Font Size: a A A

Research On User's Requirement Mining Method For Crowdsourcing Service Market

Posted on:2018-12-06Degree:MasterType:Thesis
Country:ChinaCandidate:Q ZhangFull Text:PDF
GTID:2348330536481938Subject:Software engineering
Abstract/Summary:PDF Full Text Request
With the sweeping progress of service technology and crowdsourcing,people put more attention on the crowdsourcing market.And the crowdsourcing market's develop pattern has changed from traditional demand-driven market to both demanddriven and supply-driven market.The paper proposed a novel method to solve the problem of imbalance and difficult demand matching between the supply and demand,so as to provide a convincing basis for the next accurate recommendation?The paper constructs the supply-demand requirements graph which expresses the relationship between supply and demand.The graph is built by the commerce data from the crowdsourcing market.It provides the foundation for requirements query and service matching.In the new service market,the requir ements of service from user are mostly vague text description.The traditional search can only match the string,but not really understand the user's requirements.This paper proposes a method which using locality sensitive hashing algorithm,keyword match ing and guide question method to analyze user's multiple query needs to solve the fuzzy users' requirements clearly.In this paper,GSR4 FR sub-graph matching algorithm and filter rule algorithm of composite service are proposed,and the quality of service is measured uniformly.In contrast to Nema algorithm,GSR4 FR sub-graph matching algorithm improves the searching time greatly,and average service search quality is good.Filter rule algorithm of composite service is obtained by mining user's historical purchase behavior information,which has been proved to be effective by experimental results.In the new crowdsourcing service market,the needs of users are vague and com-plex.Because human language is fuzzy,the process of matching requirements is also difficult.This paper tries to solve the problem of processing the fuzzy users' require-ments and build demand patterns to match the needs of users,put forward the corres-ponding sub-graph matching algorithm which matches the right service and filter rule algorithm of composite service which filter the priority of composite services.It is worth to solve these problems in the new filed.At last,a tool of service matching ba-sed on supply-demand requirements graph is developed based on the research content of this thesis,which could verified the research content in graphical interface.
Keywords/Search Tags:Crowdsourcing, fuzzy requirements, sub-graph matching, knowledge graph
PDF Full Text Request
Related items