Integration of Information Retrieval Techniques into a Referral System for Knowledge Management

information retrieval;MARS;referral system;tfidf;document frequency indexing;knowledge management;expertise location;referral systems

Palathingal, Paul Jose ; Dr Munindar Singh, Committee Chair,Dr Peng Ning, Committee Member,Dr Mona Singh, Committee Member,Palathingal, Paul Jose ; Dr Munindar Singh ; Committee Chair ; Dr Peng Ning ; Committee Member ; Dr Mona Singh ; Committee Member

University:North Carolina State University

关键词: information retrieval;    MARS;    referral system;    tfidf;    document frequency indexing;    knowledge management;    expertise location;    referral systems;   

Others  :  https://repository.lib.ncsu.edu/bitstream/handle/1840.16/793/etd.pdf?sequence=1&isAllowed=y

美国|英语

来源: null

【 摘 要 】

Social networks potentially form a huge repository of private knowledge that is unavailable on any search engine. The Multi Agent Referral system (MARS) prototype is a tool for building and exploiting social networks. Referral chains represent social networks and are used to request for experts in a particular field. The MARS system uses software agents to automate the search of information and expertise.However information about a user's expertise is not readily available. Also currently most systems do not exploit the fact that the user's document repository is a good pointer to other user's interests. By contrast this paper shows how to use Information Retrieval techniques to be able to answer queries and also make referrals on behalf of the user. TFIDF (Term Frequency Inverse Document Frequency) indexing is implemented on user documents and messages to determine user interests. A simulation has been carried out and results of the simulation are presented.This thesis studies the effects of using term and document frequency indexing for information retrieval on user documents and messages in the MARS prototype.