Research on Clustering Method of Related Cases BasedOn Chinese Text
Journal Title: IOSR Journals (IOSR Journal of Computer Engineering) - Year 2015, Vol 17, Issue 1
Abstract
Abstract: With the movement of population, crimes are showing mobility and professionalism characteristics,the cases which the same suspects committed have the same or similar characteristics, it is helpful to find therelated cases. In this article, firstly, analyzes the factors associated with the cases, discusses how to form vectorspace of Chinese cases, then proposes a method to associate cases by joint use of FKM and Canopy clusteringalgorithm.
Authors and Affiliations
Gang Zeng
Keywords
Related Articles
- EP ID EP94858
- DOI -
- Views 103
- Downloads 0
Trust Modeling Scheme using Cluster Aggregation of Messages for Vehicular Ad hoc Networks
Abstract: A VANET is a decentralized network as every node performs the functions of both host and router. The main benefit of VANET communication is the enhanced passenger safety by virtue of exchanging warning messages...
Comparative Analysis of Collaborative Filtering Technique
Today it is almost impossible to retrieve information with a keyword search when the information is spread over several pages. The Semantic Web is an extension of the current web in which information is given &nb...
A Survey on Word Sense Disambiguation
Ambiguity has been always interwoven with human language and its evolution. Some argue that ambiguity of the human languages is a byproduct of its complexity, with words that are frequently used in language of...
A web-based decision support system for lower back pain diagnosis
Abstract: In the era of science and technology, computer is the most spirited constituent in our today’s health care sector with variety of facilities. The use of Artificial Intelligent (AI), Knowledge Management, is the...
Processing of Top-k Selection Queries in Relational Database System
Abstract: In many applications, users specify target values for certain attributes, without requiring exact matches to these values in return. Instead, the result to such queries is typically a rank of the top-k tu...