发明名称 Method, apparatus, and system for mobile search
摘要 A method, an apparatus, and a system for mobile search are provided. The method for mobile search includes: receiving a search request, where the search request carries keyword information to be searched for, and an instant interest model and a long-term interest model acquired by a search application server; calculating a relevance score value of each member engine according to the search request, meta index information of the member engine, the instant interest model, and the long-term interest model; and selecting one or more member engines according to the relevance score value to search for the keyword information. The searching for the keyword information may integrate an instant interest model and a long-term interest model of a user, so as to provide users with personalized and highly accurate search result information that meets demands.
申请公布号 US8983971(B2) 申请公布日期 2015.03.17
申请号 US201213464391 申请日期 2012.05.04
申请人 Huawei Technologies Co., Ltd. 发明人 Hu Hanqiang;Du Jiachun;Gu Chong;Jia Jiangtao
分类号 G06F17/30;G06Q30/02 主分类号 G06F17/30
代理机构 Huawei Technologies Co., Ltd. 代理人 Huawei Technologies Co., Ltd.
主权项 1. A method for performing mobile search utilizing a plurality of member engines, comprising: receiving a search request by a search server, wherein the search request carries: keyword information to be searched for, and an instant interest model of a user associated with the search request and a long-term interest model of the user which are acquired by a search application server, wherein: the instant interest model of the user is an instant interest model vector formed by score values of N dimensions, wherein: the score value of each of the N dimensions indicates an interest level of the user on said each of the N dimensions, and is calculated according to relevant data of a query sequence q(1), . . . , q(t-1), q(t) in a same search context session initiated by the user, wherein: the same search context session corresponds to the search request, and the search context session comprises a sum of a period of preset time before the current query q(t) occurs and a time when the current query q(t) occurs; calculating by the search server, a relevance score value of each member engine utilizing: the search request, meta index information describing stored contents of the member engine, the instant interest model of the user, and the long-term interest model of the user; and selecting by the search server, one or more member engines to search for the keyword information according to the relevance score value.
地址 Shenzhen CN