Home

Introduction
  Boolean Retrieval
  - Word Association
  - Document Representation
  - Vector Space Model
  - Probabilistic Retrieval 
  - Latent Semantic Indexing
  - Document Classification

Related Sites
Research
  - Projects
  - People
  - Publications
  - Software

Internal
Back to THOR

The interest for information retrieval has existed long before the Internet. The boolean retrieval is the most simple of these retrieval methods and relies on the use of Boolean operators. The terms in a query are linked together with AND, OR and NOT [3]. This method is often used in search engines on the Internet because it is fast and can therefore be used online. This method has also its problems. The user has to have some knowledge to the search topic for the search to be efficient, e.g., a wrong word in a query could rank a relevant document non relevant. The retrieved documents are all equally ranked with respect to relevance and the number of retrieved documents can only be changed by reformulating the query [2].

The Boolean retrieval has been extended and refined to solve these problems. Expanded term weighting operations make ranking of documents possible, where the terms in the document could be weighted according to their frequency in the document [4]. Boolean information retrieval has been combined with content-based navigation using concept lattices, where shared terms from previously attained documents are used to refine and expand the query [5]. The Boolean operators have been replaced with fuzzy operators [6]. Weighted query expansion using a thesaurus [7]. A model based on fuzzy set theory allows the interpretation of a user query with a linguistic descriptor for each term [8,9].