Unit-2: AI Project Life Cycle

1. Multiple-Choice Questions (MCQs)

1. Which of the following is the first step in the AI project cycle?

o A) Data Collection

o B) Problem Identification

o C) Model Evaluation

o D) Deployment

o Answer: B) Problem Identification

2. What is the primary goal of the data preprocessing stage in the AI project cycle?

o A) To analyze the results

o B) To collect data from various sources

o C) To clean and format the data for the model

o D) To deploy the AI model

o Answer: C) To clean and format the data for the model

3. Which stage of the AI project cycle involves selecting and training the machine learning
model?

o A) Data Collection

o B) Model Development

o C) Model Evaluation

o D) Problem Identification

o Answer: B) Model Development

4. After which stage of the AI project cycle do we deploy the AI system for use in real-world
applications?

o A) Data Collection

o B) Model Evaluation

o C) Problem Identification

o D) Deployment

o Answer: D) Deployment
5. Which of the following is typically part of the evaluation phase in an AI project?

o A) Testing the model's performance on unseen data

o B) Collecting data for training

o C) Creating algorithms for the model

o D) Deploying the system in a production environment

o Answer: A) Testing the model's performance on unseen data

2. Fill in the Blanks

1. The process of converting raw data into a format suitable for modeling is called ___________.

o Answer: Data Preprocessing

2. In the AI project cycle, ___________ involves defining the problem and setting the objectives
before moving forward with the project.

o Answer: Problem Identification

3. __________ is a process of checking how well the AI model performs on data that it has not
seen before.

o Answer: Model Evaluation

4. Once an AI model is developed and evaluated, the next phase is ___________, where it is
made available for real-world use.

o Answer: Deployment

5. The main goal of the ___________ phase is to assess whether the AI model meets the project
requirements and performs as expected.

o Answer: Model Evaluation

3. True/False

1. The AI project cycle begins with the deployment of the AI model.

o Answer: False

2. Data collection involves gathering raw data from various sources to train the AI model.

o Answer: True

3. Model evaluation is the phase where the model is tested using real-world data to determine
its accuracy and effectiveness.
o Answer: True

4. Deployment of the AI system can only occur after thorough evaluation and successful testing
of the model.

o Answer: True

5. In the AI project cycle, the problem identification phase is skipped if the data is already
available.

o Answer: False

4. 2-Mark Questions

1. What is the importance of the problem identification phase in the AI project cycle?

o Answer: The problem identification phase is crucial because it defines the scope and
objectives of the AI project. Without a clear understanding of the problem to be solved,
the project may fail to meet its goals or address the right challenges.

2. What is the role of data preprocessing in the AI project cycle?

o Answer: Data preprocessing plays a key role in cleaning and transforming raw data into a
format that can be easily used by machine learning models. This includes tasks like
handling missing values, removing noise, and normalizing or standardizing data.

3. How does model evaluation help in an AI project?

o Answer: Model evaluation helps determine the performance of an AI model by testing it

on unseen data. It helps assess whether the model can generalize well to real-world
situations and meet the desired accuracy, precision, and recall metrics.

4. Explain the purpose of the deployment phase in the AI project cycle.

o Answer: The deployment phase involves putting the trained and evaluated AI model into
production, where it can be used by real users or systems to perform tasks and make
predictions. It is the phase where the model transitions from testing to actual use.

5. What does model testing involve during the model evaluation phase?

o Answer: Model testing involves running the trained model on a separate testing dataset
to measure how accurately it predicts or classifies unseen data. This helps assess the
model's generalization ability and its readiness for deployment.

5. 4-Mark Questions

1. Describe the AI project cycle and explain the importance of each phase.

o Answer: The AI project cycle consists of several key phases:

1. Problem Identification: This is the first phase, where the problem to be solved is
clearly defined, and the project’s objectives are set. It’s essential for ensuring
the project is aligned with business goals.

2. Data Collection: In this phase, relevant data is gathered from various sources,
ensuring that the data covers all aspects of the problem. High-quality data is
crucial for building an effective model.

3. Data Preprocessing: Raw data often needs cleaning and transformation. In this
phase, missing values are handled, data is normalized, and noise is removed.
This step is essential for improving the quality of the data before feeding it into
the model.

4. Model Development: This is where machine learning algorithms are selected,

trained, and tested using the preprocessed data. The goal is to create a model
that can effectively solve the identified problem.

5. Model Evaluation: Once a model is trained, it must be evaluated to ensure that

it meets the project’s goals. This phase involves testing the model on unseen
data to assess its accuracy and reliability.

6. Deployment: Finally, once the model is evaluated and performs well, it is

deployed into a production environment where it can be used in real-world
applications. Monitoring and maintenance follow to ensure its continued
effectiveness.

