An inferential approach to Information Retrieval and its implementation using a manual thesaurus

Abstract

Most inferential approaches to Information Retrieval (IR) have been investigated within the probabilistic framework. Although these approaches allow one to cope with the underlying uncertainty of inference in IR, the strict formalism of probability theory often confines our use of knowledge to statistical knowledge alone (e.g. connections between terms based on their co-occurrences). Human-defined knowledge (e.g. manual thesauri) can only be incorporated with difficulty. In this paper, based on a general idea proposed by van Rijsbetgen, we first develop an inferential approach within a fuzzy modal logic fmmework. Differing from previous approaches, the logical component is emphasized and considered as the pillar in our approach. In addition, the flexibility of a fuzzy modal logic framework offers the possibility of incorporating human-defined knowledge in the inference process. After defining the model, we describe a method to incorporate a human-defined thesaurus into inference by taking user relevance feedback into consideration. Experiments on the CACM corpus using a general thesaurus of English, Won&ret, indicate a significant improvement in the system's performance.

An Object-Based Approach to Managing Domain Specific Thesauri: Semiautomatic Thesaurus Construction and Query-Based Browsing

“…However, it is inevitable that such a statistical co-occurrence produces a considerable number of meaningless relationships between terms, since it fails not only to reflect their conceptual relationships but also coincide with our intuition [16,27,28]. Another automated thesaurus construction strategy is semantic networks.…”

“…In the automatic query expansion, once users to formulate queries with their own terms, the system automatically appends to the queries thesaurus terms which are conceptually close to the terms [3,10,28]. However, performance analyses from [9,20,27] indicate that this approach hardly achieves acceptable retrieval effectiveness due to inconsistency in the thesaurus. The query-based browsing solves this problem by allowing users to directly search terms appropriate to queries without tediously browsing the thesaurus.…”

“…In general, the retrieval effectiveness of information retrieval systems could be considerably enhanced with thesauri [14,18,27]. Recently numerous researches have been made to develop thesauri, which can fully specify various relationships between domain dependent terms as well as exploit them in a query evaluation [4,17,28].…”

Relevance as resonance: a new theoretical perspective and a practical utilization in information filtering

“…D'', Dempster-Shafer theory is used in Lalmas and Ruthven (1998) to cope with uncertainty, and logical imaging is used in Crestani and van Rijsbergen (1995). Nie and Brisebois (1996) use a fuzzy modal logic in which the calculation of the uncertainty is carried out in a fuzzy logic framework.…”

Evaluation of semantic similarity for sentences in natural language by mathematical statistics methods

“…Ча-ще всего такой словарь формируется путем ручного занесения данных [2] и семантических связей [3]. Но этот подход по определению является малоэффективным и весьма затруднительным -создание словаря достаточно большого объема и постоянная актуализация его содержимого потребует огромных времен-ных и финансовых затрат.…”