| Summary: | 碩士 === 國立中正大學 === 會計與資訊科技研究所 === 103 === ABSTRACT
Data mining general search for information from a large number of special relevance of information that behind in it. In this thesis, uses qualitative and quantitative research design to study 100 foreign spouse interviews in Taiwan 2014. Applications Grounded Theory" to summed up the the spindle topics of interview records, may be "fake marriage", then processing and analysis data with Data Mining techniques ─ Association Analysis and Decision Tree algorithm. Mainly in research of … Using IT technology as "decision support tool" to reduce the "system failure" problem that caused by human burden and inexperienced and immigration officers in immigrant traffic substantial growth and manpower shortage situation.
The results of this study, identify three association rules─「Comply with relevant information」、「Rhetoric flawless」and 「Couples living together」, and got similar results with the "Decision Tree Model". The correctly rate of prediction up to 95% in this model. Decision Tree analysis of this study can be used as an interview reference model for interview management authorities and interviewing officers. Policy makers could understand overall status of foreign spouses by the distribution results of Descriptive Statistics section, and let us learned the clear Association Rules by the Association analysis section, that are highly compatible with current essential of authorities Standard Operating Procedures.
Keywords: Foreign Spouses Interview, Association Analysis, Decision Tree, Data Mining.
|
|---|
