Reference no: EM132218849
Purpose
In this assignment, you need to demonstrate your skills for data clustering and dimensionality reduction. There are two parts of this assignment
Instructions
This is an individual assessment task of maximum 20 pages including all relevant material, graphs, images and tables. Students will be required to provide responses for series of problem situations related to their analysis techniques.
They are also required to provide evidence through articulation of the scenario, application of programming skills, analysis techniques and provide a rationale for their response
Task A - Clustering
Download BBC sports dataset from the Cloud. This dataset consists of 737 documents from the BBC Sport website corresponding to sports news articles in five topical areas from 2004-2005. There are 5 class labels: athletics, cricket, football, rugby, tennis. The original dataset and raw text files can be downloaded from here
1. There are 3 files in the dataset corresponding to the feature matrix, the class labels and the term dictionary. You need to read these files in Python notebook and store in variables X, trueLabels, and terms.
2. Next perform K-means clustering with 5 clusters using Euclidean distance as similarity measure. Evaluate the clustering performance using adjusted rand index and adjusted mutual information. Report the clustering performance averaged over 50 random initializations of K-means
3. Repeat K-means clustering with 5 clusters using a similarity measure other than Euclidean distance. Evaluate the clustering performance over 50 random initializations of K-means using adjusted rand index and adjusted mutual information. Report the clustering performance and compare it with the results obtained in step 2
4. For clustering cases (Euclidean distance and the other similarity measure), visualize the cluster centres using Tag cloud using Python package WordCloud.
Task B - (Dimensionality Reduction using PCA/SVD)
For the provided BBC sports dataset, perform PCA and plot the captured variance with respect to increasing latent dimensionality. What is the minimum dimension that captures (a) at least 95% variance and (b) at least 98% variance?