Improvement of k-means clustering using patents metadata

Authors:
Mihai Vlase;Dan Munteanu;Adrian Istrate
Affiliations:
Department of Computer Science and Engineering, "Dunarea de Jos" University of Galati, Galati, Romania;Department of Computer Science and Engineering, "Dunarea de Jos" University of Galati, Galati, Romania;Department of Computer Science and Engineering, "Dunarea de Jos" University of Galati, Galati, Romania
Venue:
MLDM'12 Proceedings of the 8th international conference on Machine Learning and Data Mining in Pattern Recognition
Year:
2012

Citing 7
Cited 0

Term-weighting approaches in automatic text retrieval

Information Processing and Management: an International Journal
A vector space model for automatic indexing

Communications of the ACM
Clustering Algorithms

Clustering Algorithms
Feature Weighting in k-Means Clustering

Machine Learning
Building a scientific knowledge web portal: the NanoPort experience

Decision Support Systems
Introduction to Information Retrieval

Introduction to Information Retrieval
Patent it yourself, 11th edition

Patent it yourself, 11th edition

Quantified Score

Hi-index	0.00

Visualization

Abstract

Over time, many clustering methods were proposed, but there are many specific areas where adaptations, customizations and modifications of classical clustering algorithms are needed in order to achieve better results. The present article proposes a technique which uses a custom patent model, aiming to improve the quality of clustering by emphasizing the importance of various patent metadata. This can be achieved by computing different weights for different patent metadata attributes, which are considered to be valuable information.