Explore the vast range of titles available.

Business Modeling and Data Mining demonstrates how real world business problems can be formulated so that data mining can answer them.The concepts and techniques presented in this book are the essential building blocks in understanding what models are and how they can be used practically to reveal hidden assumptions and needs, determine problems.

This chapter introduces a series of instructions that guides to take the next step depending on the results. This is called a methodology. Methodologies are made up of four types of “boxes” - Action Boxes, Discovery Boxes, Technique Boxes, and Example Boxes. Each box type has a different role in the methodology. Each box is exclusively labeled for reference and is grouped with similar boxes. Therefore, all the Action Boxes come first followed by the Discovery Boxes, then Technique Boxes, and finally Example Boxes. However, neither methodology can be read sequentially. This point bears repeating. The methodology is not a sequential document. In case anyone attempts to go through it in a sequential order, its consequences will be a complete chaos. There is no one path through the boxes. The essence of these methodologies is for a modeler or miner to take action, evaluate results, and take decisions on the basis of the results.

This chapter focuses on the issues that are important for the modeling effort. People always maintain the practice of any business process. It is supported by written rules, implicit and explicit expectation, corporate culture, tradition, explicit incentives, familiarity, and emotional motivation and it is entangled in formal and informal web of internal interactions that serve to maintain and alter it. As a result, it has an \"inertia\" that results from the influences just mentioned, which is to say that changing any system takes effort. In case the existing system is left to itself, it tends to react so as to achieve itself to an unchanged state. Successful model deployment has to recognize this from the outset; recruiting and maintaining support from all of the stakeholders from the beginning of the project through deployment is a key to success. A modeler is responsible for delivering the model in an appropriate form for implementation. By some considerations, the role of the modeler comes to an end with the delivery of the model, although the person, who is a modeler, may take another role.

This chapter focuses on practical examples mainly on one data set, and they are explored in some details using numerous ways. An interested reader can duplicate most of the example explorations. In addition, the chapter discusses and provides downloads of some of the actual models created to get the appropriate results. Thus, this chapter offers the opportunity for an interested reader to duplicate and extend on the explorations presented. Even without active participation, the reader should find that the continued focus on exploring a single data set leads to some degree of familiarity with its content. The repeated variety of explorations will provide same information in different layouts, showing merits and demerits of each tool and technique. One of the biggest problems in mining data, particularly for an inexperienced miner, is that the miner can easily deviate from the actual goal. To avoid this problem, one should make sure that the results achieved from mining are actually worth having. This means that the results, in whatever form they appear, should be applicable to the business problem.

This chapter is an approach to get raw data into a state that is appropriate for mining purposes. Any miner uses these basic techniques for addressing almost any business problem. Miners need to note preparation techniques when mining data in other domains, such as biomedical data, industrial automation data, telemetry data, geophysical data, time domain data, and so on. Developing mined models requires a hard work. It involves working with data, including making changes and taking actions that depend on the needs of the business problem and the miner's data discovery. Mining data is not magic, and it is not something that computer software will do for any one. Basically, data mining is an organized way of working with data, digging out useful information, and application of that useful information in solving the business problems. Most of the tools that are currently in use were developed from three main areas including statistics, artificial intelligence, and machine learning. Despite apparently different roots, these tools essentially do only one thing that is related to the discovery of a relationship, which more or less maps measurements in one part of a data set to measurements in another linked part of the data set.