Stemming as a basis for some non-conventional methods of information retrieval
Abstract The article presents various techniques of stemming, arguing that they are the most important phase in preparing the text for inclusion into full-text databases, especially those using non-Boolean search models. Stemming is a process of text processing us¬ing stemming algorithms, the purpo...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Slovenian Library Association & University of Ljubljana Press (Založba Univerze v Ljubljani)
2000-10-01
|
| Series: | Knjižnica |
| Subjects: | |
| Online Access: | https://journals.uni-lj.si/knjiznica/article/view/13960 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Abstract
The article presents various techniques of stemming, arguing that they are the most important phase in preparing the text for inclusion into full-text databases, especially those using non-Boolean search models. Stemming is a process of text processing us¬ing stemming algorithms, the purpose of which is an automated selection of indexing terms used for content description. The article presents a statistic approach to stemming, morphological and semantical aspects of stemming, and several stemming algorithms. The authors also speak about evaluation criteria and linguistic dependence of such algorithms. At the end, they give more detailed descriptions of some stemming algorithms developed for English, Slovene, French, Japanese and Arabic languages. |
|---|---|
| ISSN: | 0023-2424 1581-7903 |