Real World Data Mining Applications

ISBN-13: 9783319078113 / Angielski / Miękka / 2014 / 418 str.

Mahmoud Abou-Nasr; Stefan Lessmann; Robert Stahlbock

Opis

Data mining applications range from commercial to social domains, with novel applications appearing swiftly; for example, within the context of social networks. The expanding application sphere and social reach of advanced data mining raise pertinent issues of privacy and security. Present-day data mining is a progressive multidisciplinary endeavor. This inter- and multidisciplinary approach is well reflected within the field of information systems. The information systems research addresses software and hardware requirements for supporting computationally and data-intensive applications. Furthermore, it encompasses analyzing system and data aspects, and all manual or automated activities. In that respect, research at the interface of information systems and data mining has significant potential to produce actionable knowledge vital for corporate decision-making. The aim of the proposed volume is to provide a balanced treatment of the latest advances and developments in data mining; in particular, exploring synergies at the intersection with information systems. It will serve as a platform for academics and practitioners to highlight their recent achievements and reveal potential opportunities in the field. Thanks to its multidisciplinary nature, the volume is expected to become a vital resource for a broad readership ranging from students, throughout engineers and developers, to researchers and academics.

Spis treści

Introduction.- What Data Scientists can Learn from History.- On Line Mining of Cyclic Association Rules From Parallel Dimension Hierarchies.- PROFIT: A Projected Clustering Technique.- Multi-Label Classification with a Constrained Minimum Cut Model.- On the Selection of Dimension Reduction Techniques for Scientific Applications.- Relearning Process for SPRT In Structural Change Detection of Time-Series Data.- K-means clustering on a classifier-induced representation space: application to customer contact personalization.- Dimensionality Reduction using Graph Weighted Subspace Learning for Bankruptcy Prediction.- Click Fraud Detection: Adversarial Pattern Recognition over 5 years at Microsoft.- A Novel Approach for Analysis of 'Real World' Data: A Data Mining Engine for Identification of Multi-author Student Document Submission.- Data Mining Based Tax Audit Selection: A Case Study of a Pilot Project at the Minnesota Department of Revenue.- A nearest neighbor approach to build a readable risk score for breast cancer.- Machine Learning for Medical Examination Report Processing.- Data Mining Vortex Cores Concurrent with Computational Fluid Dynamics Simulations.- A Data Mining Based Method for Discovery of Web Services and their Compositions.- Exploiting Terrain Information for Enhancing Fuel Economy of Cruising Vehicles by Supervised Training of Recurrent Neural Optimizers.- Exploration of Flight State and Control System Parameters for Prediction of Helicopter Loads via Gamma Test and Machine Learning Techniques.- Multilayer Semantic Analysis In Image Databases.

O autorze

Dr. Abou-Nasr is a Senior Member of the IEEE and Vice Chair of the Computational Intelligence & Systems Man and Cybernetics, Southeast Michigan Chapter.  He has received the B.Sc. degree in Electrical Engineering in 1977 from the University of Alexandria, Egypt, the M.S. and the Ph.D. degrees in 1984 and 1994 respectively from the University of Windsor, Ontario, Canada, both in Electrical Engineering.  Currently he is a Technical Expert with Ford Motor Company, Research and Advanced Engineering, Modern Control Methods and Computational Intelligence Group, where he leads research & development of neural network and advanced computational intelligence techniques for automotive applications.   His research interests are in the areas of neural networks, data mining, machine learning, pattern recognition, forecasting, optimization and control.  He is an adjunct faculty member of the computer science department, Wayne State University, Detroit, Michigan and was an adjunct faculty member of the operations research department, University of Michigan Dearborn. Prior to joining Ford, he held electronics and software engineering positions with the aerospace and robotics industries in the areas of real-time control and embedded communications protocols.  He is an associate editor of the DMIN'09-DMIN'14 proceedings and a member of the program and technical committees of IJCNN, DMIN, WCCI, ISVC, CYBCONF and ECAI. He is also a reviewer for IJCNN, MSC, CDC, Neural Networks, Control & Engineering Practice and IEEE Transactions on Neural Networks & Learning Systems.  Dr. Abou-Nasr has organized and chaired special sessions in DMIN and IJCNN conferences, as well as international classification competitions in WCCI 2008 in Hong Kong and IJCNN2011 in San Jose CA.

