Improved IR in Cohesion Model for Link Detection System

摘要: Given two stories, Story Link Detection System identifies whether they are discussing the same event. Standard approach in link detection system is to use cosine similarity measure find documents linked. Many researchers applied query expansion technique successfully system, where models built from relevant retrieved collection using expansion. In this approach, success depends on quality of information retrieval system. current research, we propose a new for that uses intra-cluster addition with respect document. Our enhances thus improving performance System. Combining improved IR our Cohesion Model provides excellent result detection. Experimental results confirm effect technique.