With a title such as “Data Mining in Finance: Advances in Relational and Hybrid Methods”, one can think that this book is a set of research papers in this topic. However, it is not the case. The authors, Kovalerchuk and Vityaev, have written more than 300 pages about applying data mining techniques in finance.
Although the main topic of the book is relational data mining and its financial applications, there are several other topics:
- Statistical models
- Autoregression models
- Neural networks
- Decision trees
- Naive Bayes
- Fuzzy logic
There are also three other interesting chapters in the book. First, the introduction explains both the data mining (existing techniques) and the financial fields (technical and fundamental analysis). Second, a chapter about application of relational data mining to finance. Third, a chapter with comparison of techniques presented in the book. It can be noted that the foreword is written by Gregory Piatetsky-Shapiro from KDnuggets.
An advantage of the book is that both finance and data mining concepts are explained. Thus, one can read the book without the need of other resources. Before applying a given technique, the authors describe it, explain the data as well as the financial application need. Maybe more attention should have been paid to results and their explanations. They are sometimes confusing and lack some descriptions. Also the book is a little old now (2000) and due to this, one can be disappointed not to see SVM for example. Who knows, maybe in a next edition of the book. Finally, although not up to date, this book should be read by anybody applying data mining in finance.