Dr. Lessmann received a M.Sc. and a Ph.D. in Business Administration from the University of Hamburg (Germany) in 2001 and 2007, respectively. He is currently employed as a lecturer in Information Systems at the University of Hamburg. Stefan is also a member of the Centre for Risk Research at the University of Southampton, where he teaches courses in Management Science and Information Systems. His research concentrates on managerial decision support and advanced analytics in particular. He is especially interested in predictive modeling to solve planning problems in marketing, finance, and operations management. He has published several papers in leading scholarly outlets including the European Journal of Operational Research, the ICIS Proceedings or the International Journal of Forecasting. He is also involved with consultancy in the aforementioned domains and has completed several technology-transfer projects in the publishing, the automotive and the logistics industry.

Dr. Stahlbock holds a diploma in Business Administration and a PhD from the University of Hamburg (Germany). He is currently employed as a lecturer and researcher at the Institute of Information Systems at the University of Hamburg. He is also lecturer at FOM University of Applied Sciences (Germany) since 2003. His research interests are focused on managerial decision support and issues related to maritime logistics and other industries as well as operations research, information systems and business intelligence. He is author of research studies published in international prestigious journals as well as conference proceedings and book chapters and serves as reviewer for international leading journals as well as a member of conference program committees. He is General Chair of the International Conference on Data Mining (DMIN) since 2006.

Dr. Gary Weiss is an Associate Professor in the Computer and Information Science Department at Fordham University in New York City. His current research involves the mining of sensor data from smartphones and other mobile devices in support of activity recognition and related applications. His Wireless Sensor Data Mining (WISDM) Lab recently released the actitracker activity tracking app (actitracker.com). Prior to coming to Fordham, Dr. Weiss worked at AT&T Labs as a software engineer, expert system developer, and as a data scientist. He received a B.S. degree in Computer Science from Cornell University, an M.S. degree in Computer Science from Stanford University, and a Ph.D. degree in Computer Science from Rutgers University. He has published over fifty papers in machine learning and data mining and his research is supported by funding from the National Science Foundation, Google, and Citigroup.

Dr. Lessmann received a M.Sc. and a Ph.D. in Business Administration from the University of Hamburg (Germany) in 2001 and 2007, respectively. He is currently employed as a lecturer in Information Systems at the University of Hamburg. Stefan is also a member of the Centre for Risk Research at the University of Southampton, where he teaches courses in Management Science and Information Systems. His research concentrates on managerial decision support and advanced analytics in particular. He is especially interested in predictive modeling to solve planning problems in marketing, finance, and operations management. He has published several papers in leading scholarly outlets including the European Journal of Operational Research, the ICIS Proceedings or the International Journal of Forecasting. He is also involved with consultancy in the aforementioned domains and has completed several technology-transfer projects in the publishing, the automotive and the logistics industry.

Dr. Stahlbock holds a diploma in Business Administration and a PhD from the University of Hamburg (Germany). He is currently employed as a lecturer and researcher at the Institute of Information Systems at the University of Hamburg. He is also lecturer at FOM University of Applied Sciences (Germany) since 2003. His research interests are focused on managerial decision support and issues related to maritime logistics and other industries as well as operations research, information systems and business intelligence. He is author of research studies published in international prestigious journals as well as conference proceedings and book chapters and serves as reviewer for international leading journals as well as a member of conference program committees. He is General Chair of the International Conference on Data Mining (DMIN) since 2006.

Dr. Gary Weiss is an Associate Professor in the Computer and Information Science Department at Fordham University in New York City. His current research involves the mining of sensor data from smartphones and other mobile devices in support of activity recognition and related applications. His Wireless Sensor Data Mining (WISDM) Lab recently released the actitracker activity tracking app (actitracker.com). Prior to coming to Fordham, Dr. Weiss worked at AT&T Labs as a software engineer, expert system developer, and as a data scientist. He received a B.S. degree in Computer Science from Cornell University, an M.S. degree in Computer Science from Stanford University, and a Ph.D. degree in Computer Science from Rutgers University. He has published over fifty papers in machine learning and data mining and his research is supported by funding from the National Science Foundation, Google, and Citigroup.

Dr. Lessmann received a M.Sc. and a Ph.D. in Business Administration from the University of Hamburg (Germany) in 2001 and 2007, respectively. He is currently employed as a lecturer in Information Systems at the University of Hamburg. Stefan is also a member of the Centre for Risk Research at the University of Southampton, where he teaches courses in Management Science and Information Systems. His research concentrates on managerial decision support and advanced analytics in particular. He is especially interested in predictive modeling to solve planning problems in marketing, finance, and operations management. He has published several papers in leading scholarly outlets including the European Journal of Operational Research, the ICIS Proceedings or the International Journal of Forecasting. He is also involved with consultancy in the aforementioned domains and has completed several technology-transfer projects in the publishing, the automotive and the logistics industry.

