An Extended Model of Local Pattern Analysis
The model of local pattern analysis provides sound solutions to many multi-database mining problems. In this chapter, we will discuss different types of extreme association rules in multiple databases viz., heavy association rule, high-frequency associati
- PDF / 382,848 Bytes
- 21 Pages / 439.37 x 666.142 pts Page_size
- 5 Downloads / 246 Views
An Extended Model of Local Pattern Analysis
The model of local pattern analysis provides sound solutions to many multi-database mining problems. In this chapter, we will discuss different types of extreme association rules in multiple databases viz., heavy association rule, high-frequency association rule, low-frequency association rule and exceptional association rule. Also, we show how one can apply the model of local pattern analysis more systematically and effectively. For this purpose, we introduce an extended model of local pattern analysis. We apply the extended model to mine heavy association rules in multiple databases. Also, we justify why the extended model works more effectively. We develop an algorithm for synthesizing heavy association rule in multiple databases. Furthermore, we show that the algorithm identifies whether a heavy association rule is high-frequency rule or exceptional rule. We have provided experimental results obtained for both synthetic and real-world datasets and carried out detailed error analysis. Furthermore, we bring a detailed comparative analysis by contrasting the proposed algorithm with some of those reported in the literature. This analysis is completed by taking into consideration the criteria of execution time and average error.
2.1 Introduction In the previous chapter, we have discussed limitations of using a conventional data mining technique for mining multiple large databases. Also we have discussed challenges involved in mining multiple large databases. In many decision support applications, an approximate knowledge stemming from multiple large databases might result in significant savings when being used in decision-making. Hence the model of local pattern analysis (Zhang et al. 2003) used for mining multiple large databases can constitute a viable solution. In this chapter, we show how one can apply the model of local pattern analysis in a systematic and efficient manner for mining non-local patterns in multiple databases. For mining multiple large databases, careful preparation of data collected at the respective branches is of significant importance. In fact, data preparation can be divided into several sub-tasks, so that it makes the overall data mining easy to perform. We divide the overall data mining task into a hierarchy of sub-tasks to be A. Adhikari et al., Developing Multi-database Mining Applications, Advanced Information and Knowledge Processing, DOI 10.1007/978-1-84996-044-1_2, C Springer-Verlag London Limited 2010
15
16
2 An Extended Model of Local Pattern Analysis
performed at each branch, and finally an application could be developed using local patterns at different branch databases. A non-local application might aim at mining non-local interesting patterns in multiple databases, or making a non-local decision based on findings realized in multiple databases. For determining a solution to the latter problem, sometimes we need to compute appropriate statistics based on the patterns discovered in multiple databases. An appropriate statistic t
Data Loading...
