Page 245 - Ai Book - 10

F. Long answer type question.
1. Can you elaborate the 4W’s Canvas framework using the college mess scenario? How does it assist in
defining a problem and scoping an AI project?
Ans. The 4 W’s (Who, What, Where, Why) framework is used to identify stakeholders, understand the problem’s
nature, its location, and why it needs a solution. In the college mess context, it helps in defining the
problem and setting project goals effectively.
2. In the AI project life cycle, why is Data Acquisition considered a crucial step? Explain how authentic data
acquisition contributes to the success of the project.
Ans. Data Acquisition is vital as it lays the foundation for AI models. Authentic data ensures the model’s
reliability and accuracy, preventing conflicts. It forms the basis for meaningful analysis and successful
predictions.

3. Explain the significance of Python packages like NumPy, Pandas, and Matplotlib in the context of Data
Science. How do these packages contribute to data analysis and visualization?

Ans. Python packages such as NumPy handle numerical operations, Pandas facilitate data manipulation and
analysis with flexible structures, and Matplotlib aids in data visualization. Together, they make data
analysis and interpretation easier for data scientists.
4. Can you describe the role of statistical measures like Mean, Median, and Variance in data analysis for AI
projects? How do these measures contribute to understand and interpret data patterns?
Ans. Mean gives the average, Median represents the middle value, and Variance measures data spread. These
statistics help analyze data patterns, understand central tendencies, and gauge data variability, forming
the basis for making informed decisions in AI projects.
5. How does the Evaluation phase in the AI project life cycle play a critical role? Explain the steps involved
and why it is necessary to assess the model’s accuracy before deployment.
Ans. The Evaluation phase involves feeding data into the trained model, predicting outcomes, comparing
predictions with actual values, and checking accuracy. This step ensures the model produces desired
results and meets project objectives before real-time deployment.

6. Discuss the importance of data types in Python for Data Science. How do data formats like Spreadsheet,
CSV, SQL, and ZIP contribute to handling and storing data efficiently in Python?

Ans. Data types in Python are crucial for effective data manipulation. Formats like Spreadsheet, CSV, SQL,
and ZIP provide diverse options for storing and handling data. For instance, CSV simplifies tabular data
storage, while SQL facilitates efficient database management, contributing to streamlined data processes
in Python.

G. Application based questions.
1. How can data science be applied in the banking industry? Provide an example.
Ans. Data science is utilized in the banking industry for fraud detection. By analysing patterns in transactions
and user behaviour, AI algorithms can identify anomalies and potential fraudulent activities, ensuring
the security of financial transactions.
2. Describe the application of data science in a shopping comparison website.

Ans. Data science is integral in the functioning of shopping comparison websites. These platforms gather data
from various e-commerce sites, including prices, features, reviews, etc. By analyzing and combining this
information, they provide users with tailored results for product comparisons.

119
119