Detection of traffic congestion based on twitter using convolutional neural network model

<span lang="EN-US">Microblogging is a form of communication between users to socialize by describing the state events in real-time. Twitter platform for microblogging. Indonesia one countries with largest users, people can share information about traffic jams. This research aims detect jams extracting tweets vectors and then inserting them into Convolution neural network (CNN) model getting best from CNN+Word2Vec, CNN+FastText, support vector machine (SVM). Data retrieval was conducted using Rapidminer application. Then, context checked so that there were 2777 data consisting 1426 congestion road 1351 smooth data. The taken certain coordinate points around Jakarta, Indonesia. preprocessing changes carried out Word2Vec FastText methods, inserted CNN model. results CNN+Word2Vec CNN+FastText compared SVM method. evaluation done manually actual conditions. highest result obtained test method are 86.33% while 85.79% 67.62%.</span>