Research On Multidimensional Analysis Method Of Search Engine User Satisfaction

With the development of the Internet,the arrival of big data age,the search engine has become the main entrance for people to get information.Search engine system has been the pursuit of better search results will be displayed to the user,so the evaluation of the quality of search engine results has become the focus of search engine companies focus.Search engine companies according to their own search engine features and business needs,have a set of their own evaluation system,to evaluate a improvement is good or bad before a search engine improvement to online.User satisfaction is a very important indicator of search engine evaluation,it directly reflects the use of search engines in the search engine to return the results of satisfaction.This paper is in this context,by analyzing the existing user behavior click data,the establishment of a search engine for the full analysis of users of multi-dimensional analysis method.First of all,the existing user click behavior log for cleaning,conversion,analysis of the characteristics of the log,selected 71 dimension attributes to build a multidimensional data model.Then,multidimensional analysis is carried out on the basis of multidimensional data model.In the multidimensional analysis,the results may be difficult to explain,not enough confidence,and because of the larger dimension,in the multi-dimensional cross analysis,there will be a problem of dimensional explosion.Therefore,the association rule mining is introduced into the multidimensional analysis and analysis to solve the problem of insufficient confidence and multi-dimensional explosion.Finally,through the design experiment,to verify the feasibility of multi-dimensional analysis methods and association rules,and to determine the association rules mining some of the parameters used.Through the research of this paper,on the one hand,we set up a multidimensional analysis method for the analysis of user satisfaction,so that the analysis of the user satisfaction can be carried out from multiple dimensions and multiple angles.On the other hand,it also provides a research idea for the future multidimensional analysis method design,and the application of association rules mining to multidimensional analysis method.