Prediction or Guess? Decide by Looking at Two Images Generated by a “MATLAB MySQL” Algorithm
In the field of data mining, predictive modeling refers to the usage of a statistical model built on a training data set in order to make predictions about new prospects contained in the scoring data set. A model should not be used to predict when it encounters unseen data in the scoring set because such predictions would be a guess or a speculation. This paper proposes an algorithm that will produce two simple images and a “level of guessing” (LOG) pie chart. These images will tell the analyst whether or not it is appropriate to use a statistical predictive model to make predictions on a particular scoring set. The proposed algorithm will offer a solution to the scoring adequacy problem based on subsets of the original data. The algorithm will be implemented with a user interface built with MATLAB code, which acts on MySQL databases that contain the data.
Keywordspredictive modeling data mining scoring set supervised learning MATLAB MATLAB GUI MySQL
Unable to display preview. Download preview PDF.
- 1.Kutner, M., Nachtsheim, C., Neter, J.: Applied linear statistical models. Irwin, Chicago (2004)Google Scholar
- 3.Attaway, S.: MATLAB: A practical introduction to programming and problem solving. Butterworth-Heinemann, Waltham (2011)Google Scholar
- 4.MySQL, A., MySQL Administrator’s Guide and Language Reference. MySQL, Indianapolis (2006)Google Scholar
- 5.Coulson, L.: MATLAB Programming (e-book). Global Media, Chandni Chowk (2009), Available from: eBook Collection (EBSCOhost), Ipswich, MA (accessed October 29, 2013)Google Scholar
- 6.Cerrito, P.: SAS I. Introduction to Data Mining Using SAS Enterprise Miner (e-book). SAS Institute, Cary (2006); Available from: eBook Collection (EBSCOhost), Ipswich, MA (accessed October 30, 2013)Google Scholar