Description

Foundations of Machine Learning (CS564 )
Department of CSE, IIT Patna

Marks:- 25 Marks

Instructions:
3. Code should be done in Python or R .
4. You should zip all the required files and name the zip file as roll_no.zip, eg.
1501cs11.zip.
5. Upload your assignment (the zip file) in the following link: https://www.dropbox.com/request/sPldNavubkUTaFhQzaP Y

____________________________________________________________________________ • This assignment is related to clustering algorithm.
____________________________________________________________________________

Dataset Description: Gene Expression Data Set Dataset link:
https://drive.google.com/file/d/1FMlg8OpfHEFwWiso0sdLrmv0IQjK1wcl/view?usp=shari ng
Number of instances (gens) :- 4,655 (N )
Number of samples/features for each gene:- 21 (GSM45021, GSM45022, GSM45023,
GSM45024, GSM45025, GSM45066, GSM45067, GSM45068, GSM45069, GSM45070,
GSM45071, GSM45072, GSM45073, GSM45074, GSM45075, GSM45076, GSM45077, GSM45078, GSM45079, GSM45080, GSM45081)

Implement k-mean clustering step by step (do not use in build packages). In the clustering algorithm, the value of K (number of cluster centers) varies from 2 to √N . Finding a suitable K for any clustering algorithm is very much important. In this assignment, you run your implemented k-means clustering algorithm for 10 times where each time the value of K is randomly selected from 2 to √N .
For each iteration, save your output in a folder named “cluster_number_K” and inside the folder create K text file (cluster0.txt, cluster1.txt, …, clusterK.txt) where each text file contains gene IDs(column 1 of the dataset) of particular cluster.

Example:- Suppose, the value of K in one iteration is 14. So create a folder(using code) named “cluster_number_14” and generate total 14 text file (using code) named cluster0.txt, cluster1.txt, …, cluster14.txt. Each text file contains the set of gene IDs of a particular cluster.