AUTHOR=Szymański Julian , Dziubich Tomasz 

TITLE=Spectral Clustering Wikipedia Keyword-Based Search Results

JOURNAL=Frontiers in Robotics and AI

VOLUME=Volume 3 - 2016

YEAR=2017

URL=https://www.frontiersin.org/journals/robotics-and-ai/articles/10.3389/frobt.2016.00078

DOI=10.3389/frobt.2016.00078

ISSN=2296-9144

ABSTRACT=The paper presents an application of spectral clustering algorithms used for grouping Wikipedia search results.
The main contribution of the paper is a representation method for Wikipedia articles that has been based on combination of words and links and it has been used to categorize search result in this repository. 
We evaluate the proposed approach with Primary Component Analysis and show, on the test data, how usage of cosine transformation to create combined representations influence   data variability.  On  sample test datasets we also show how combined representation improves the data separation that increases overall results of data categorization.
The paper reviews the three main spectral clustering methods and we test their usability for text categorization
comparing them using external validation criteria with standard clustering quality measures. 
Discussion on descriptiveness of evaluation measures and performed experiments on test datasets
allows us to select the one spectral clustering algorithm that has been implemented 
in our system.
We give a brief description of the  system architecture 
 that groups on-line Wikipedia articles 
retrieved with user-specified keywords.
Using the system we show how clustering increases information retrieval effectiveness for Wikipedia data repository.