2. Explain how data collection and preprocessing affect the performance of an AI model.

o Answer: Data collection and preprocessing significantly impact the performance of an AI

model. During data collection, it is crucial to gather high-quality, relevant data that
accurately represents the problem domain. Poor data can lead to inaccurate models.
Preprocessing ensures that the data is in the correct format, free from noise, and
normalized for consistency, which helps the model learn better patterns. If the data is
not preprocessed properly, the model might learn from irrelevant or misleading
information, which can result in poor performance.

3. What are the challenges faced in the evaluation and deployment phases of an AI project?

o Answer: During the evaluation phase, one challenge is ensuring that the model is tested
on a diverse set of data to avoid overfitting. If the model performs well on training data
but poorly on test data, it may not generalize well. In the deployment phase, challenges
include integrating the AI model with existing systems, ensuring it performs under real-
world conditions, and maintaining its performance over time. Additionally, issues like
model drift, where the model's performance deteriorates due to changes in the
underlying data, must be managed.

4. What factors should be considered when transitioning an AI model from development to

deployment?
o Answer: Several factors should be considered when transitioning an AI model to
deployment, including:

1. Scalability: The model should be able to handle real-time data and process large
volumes of data without performance degradation.

2. Robustness: The model must be able to function correctly under different

conditions, including data variability and unexpected inputs.

3. Integration: The model needs to be integrated into existing software and

infrastructure, ensuring smooth interaction with other systems.

4. Monitoring: Once deployed, the model’s performance should be continuously

monitored to detect any decline in accuracy or emerging issues.

5. Ethical considerations: The deployment must ensure that the AI model is fair,
transparent, and does not perpetuate bias.

5. Why is model evaluation an ongoing process, and how is it performed?

o Answer: Model evaluation is an ongoing process because an AI model's performance

can degrade over time due to changes in the data it is exposed to (e.g., new trends or
behaviors). It is performed using various evaluation metrics such as accuracy, precision,
recall, F1-score, or area under the ROC curve. Models are tested on data that was not
used in training to check how well they generalize to new, unseen situations. Continuous
evaluation allows for improvements and updates to be made to the model based on
real-world performance.

Reason Assertion Questions:

1.

Assertion: In the AI project life cycle, model evaluation is done after the model has been trained and
tested.
Reason: Model evaluation helps determine the performance of the AI model, ensuring it meets the
required accuracy and performs well on unseen data.

Options:
A) Both assertion and reason are true, and the reason is the correct explanation of the assertion.
B) Both assertion and reason are true, but the reason is not the correct explanation of the assertion.
C) The assertion is true, but the reason is false.
D) The assertion is false, but the reason is true.

Answer: A) Both assertion and reason are true, and the reason is the correct explanation of the
assertion.

2.
Assertion: Data collection is the first step in the AI project life cycle.
Reason: The quality and quantity of data collected directly influence the performance and accuracy of
the AI model.

Options:
A) Both assertion and reason are true, and the reason is the correct explanation of the assertion.
B) Both assertion and reason are true, but the reason is not the correct explanation of the assertion.
C) The assertion is true, but the reason is false.
D) The assertion is false, but the reason is true.

Answer: A) Both assertion and reason are true, and the reason is the correct explanation of the
assertion.

3.

Assertion: AI model training involves providing the model with data to learn patterns and make
predictions.
Reason: During model training, the AI system adjusts its internal parameters to reduce the error
between its predictions and the actual outcomes.

Options:
A) Both assertion and reason are true, and the reason is the correct explanation of the assertion.
B) Both assertion and reason are true, but the reason is not the correct explanation of the assertion.
C) The assertion is true, but the reason is false.
D) The assertion is false, but the reason is true.

Answer: A) Both assertion and reason are true, and the reason is the correct explanation of the
assertion.

4.

Assertion: The deployment phase of an AI project occurs after the model has been fully trained and
evaluated.
Reason: In the deployment phase, the AI model is integrated into real-world applications where it can
make predictions on new data.

Options:
A) Both assertion and reason are true, and the reason is the correct explanation of the assertion.
B) Both assertion and reason are true, but the reason is not the correct explanation of the assertion.
C) The assertion is true, but the reason is false.
D) The assertion is false, but the reason is true.

Answer: A) Both assertion and reason are true, and the reason is the correct explanation of the
assertion.
5.

Assertion: Feature engineering is an essential step in the AI project life cycle, where raw data is
transformed into a format suitable for modeling.
Reason: Feature engineering helps in selecting the most relevant features of the data that will improve
the model’s accuracy and efficiency.

Options:
A) Both assertion and reason are true, and the reason is the correct explanation of the assertion.
B) Both assertion and reason are true, but the reason is not the correct explanation of the assertion.
C) The assertion is true, but the reason is false.
D) The assertion is false, but the reason is true.

Answer: A) Both assertion and reason are true, and the reason is the correct explanation of the
assertion.