Recovery Of Documents Using Extended Text-Probabilistic Models
Keywords:
Information Retrieval, Extended Probabilistic Model, Extended Exponential Probabilistic Model
Abstract
This article presents strategies used for retrieval information, based on probabilistic model of information retrieval. These strategies have been adopted the probabilistic model and exponential probabilistic model and were combined with vector features, now are called probabilistic model of extended and stretched exponential probabilistic model. The information retrieval community values of the probability of relevance and non-relevance for the classification of the resulting documents. We present experiment results showing that the combination of probabilistic models with the vector model allows a more efficient recovery, bringing in response to relevant documents that would not be recovered using only one of the models.
Issue
Section
Artigos Originais
The Editorial Board is authorized to make certain changes in the original text to comply with bibliographical norms, orthography and grammar, so that standards in language and style may be maintained. However, the author´s style will be preserved. The concepts developed by the authors are their exclusive concern and responsibility.
In case of approval, the e-mail of acceptance will contain a Letter of Copyright with authors´ name, personal data, and signature of all authors.
In case of approval, the e-mail of acceptance will contain a Letter of Copyright with authors´ name, personal data, and signature of all authors.