An optimized model for classification of appropriate technology products using neural networks and genetic algorithms

Abstract

The background of this research is the difficulty faced when classifying the result of the assesment in the event Competition Appropriate Technology Tegal City. In this paper, the classification result of the assesment is carried out by categorizing the result of technology product that are suitable for implementation in small to medium industries by using the model. The model proposed in this study is using the Neural Network (NN) method to be able to classify products whether into appropriate technology for small/medium industries. Neural Network can produce a low level of accuracy, but Neural Network optimized based on Genetics Algoritm can produced a level of accuracy more higher than without optimation. The genetic algorithm (GA) is proposed to optimize the value of Neural Network parameters such as the number of hidden layers and learning rates. Based on the results of the experiments that have been produced, and system validation using the 10 Fold cross-validation trial shows that the Neural Network method that optimizes the genetic algorithm produces an accuracy rate of 91.37%, better than the Neural Network method without optimization which is equal to 90, 14%.