SEMANTIC BASED ALGORITHM FOR LANGUAGE COMPONENT ANALYSIS OF SPEECH
Article Sidebar
Main Article Content
Abstract
The presented study describes an algorithm for analysis of the language component of speech to detect a certain situation which is connected with some expressed meaning. The basis of the proposed algorithm is a semantic graph which helps to model the semantic context of the detected situation. To construct this graph lexical-semantic database of English WordNet is used which allows to search for related concepts, and to identify the necessary semantic relations between them. The proposed approach has several advantages like the easiness of describing and modelling of target situations, there is no need to train the system, and also taking into account semantic relations within natural language. The obtained test results allow to say that the suggestions made are justifiable and the proposed algorithm outperforms the traditional key-word based approach.
Article Details
This work is licensed under a Creative Commons Attribution 4.0 International License.
References
Introduction to WordNet: An On-line Lexical Database [Electronic resource].
Toutanova, K. Enriching the Knowledge Sources Used in a Maximum Entropy: Part-of-Speech Tagger / K.Toutanova, Ch.D. Manning // In Proceedings of the Joint SIGDAT Conference on Empirical Methods in Natural Language Processing and Very Large Corpora. USA. – 2000. – P. 63 – 70.
Toutanova, K. Feature-Rich Part-of-Speech Tagging with a Cyclic Dependency Network: In Proceedings of HLT-NAACL / K. Toutanova, D. Klein, Ch. Manning. – USA. – 2003. – P. 252 – 259.
Machine Learning and Intelligent Systems, University of California [Electronic source].
Holmes, G. WEKA: a machine learning / G. Holmes // Intelligent Information Systems. – 1994. – № 29. – P. 357 – 361.
Jain, A.K. Data Clustering: 50 Years Beyond K-Means / A.K. Jain // Pattern Recognition Letters. – 2010. – № 8. – Vol. 31. – P. 651 – 666.
Clustering With EM and K-Means [Electronic resource].
Most read articles by the same author(s)
- D. LIPNITSKI, V. SADOV, PROTECTION OF POLIGRAPHIC PRODUCTS USING DIGITAL WATERMARKS, Vestnik of Polotsk State University. Part C. Fundamental Sciences: No. 2 (2023)