Cyrillic Mongolian and traditional Mongolian bilingual knowledge map construction method

Abstract

The invention discloses a Cyrillic Mongolian language and traditional Mongolian language bilingual knowledge map construction method, which comprises the following steps: capturing and preprocessing the open source knowledge map and Mongolian web page resources; Converting the preprocessed Cyrillic Mongolian text into the traditional Mongolian text; Establish the traditional Mongolian knowledge map data model; Identification and Resolution of Named Entities in Traditional Mongolian; Traditional Mongolian fact extraction; Traditional Mongolian knowledge map integration; Establish the knowledgemap of Cyrillic Mongolian and traditional Mongolian. The knowledge map of Cyrillic Mongolian language and traditional Mongolian language is obtained, which solves the problem that the Mongolian language knowledge map which has a certain scale and meets the application requirements and is not open in the prior art, and restricts the research and application development related with the Mongolian language intelligent information processing.

Description

technical field [0001] The invention belongs to the technical field of minority language processing, and relates to a bilingual knowledge map construction method of Cyrillic Mongolian and traditional Mongolian, which is mainly used in semantic analysis, intelligent question and answer, knowledge reasoning, analysis and decision-making and other fields. Background technique [0002] The Knowledge Graph describes the concepts, entities, events, and relationships in the objective world, presents Internet information in a form closer to human cognition, realizes the efficient organization and management of massive information, and provides information for deep processing and utilization. Foundation. As one of the driving forces for the development of natural language processing and artificial intelligence in the information age, knowledge graphs are widely used in fields such as semantic analysis, intelligent question answering, knowledge reasoning, and analysis and decision-mak...

AI Technical Summary

Problems solved by technology

The difficulty in identifying named entities in traditional Mongolian lies in five aspects: first, traditional Mongolian has no initial capitalization similar to that in English text, which makes it difficult to determine the left boundary of named entities; second, there are many concurrent words in traditional Mongolian, Many common nouns or adjectives with good meanings are used as proper nouns at the same time, and it is very difficult to judge whether these words are proper nouns; third, Mongolian word order is in the form of "subject + object + predicate", lacking the ability to distinguish the subject The obvious boundary with the object increases the difficulty of finding named entities from the subject and object; fourth, Mongolian is an agglutinative language, and the case suffix behind the word increases the difficulty of word matching; fifth, with the development of society , there are more and more new and foreign words in Mongolian, which further increases the difficulty of entity recognition

Images