Research of demand recognition based on template matching
Weiqiang Liu, Beizhan Wang, Lingyu Chen, Kang Chen, Jiajun Wang · 2016
The search way of traditional search engines is the full-text retrieval based on inverted index, which is also the retrieval way used to query the index database based on the search statement, and this doesn't make good use of the meaning expressed by the search statements, thereby failing to accurately identify the specific demand of users, and this is bound to give users greater search costs. The introduction of vertical search is a solution to this problem of traditional search engines, while the first is to identify the meaning of users' search statements in order to achieve vertical search, which is the problem needs to be solved in natural language processing. This paper designs the demand recognition algorithm based on the template matching, and introduces specific design ideas, and algorithms. The experimental results show that the demand recognition algorithm designed in this paper may well be used to identify the specific demand of the users' search statements.