Research On Multi-Relational Data Mining With Bayesian Method

Multi-Relational Data Mining (MRDM) is the multi-disciplinary field dealing with knowledge discovery from relational databases consisting of multiple tables. Mining data which consists of complex/structured objects also falls within the scope of this field, since the normalized representation of such objects in a relational database requires multiple tables. The field aims at integrating results from existing fields such as inductive logic programming, KDD, machine learning and relational databases; producing new techniques for mining multi-relational data; and practical applications of such techniques.Typical data mining approaches look for patterns in a single relation of a database. But for most real-word applications, squeezing data from multiple relations into a single table requires much thought and effort and can lead to loss of information. And now, MRDM is the field whose time has come.On the basis of research upon traditional data mining arithmetic, this dissertation introduces and compares some main techniques of multi-relational data mining, and then from the angle of efficiency and accuracy, it researches into various technologies based on Bayesian method and applies them to MRDM.Chapter 1 is a preface, in which the definition of MRDM will briefly be introduced above all, and then according to the research fields, and the significance of MRDM, we will analysis its application status in quo. In the end, it will give the framework of this dissertation.Chapter 2 summarizes traditional data mining techniques' conceptions, procedures and some analysis ways and means.In the third chapter, we will present techniques in common of MRDM, including ILP, multi-relational association rules, multi-relational classification, multi-relational clustering, etc.From chapter 4 on, we will do some specific and detailed research on classification of MRDM. Chapter 4 mainly applies various Bayesian techniques to...