Intelligent word grouping method and apparatus, and apparatus for intelligent word grouping

Abstract

Embodiments of the present invention provide an intelligent word grouping method and apparatus, and an apparatus for intelligent word grouping. The method specifically comprises: obtaining user inputcontent; obtaining the to-be-grouped vocabulary corresponding to the input content, and the part-of-speech of each to-be-grouped vocabulary; according to a preset part-of-speech matching rule and thepart-of-speech of each to-be-grouped vocabulary, determining the part-of-speech matching score between the adjacent vocabulary in a word-grouping path corresponding to the to-be-grouped vocabulary, wherein the preset part-of-speech matching rule is used to describe the matching relationship between the part-of-speech; according to the part-of-speech matching score between the adjacent vocabulary contained in the word-grouping path, determining the path score of the word-grouping path; and according to the path score, selecting word grouping candidates from the word-grouping path. According tothe embodiments of the present invention, the rationality and quality of the word grouping candidates can be improved, so that even in the case of the failure of intelligent word grouping, relativelyreasonable word grouping candidates can be provided, and input efficiency of the user can be improved.

Description

technical field [0001] The invention relates to the technical field of computer information input, in particular to a method and device for intelligent word formation, and a device for intelligent word formation. Background technique [0002] At present, devices involving interaction usually require users to recognize their operation intentions to interact with the device through an input method system. For example, the user can input an input string, and then the input method system converts the input string into candidates in the corresponding language according to its preset standard mapping rules and displays them, and then displays the candidates selected by the user on the screen. [0003] When there is no entry directly matched by the input string in the thesaurus, the input method system can trigger the intelligent word formation function. The existing intelligent word formation scheme is as follows: find the binary relationship in the binary library, calculate the ...

AI Technical Summary

Problems solved by technology

However, on the one hand, limited by the storage space, the stored binary relations are limited; on the other hand, the binary relations stored in the binary library are often obtained through statistical learning, and it is difficult to ensure that the stored binary relations can Cover all cases; in this way, if the binary relationship in the binary library cannot be hit during the intelligent word formation process, the intelligent word formation will fail

For example, if "90,000-8,000" and "8,000-yuan" are not stored in the binary library, then the words "90,000" and "8,000" corresponding to the input string "jiuwanliangqianyuan", and "8,000" and " Yuan" will not be able to hit the binary relationship in the binary library, which will lead to the failure of intelligent word formation

When the intelligent word formation fails, existing schemes often select the words with the highest word frequency to combine to obtain the corresponding word formation candidates. "Playing Liangqianyuan" is obviously a low-quality and unreasonable candidate, and the probability of meeting the user's input intention is low

Images