ANALISIS SENTIMEN MASYARAKAT TERHADAP PELAKSANAAN P3K GURU DENGAN ALGORITMA NAIVE BAYES DAN DECISION TREE

Abstract

The implementation of education in Indonesia is still inseparable from the problems of teacher management, honorary teachers, and bureaucratic reforms that affect the quality of education and the work climate in it. In an effort to improve the quality of public services by the State Civil Apparatus (ASN), the Ministry of Education and Culture agreed with the Ministry of Empowerment of State Apparatus and Bureaucratic Reform and the Ministry of Finance to change the government employee teacher recruitment system from accepting Civil Servant Candidates (CPNS) to Government Employees by Work Agreement (PPPK) which in its implementation still leaves several problems and pros and cons. Therefore, the researchers conducted a sentiment analysis in the field of data mining on the Implementation of Teacher Teacher Training on social media Twitter as many as 871 data which were then filtered and cleaned into 519 data due to duplicate data, empty data and data cleaning. The author uses the Naive Bayes and Decision Tree methods to compare the accuracy of the two methods. The researcher uses RapidMiner version 9.10.1 tools. The results showed that the sentiment analysis of Twitter data on teacher PPPK using the Naive Bayes method achieved an accuracy rate of 100.00%. Where is the class precision for pred. negative is 100.00% and pred positive is 100.00%, in the Decision Tree method the accuracy rate reaches 53.95%. Where is the class precision for pred. negative was 0.00% and pred positive was 53.95%. In this study, it can be seen that the Naive Bayes method is a method that has a higher accuracy rate than other methods with an accuracy rate of 100.00%.

Keywords: Sentiment Analysis; PPPK; Indonesia; Data Mining