Data mining and knowledge discovery in real life applications

Published by InTech. Janeza Trdine 9, 51000 Rijeka, Croatia. 2009. ISBN 978-3-902613-53-0 436 p. Первоисточник: http://www.intechopen.com. This book presents four different ways of theoretical and practical advances and applications of data mining in different promising areas like Industrialist, Biological, and Social. Twenty six chapters cover different special topics with proposed novel ideas. Each chapter gives an overview of the subjects and some of the chapters have cases with offered data mining solutions. We hope that this book will be a useful aid in showing a right way for the students, researchers and practitioners in their studies. Contents: A data mining & knowledge discovery process model Knowledge discovery on the grid Rough set theory — fundamental concepts, principals, data extraction, and applications Robust data mining: an integrated approach On the selection of meaningful association rules Hybrid clustering for validation and improvement of subject-classification schemes Automatic product classification control system using RFID tag information and data mining Hyperspectral remote sensing data mining using multiple classifiers combination Content-based image classification via visual learning Clustering parallel data streams Mining multiple-level association rules based on pre-large concepts Data mining applications: promise and challenges Mining spatio-temporal datasets: relevance, challenges and current research directions Benchmarking the data mining algorithms with adaptive neuro-fuzzy inference system in GSM churn management Using data mining to investigate the behavior of video rental customers A novel model for global customer retention using data mining technology Data mining in web applications Application of data mining techniques to the data analyses to ensure safety of medicine usage Data mining in the molecular biology era — a study directed to carbohydrates biosynthesis and accumulation in plants Microarray data mining for biological pathway analysis Development of microsatellite markers by data mining from DNA sequences Quality improvement using data mining in manufacturing processes The deployment of data mining into operational business processes Data mining applied to the instrumentation data analysis of a large dam A data mining algorithm for monitoring PCB assembly quality An overview of data mining techniques applied to power systems 5.1 Semantic analysis Association rules can be classified by their meanings or semantics. A rule is meaningful if its whole content describes something which exists in the domain. In contrast, it is meaningless if its whole content describes something which does not exist. A partially meaningful rule is somewhere in the middle. It is comparable to a negative association rule in "A ~B" or "~A B" forms. As mentioned in Section 2, this type of association is useful in marketing research. It helps identify conflicting items that should not be promoted together, or a replacement item in case that the other is in short supply. Wu et al. (2004) gave another example. Suppose that A normally triggers an alert of event B. Rule "A ~B" suggests that the alert can be postponed because B has not yet happened. This chapter has also demonstrated how such rules were used to gain a better understanding about the domain. However, not all of them are useful for the analysis. Consider the following: 1. {V3 = 1, C5 = 1} {S2 = 2, V1 = 1, V2 = nil, V4 = nil, V5 = nil, V6 = 1} 2. {V3 = 1, C5 = 1} {S2 = 2, C1 = nil, C3 = nil, C4 = nil, C8 = nil} Both rules say that accidents involving bus and being caused by swerving in close distance is likely to occur at the intersection. They give further detail about vehicles involved and causes of accidents, respectively. The extra detail regarding vehicles involved is interesting because it is normal that an accident involves more than one vehicles. On the other hand, it is unlikely (but sometimes possible) that an accident is caused by so many reasons. Hence, adding that there is no other cause of accident is unnecessary. The above example shows different natures of the subjects. In some subjects, multiple or all the items can exist at the same time. But in the others, one or only a couple of items can coexist. Further refinement should be made to the semantic analysis to handle this. A few works have mentioned negative association rules in "~A ~B" form (Antonie & Zaiane, 2004; Wu et al., 2004; Yuan et al., 2002). None explained how to exploit such rules. Only Yuan et al. (2002) suggested that "~A ~B" is equivalent to "B A", but did not elaborate any further. To make sense of this, "~A ~B" is interpreted as that the absence of A causes the absence of B. Therefore, the presence of B would imply the presence of A. It is probable if the association () is perceived as cause-and-effect relationship It is evident that a number of challenges and issues are faced during data mining applications. Some of them include:? How do we determine goals for a DM application?? How do we select the data that will result in achieving the goal?? What type of DM technique(s) should be considered?? How complete is the exploration? Did we miss any useful "nuggets"? The responses to these challenging questions have a major impact on the direction taken and results obtained in a DM initiative. At present, there aren't definitive answers or processes to determine answers to these challenging questions. It is more-or-less a judgement made by the DM team. Today, to address this uncertainty, data mining applications are conducted in an iterative manner. This process allows the DM team to develop a better understanding of the data, domain and data mining technqiues and gain insights with each iteration. The insights gained assists in decision making. The iterative nature of data mining application is reflected in the KDDM process models as well. However, a disadvantage of this approach is that the iterative nature results in a trial-and-error process to data mining applications, which is resource-intensive. Finding approaches that enable a more predictive, controlled, and risk-averse methodology to data mining applications is a challenge that remains to the DM research community. The authors believe that such methodologies require addressing the challenging questions presented in section 3. At the beginning of the chapter, the authors described Data Mining Application as an "experiment" where the goal is discovery of knowledge from large data sets. The analogy between a research experiment and application of DM is evident. Similar to a scientific experiment, where researchers explore the unknown, data mining applications require the DM team to explore for unknown knowledge hidden in the vast data sets. This analogy assists us in determining conducive environments for conducting a DM experiment. Based