Application of the nearest neighbor algorithm for classification of online taxibike sentiments in indonesia in the google playstore application

Abstract

Google playstore is an official platform that provides applications to services, and not only that, user can perform assessments and reviews after using the Google playstore. One of the applications and services provided by Google playstore is online transportation. There's a couple of online transports available but at this time's research just scanned the official transport account of gojek and grab online.The data used in this study were application review data and service review data from the two online transportations which are processed using the K-Nearest Neighbor and Improved K-Nearest Neighbor algorithms by comparing the accuracy values. An experiment was conducted to divide the data with 5 K on the K-Fold Cross-Validation. The results of the comparison of the KNN application data were that Grab had a higher accuracy of 77.77% compared to Gojek which had an accuracy value of 66.24%. In terms of the Data Services using the IKNN algorithm, Grab had a higher accuracy value of 78.23% while Gojek had an accuracy value of 48.19%. Meanwhile, from the time complexity of the process, the IK-NN algorithm is better than the K-NN.