---

With the rapid growth of computer networks and data storage technologies, the volume and diversity of data has become a major challenge for data exploitation. These large volumes of heterogeneous data types need to be accessed and treated in a uniform manner for richer information extraction. However, mining heterogeneous data types is complex because of the unique nature of the heterogeneous data types, different ways of representing heterogeneous data types, and particular complexity of combining heterogeneous data mining results. This is the first book dedicated to heterogeneous data mining. A unified vectorization method is proposed for simultaneous processing, classification and clustering of heterogeneous data types, which allows a unified visualization of the data. The applicability of the proposed methodologies is demonstrated by the development of a data integration (schema matching) tool for numerical and textual data types. This book will interest data mining researchers and professionals involved in processing, classification or clustering of heterogeneous data types.

---

Farid Bourennani is Lecturer and Research Associate at the University of Ontario Institute of Technology, Canada.

---