Prerana Educational and Social Trust®
PES Institute of Technology and
Management
NH-206, Sagar Road, Shivamogga-577204
Department of Computer Science and Engineering
Problem Statements Machine Learning
1. Using the Iris Dataset, implement the K-Nearest Neighbors (KNN) algorithm. Train the model
and evaluate it for different values of K (1, 3, 5, 7). Compare the accuracy obtained for each
value of K and present the results using a table or graph.
2. Using the Wine Dataset, implement the KNN algorithm using different distance metrics. Apply
both Euclidean and Manhattan distance measures and compare their performance.
3. Using the Pima Indians Diabetes Dataset, implement KNN classification. Train the model on the
original dataset and then on normalized data. Compare the performance before and after
normalization.
4. Using the Breast Cancer Dataset, implement KNN classification. Train the model using the full
dataset and then using a reduced subset of data. Compare the classification results.
5. Using the Iris Dataset, implement KNN classification. Introduce noise into the dataset and
analyze its effect on model performance.
6. Using the Digits Dataset, implement KNN for multi-class classification. Evaluate the model
using a confusion matrix and accuracy.
7. Using the Wine Dataset, implement both standard KNN and weighted KNN algorithms.
Compare their performance in terms of classification accuracy.
Prerana Educational and Social Trust®
PES Institute of Technology and
Management
NH-206, Sagar Road, Shivamogga-577204
Department of Computer Science and Engineering
8. Using the Salary Dataset, implement Linear Regression. Fit the regression model, plot the
regression line, and predict values for new inputs.
9. Using the Boston Housing Dataset, implement Multiple Linear Regression. Train the model
using multiple features and analyze the contribution of each feature.
10. Using the Auto MPG Dataset, implement both Linear Regression and Polynomial Regression.
Compare the performance of both models using graphical representation.
11. Using the Auto MPG Dataset, implement Polynomial Regression models with degrees 2, 3, and
4, Compare their performance and identify the best model.
12. Using the California Housing Dataset, implement a regression model. Evaluate the performance
using error metrics such as MSE and RMSE.
13. Using the Breast Cancer Dataset, implement Logistic Regression. Train the model, evaluate
accuracy, and display the confusion matrix.
14. Using the Titanic Dataset, implement Logistic Regression. Identify and analyze misclassified
samples.
15. Using the Heart Disease Dataset, implement Logistic Regression. Train the model with and
without feature scaling and compare results.
16. Using the Salary Dataset, implement both Linear Regression and Locally Weighted Regression
(LWR). Compare their performance using graphical representation.
Prerana Educational and Social Trust®
PES Institute of Technology and
Management
NH-206, Sagar Road, Shivamogga-577204
Department of Computer Science and Engineering
17. Using the Auto MPG Dataset, implement LWR. Vary the bandwidth parameter (τ) and analyze
its effect on the regression curve.
18. Using the Iris Dataset, implement the Decision Tree algorithm. Train the model and visualize the
tree structure.
19. Using the SMS Spam Collection Dataset, implement Naive Bayes classification. Train the model
to classify messages as spam or not spam and evaluate accuracy.
20. Using the Iris Dataset, implement both Naive Bayes and KNN algorithms. Compare their
classification performance.
21. Using the Pima Indians Diabetes Dataset, implement Naive Bayes classification. Evaluate model
accuracy and analyze results.
22. Using the Iris Dataset, implement a simple Artificial Neural Network (ANN) for classification.
Train the model and evaluate its performance.
23. Using the MNIST Dataset, implement an ANN model for digit recognition. Train and test the
model and display predictions.
24. Using the Boston Housing Dataset, implement both ANN and Linear Regression models.
Compare their performance on the same dataset.
Prerana Educational and Social Trust®
PES Institute of Technology and
Management
NH-206, Sagar Road, Shivamogga-577204
Department of Computer Science and Engineering
25. Using the Wine Dataset, implement ANN models with different hidden layer configurations.
Compare their performance.
26. Using the Breast Cancer Dataset, implement both ANN and Logistic Regression models.
Compare their classification accuracy.
27. Using the Mall Customers Dataset, implement K-Means clustering. Visualize the clusters and
analyze the grouping.
28. Using the Iris Dataset, apply clustering and compare the obtained clusters with actual class
labels.
29. Using the Wholesale Customers Dataset, implement clustering techniques. Analyze the resulting
customer segments.
30. Using the Mall Customers Dataset, implement K-Means clustering for different values of K. Use
the elbow method to determine the optimal number of clusters.
31. Implement a Multi-Armed Bandit problem. Simulate multiple actions, apply a selection strategy,
and analyze the reward obtained over iterations.
32. Design a Grid World environment. Assign rewards to different states and simulate an agent
learning the optimal path.
33. Implement a basic Q-Learning algorithm. Train the agent and analyze its learning performance
over time.
Prerana Educational and Social Trust®
PES Institute of Technology and
Management
NH-206, Sagar Road, Shivamogga-577204
Department of Computer Science and Engineering