As data scientist/analyst, we will tend to work with large sets of data. It could be overwhelming, this is why we have this technique, Dimension Reduction. In a dataset, we might have some variables that their presence or absence does not have any meaningful effect on our output, thus, the need for dimension reduction comes up. In this analysis, instead of regular datasets, we will be performing the dimension reduction on an image to further compress the size. This is to show that dimension reduction can be done on various datasets.

This analysis is based on using an Unsupervised Learning Technique, Association Rules. The aim was to find association or relation that exists between items bought in a store. In summary, what are the chances that customer A buying item A will also buy item B

This analysis is based on using an Unsupervised Learning Technique, Clustering, to find similar groups of features based on a heart disease dataset. Kmeans and Hierarchical clustering are used to identify these clusters. The results are compared to select the most efficient technique.