Method for constructing personalized dictionary applicable to mobile search

Abstract

The invention provides a method for constructing a personalized dictionary applicable to mobile search. A secondary word zone bit code segment hash mechanism introduced in a dictionary structure is used for dividing entries with same first words into a plurality of subzones according to secondary word zone bit codes, so that secondary words can be fast searched in a small range by using a dichotomy, and the dictionary search efficiency is effectively improved. At the same time, relative information comprising classification and use frequency is introduced into each entry structure of the personalized dictionary, classification information of content searched by a user can be directly acquired after segmenting words, so that the mobile search personalized requirement is met; relative treatment of query expansion and query suggestion is performed by a system conveniently.

Description

technical field [0001] The invention relates to the technical field of Chinese information processing in mobile search, and specifically relates to a method for constructing a personalized dictionary in mobile search. Background technique [0002] A word is the smallest unit with certain semantics. In order to realize the machine's understanding of Chinese sentences, it is first necessary to perform word segmentation to determine each word in the sentence. The so-called word segmentation is to segment a sentence according to the meaning of the words in it. Automatic word segmentation is the basic link of Chinese information processing. The dictionary mechanism and processing efficiency referred to by word segmentation directly affect the system processing efficiency and the information that can be provided after word segmentation. [0003] At present, the commonly used dictionary mechanism is mainly based on the three methods of whole word dichotomy, TRIE index tree method ...

AI Technical Summary

Problems solved by technology

[0003] At present, the commonly used dictionary mechanism is mainly based on the three methods of whole word dichotomy, TRIE index tree method and word-for-word dichotomy. Since these methods realize the search for subwords through dichotomy, as the number of entries in the word segmentation dictionary grows, It will cause the query range of subwords to be too large, and the degree of efficiency improvement is very limited

There are also double-word or multi-word hashing mechanisms derived on this basis, but these methods will make the length of the sub-word hash table too long or make it difficult to construct the sub-word hash table, resulting in a complex storage structure of the dictionary that is difficult to manage

At the same time, in view of the fact that the current word segmentation dictionary based on the conventional word segmentation cannot obtain the interests of the user's query content after the word segmentation, so it cannot meet the high-precision and personalized query requirements in mobile search.

Images