Automatic subject indexing using an associative neural network

Abstract

The global growth in popularity of the World Wide Web has been enabled in part by the availability of browser based search tools which in turn have led to an increased demand for indexing techniques and technologies. As the amount of globally accessible information in community repositories grows, it is no longer cost-effective for such repositories to be indexed by professional indexers who have been trained to be consistent in subject assignment from controlled vocabulary lists. The era of amateur indexers is thus upon us, and the information infrastructure needs to provide support for such indexing if search of the Net is to produce useful results.

The influence of indexing practices and weighting algorithms on document spaces

“…Concepts extracted from a document were used as the input pattern to a concept space represented as a Hopfield network. From this, a network parallel spreading activation process yielded a set of concepts (Chung, Pottenger, & Schatz, 1998). Wacholder et al (2001) introduced an evaluation standard that included three criteria to assess the quality of indexing terms: coherence, thoroughness of the coverage of document content, and usefulness of the index terms.…”

Robust document image understanding technologies

“…In previous work we have developed techniques for classification of and automatic assignment of keywords to documents [14]; this work can be merged and extended with our work on information extraction techniques (e.g., [52]) to aid in the automatic assignment of various types of textual metadata to document images.…”

Improving browsing in digital libraries with keyphrase indexes

“…The related terms may come from a thesaurus, which gathers and records related terms in the collection (e.g. [10]), or from other lexical means (e.g. [1]).…”

“…[43]). In most document collections, however, author-specified keyphrases are not available for the documents; and it is laborious to manually determine and enter terms for each document in a large collection [10]. A keyphrase index, therefore, is usually feasible only if index terms can be determined automatically.…”

“…[1,38]), and to provide facilities for subject browsing (e.g. [10,29]). In addition, there are a variety of projects that have extracted different kinds of information from text, such as document summaries (e.g.…”