<![CDATA[Detection of computer informatics student competence is indispensable for anticipating students who have a very poor performance in following the learning process in an educational institution for the purpose of all educational institutions are creating a qualified student. It can be seen in the results of the 5th and 6th semester students who have gained employment. Polytechnic LP3I Jakarta Depok one vocational education institution founded to create a human being who has the ability / skills required by the company so that the concept is to offer education that have Link and Match. Competitors who have the same goals is one of the challenges to be faced by the agency so we need a solution to overcome it. One solution is the detection of computer informatics student competence of students. This can be done by using data mining techniques. One data mining techniques used are support vector machines (SVM). Support vector machine method is able to overcome the problem of high-dimensional, addressing the problem of classification and regression with linear or nonlinear kernel that can be the ability of learning algorithms for classification and regression, but the support vector machine has a problem in the appropriate parameters. To overcome these problems required method of decision tree as a comparison, for the selection of appropriate parameters. Several experiments were conducted to obtain optimum accuracy. Experiments using support vector machine and decision tree which is used to optimize the parameters C, and ε population. Training data used computer informatics student data from 2012 to 2014 academic year. The experimental results show the decision tree method of data that is equal to 92.50% with a ratio of 60 training data were compared with data vector machine that is equal to an accuracy of 92.56% and the second T-Test metod done that method has a probability value of < 0.05 which algorithm C4.5.Keywords: Detection, Competence, Support Vector Machine, Decision Tree]]>
Copyrights © 2019
