| Scientific and technological resources play a very important role in academic research and technology development.With the development of Internet search engines,document retrieval becomes more and more convenient.However,massive scientific and technological resources contain a lot of implicit knowledge and information,and only being mined and processed deeper can they meet the needs of specific industries and fields.SMEs have a large demand for searching for experts and scientific and technological resources during their quick development.However,many existing science and technology search engines have limited knowledge of scientific and technological resources,the search results are not ideal and cannot meet the search needs of enterprises.This paper has done some research focused on optimizing the full-text search algorithm for scientific and technological resources and improving the search results of finding experts in specific fields.This paper mainly put forward a scientific and technological resources search and ranking algorithm based on document scoring model and relevance,which is used to improve the full-text search results of scientific and technological resources;and proposes a method for searching experts in specific fields,which can help enterprises efficiently find outstanding experts.The concrete research work of this paper is as follows:(1)Collection and preprocessing of big data of science and technology resources.Use distributed web crawler to collect the open scientific and technological resources and experts' data on the Internet.Then clean and preprocess the crawled raw data to make it as structured data so that they can be operated and indexed for the search engine.(2)Proposed a sort and search algorithm of scientific and technological resources based on document scoring model and relevance.The algorithm constructs the document scoring model according to some characteristics of the document itself,and then calculates the comprehensive score according to the document scoring model and the search relevancy.Search engine uses this algorithm to rank search results during search.Experiments show that the algorithm improves the search results and can better meet the search needs of enterprises.(3)Proposed a method for searching experts in specific fields.According to the field keyword and word segmentation matching method,this method carries out intelligent annotation of mass scientific and technological resources documents and at the same time completes the field affiliation annotation of the corresponding experts.And proposed a model of professional competence for expert in specific field,which is used to estimate the ability value of experts in various fields by means of statistics and other methods.It also achieved the function of searching expert in specific field by offline computing.Experiments show that this method can intelligently finish the ranking of experts in specific field by big data computation,and achieve better search results for expert searching.Based on the above research results,a search engine for enterprise innovation was developed and put into practical application.It effectively solved the problem of searching experts and scientific and technological resources in the process of enterprise innovation. |
PDF Full Text Request