This paper surveys the use of Spreading Activation techniques on Semantic Networks in Associative Information Retrieval. The major Spreading Activation models are presented and their applications to IR is surveyed. A number of works in this area are critically analyzed in order to study the relevance of Spreading Activation for associative IR. Key words: spreading activation, information storage and retrieval, semantic networks, associative information retrieval, information processing, knowledge representation.

Semantic search has been one of the motivations of the Semantic Web since it was envisioned. We propose a model for the exploitation of ontology-based knowledge bases to improve search over large document repositories. In our view of Information Retrieval on the Semantic Web, a search engine returns documents rather than, or in addition to, exact values in response to user queries. For this purpose, our approach includes an ontology-based scheme for the semiautomatic annotation of documents, and a retrieval system. The retrieval model is based on an adaptation of the classic vector-space model, including an annotation weighting algorithm, and a ranking algorithm. Semantic search is combined with conventional keyword-based retrieval to achieve tolerance to knowledge base incompleteness. Experiments are shown where our approach is tested on corpora of significant scale, showing clear improvements with respect to keyword-based search.

The paper addresses the problem of the automatic construction of a hypertext to be used for Information Retrieval (IR) purposes. The starting point is modelling a set of raw IR data by means of a conceptual schema. The conceptual schema produced for a specific IR application domain provides the user with a frame of reference in the query formulation process. This schema can be implemented by means of a hypertext tool using a set of automatic techniques for the identification of nodes and links. The hypertext tool can be used for querying processing, as an automatic tool of query expansion, and for browsing, enabling the user to query and move through the IR data using different levels of abstraction. This paper describes the modelling and design process and concentrates on the set of automatic techniques to be used for the automatic construction of the IR hypertext. 1 Introduction Information Retrieval (IR) systems are typically implemented as stand-alone systems for efficient storage...

The paper describes the design and implementation of TACHIR, a tool for the automatic construction of hypertexts for Information Retrieval. Through the use of an authoring methodology employing a set of well known Information Retrieval techniques, TACHIR automatically builds up a hypertext from a document collection. The structure of the hypertext re ects a three level conceptual model that has proved to be quite e ective for Information Retrieval. Using this model it is possible to navigate among documents, index terms, and concepts using automatically determined links. The hypertext is implemented using the HTML hypertext mark up language, the mark up language of the World Wide Web project. It can be distributed on di erent sites and di erent machines over the Internet, and it can be navigated using any of the interfaces developed in the framework World Wide Web project, for example NetScape. Aversion of this paper was presented at RIAO94 conference in New York [6]. This paper supersedes the conference paper in critically relating the work of the authors to previous work in the area of automatic construction of hypertexts for information retrieval and in the completeness of the presentation of both methodological and implementation characteristics of the developed tool. The TACHIR prototype can be seen at work at the following

In this paper we present a network model that can be used to produce conceptual and logical schemas for Information Retrieval applications. The model has interesting adaptabilitycharacteristics and can indeed be instan# tiated in various e#ectiveways. The paper also reports the results of an experimental investigation into the e#ectiveness of associative and adaptive retrieval on the model by means of Neural Networks. The implementation makes use of the learning and generalisation capabilities of the Backpropagation learning procedure to build up and use application domain knowledge in the form of a sub#symbolic knowledge representation. The knowledge is acquired from examples of queries and relevant documents. Three di#erent learning strategies are introduced# their performance analysed and compared with the performance of a traditional Information Retrieval system. 2 Contents 1

, CONTENT-INDEX, PREFACE. And EXAMPLES OF auxiliary data are a classification system and a thesaurus. An example of the representation of an IR document is reported in Figure 1.This representation is ready to be stored and managed by an operational information retrieval system. This IR document has been extracted from the INSPEC on-line database by the ESA-QUEST information retrieval system. The various different parts of the document have been named according to the terminology previously introduced. This figure gives a clearer understanding of what we mean by structured and unstructured data and the differences existing between these two kinds of data. The information retrieval models which have been proposed for application in the information retrieval area (see References [6] and [7] for a general introduction to the subject) differ considerably from one another, basically due to the different nature and structure of the auxiliary data used in the operations of the IR system bas...