple) and initially makes an incorrect prediction (“It’s a mango”). Feedback is
provided to correct the error (“Wrong! It’s an apple”), and the system updates its model based on this feedback.
Over time, it learns to respond correctly (“It’s an apple”) when encountering similar inputs, improving accuracy through trial,
error, and feedback.
Benefits of Machine Learning
Data-Driven Insights: ML can analyze vast amounts of data to identify patterns and trends that humans might miss. This allows
for better decision-making based on real-world data.

Improved Personalization: ML personalizes user experiences across various platforms. From recommendation systems to
targeted advertising, ML tailors content and services to individual preferences.

Enhanced Efficiency and Automation: ML automates repetitive tasks, freeing up human resources for more complex work. It
also streamlines processes, leading to increased efficiency and productivity.

Advanced Automation and Robotics: ML empowers robots and machines to perform complex tasks with greater accuracy and
adaptability. This is revolutionizing fields like manufacturing and logistics.

Challenges of Machine Learning

Data Bias and Fairness: ML algorithms are only as good as the data they are trained on. Biased data can lead to discriminatory
outcomes, requiring careful data selection and monitoring of algorithms.

Security and Privacy Concerns: As ML relies heavily on data, security breaches can expose sensitive information. Additionally,
the use of personal data raises privacy concerns that need to be addressed.

Interpretability and Explainability: Complex ML models can be difficult to understand, making it challenging to explain their
decision-making processes. This lack of transparency can raise questions about accountability and trust.
THANK YOU