Seminar-Probabilistic Retrieval Model

Probabilistic Retrieval Model

Probabilistic theory has been used for modelling the retrieval process in mathematical term.In conventional retrieval situation a document is retrieved whenever the keywod set is attached to document set apears similar to the keyword query.In that case the document is consider as relevant to the query. So relevance is a matter of degree and it can be postulated that when the document and query vector are sufficiently similar. the basic tenet of this approach is that for optimal performances. documents should be ranked in order of decreasing probability of relevance to the user. Probabilistic aproaches attempt to calclate the probability that a document will be relevant to the user.Several modes are there for this purpose., here we shall breifly look ninto 3 such models

Marons and Kuhns model

they proposed that the probability that a document relevant to a user can be assesed by the calculaton of Probability, for each document in collection,that user submitting perticular query term would judjed the document relevant

Robertson and Sparck jone Model

In this model the probability that a document can be calculated not for set of user employing perticular query term, but for a set of document having the property in relation to a given user.

Salton and McGill Model

This model takes a diferent aproach, the essence of this model is that it tells us the retieved item whether it is relevant or not, it depends on two basic parameter . it has 2 cost parameter and loss factor . it gives the total loss for a retrieval process..

Reference:

Chowdary.C. G.1999.Introduction to Modern Information Retreval.London:Library Association Publication.