wiki:GSoC2016/Clustering

PROJECT TITLE

Name Richhiey
IRC nick richhiey1996
Timezone UTC +0530
Work hours 09:00 - 17:00 or 20:00 - 02:00 IST
Official mentor Parth
Code repository https://github.com/xapian/xapian/tree/richhiey1996/cluster && https://github.com/richhiey1996/xapian/tree/kmeans
Proposal https://summerofcode.withgoogle.com/dashboard/project/6042029100367872/overview/
Public GSoC page https://summerofcode.withgoogle.com/projects/#6042029100367872
  • PROJECT DESCRIPTION

I will be working on the Clustering of Search Results project for GSoC 2016.

The clustering functionality that we aim to implement will be a hybrid of the standard K-means clustering algorithm called as PSO K-means clustering. Particle Search Optimization is mainly used to find the initial clustering centroids to help K-means derive the most optimal clustering result.

The main goal of this project would be to provide a clustering functionality that is fast and provides quality clusters.

Currently, upto the mid term evaluation, we have successfully set up an API and a test clusterer which uses the API. They seem to work fine. The timeline from now onwards shows the work that has been planned ahead of the mid term evaluations :

Last modified 3 years ago Last modified on 25/08/16 10:17:12