WORD SENSE DESAMBIGUATIONBY HIGH UTILITY PATTERNS IN SENTENCES
No Thumbnail Available
Date
2022-06-27
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Frequent Pattern Mining aims to identify recurrences, called frequent patterns,
among records in databases. These frequent patterns can be sets of frequent items, frequent
sub-sequences or any other type of sub-structures such as graphs, trees, etc.
Moreover, high-utility pattern mining is an emergent task of data mining, which consists in
discovering patterns with high importance in databases. The usefulness of a pattern can be
measured based on various objective criteria such as its profit, frequency, and weight.
On the other hand, disambiguation improves many applications in automatic language
processing (TAL) such as information retrieval, information extraction, machine translation,
or lexical simplification of texts. Schematically, it is a question of choosing what is the most
appropriate meaning for each word of a text.
In this topic, we propose to apply one of the highly useful pattern mining methods for word
disambiguation in a textual corpus.