Exploring Practical Data Mining Techniques at Undergraduate Level

Abstract: Data mining is referred to as the process of analyzing and extracting patterns embedded in large amounts of data by using various methods from machine learning, pattern recognition, statistics and database management. With the rapid proliferation of the Internet and advances of computing technology, data mining has become an increasingly important tool of transforming large quantities of digital data into previously unknown and meaningful information and has been applied in many areas that include business and finance, health care, telecommunication, science and higher education. Data mining is also a relatively new field of computer science, and there are only a few undergraduate data mining courses are currently offered in institutions of higher education. In this paper, we describe the design, implementation and evaluation of a data mining course that we have developed and offered as an undergraduate computer science upper-division elective at the University of San Diego. The course combines lectures on a number of key data mining principles and applications, mini student lecture sessions, programming projects and research activities to engage students in active learning. Our experience has shown that data mining can be taught successfully at the undergraduate level and students can learn a great deal of data mining techniques and are able to apply them to solve many real world problems.