Data Mining Practice

Author: Olivia Parr Rud
Publisher:
Publish Date: 2003-09-01
Features: This book teaches how to create and implement models in commonly used data mining areas such as marketing, sales, risk analysis, and customer relationship management and support. It discusses validated modeling techniques in relevant fields, helping readers discover new methods to increase profits and reduce costs. Starting from the basics, the book explains how to plan and select the right materials or ask the right questions before seriously considering business needs, obtain data, and prepare for data mining. The book includes numerous case studies, detailing effective data sources used for developing target models. It then demonstrates how to process, evaluate, and implement models through an in-depth case study of the life insurance direct mail campaign's lifetime value. The book introduces a new data mining technique—data modeling—and emphasizes the details of the entire model development process. It is divided into three parts. Part one covers the basics, including the importance of defining goals and objectives from business predictions, along with examples of data collection and dataset creation. Part two elaborates on the entire model development process through a detailed case study. Part three applies data modeling to the insurance, banking, and telecommunications industries, illustrating key steps in the process for different objectives. The book interprets the art of data mining through the metaphor of cooking, making it easy to understand and accept. The numerous examples provided fully reflect the author's years of industry experience and offer excellent insights for current marketing and customer relationship management modeling. This book is suitable for readers with a basic foundation in statistics and analytical modeling. It can serve as a reference manual for analysts, data miners, and marketing managers, as well as a textbook or supplementary reading for undergraduate and graduate students in computer-related fields.