Linking Twitter Sentiment Knowledge with Infrastructure Development

Abstract

Sentiment analysis is related to the automatic extraction of positive or negative opinions from the text. It is a special text mining application. It is important to classify implicit contents from citizen’s tweet using sentiment analysis. This research aimed to find out the opinion of infrastructure that sustained urban development in Surabaya, Indonesia’s second largest city. The procedures of text mining analysis were the data undergoes some preprocessing first, such as removing the link, retweet (RT), username, punctuation, digits, stopwords, case folding, and tokenizing. Then, the opinion was classified into positive and negative comments. Classification methods used in this research were support vector machine (SVM) and neural network (NN). The result of this research showed that NN classification method was better than SVM.