The following invited talks will be given in the DMSS2009:
Recent Topics on BDDs/ZDDs for Data Mining and Knowledge Discovery
July 7, 15:15 - 16:15
Binary Decision Diagram (BDD) is an efficient data structure for representing Boolean functions. Techniques of BDD manipulation have been developed in the area of VLSI logic design since 1990's. We found that BDD-based techniques can also be applied effectively to problems of data mining and knowledge discovery. Especially, Zero-suppressed BDDs (ZDDs), a special type of BDDs, are suitable for handling sets of sparse combinations that often appear in many practical database analysis. In this talk, we show our recent results of BDD/ZDD-based knowledge discovery techniques, including fast algorithms of frequent itemset mining, various query processing for itemsets, and fast algorithms for extracting hidden item correlations from itemset databases. We also present another interesting applications for dealing with uncertainty, such as probability calculation for large-scale Bayesian Networks.
Data Mining in Marketing
July 8, 13:30 - 14:30
In recent years, data mining has been applied to a wide range of fields, including distribution, production management and marketing. Many businesses have used data mining technology to build their own efficient operational processes. A variety of problems can arise in the application process, because the features and the contents of data which data-miner uses have changed considerably, from "result" to "process." In this talk, I will introduce some case studies and explain current and future research topics under such a circumstance.