This book introduces the concept of Event Mining for building explanatory models from analyses of correlated data. Such a model may be used as the basis for predictions and corrective actions. The idea is to create, via an iterative process, a model that explains causal relationships in the form of structural and temporal patterns in the data. The first phase is the data-driven process of hypothesis formation, requiring the analysis of large amounts of data to find strong candidate hypotheses. The second phase is hypothesis testing, wherein a domain expert's knowledge and judgment is used to…mehr
This book introduces the concept of Event Mining for building explanatory models from analyses of correlated data. Such a model may be used as the basis for predictions and corrective actions. The idea is to create, via an iterative process, a model that explains causal relationships in the form of structural and temporal patterns in the data. The first phase is the data-driven process of hypothesis formation, requiring the analysis of large amounts of data to find strong candidate hypotheses. The second phase is hypothesis testing, wherein a domain expert's knowledge and judgment is used to test and modify the candidate hypotheses. The book is intended as a primer on Event Mining for data-enthusiasts and information professionals interested in employing these event-based data analysis techniques in diverse applications. The reader is introduced to frameworks for temporal knowledge representation and reasoning, as well as temporal data mining and pattern discovery. Also discussed are the design principles of event mining systems. The approach is reified by the presentation of an event mining system called EventMiner, a computational framework for building explanatory models. The book contains case studies of using EventMiner in asthma risk management and an architecture for the objective self. The text can be used by researchers interested in harnessing the value of heterogeneous big data for designing explanatory event-based models in diverse application areas such as healthcare, biological data analytics, predictive maintenance of systems, computer networks, and business intelligence.
Laleh Jalali is a computer scientist with an extensive background in applying advanced machine learning, deep learning, and data mining techniques in different domains such as Healthcare and Industrial Analytics. She is currently a senior scientist at Hitachi America Ltd. R&D. Her professional interests focus on event-based frameworks, big data analytics, healthcare analytics, natural language understanding, and context-aware mobile systems. At Hitachi, Laleh has been involved in many customer co-creation projects where the team owns the end-to-end process, from research to production models. Before Hitachi, Laleh graduated with a PhD in Computer Science from the University of California, Irvine. With Prof. Ramesh Jain as her advisor, she completed her doctoral thesis focusing on designing and developing an event mining framework, called EventMiner, with an emphasis on interactivity and effective integration of techniques from data mining, event processing, and human-computer interaction. Key elements of her work were published in International Conference on Multimedia & Expo (ICME), and the brave new ideas track in Association for Computing Machinery Multimedia (ACC MM), and International Conference on Multimedia Retrieval (ICMR).
Es gelten unsere Allgemeinen Geschäftsbedingungen: www.buecher.de/agb
Impressum
www.buecher.de ist ein Shop der buecher.de GmbH & Co. KG Bürgermeister-Wegele-Str. 12, 86167 Augsburg Amtsgericht Augsburg HRA 13309