Lin K, Chien C and Kerh R (2016). UNISON framework of data-driven innovation for extracting user experience of product design of wearable devices, Computers and Industrial Engineering , 99 :C , (487-502), Online publication date: 1-Sep-2016 .
Joglekar N, Anderson E and Shankaranarayanan G (2013). Accuracy of aggregate data in distributed project settings, Journal of Data and Information Quality , 4 :3 , (1-22), Online publication date: 1-May-2013 .
Hu J, Peng Y, Li D, Yin J and Xiong G Multidisciplinary knowledge modeling from simulation and specification to support concurrent and collaborative design Proceedings of the 10th international conference on Computer supported cooperative work in design III, (570-578)
Tang M, Zhang Y and Zhang G Type-2 Fuzzy Web Shopping Agents Proceedings of the 2004 IEEE/WIC/ACM International Conference on Web Intelligence, (499-503)
Huang C, Chen Y and Chen A An association mining method for time series and its application in the stock prices of TFT-LCD industry Proceedings of the 4th international conference on Advances in Data Mining: applications in Image Mining, Medicine and Biotechnology, Management and Environmental Control, and Telecommunications, (117-126)
Reviewer: A.K. Krishna Menon
Data mining (DM) can be defined as the application of computer algorithms to discover useful knowledge from large databases. The process of mining a database is known as knowledge discovery in databases, and DM is one of the many steps in this process. The potential of the Web as a rich source of distributed data for a myriad of applications has boosted the value of DM in massive computing applications ranging from aerospace design, product manufacturing, tight-tolerance process control, and so on. This book has 22 chapters, organized into four parts. Part 1 provides an overview of data mining in two chapters, which together render an introduction to DM. Part 2 covers application of DM in product design. The seven chapters describe how DM helps a designer to extract patterns from customer needs, to establish viable relations between these patterns and the design specifications, and to group the proposed prototypes based on functional similarity. Such groupings enable benchmarking, modular design, reuse of designs, and mass customization. In designing a prototype, it is desirable to fall back on information gathered from past design sessions. The training data for the learning process would be the design goals and designs that satisfy those goals. The learning algorithm would learn a function that maps a design goal into a viable design. Inductive learning methodology can help in selecting the initial prototypes, after establishing that the goals are achievable. Chapters 3, 4, and 6 show how DM augments organizational learning through transformation of engineering data, hidden in various databases over the years, into elegant mathematical models. Part 3, "Data Mining in Manufacturing," has 11 chapters, and makes up nearly one-half of the book. As the ubiquitous semiconductor chips rush into every nook and cranny of our lives, chip-fabrication attracts much attention. Intelligent process control and shortest manufacturing cycle time have become more crucial than ever. Chapters 8, 9, and 13 deal with issues related to the quick-response quality control of wafers. The methodology-employing neural networks, optimal sampling, and statistical homogeneity tests to a wafer-bin map-has proven to be effective in predicting the defect population on whole wafers. Chapter 8 has the details. Chapter 9 takes a theoretical approach to mining the data on process and quality in wafer fabrication. The complexity of the wafer manufacturing process generates a large number of attributes. The methodology described in chapter 13 automatically decomposes the input attributes into smaller subsets. A classification model is built for each subset after which they are recombined using Bayesian combination. The technique has been verified using a variety of semiconductor databases. Database, pattern recognition, machine learning, statistics, and visualization are said to be the core technologies for DM. However, chapter 16 is the only one in this book that directly covers visualization. Industrial monitoring involves signals from multiple sensors, for example, spatial, temporal, and spectral. Integration of these signals, to yield the status or progress of a process, reduces the need for repeated interpretation of low-level sensor data, thereby increasing the productivity of the processing plant. The chapter discusses the effectiveness of transforming a variety of monitored signals into color information, and processing the composite data with DM technology. Part 4, "Enabling Technologies for Data Mining in Design and Manufacturing," has four chapters. Knowledge engineering is vital to DM. The two challenging issues of knowledge engineering are accumulating huge volumes of knowledge, and supporting heterogeneous knowledge and processing. Reusing legacy knowledge systems, integrating knowledge systems with legacy databases, and enabling the sharing of the databases by multiple knowledge systems can resolve the first issue. The architecture for realizing this idea is described in chapter 19. The book has some minor flaws. There is so much overlap in the contents of chapters 1 and 2 that most readers would wish the two were merged. There are also several grammatical mistakes and flaws in sentence construction in several chapters. For the first time, this book has brought together the reports on DM from the research organizations and the manufacturing sector. It has succeeded in clarifying, for the benefit of practitioners and graduate students in industrial and mechanical engineering, the integration of DM into design and manufacturing activities, in presenting a wide range of domains to which DM can be applied, and in illustrating how to overcome problems in a design or manufacturing environment. The book is certain to become a significant milestone in the field of data mining. Online Computing Reviews Service
Become a reviewer for Computing Reviews.
As an important data mining and knowledge discovery task, association rule mining searches for implicit, previously unknown, and potentially useful pieces of information—in the form of rules revealing associative relationships—that are embedded in the .
