The Research Of Behavior And Information-based Recommendation Method In Meta Search Engine

With the explosive growth of content on the Internet,more than 40 thousand GB per second will be generated.In this case,the traditional search engine index coverage is decreasing,and it is difficult for users to provide comprehensive and accurate retrieval services.The meta search engine can effectively improve the coverage rate of the search results by scheduling several different search engines.At the same time,through a combination of search engine and recommendation technology,mining search engine similar relationship between users,make full use of the rich data resources of search engines,to provide more personalized search services for users.However,in the current stage,the research on meta search technology and its recommendation technology is still imperfect,and there are still many problems that need to be solved urgently.This paper mainly studies the design,implementation and application of meta search engine technology and recommendation technology.First of all,this paper analyzes and summarizes the research progress of meta search engine technology and recommendation technology,and on this basis puts forward the design proposal and process of meta search engine recommendation method.Secondly,the key technologies in meta search engine are deeply studied.Including the full analysis of the user in the meta search engine of implicit information and explicit information,complete the construction of user model of meta search engine in the standardized description of the meta search engine users in a gap in this grouping mechanism;recommendation method in the same dimension,never of user information analysis,design a meta search user group classification method of fine-grained engine,independent classification of the meta search engine users;click through the user model of user click log filtering on the recommendation,consider the comprehensive degree of similarity between users to measure correlation content recommendation,to complete the screening and ranking of recommended content.Finally,this paper implements the recommendation method based on user information and behavior in meta search engine,and achieves the effective combination of meta search engine technology and recommendation technology.In order to illustrate the effect of the recommendation method proposed in this paper,the corresponding experimental verification is carried out.Not only the validity of the method recommended in all the key technologies are verified,and the accuracy was evaluated by the method of contrast experiment on the overall recommendation method,the experimental results show that the recommended method proposed in this paper can efficiently provide personalized search services for users,and enhance the user search experience.