Basic-K means, Bisect K-means and Agglomerative Hierarchical Clustering algorithms
K-means clustering techniques can cluster the data into K clusters depending on the attributes of the data
The module used
- Sum of Squared Error ( SSE ) which is based on Euclidean distance
- can cluster with or without normalizing the data
- basic k-means, bisect k-means and Agglomerative Hierachical Clustering and compared result
- could allow proximity options for either single Link, Complete Link and Group Average
- can measure performance based on Silhouette Coefficient and Rand Statistics on the data sets
Programming Language : Java in Model View Controller design pattern
Module was developed only using standard Java libraries under JDK 8 and not by the use of any predefined data mining libraries