OPTIMASI ALGORITMA C4.5 BERBASIS PARTICLE SWARM OPTIMIZATION (PSO) UNTUK MENENTUKAN WHOLESALES PENJUALAN

Abstract

Algorithm C4.5 is an algorithm used to form a Decision Tree. The C4.5 Algorithm definitely has advantages and disadvantages. The features of the C4.5 algorithm can create decision trees that are easy to version, dominate the level of acceptable accuracy, efficient in managing effective category attributes and can set attributes of discrete and numeric types, and in an advantage there are definitely drawbacks. The weakness of the C4.5 algorithm is the instability in determining accuracy. The amount of data used is 1000 with 7 attributes. Data were analyzed using Particle Swarm optimization in C4.5. Because the accuracy produced by C4.5 is still low, it is optimized with Particle Swarm optimization. Accuracy on C4.5 is 81% after using optimization the accuracy increases by 86%. Data processing uses Python programming and accuracy testing uses the Confusion Matrix to compare accuracy results.