This repository consists of the unsupervised clustering techniques to used to develop a social networking platform in an educational institute.
The following repository depicts the way in which self-organizing maps can be used to connect people based on similar mindsets. The differences between another clustering method – K means clustering and self-organising maps was also brought out. Using self-organising maps we can also propose an architecture to develop an application that can bring about social networking on a professional scale. A data-set was simulated which consists of various input attributes such as student grades in 5 different subjects and student's profeciency ratings in 4 different fields. The .csv file - example1.csv shows the dataset. The file stud_som.py is the code which generates a self-organising map which shows the gradual differentiation in coloration between vaious groups of students. The file Kmeans.py is the code which generates a k-means clustering of the students, which consists of definate boundaries between different groups of students