IR

We describe a framework for the development of a mixed-mode message system for an office environment. Messages may be composed of attributes, text, images, and voice. Message retrieval is based on content. We discuss several issues related to the ...

This paper compares two different approaches for indexing archived text documents. The first approach is based on inversion of words in the text, the second on the generation of a signature file representing the text content. A system reflecting the ...

A system is presented which makes possible the definition and manipulation of views by users who are not familiar with the overall organization of the database. The method used is interactive "conversation" and the user's view of data is an "abstract ...

The possibility to put an Information Retrieval System (IRS)on top of a Data Base Management System (DBMS) is investigated with respect to concurrency control and recovery. The simple mapping of one IRS transaction to one DBMS transaction is analyzed ...

Logical models or schema are used to represent the entities, relationships, and transformations of an information system. The relational model and the relational algebra have been developed to perform this function for database systems, but until ...

A document retrieval system can incorporate many types of flexibility. One example of this is the ability to choose a search strategy that is appropriate for a particular user and query. This paper investigates the use of adaptive mechanisms to control ...

We describe a theoretical model and an on-going series of experiments aimed at a priori query enhancement. The model presents a synthesis of concepts from retrospective and current awareness retrieval systems, employing the user profile as a factor in ...

The Utah Text Retrieval Project addresses a number of areas in information retrieval, including basic system structure, user interfaces integrating information retrieval with word processing, indexing techniques, and the use of specialized backend ...

The Semantic Binary Relationship Model (SBRM) is a first-order formalism which combines an organisationally simple basis (i.e. binary relationships) with the capabilities of semantic networks and logical integrity and deduction rules. The aim is to ...

We have built a prototype distributed information retrieval system known as TBIRD, based on an inverted file and shared between a personal computer, acting as an advanced intelligent terminal, and a timeshared mainframe. It was developed to study the ...

In this paper we, in essence, point out that the methods used in the current vector based systems are in conflict with the premises of the vector space model. The considerations, naturally, lead to how things might have been done differently. More ...

Beijing Document Service(BDS) is a new-type information service providing online information retrieval service mainly in Beijing and other cities besides. This paper presents the background, goals, design considerations of the BDS system and what have ...

We present an integrated method for record clustering and reorganization which can be applied to any set of queries whose frequencies of request are known. The clustering algorithm works by splitting and merging current clusters and, furthermore, ...

The use of cited title terms of a scientific document for automatic indexing is explored. It offers a means of index term selection as well as term relevance weighting, based on author-provided relevance information and Bayes Theorem as in probabilistic ...

In this paper evaluation measures for information retrieval system outputs are investigated from a measurement theoretic point of view. Two axioms are introduced: the axiom of monoto-nicity and the Archimedian axiom. It is shown that the measures ...

Transaction log analysis represents a powerful methodology which allows examination of both user commands and system responses when conducting an online information search. Machine-readable transaction log tapes from online catalogs are obtained and ...

Information retrieval, in the broadest sense of the term, includes a concern with 'expert' or 'knowledge-based' systems and their potential future successors. It is unlikely that sophisticated systems of this sort can be developed in such a way as to ...

The use of expert systems (ES) within information retrieval systems (IRS) seems to be an interesting way, particularly for the query process. Nevertheless we must examine what knowledge we need. We think that the thesaurus may be the kernel of which ...

The aim of this paper is to provide some details about the inference procedures of RESEDA, an "intelligent" Information Retrieval system using techniques borrowed from Knowledge Engineering. A RESEDA prototype has been operational for over a year : ...

The notions of a situation and a distance between situations in a narrow subject domain are introduced. An approach to the construction of an intellectual data base is suggested. The main task of this data base is to find situations which are near to ...

This paper describes the development of a parser based on the Moulton and Robinson (1981) dependency theory of syntax, and several strategies by which we are attempting to apply the outputs of this parser to the processes of Information Retrieval. We ...

The paper starts with a theoretical consideration of the requirements for a computerised information retrieval system to aid open learning within an educational establishment. The requirements for such a system include consideration of the need to ...

An algorithm for text summarization (automatic abstracting) is presented which constitutes the text condensation component of TOPIC, a knowledge-based text information system. Based on the results of text parsing knowledge representation structures of ...

In this paper we present the system MARS which has been designed for potential database users working with retrieval systems. It is an instrument that is to assist the retriever in specifying the search queries. MARS is an aid for increasing ...

This paper describes two experiments concerned with term conflation for information retrieval, and the CATALOG retrieval system designed utilizing the results of the experiments. The experiments performed here has as their aim 1) finding a theoretical ...

The automatic indexing system AIR/PHYS and its evaluation by means of a retrieval test with 309 requests and 15,000 documents is described. First, the underlying conception of a rule based approach is given which is suited to the task of a controlled-...

The automatic extraction of words from texts to form the input for information retrieval systems based on inverted files is partly considered on a theoretical basis, and partly in relation to experience gained from developing what has become an ...

The formal modeling of information storage and retrieval systems has been an important element in the analysis and design of these systems. The retrieval mechanism has been viewed as a probablistic decision problem, often involving utilities. One key ...