Word segmentation method and apparatus

Abstract

The invention provides a word segmentation method and apparatus. The word segmentation method comprises: establishing an initial word segmentation dictionary according to existing entries; obtaining a first entry set, selecting entries meeting a preset condition from the first entry set, obtaining the word segmentation dictionary, performing word segmentation on the entries meeting the preset condition by adopting the obtained word segmentation dictionary, and updating the obtained word segmentation dictionary by using the entries subjected to the word segmentation, wherein the initial first entry set consists of the existing entries, and the initially obtained word segmentation dictionary is the initial word segmentation dictionary; performing word segmentation on the entries in the first entry set by adopting the updated word segmentation dictionary a second entry set is obtained according to the entries subjected to word segmentation; and when determining that a convergence condition is met, obtaining a word segmentation result according to the second entry set. The method is capable of improving word segmentation effect.

Description

technical field [0001] The present invention relates to the technical field of data processing, in particular to a word segmentation method and device. Background technique [0002] In map navigation products, it is often necessary to search for information points (PointOfInterest, POI). In a geographic information system, a POI can be a house, a shop, a mailbox, a bus stop, and so on. Since the place name input by the user is usually not a standard POI, it is difficult to obtain the query result required by the user by directly matching the POI in the database. In order to obtain the query results required by the user, post-processing matching is usually performed to obtain a fuzzy approximate result as the query result. POIs in the database need to be segmented during post-processing matching, and the performance of word segmentation directly affects the result of post-processing matching. [0003] The existing word segmentation method usually uses a third-party word seg...

AI Technical Summary

Problems solved by technology

[0003] The existing word segmentation method usually uses a third-party word segmentation dictionary to segment POI, but the word segmentation effect is not ideal

Images