AL-702-C Predictive Analytics - Nov 2023 Question Paper

Under each question, How to answer lists the points an answer should cover, outlined by CampusPrep from the syllabus. Check the details against your textbook.

1. 7 Marks
   aExplain the steps for building a successful data product strategy.
   How to answer

   • Define data product strategy and its goal
   • List steps: identify problem/use-case, data discovery, design, build, deploy, monitor/feedback
   • Mention stakeholders, metrics and iteration
   • Business value and success criteria
2. 7 Marks
   bExplain the different types of data products based on their functionality with suitable example.
   How to answer

   • Define data product types by functionality
   • Classify e.g. dashboards, recommendation, prediction, automated decision
   • Give suitable example for each type
   • Compare functionality and use-case
3. 7 Marks
   aWhat steps are involved in converting a string to a JSON Array? Explain with a code.
   How to answer

   • Define JSON array and json.loads
   • Steps: import json, validate string, parse with json.loads
   • Handle exceptions and show resulting list
   • Provide sample Python code and output
4. 7 Marks
   bWhat are CSV and JSNO files in Python? Explain how you will read these files in python?
   How to answer

   • Define CSV and JSON files in Python context
   • Explain csv module and json module
   • Steps/code to read CSV using csv.reader/pandas
   • Steps/code to read JSON using json.load
5. 7 Marks
   aWhat is Data Extraction? Explain Data Extraction Tools and Techniques available in python.
   How to answer

   • Define data extraction
   • List tools/techniques in Python: BeautifulSoup, Scrapy, requests, pandas
   • Explain web scraping vs API vs file extraction
   • Example workflow
6. 7 Marks
   bExplain the different libraries available for Text processing in Python.
   How to answer

   • Define text processing need
   • List libraries: NLTK, spaCy, TextBlob, Gensim, re
   • Explain key functions: tokenization, stemming, lemmatization
   • Comparison of use-cases
7. 7 Marks
   aWrite a Python code to inverse the matrix using Python libraries.
   How to answer

   • Import numpy
   • Create matrix with np.array
   • Compute inverse with np.linalg.inv
   • Print result and handle singular matrix
8. 7 Marks
   bWrite a Python program to display a bar chart for the below marks scored by the student:

   

   How to answer

   • Import matplotlib.pyplot
   • Define subjects list and marks list [69,90,76,88,91]
   • Use plt.bar, xlabel, ylabel, title and show
   • Expected bar chart output
9. 7 Marks
   aCalculate the average, variance and standard deviation in Python using NumPy.
   How to answer

   • Import numpy and create array
   • Use np.mean, np.var, np.std
   • Explain formulas for average, variance, std deviation
   • Show code and sample output
10. 7 Marks
    bWhat is the main purpose of Matplotlib? Explain its application in data visualization.
    How to answer

    • Define Matplotlib as plotting library
    • Explain pyplot, figures, axes
    • Applications: line, bar, scatter, histogram for EDA
    • Example plot code
11. 7 Marks
    aExplain how you will implement Linear Regression using Gradient Descent.
    How to answer

    • Define linear regression hypothesis and MSE cost
    • Derive gradient descent update rules
    • Steps: initialize, compute gradients, update iteratively
    • Python implementation outline
12. 7 Marks
    bExplain the steps involved in using a gradient descent technique in TensorFlow.
    How to answer

    • Define TensorFlow GradientTape and optimizers
    • Steps: define model, loss, compute gradients
    • Apply optimizer.minimize / apply_gradients loop
    • Training iteration and convergence
13. 7 Marks
    aExplain precision and recall classification with help of example.
    How to answer

    • Define TP, FP, FN and formulas for precision and recall
    • Give classification example with confusion matrix
    • Explain trade-off
    • Interpretation for imbalanced data
14. 7 Marks
    bExplain different methods for evaluating classifiers with help of example.
    How to answer

    • List metrics: accuracy, precision, recall, F1, ROC-AUC
    • Explain each with formula and example
    • Confusion matrix based evaluation
    • When to use which metric
15. 14 Marks
    Write a short note on any two:

    1. i)BeautifulSoup Library
    2. ii)JSON Parser
    3. iii)Mini-Batch Gradient Descent
    4. iv)Overfitting and Underfitting
    How to answer

    • Select any two topics from four options
    • Define each with key features/formula
    • Give example/use-case and diagram if relevant
    • Advantages/limitations