Selecting method and system of participle lexicon

Abstract

The embodiment of the invention provides a selecting method of a participle lexicon and relates to the technical field of information. The method comprises the following steps: setting an evaluation processing device; leading a plurality of participle lexicons into the evaluation processing device and generating a plurality of evaluation values corresponding to the plurality of participle lexicons; and selecting the maximum evaluation value from the plurality of evaluation values, and taking the participle lexicon corresponding to the maximum evaluation value as a participle lexicon to be selected. The embodiment of the invention further provides a selecting system of the participle lexicon. The problem in the prior art that the participle lexicons cannot be directly and accurately selected is solved; compared with the prior art, the selecting method and system are relatively accurate and convenient and more time is saved; and long-time counting of user behavior characteristics is not needed and a verifiable method is provided for a pertinent search group.

Description

technical field [0001] The embodiment of the present invention relates to the field of information technology, in particular to a method and system for selecting a word segmentation lexicon. Background technique [0002] The word segmentation ability of search text is an important factor affecting the Chinese retrieval quality of search engines. Whether accurate and effective word segmentation is crucial to improving search results and user satisfaction. The commonly used word segmentation method is based on the thesaurus, and by adding a custom thesaurus to improve the word segmentation effect, so the vocabulary of the thesaurus will greatly affect the search effect. [0003] Furthermore, the inventors found that if the search results can be accurately positioned, the user's search experience will be optimized. Judging from the current situation, the search system will first segment the query sentence entered by the user with the help of a tokenizer, and then search. Ther...

AI Technical Summary

Problems solved by technology

But for targeted search clusters (such as video sites), there is no standard document for comparison, so it is difficult to select thesaurus

[0006] Neither of these two methods can directly make accurate judgments on the lexicon for selection.

Images