Concept Extraction Based on Hybrid Approach Combined with Semantic Analysis

Xian-ming YAO, Jian-hou GAN, Jian XU

Abstract


This paper proposed a hybrid approach for concept extraction, which combined linguistic, statistical and semantic analyses. Similar to traditional methods, linguistic and statistical analyses were reserved; this paper innovatively adopted semantic analysis to analyze the semantic relationships of concepts in order to get a fine grained result. Another creative work is the combination of several statistical approaches from different perspective to filter the candidate concepts. Experiments had shown a promising result.

Keywords


Concept extraction, Domain relevance, Domain consensus, Semantic analysis


DOI
10.12783/dtetr/amma2017/13380

Full Text:

PDF

Refbacks

  • There are currently no refbacks.