Little words can make a big difference for text classification

Published 1 January 1995

Ellen Riloff

Citations117

TL;DR

This work presents results from text classification experiments that compare relevancy signatures, which use local linguistic context, with corresponding indexing terms that do not, and suggests that stopword lists and stemming algorithms may remove or conflate many words that could be used to create more effective indexing Terms.

Abstract

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.

Keywords

Computer Science

Information Retrieval: Data Structures and Algorithms

2,428 Citations1992William B. Frakes, Ricardo Baeza‐Yates

For programmers and students interested in parsing text, automated indexing, its the first collection in book form of the basic data structures and algorithms that are critical to the storage and retrieval of documents.

Viewing morphology as an inference process

661 Citations1993Robert Krovetz

The role of morphological analysis in word sense disambiguation, and in identifying lexical semantic relationships in a machine-readable dictionary, is described.

Automatically constructing a dictionary for information extraction tasks

453 Citations1993Ellen Riloff

Using AutoSlog, a system that automatically builds a domain-specific dictionary of concepts for extracting information from text, a dictionary for the domain of terrorist event descriptions was constructed in only 5 person-hours and the overall scores were virtually indistinguishable.

Journal of the American Society for Information ScienceHow effective is suffixing?

388 Citations1991Donna Harman

Three measures were selected which evaluate performance at given rank cutoff points, such as those cor- responding to a screenful of document titles,such as those responding to the lists of the top ranked documents.

The use of phrases and structured queries in information retrieval

265 Citations1991W. Bruce Croft, Howard R. Turtle +1 more

The results show that using phrases in this way can improve performance, and that phrases that are automatically extracted from a natural language query perform nearly as well as manually selected phrases.

View PDF

ACM Transactions on Information SystemsInformation extraction as a basis for high-precision text classification

206 Citations1994Ellen Riloff, Wendy G. Lehnert

An approach to text classification that represents a compromise between traditional word-based techniques and in-depth natural language processing and an automated method for empirically deriving appropriate threshold values is described.

Journal of the American Society for Information ScienceThe effectiveness of a nonsyntactic approach to automatic phrase indexing for document retrieval

132 Citations1989Joel L. Fagan

It is not likely that phrase indexing of this kind will prove to be an important method of enhancing the performance of automatic document indexing and retrieval systems in operational environments, and a general syntactic analysis facility may be required.

Journal of the American Society for Information ScienceFASIT: A fully automatic syntactically based indexing system

99 Citations1983Martín Dillon, Ann Gray

FASIT (Fully Automatic Syntactically based Indexing of Text) identifies content bearing textual units without a full parse, and, without using semantic criteria, groups these units Into quaslsynonymous sets.

ACM SIGIR ForumThe DARPA TIPSTER project

65 Citations1992Donna Harman

This note is the first of four papers in this issue describing the ongoing work connected with the DARPA TIPSTER Project, and the next papers by three of the contractors involved in the project provide some details on the systems involved, and some of the initial results.

Scholarworks (University of Massachusetts Amherst)Information extraction as a basis for portable text classification systems

22 Citations1994Ellen Riloff

This dissertation addresses the knowledge-engineering bottleneck for a natural language processing task called "information extraction" and presents a system called AutoSlog, which automatically constructs dictionaries for information extraction, given an appropriate training corpus.

View PDF