English synonym recommendation method fusing dictionary and context information

Abstract

A dictionary and context information fused English synonym recommendation method comprises the following steps: inputting a query term word, and obtaining the part-of-speech, English definition and synonyms of the word; calculating a context similarity degree between context information provided by the dictionary definition and a text in which the queried word is located by using the context information provided by the dictionary definition, and carrying out descending sort on the dictionary definition; covering a word to be queried, constructing a text with a mask, inputting the text into the mask language model, and obtaining a context-based candidate word; calculating the semantic similarity between the candidate words and the words to be queried, and screening out qualified candidate words; and matching the qualified candidate words with the dictionary definitions based on the reordered dictionary definitions and dictionary synonyms, and combining the synonyms to generate a new recommended synonym list. According to the method, the defects in the prior art are overcome, after the to-be-queried word and the sentence where the to-be-queried word is located are input, the definition and the synonym of the to-be-queried word most conforming to the current sentence context can be output, and the diversity of the synonym under each definition is enriched.

Description

technical field [0001] The invention relates to the technical field of data processing, in particular to an English synonym recommendation method that integrates dictionaries and context information. Background technique [0002] In English writing scenarios, the synonym recommendation function can help authors write more authentic articles. The existing synonym recommendation methods are divided into the following two types: [0003] First, query various English dictionaries to return the part of speech, definition and synonyms of the word. The advantage of this method is that the part of speech, definition, and synonyms of a word correspond. Users can find the corresponding word definition and synonyms according to the specific text. The disadvantage is that the synonyms in the dictionary are limited and fixed, and context information, multi-speech and polysemous words are not used. , which increases the user's query time. For example, "good" has 3 parts of speech and 28...

AI Technical Summary

Problems solved by technology

The advantage of this method is that the candidate words are diverse and conform to the context of the text to be queried. The disadvantage is that such candidate words are actually co-occurrence words, which are not necessarily synonyms with the word to be queried, and may even be antonyms, especially in the text When the text is short and the context information is insufficient, for example, the text to be queried is "I am fine.", the word to be queried is "fine", and "fine" is masked. Input the masked language model, and the first five candidate words returned are "here", "sorry", "ready", "alive", "not", these words and "fine" are not synonyms, and are not qualified recommended replacements

Images