The popularity and usefulness of data mining and knowledge discovery is increasing at an exponential rate. These fields are making steady advances that make them more and more useful to people dealing with huge volumes of data. These techniques are becoming invaluable for their ability to analyze huge volumes of complex data and automatically come up with useful information on trends and patterns (automatic information extraction). Today, we use data mining and knowledge discovery in commercial, medical, scientific, geographical, meteorological, and other areas that generate large volumes of data. We require automatic information processing methods to be of real use.

This book collects and collates the latest developments in data mining and knowledge discovery for big data from diverse fields so that one can read about them in one place. The book contains nine papers from different fields, on topics such as opinion mining, spatiotemporal data mining, discriminative subgraph patterns, path knowledge discovery, social media mining, and binary matrix factorization.

Opinion mining (or sentiment analysis) is the extraction of opinions, valuations, appraisals, attitudes, and emotions concerning entities, products, services, or events. Spatiotemporal data mining (STDM) deals with mining data about moving objects (people, animals, birds, and so on.) and climate changes. Discriminative subgraph mining is used to find patterns in complex structural information represented as graphs. Path knowledge discovery is mainly used in the field of neuropsychiatry, where there is a need to discover relations among concepts at multiple levels. Social media mining deals with many different aspects of social networking, such as mining information from sites, using social media in emergencies and natural disasters for humanitarian assistance and relief work, generating useful information while protecting the privacy of users, and so on. Binary matrix factorization (BMF) is used to reduce the dimension of high-dimensional datasets with binary attributes.

Each chapter deals with one of the above topics extensively, with clear, thorough, and comprehensive explanations of the methodologies, algorithms, and techniques used and the obstacles and challenges faced. The chapters also illustrate the practical applications, opportunities, and usefulness of these techniques using real-life examples and cases. The extensive references at the end of each chapter provide valuable material for further reading and research.

This book is primarily for practicing professionals and researchers. It explains state-of-the-art methodologies, techniques, and developments in many fields of data mining. The compilation of the latest developments from diverse fields into one volume gives professionals an opportunity to learn what is happening in other fields and gain insights and knowledge that can be used in their own fields.