Twitter Opinion Sentiment Analysis Based on New Student Admission Zoning Issues Using the Naïve Bayes and TensorFlow Methods

Education is a reflection of the quality of a nation. Therefore, this matter is highly prioritized. The Indonesian government is currently implementing a zoning policy for new student admission with the aim of equal distribution of educational resources. Government policy regarding zoning has been a controversial topic. Twitter has become a platform for people to express their views and opinions. As a result of this zoning issue, pros and cons arise regarding implementing this policy. Therefore, researchers conducted sentiment analysis to understand in depth the opinions of the Twitter user community, which can be used as material for evaluating government policies related to education. Qualitative data collection techniques: data resulting from the process of crawling tweets from Twitter social media users. Next, data preprocessing is carried out. The preprocessing data was analyzed for sentiment using the Naïve Bayes Classifier Method and the TensorFlow Machine Learning Framework by producing three sentiment analysis classifications, namely positive, negative, and neutral, and measuring the accuracy of implementing the performance of these two methods. Apart from providing an overview of public opinion regarding zoning policies, this research also provides an overview of the effectiveness of the two methods in Sentiment Analysis. From the results of this research, the results of the sentiment classification of PPDB zoning problems related to public opinion tend to be negative. A comparison of the application of the method between the TensorFlow Machine Learning Framework and the Naïve Bayes Classifier is that the accuracy obtained by the TensorFlow Machine Learning Framework is 0.78, and the accuracy obtained by the Naïve Bayes classifier is 0.95. From the accuracy results, it is known that the Naïve Bayes method is better than the TensorFlow Machine Learning Framework in terms of Sentiment Analysis.