Scribd
Upload
3 views

CS4038D Data Mining Assignment 2 - 2024 (1)

The assignment for CS4038D Data Mining at NIT Calicut requires students to complete three tasks involving data analysis of Iris flower species. Students must implement classifiers and clustering algorithms, submit their code and documentation in specified formats, and ensure team collaboration for submission. The deadline for submission is November 10, 2024, and evaluation will include a viva/quiz to assess individual contributions.

Uploaded by

SHREE KRIPA SHANKARI S
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
3 views

CS4038D Data Mining Assignment 2 - 2024 (1)

The assignment for CS4038D Data Mining at NIT Calicut requires students to complete three tasks involving data analysis of Iris flower species. Students must implement classifiers and clustering algorithms, submit their code and documentation in specified formats, and ensure team collaboration for submission. The deadline for submission is November 10, 2024, and evaluation will include a viva/quiz to assess individual contributions.

Uploaded by

SHREE KRIPA SHANKARI S
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 2

National Institute of Technology Calicut

Department of Computer Science and Engineering


CS4038D DATA MINING MONSOON 2024 - Assignment 2

Submission deadline (on or before):


10th November, 2024 10:00:00 PM

Complete the following Assignment Questions, and submit your assignment in the moodle (Eduserver)
course page, on or before the submission deadline. Only one member among your team should make a
submission in Eduserver on behalf of the entire team. Include a README.PDF which contains the name
and roll number of the group members. Total of 7 files (3 Zip files, 3 PDF files related with each of the 3
following tasks and the README file) is expected to be submitted as part of this assignment.

During evaluation, the genuinity of the submission and contribution of each member will be checked
either through viva/quiz. The total marks for the assignment is 12. The marks awarded will be based on
the uploaded documents and the viva/quiz.

Assignment Questions

Perform the following tasks and submit the outcomes described for each task.

Dataset: It consists of 150 samples of three species of Iris flowers (Iris setosa, Iris versicolor, and Iris
virginica). From each class, take 75% for training (including 5% validation) and 25% for testing
randomly. Store it in train.csv, train-valid.csv and test.csv. These files should be included in the respective
python code folders.

1. Task 1: Implement Naive Bayes classifier from scratch (do not use libraries) by discretizing the
numeric feature into three equal-width bins. You should use Laplacian correction. After
calculating the likelihood and prior probabilities in trainNBayes.py, save the probability values to
a txtfile. The testNBayes.py should read the probabilities from the txtfile, and can use it for
predicting the class labels for test dataset, and finally find accuracy, precision and recall by
creating the confusion matrix.
Outcome: Zip file (T1CODE_<TeamNumber>.zip) containing all the python codes. Document
(T1_<TeamNumber>.PDF) which describes the experiment setup (if needed) and tabulates the
evaluation measures - Accuracy, Precision and Recall, obtained for the task. Also, write down
your inference on the misclassified samples.
2. Task2.1: Implement k-Nearest Neighbor classifier from scratch without using libraries for ‘k’ =
5, and Euclidean distance measure.
Task2.2: Use the scikit-learn library to create k-Nearest Neighbor classifier. Experiment with
different distance measures, and different ‘k’ values (Use grid search for fine tuning parameters).
You are allowed to do any sort of tuning on the parameters.
Outcome: Zip file (T2CODE_<TeamNumber>.zip) containing python codes. Document
(T2_<TeamNumber>.PDF) which describes the experiment setup (if needed) and tabulates the
evaluation measures - Accuracy, Precision and Recall, obtained for tasks. Also, write down your
inference on the misclassified samples. The optimized model parameters and the performance of
the model should be tabulated well.

3. Task3: Implement k-means clustering using scikit-learn library. Make use of the ground truth
class labels only for cluster evaluation. Find the optimal number of clusters using any automated
method. Experiment with different distance measures.
Outcome: Zip file (T3CODE_<TeamNumber>.zip) containing python codes. Document
(T3_<TeamNumber>.PDF) which describes the experiment setup (if needed) and tabulates the
extrinsic and intrinsic cluster evaluation measures obtained for the task. The optimized model
parameters and the performance of the model should be tabulated well.

You might also like