Abstract
Data mining technology is the technique through which we make all kind of analysis on large amount of data. Applying Data Mining in Educational sector has become an emerging research field used to determine information and find out patterns from E-learning systems. The educational system in various universities are presently facing lots of issues like identifying students' needs, personalization of training and announcing the excellence of scholar communications. Educational Data Mining provides a set of ways, which can help the learning system to overcome these issues. The idea of this work is to explain Educational Data Mining examination, by telling a step-by-step development using a range of techniques such as Clustering (K-Means), Organization (Tree Induction), Association Mining (Apriori, FPGrowth, Create Association Rule, GSP) in order to get the goal to find out obliging information from the Moodle LMS. Analyzing mining results enables learning institutions to superior assign resources and systematize the education process in order to look up the learning experience of students as well as enhance their profits. In this study Data for the primary set of experiments was derived from the Moodle logs. These data were analyzed from different levels and perspectives, in order to provide more approaching on the whole educational system. From the point of an entity course user action is measured, accomplishing coursework, quizzes development coursework, membership in forum and chat. By applying clustering methods, the goal was to split data set in groups of data points that logically group together. By means of prediction techniques, the objective was to build up a representation that can deduce predicted variables from predictors' variables. By applying Society Mining techniques, the objective was to find out interaction between variables. The mining outcome can be used by the class teacher to examine the impact of a number of e-learning behaviors on the students' knowledge growth. This paper concentrated on the overall LMS performance at our college and the mining development of Moodle data. Mining the Moodle data permitted to identify the most successful ways to the teaching development that can be used to improve the education development. The investigational outcome have exposed that the data mining representation presented in this paper was capable to attain understandable and reasonable response from the LMS data telling students' knowledge performance patterns. To test the efficiency of the planned representation and to increase the simplification of this research, more broad experiments should be conducted by using larger LMS data sets.