Data Mining in Economic Science
msra(2001)
摘要
Data mining is commonly defined as the computer-assisted search for interesting patterns and relations in large databases. It is a relatively young area of research that builds on the older disciplines of statistics, databases, artificial intelligence (machine learning) and data visualization. The emergence of data mining is often explained by the ever increasing size of databases together with the availability of computing power and algorithms to analyse them. Data mining is usually considered to be a form of secondary data analysis. This means that it is often performed on data collected and stored for a different purpose than analysis; usually for administrative purposes. In this chapter we consider the possibilities of applying data mining in economic science. In doing so, we must naturally be aware of the considerable amount of research that has already been done in economic data analysis. To what extent can data mining contribute to the analysis of economic data? In answering this question we could consider data mining as a collection of techniques and algorithms that have been developed in this area of research. In doing so we could compare data mining algorithms to data analysis techniques more commonly used in economics, and see if they allow us to answer different questions, or to answer existing questions in a better way. Alternatively, we can also consider data mining as a highly exploratory form of data analysis that is data driven rather than theory driven. The latter aspect of data mining is most important in this contribution. This chapter is organized as follows. In section 2 we give a brief description of the object of study of economics. Then we consider economic modelling as a way to apply economic theory to particular problems and as a tool to deduce the consequences of particular assumptions. In order to give empirical content to
更多查看译文
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络