Abstract:
|
The concepts and practices of data mining when presented in a business setting can often present complex challenges when time and course sequencing prevent full discussions of prevailing mathematical principals behind each analysis. Students struggle with which types of analysis to choose. Over the course of two years (four semesters), data has been collected on multiple classes considering the use of real, messy, and varying data sets to immerses students in problem solving. They see that despite each group using diverse approaches, when applied properly and with logic, similar results occur. Using survey data from Fe y AlegrÃa Bolivia, the students worked using data mining techniques to identify the most impoverished students across varying regions; data from an Amazon entrepreneur presented inventory and branding procedures issues; and an electric and light company optimizing fan displays across multiple venues while minimizing inventory are a few examples. Immersing the students in problems and encouraging the use data mining tools and visualization to work through them creates confidence and understanding, a challenge in traditional testing. Results will be discussed.
|