Font Size: a A A

Web Mashup Pattern Mining Based On Frequent Clustering

Posted on:2012-02-21Degree:MasterType:Thesis
Country:ChinaCandidate:X M CaiFull Text:PDF
GTID:2248330395958241Subject:Computer application technology
Abstract/Summary:PDF Full Text Request
Mashup is a kind of facing the end user’s web application mode, it through the polymerization of the existing web content or service to create new web applications, and to meet the needs of the user individuation and diversification. However, along with the rapid increase of application data sources, users want to look for to the required of the Web application became more and more difficult, and lack of understanding of relationships between data sources. How to help the end users looking for novel and the suitable Mashup application service becomes an urgent problem.In order to help users convenient building Web Mashup application, this paper proposes a Web Mashup pattern mining based on frequent clustering algorithm and on the basis of this algorithm offers a face the end users Mashup model recommended system. Firstly, this paper extracts the web application interface parameters information, and carries on the parametric modeling and matching, then forms the candidate itemsets application parameters. And then based on the analysis of the frequent itemsets parameters space to get the currently theme of the frequent. The theme clustering analysis finally gets Mashup core data according to the relevant application service function of the core application in frequently set novelty, and gets the results. This algorithm is tested and analyzed on many actual data sets. And compared with the results of the actual, it found that dig up related services with high accuracy Mashup results and credibility.On the basis of the Mashup mode mining algorithm, this paper builds up a Mashup model recommended system. The system offers different mining methods for different users. The system can not only help to provide users with information to create Mashups, enable them to create richer Web Mashups, but also can correct the Mashupand add new Mashup. And make the system can provide more innovative more realistic Mashup to users. This paper proposes the algorithm considering the construction characteristics, and designs the system of Mashup can provide intelligence support for the user to create Mashup.
Keywords/Search Tags:Web application, Mashup application, frequent clustering
PDF Full Text Request
Related items