Clustering Techniques for Data Analysis

Microsoft Excel is omni present. It is also an excellent vehicle for implementing many algorithms in their basic form to gain a better understanding of the underlying concepts. In my previous post, I had provided an Excel implementation of a classification problem to help understand naive Bayes classifier. In this post I want to demonstrate K-means clustering via an Excel implementation.Clustering implies partitioning data into meaningful groups such that data items within each group are similar but are dissimilar across different groups. One of the most widely used clustering procedure is the K-means clustering algorithm, which seeks to group data into K clusters. The K-means algorithm is meant for situations where data items are represented as vectors of continuous attributes, i.e. numeric attributes, and each cluster is represented by the mean/centroid of respective cluster members. The similarity between different data items in such a case is measured by the Euclidean distance. The steps of the K-means algorithm are:Step 1: Randomly assign every data item to one of the K clusters. (K is a user specified parameter)

Deep Learning for Advanced Analytics

Deep Learning for Advanced Analytics