Knowledge Discovery based Research Papers Recommender System using Improved K-means Techniques
Sandip S. Rabade,  Shweta A. Joshi
The main objective of recommender system is to provide correct and useful recommendations that makes user happy and satisfied. The users are interesting in accessing the document collection which contains the available information. Clustering is the main analytical method used in data mining. For data clustering the generally accepted algorithm is k-means. The similar kind of data presented in the large data sets are tried to be clustered together using k-means. The one of the limitation of the traditional K-means algorithm is that it require a large computational time. Searching and retrieving also reading research documents is more time consuming. To overcome this problem we develop a search engine for recommending research papers which is based on improved K-means algorithm that provide best results with reduced time complexity. To store the papers the MongoDB database is used which can support large number of simple read/write operations per second. MongoDB is NoSQL database. Search engine used is based on clustering and text mining.
Keywords- Clustering, Text mining, K-means algorithm, MongoDB, NoSQL
Unique Identification Number - IJEDR1602319Page Number(s) - 1824-1829Pubished in - Volume 4 | Issue 2 | June 2016DOI (Digital Object Identifier) -    Publisher - IJEDR (ISSN - 2321-9939)
Cite this Article
Sandip S. Rabade,  Shweta A. Joshi,   "Knowledge Discovery based Research Papers Recommender System using Improved K-means Techniques"
, International Journal of Engineering Development and Research (IJEDR), ISSN:2321-9939, Volume.4, Issue 2, pp.1824-1829, June 2016, Available at :http://www.ijedr.org/papers/IJEDR1602319.pdf