Lexicon construction method and computing device

Abstract

The invention discloses a lexicon construction method which comprises the following steps: acquiring professional words related to the field of constructed lexicons, and generating an initial lexicon;utilizing the word vector model to process each word in the initial word bank to generate a word to be added, the word to be added being a word similar to each word in the initial word bank; adding words to be added into an initial word bank to generate a new word bank; and for each word in the new word bank, repeating the step of generating the word to be added and the step of generating the newword bank until the number of repetitions is reached, and taking the generated new word bank as the constructed word bank. The invention also discloses a corresponding computing device.

Description

technical field [0001] The invention relates to the technical field of information processing, in particular to a method for constructing a thesaurus and a computing device. Background technique [0002] Domain professional thesaurus refers to a collection of specialized vocabulary reflecting knowledge, terminology, and institutions in a specific field. It is the basis and premise for domain knowledge discovery, semantic analysis, and feature extraction. Therefore, more and more researches have begun to pay attention to the construction of domain-specific thesaurus. [0003] A commonly used solution for constructing a domain thesaurus is to segment sentences based on domain text, and filter through word segmentation and part-of-speech tagging, and use keywords with specified parts of speech as candidate keywords. Then, weight calculation is performed on each candidate keyword through word co-occurrence or TF-IDF (term frequency-inverse text frequency index) algorithm. Fina...

AI Technical Summary

Problems solved by technology

[0005] Taking the vehicle field as an example, due to the complex system structure and various functional features of the vehicle, and the vehicle field involves many upstream and downstream industries such as OEMs, spare parts suppliers, dealers, after-sales and maintenance, etc. , the constructed lexicon cannot accurately and completely describe the field of vehicles. At the same time, with the development of network technology and the increasing number of car users, more and more nicknames, abbreviations, joking words

The traditional professional thesaurus built based on the structure and function of the car often lacks attention to the emerging vocabulary of the Internet, which brings great difficulties to high-level requirements such as feature extraction and knowledge discovery.

Images