Recognition of a person named entity from the text written in a natural language

International Conference on Information Technologies in Business and Industry 2016

计算机科学;经济学;工业技术

Dolbin, A.V.^1 ; Rozaliev, V.L.^1 ; Orlova, Y.A.^1

Volgograd State Technical University, 28, Lenina ave., Volgograd

400131, Russia^1

关键词: Efficiency criteria;    Information Modeling;    Latent Dirichlet allocation;    Latent Semantic Analysis;    Natural languages;    Person named entities;    Recognition rates;    Semantic analysis;   

Others  :  https://iopscience.iop.org/article/10.1088/1742-6596/803/1/012033/pdf
DOI  :  10.1088/1742-6596/803/1/012033

来源: IOP

【 摘 要 】

This work is devoted to the semantic analysis of texts, which were written in a natural language. The main goal of the research was to compare latent Dirichlet allocation and latent semantic analysis to identify elements of the human appearance in the text. The completeness of information retrieval was chosen as the efficiency criteria for methods comparison. However, it was insufficient to choose only one method for achieving high recognition rates. Thus, additional methods were used for finding references to the personality in the text. All these methods are based on the created information model, which represents person's appearance.