Dr. Stahlbock holds a diploma in Business Administration and a PhD from the University of Hamburg (Germany). He is currently employed as a lecturer and researcher at the Institute of Information Systems at the University of Hamburg. He is also lecturer at FOM University of Applied Sciences (Germany) since 2003. His research interests are focused on managerial decision support and issues related to maritime logistics and other industries as well as operations research, information systems and business intelligence. He is author of research studies published in international prestigious journals as well as conference proceedings and book chapters and serves as reviewer for international leading journals as well as a member of conference program committees. He is General Chair of the International Conference on Data Mining (DMIN) since 2006.

Dr. Gary Weiss is an Associate Professor in the Computer and Information Science Department at Fordham University in New York City. His current research involves the mining of sensor data from smartphones and other mobile devices in support of activity recognition and related applications. His Wireless Sensor Data Mining (WISDM) Lab recently released the actitracker activity tracking app (actitracker.com). Prior to coming to Fordham, Dr. Weiss worked at AT&T Labs as a software engineer, expert system developer, and as a data scientist. He received a B.S. degree in Computer Science from Cornell University, an M.S. degree in Computer Science from Stanford University, and a Ph.D. degree in Computer Science from Rutgers University. He has published over fifty papers in machine learning and data mining and his research is supported by funding from the National Science Foundation, Google, and Citigroup.

Blurb

Introduction

Mahmoud Abou-Nasr, Stefan Lessmann. Robert Stahlbock, Gary M. Weiss

 

What Data Scientists can Learn from History

Aaron Lai

 

On Line Mining of Cyclic Association Rules From Parallel Dimension Hierarchies

Eya Ben Ahmed, Ahlem Nabli, Faıez Gargouri

 

PROFIT: A Projected Clustering Technique

Dharmveer Singh Rajput, Pramod Kumar Singh, Mahua Bhattacharya

 

Multi-Label Classification with a Constrained Minimum Cut Model

Guangzhi Qu, Ishwar Sethi, Craig Hartrick, Hui Zhang

 

On the Selection of Dimension Reduction Techniques for Scientific Applications

Ya Ju Fan, Chandrika Kamath

 

Relearning Process for SPRT in Structural Change Detection of Time-Series Data

Ryosuke Saga, Naoki Kaisaku, Hiroshi Tsuji

 

K-means clustering on a classifier-induced representation space: application to customer contact personalization

Vincent Lemaire, Fabrice Clerot, Nicolas Creff

 

Dimensionality Reduction using Graph Weighted Subspace Learning for Bankruptcy Prediction

Bernardete Ribeiro, Ning Chen

 

Click Fraud Detection: Adversarial Pattern Recognition over 5 Years at Microsoft

Brendan Kitts, Jing Ying Zhang, Gang Wu, Wesley Brandi, Julien Beasley, Kieran Morrill, John Ettedgui, Sid Siddhartha, Hong Yuan, Feng Gao, Peter Azo, Raj Mahato

 

A Novel Approach for Analysis of ’Real World’ Data: A Data Mining Engine for Identification of Multi-author Student Document Submission

Kathryn Burn-Thornton, Tim Burman

 

Data Mining Based Tax Audit Selection: A Case Study of a Pilot Project at the Minnesota Department of Revenue

Kuo-Wei Hsu, Nishith Pathak, Jaideep Srivastava, Greg Tschida, Eric Bjorklund

 

A nearest neighbor approach to build a readable risk score for breast cancer

Emilien Gauthier, Laurent Brisson, Philippe Lenca, Stephane Ragusa

 

Machine Learning for Medical Examination Report Processing

Yinghao Huang, Yi Lu Murphey, Naeem Seliya, Roy B. Friedenthal

 

Data Mining Vortex Cores Concurrent with Computational Fluid Dynamics Simulations

Clifton Mortensen, Steve Gorrell, Robert Woodley, Michael Gosnell

 

A Data Mining Based Method for Discovery of Web Services and their Compositions

Richi Nayak, Aishwarya Bose

 

Exploiting Terrain Information for Enhancing Fuel Economy of Cruising Vehicles by Supervised Training of Recurrent Neural Optimizers

Mahmoud Abou-Nasr, John Michelini, Dimitar Filev

 

Exploration of Flight State and Control System Parameters for Prediction of Helicopter Loads via Gamma Test and Machine Learning Techniques

Catherine Cheung, Julio J. Valdes, Matthew Li

 

Multilayer Semantic Analysis In Image Databases

Ismail El Sayad, Jean Martinet, Zhongfei (Mark) Zhang, Peter Eisert