FASIT: A fully automatic syntactically based indexing system

Dillon, Martin; Gray, Ann

doi:10.1002/asi.4630340202

Cited by 80 publications

(39 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Researchers in the information retrieval community have experimented with phrase-based indexing to create more effective indexing terms (e.g., [Croft et al, 1991;Dillon, 1983;Fagan, 1989]). However, most of these systems build complex phrases from combinations of nouns, noun modifiers, and occasionally verbs.…”

Section: Discussionmentioning

confidence: 99%

Little words can make a big difference for text classification

Riloff

1995

Proceedings of the 18th Annual International ACM SIGIR Conference on Research and Development in Information Retrieval - SIGIR

View full text Add to dashboard Cite

Most information retrieval systems use stopword lists and stemming algorithms. However, we have found that recognizing singular and plural nouns, verb forms, negation, and prepositions can produce dramatically different text classification results. We present results from text classification experiments that compare relevancy signatures, which use local linguistic context, with corresponding indexing terms that do not. In two different domains, relevancy signatures produced better results than the simple indexing terms. These experiments suggest that stopword lists and stemming algorithms may remove or conflate many words that could be used to create more effective indexing terms.

show abstract

Section: Discussionmentioning

confidence: 99%

Little words can make a big difference for text classification

Riloff

1995

Proceedings of the 18th Annual International ACM SIGIR Conference on Research and Development in Information Retrieval - SIGIR

View full text Add to dashboard Cite

show abstract

“…Une autre approche consiste à fonder l'analyse sur des critères linguistiques (syntaxiques et morphologiques) pour identifier seulement les suites de mots susceptibles de constituer des expressions intéressantes (par exemple, Dillon et Gray 1983, Salton 1988, ou Smart 1992. Avec cette approche, la fréquence d'occurrence peut jouer un rôle secondaire puisque l'on considère que la forme des expressions est davantage garante de leur intérêt pour l'indexation.…”

Section: La Méthode D'identification D'expressions Intéressantesunclassified

Nouveaux horizons en indexation automatique de monographies

Sylva

2015

documentation

View full text Add to dashboard Cite

Quel est l’état de la question en indexation automatique de monographies ? Bien que les premières tentatives d’indexation automatique datent du début des années 1960, elles n’ont toujours pas abouti à des systèmes satisfaisants du point de vue des indexeurs professionnels. Pourtant il y a lieu de s’interroger sur les possibilités actuelles d’indexation automatique, compte tenu du nombre croissant de documents numériques pour lesquels il serait intéressant de fournir un index comme celui qu’on trouve à la fin d’un livre (back-of-the-book index). En outre, les quinze dernières années ont vu des innovations importantes dans le domaine du traitement automatique des langues (TAL), qui pourraient avoir des applications avantageuses pour l’indexation automatique de monographies. Cet article propose de définir la problématique et d’identifier les nouvelles pistes de solutions à explorer afin de dépasser les performances des systèmes actuellement offerts pour l’indexation automatique de monographies.

show abstract

“…Several techniques are described in the literature to obtain them. One of the most frequently used is text simplification [5]: as a first step, we make a single word stemming, after which stopwords are deleted; in the final step, terms are extracted and conflated by means of pattern matching [3], statistical criteria [4], etc. As we can see, most operations lack solid linguistic grounding 6 , which often results in incorrect conflations.…”

Section: Multi-word Term Conflationmentioning

confidence: 99%

Using Syntactic Dependency-Pairs Conflation to Improve Retrieval Performance in Spanish

Vilares

Barcala

Alonso

2000

Computational Linguistics and Intelligent Text Processing

View full text Add to dashboard Cite

Abstract. This article presents two new approaches for term indexing which are particularly appropriate for languages with a rich lexis and morphology, such as Spanish, and need few resources to be applied. At word level, productive derivational morphology is used to conflate semantically related words. At sentence level, an approximate grammar is used to conflate syntactic and morphosyntactic variants of a given multi-word term into a common base form. Experimental results show remarkable improvements with regard to classical indexing methods.

show abstract

FASIT: A fully automatic syntactically based indexing system

Cited by 80 publications

References 10 publications

Little words can make a big difference for text classification

Little words can make a big difference for text classification

Nouveaux horizons en indexation automatique de monographies

Using Syntactic Dependency-Pairs Conflation to Improve Retrieval Performance in Spanish

Contact Info

Product

Resources

About