Data Mining For Robust Flight Scheduling [chapter]

Ira Assent, Ralph Krieger, Petra Welter, Jörg Herbers, Thomas Seidl
Data Mining for Business Applications  
Preface This edited book, Data Mining for Business Applications, together with an upcoming monograph also by Springer, Domain Driven Data Mining, aims to present a full picture of the state-of-the-art research and development of actionable knowledge discovery (AKD) in real-world businesses and applications. The book is triggered by ubiquitous applications of data mining and knowledge discovery (KDD for short), and the real-world challenges and complexities to the current KDD methodologies and
more » ... chniques. As we have seen, and as is often addressed by panelists of SIGKDD and ICDM conferences, even though thousands of algorithms and methods have been published, very few of them have been validated in business use. A major reason for the above situation, we believe, is the gap between academia and businesses, and the gap between academic research and real business needs. Ubiquitous challenges and complexities from the real-world complex problems can be categorized by the involvement of six types of intelligence (6I s ), namely human roles and intelligence, domain knowledge and intelligence, network and web intelligence, organizational and social intelligence, in-depth data intelligence, and most importantly, the metasynthesis of the above intelligences. It is certainly not our ambition to cover everything of the 6I s in this book. Rather, this edited book features the latest methodological, technical and practical progress on promoting the successful use of data mining in a collection of business domains. The book consists of two parts, one on AKD methodologies and the other on novel AKD domains in business use. In Part I, the book reports attempts and efforts in developing domain-driven workable AKD methodologies. This includes domain-driven data mining, postprocessing rules for actions, domain-driven customer analytics, roles of human intelligence in AKD, maximal pattern-based cluster, and ontology mining. Part II selects a large number of novel KDD domains and the corresponding techniques. This involves great efforts to develop effective techniques and tools for emergent areas and domains, including mining social security data, community security data, gene sequences, mental health information, traditional Chinese medicine data, cancer related data, blog data, sentiment information, web data, procedures, v vi Preface
doi:10.1007/978-0-387-79420-4_19 fatcat:ap6jg24o4fc4ndcnunkepmzrhi