Multilingual news text clustering method, storage medium and terminal device

Abstract

The invention discloses a multilingual news text clustering method, which comprises the following steps: obtaining the text characteristics of each monolingual news text in the multilingual news textin advance; clustering the monolingual news text according to the text characteristics of each monolingual news text based on the keyword coincidence degree, and obtaining a monolingual clustering cluster set of each monolingual news text correspondingly; according to the monolingual clustering set of each of the monolingual news texts, a cross-linguistic clustering set of the multilingual news texts being obtained. Correspondingly, the invention also discloses a computer-readable storage medium and a terminal device. The technical proposal of the invention can realize large-scale clustering of multilingual news texts without relying on multilingual resources, meet the requirements of multilingual network public opinion analysis of Chinese, English, Indonesian and Malay, and improve clustering speed and efficiency.

Description

technical field [0001] The invention relates to the field of natural language processing in information technology, in particular to a multilingual news text clustering method, a computer-readable storage medium and a terminal device. Background technique [0002] With the increasing abundance of Internet information resources, the number of non-English text resources on the Internet is increasing day by day, the multilingual tendency of Internet information sources is increasing, and the analysis of Internet public opinion is also tending to be multilingual. How to achieve accurate and efficient transnational Language clustering has become one of the key issues in multilingual public opinion analysis. [0003] At the same time, the Chinese government and enterprises are increasingly paying attention to the analysis of online public opinion in the countries along the route to avoid risks. Among the countries along the route, Indonesia and Malaysia are the founding countries ...

AI Technical Summary

Problems solved by technology

[0004] The text clustering methods provided by the prior art mainly include "clustering first, then merging" methods, multilingual text clustering methods based on machine translation systems, multilingual text clustering methods based on multilingual dictionaries, and multilingual text clustering methods based on multilingual subject headings. Multilingual text clustering methods based on tables or multilingual ontologies, multilingual text clustering methods based on parallel corpora, and methods based on homologous named entities, etc. However, due to resource acquisition costs such as multilingual dictionaries, thesaurus and parallel corpora High, and there is a problem of incompatibility in the field of clustering text, which restricts the development of large-scale multilingual news text clustering technology

Images