Research On Implementation Method Of Search Engine Of Database Based On Cloud Platform

With the rapid development of electronic commerce information, it becomes more and moreimportant to search information in the increasingly complex network environment.The futuredevelopment trend of electronic commerce will be the combination of search engines and electroniccommerce. In the face of massive data, the traditional database search engines have been incapableof action.The three main problems are the single point failure, the efficiency index and the searchresult accuracy. Thus it is difficult to meet the requirements of application.How to improve theperformance of the system is the problem needed to be solved.Aiming at the shortcomings of traditional database search engine system, we develop our studyon the realization method of search engine based on cloud platform database. Firstly, to improve theindexing and searching efficiency, we propose the architecture of search engine based on cloudplatform database, which can implement the existing database search engine system in HBasedistributed database. Secondly, to simplify the frame of the project, we propose a novel datapretreatment method, which improves the data interaction between the relation database and HBasedistributed database. Thirdly, to improve the accurate rate of search results, we propose the scoringmechanism based on Levenshtein Distance vector distance algorithm, which judges the registrationdata duplication of users through calculating the similarity between the keywords.Finally, we implement a prototype system of database search engine based on cloud platformin the BI department of electric company. The test results show that our system has better accuracy,efficiency and scalability. Our research shows certain theoretical value and good applicationprospect in cloud computing, database search engine and their application fields.