ANALISA SENTIMEN TERHADAP BELAJAR ONLINE PADA MASA COVID-19 MENGGUNAKAN ALGORITMA SUPPORT VECTOR MACHINE BERBASIS PARTICLE SARM OPTIMIZATION

Abstract

Coronavirus 19 (COVID-19) is a contagious viral infection that has now spread to various countries, one of which is Indonesia. Monitoring the spread of COVID-19 in Indonesia, direct services by the Government of Indonesia, especially. The Indonesian government immediately followed up on the case. One of the government’s actions is to carry out Social Distancing for 14 days to minimize the spread of the virus. This online learning activity is carried out to replace direct learning activities. Online learning has weaknesses in the use of internet networks, adequate infrastructure, requires a lot of money, communication via the internet which has various networks is slow. There are quite a lot of public comments on twitter about online learning. Based on comments from the general public who are hurt, it is easy, based on very many orders to leave so that you can see the extent to which the analysis of public sentiment is based on positive and negative comments using classification techniques, namely using the Particle Swarm Optimization-Based Support Vector Machine. The test result with accuracy values and AUC values by means of SVM + PSO accuracy value = 71.39% and AUC value = 0.762. for this reason, in this study it can be stated that the use of Particle Swarm Optimization (PSO) in the Support Vector Machine (SVM) algorithm model can be a solution to improve accuracy and AUC analysis of public sentiment regarding online learning during the Covid-19 period can be used to provide solutions to problems. Sentiment analysis on public comments on twitter during the Covid-19 period.