Font Size: a A A

The Research Of The Science And Technology Public Service Platform's Satisfaction Degree Based On The Support Vector Machine

Posted on:2010-11-12Degree:MasterType:Thesis
Country:ChinaCandidate:J Y WangFull Text:PDF
GTID:2189360275462118Subject:Business management
Abstract/Summary:PDF Full Text Request
Support Vector Machine (SVM) is one new method based on the statistical learning theory(SLT), which has been one kind of the successful classified tools at present. This paper discusses the method of appraising the science and technology public service platform's satisfaction degree based on the support vector machine. It sets up the appraisal systems of the science and technology public service platform with the method of factor analysis. According to the method of multi-classification, it refines the appraisal category, which makes the appraisal to be more concrete. Through massive investigates, it takes 10 science and technology public service platforms in Qingdao as example. It chooses 8 platforms as study sample and 2 platforms as appraisal example. And the accuracy is 100%.Then it evaluates the category of the appraisal example, points out the limitation of SVM method and proposes the improvement.The research carries on the accurate and effective appraisal. Through setting up the index systems of the science and technology public service platform, it makes the theory and the method research, establishes a set of scientific appraisal model, and obtains a valuable appraisal system, which provides the auxiliary decision information for the government department, provides the innovative direction in the platform improvement, and seeks the way to optimize the platform.It's an exploration to make research of the science and technology public service platform's satisfaction degree based on the support vector machine. The paper opens a new view for appraising the science and technology public service platform, which is from the satisfaction degree. It optimizes the index systems with the method of factor analysis, which has many advantages such as accurate and simple, compared with fuzzy comprehensive evaluation.Simultaneously, it innovates to use the multi-classification method to optimize the appraisal, set up satisfaction degree index system and appraisal model and take 10 science and technology public service platforms in Qingdao as example.
Keywords/Search Tags:support vector machine, the science and technology public service platform, satisfaction degree, multi-classification, factor analysis
PDF Full Text Request
Related items