| Autor: | Peter Schäuble, Elke Mittendorf, Bojidar Mateev |
|---|---|
| Rok vydání: | 2000 |
| Předmět: |
business.industry
Computer science Search engine indexing Value (computer science) Window (computing) Computer Science::Computation and Language (Computational Linguistics and Natural Language and Speech Processing) Pattern recognition Library and Information Sciences Weighting Range (mathematics) Pattern recognition (psychology) Visual Word Artificial intelligence business Word (computer architecture) Information Systems |
| Zdroj: | Information Retrieval. 3:243-251 |
| ISSN: | 1386-4564 |
| DOI: | 10.1023/a:1026520926673 |
| Popis: | We have applied the well-known Robertson-Sparck Jones weighting to sets of indexing features that are different from word-based features. Our features describe the co-occurrences of words in a window range of predefined size. The experiments have been designed to analyse the value of features that are beyond word-based features but all used retrieval methods can be motivated strictly in the probabilistic framework. Among the several implications of our experiments for weighted retrieval is the surprising result that features that describe the co-occurrences of words in sentence-size or paragraph-size windows are significantly better descriptors than purely word-based indexing features. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Full text from SpringerLink