Data Mining
rizki rizkyatul basir

STMIK Nusa Mandiri Jakarta

Abstract

The increment of online-based transportation shows the desire of the society to find out cheap and fast alternative transportation as a response to the bad public transportation services provided by the government. This online transportation phenomenon has become popular quickly because it offers the latest innovations in transportation combined with online communication technology so that it makes it easy for people to order motorcycle taxi anywhere and anytime. By analyzing the sentiments of the people as road users, it is intended as a reference for the service units of the relevant government agencies to determine the level of public sentiment towards traffic congestion caused by online transportation so that it can be used as material for assessment and evaluation. In this study the writer conducted sentiment analysis on Twitter social media to determine the level of public sentiment towards congestion caused by Online-Based Transportation. The steps in conducting a sentiment analysis include preprocessing, extraction feature and classification. The making of the sentiment classification model uses two algorithms namely Support Vector Machine (SVM) and K-Nearest Neighbor. After testing the Support Vector Machine (SVM) algorithm has an accuracy value of 84.00% with AUC of 0.862, while the K-Nearest Neighbor (K-NN) algorithm has an accuracy of 80.00% with AUC of 0.740 with 250 sampling.

Keywords: Transportation based on online, traffic, Twitter, Support Vector Machine, K-Nearest Neighbor

Topic: International Symposium of Engineering, Technology, and Health Sciences

Link: https://ifory.id/abstract-plain/JYeVaMZpkHqt

Web Format | Corresponding Author (Rizki Rizkyatul Basir Basir)