یک رویکرد تحلیلی برای انتخاب داده کاوی برای تصمیم گیری کسب و کار

Due to the information technology improvement and the growth of internet, enterprises are able to collect and to store huge amount of data. Using data mining technology to aid the data processing, information retrieval and knowledge generation process has become one of the critical missions to enterprise, so how to use data mining tools properly is user concern. Since not every user completely understand the theory of data mining, choosing the best solution from the functions which data mining tools provides is not easy. If user is not satisfied with the outcome of mining, communication with IT employees to adjust the software costs lots of time. To solve this problem, a selection model of data mining algorithms is proposed. By analyzing the content of business decision and application, user requirements will map to certain data mining category and algorithm. This method makes algorithm selection faster and reasonable to improve the efficiency of applying data mining tools to solve business problems.

1.1. Research background Due to the information technology improvement and the growth of Internet, enterprises are able to collect and to store huge amount of data. People gradually realize that data is not equal to information that data should be further analyzed and extracted. Professionals are trained to analyze and interpret data, but the increases in data amount, data type, and analytical dimensions. Information technology has gone beyond storage, transmission, and processing. Data needs to be converted into information and knowledge in order to support decision making. 1.2. Research issue Enterprises use data mining tools to support knowledge discovery and decision making. In this research, we develop a selection model to solve the research issue. This model recommends the most suitable algorithm after marketing professionals and analysts describe the business problems using a standard procedure and format. This model provides an algorithm standard as the foundation of dynamic data mining modeling. This selection study of data mining mainly has two parts: the commercial problems analysis and the data mining algorithms analysis. Commercial problems analysis contains a general set of 22 problems. These problems relate to the banking, finance, insurance, telecommunication, retail, and manufacturing applications. They are classified into 12 business application categories according to their characteristics. Data mining algorithms analysis discuss five main types of data mining: association, classification, prediction, clustering, and profiling. In each category, several typical algorithms are listed and each of them is used to discover similar knowledge in different situations. The algorithm concepts, parameters, processes, and characters are compiled and compared to other algorithms in the same category. 1.3. Research limitation Data mining has been applied to business area for more than two decades. There are countless application cases for different industries, different information requirement, and lots more circumstances. We adopt the literature survey to summarize approaches and concepts from our literature review to formulate the selection framework and produce the processing details into broader concepts and terms. This selection model focuses on matching the applicable data mining method with the characteristics of business decision and application.

Each business decision and application is described by four parts. They are business decision and application, business problem, problem processing steps, problem characteristics. Each data mining algorithm is depicted by four parts. They are data mining method, input data unit, output data unit, and algorithmic step. After systematically mapping the features of business decision and application and the characteristic of data mining method, the selection model generates the matched set of choice. In terms of future research work, the following study will be continuously carried out. More practical selection can be achieved by considering the features of data side. Extension of data side enables precise description of mining procedure. Data side means analysis in data type, data format, data structure, and data definition.