Ramli1, Desi Yuniarti2,dan Rito Goejantoro3
FMIPA Universitas Mualwarman
Classification problems can be solved by using logistic regression and classification method of artificial neural network or artificial neural network (ANN). Classification with logistic regression method performed by transforming the dependent variable into a logit variable is the natural log of the odds ratio. In the ANN classification done by accepting a vector of input and then calculates a response or output by processing through the process elements are interrelated. The elements are arranged in several layers (layer) and the input data flow from one layer to the next in sequence. Output values can be scalar values or vectors, calculated at the output layer. The purpose of this study was to determine the results of the classification by using logistic regression and neural network analysis then compared with the classification accuracy. The data used in this study is the data the average value of report cards in the 1st half and the 2nd half of the class X for English and social studies subjects at SMAN 2 Samarinda academic year 2011/2012. The amount of data is 314 students with 2 to 4 response variables explanatory variables. Based on these results, the obtained results for the logistic regression classification accuracy of 78.34% and 80.89% for ANN analysis. From the comparison of artificial neural network classification method is a better classification method in solving the classification of English and social studies majors elections in SMAN 2 Samarinda academic year 2011/2012.
Keywords: Logistic Regression, Neural Networks, Classification, Classification