Research On Multi-relational Decision Tree Classification Algorithm

Multi-relational data mining is an important and rapid development of one of the areas of data mining in recent years. Efficiency of data mining have been an important research topic. For Multi-relational data mining algorithm, the complexity of multi-relational data mining put forwarc to higher requirements on performance of the algorithm. With the traditional data mining algorithm the search space of multi-relational data mining algorithm becomes more complex and much bigger and the main bottleneck of improving algorithm efficiency is the assumption space. In view of the above problems, this paper has proposed an improved algorithm based on multi-relational decisior tree MRDTL-2.This paper has studied deeply on data mining theory, multi-relational data mining theory Especially on multi-relational decision tree classification algorithms and latest technology of multi-relational data mining-tuple ID propogation. This paper has proposed an improved algorithm based on multi-relational decision tree MRDTL-2. it improved algorithm efficiency and user' satisfaction. The improved algorithm is completed under user'guide, and has been improved in two major areas:under user'guide, when data item is no bigger than attribute sendind sending value tuple ID propogation technologies of virtual connecting has been applied to improved algorithrr. based on multi-relational decision tree MRDTL-2;the second, under user'guide, when data item bigger than attribute sendind valve value, supposing null relation Ra,main key,background attribute and class label which is in target relation are sent to background relation,and then Ra instead of background relation perform other operation.Finally, In this paper, it has gave the theoretical proof and the experiment proving of the improved algorithm based on multi-relational decision tree MRDTL-2. this paper proposes MRDTL-2 on the PKDD CUP'99 financial data set. The experimental results show that oui algorithm improves the efficiency of MRDTL obviously.Experimental results show that the algorithms proposed in this paper are more efficient than the current ones, and the anticipated results are realized.