Sentiment Analysis of Indonesian New Capitol (IKN) on Twitter Using Classification Algorithm

The objective of this research is to examine the viewpoints of users on Twitter’s social media platform with regards to the proposed development plans for Ibu Kota Nusantara. It aims to determine whether these users hold negative, positive, or neutral opinions on the matter. The data is collected by means of data scraping, focusing specifically on information related to ’Ibu Kota Nusantara’. The chosen method for analysis is the Support Vector Machine (SVM) algorithm, utilizing both linear and sigmoid kernels, as well as the Naïve Bayes Classifier and Random Forest. The sentiment analysis is performed by comparing the performance metrics of each classification algorithm. The results indicate a positive sentiment of 38.67%, negative sentiment of 12.94%, and neutral sentiment of 48.39%. The evaluation of the system involves the use of a confusion matrix to obtain four accuracy values. Firstly, the SVM with a linear kernel achieves an accuracy of 89.77%, a precision value of 87.36%, and a recall value of 85.62%. Secondly, the SVM with a sigmoid kernel demonstrates an accuracy of 88.69%, a precision value of 87.39%, and a recall value of 83.94%. Thirdly, the Naïve Bayes classifier attains an accuracy of 79.26%, a precision value of 76.19%, and a recall value of 76.19%. Finally, the Random Forest algorithm yields an accuracy of 66.88%, a precision value of 84.42%, and a recall value of 49.33%. In conclusion, it can be observed that the linear kernel SVM outperforms the other classification algorithms in terms of accuracy.