An Integration Model of Semantic Annotation Based on Synergetic Neural Network. Issue 3 (2nd July 2016)

• Record Type: Journal Article
• Authors: Huang, Zhehuang; Chen, Yidong
• Is Part Of: Intelligent automation & soft computing. Volume 22:Issue 3(2016)
• Page Start: 525
• Page End: 532
• Publication Date: 2016-07-02
• Subjects: SRL -- WSD -- SNN -- semantic annotation -- order parameters; Artificial intelligence -- Periodicals; Intelligent control systems -- Periodicals; 003.5
• DOI: 10.1080/10798587.2016.1158498
• Languages: English
• ISSNs: 1079-8587

Abstract

Abstract: Correct and automatical semantic analysis has always been one of major goals in natural language understanding. However, due to the difficulties in deep semantic analysis, at present, the mainstream studies of semantic analysis are focused on semantic role labeling (SRL) and word sense disambiguation (WSD). Nowadays, these two issues are mostly considered as separate tasks. However, this approach ignores possible dependencies between them. In order to address the issue, an integrative semantic analysis model based on synergetic neural network (SNN) is proposed in this paper, which can easily express useful logic constraints between SRL and WSD. The semantic analysis process can be viewed as the competition process of semantic order parameters. The strongest order parameter will win by competition and desired semantic patterns will be recognized. There are three main innovations in this paper. First, an integrative semantic analysis model is proposed that jointly models word sense disambiguationand semantic role labeling. Second, integrative order parameter is reconstructed to reflect the relation among semantic patterns. Finally, integrative network parameters and integrative evolution equation are reconstructed, which can reflect the relationship of guiding and driving each other between word sense and semantic roles. The experiment results on OntoNotes 2.0 corpus shows the integrative method in this paper has a higher performance for semantic role labeling and word sense disambiguation, and provides a good practicability and a promising future for other natural language processing tasks.