Forecasting is apply because of complexity and uncertainty faced by high-dimensional data available in the fields of bioinformatics, chemometrics, banking and other applications. A process for systematically estimating what is most likely to happen in the future based on past and present data requires an appropriate forecasting model, so that the difference between what happens and the estimated results can be minimized. To get the right method, a measuring technique is needed to detect the accuracy of forecasting value. In this paper we discuss the technique of measuring forecasting accuracy with Mean Square Error (MSE) and Mean Absolute Percentage Error (MAPE) using the Random K-Nearest Neighbor (RKNN) method. With the two measuring technique for the horizontal modeling above, the smallest MSE and MAPE values are chosen (the smallest error value). From the results of the analysis of the calculation of forecasting accuracy measurement values during training with RKNN, the MAPE accuracy value is 0.728427% and MSE is 0.545751, while the smallest accuracy value is achieved using MSE which is 0.545751.
This study discusses effect of tournament selection on the way individuals compete on the performance of Genetic Algorithms so which one tournament selection is most suitable for the Traveling Salesman Problem (TSP). One algorithm in solving TSP is Genetic Algorithm, which has 3 (three) main operators, namely selection, crossover, and mutation. Selection is one of the main operators in the Genetic Algorithm, where select the best individuals who can survive (the shortest travel route). Tournament selection compares a number of individuals through a match to choose the best individual based on each fitness value, so that the winning individual (the individual going to the next generation) will be chosen. There is two way to compete in an individual in tournament selection is by tournament selection with replacement (TSWR) and without replacement (TSWOR). The final results of the study conducted TSWR gets the best fitness, even though the generation that gets the best fitness is reaching the maximum generation (takes longer to get the best fitness).
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.