Uji Komparasi Sentiment Analysis pada Opini Alumni terhadap Perguruan Tinggi

Opinion is an important part of decision making, so it takes the ability to get information from opinions. Sentiment Analysis is a branch of science from Text mining that can be used for opinion analysis in the form of text to classify opinions into 3 types of opinions, namely positive opinions, neutral opinions and negative opinions. Support Vector Machine (SVM) is one method that is widely applied for text mining because it is able to show good performance (Styawati and Mustofa, 2019). SVM works with a learning system that uses a hypothetical space in the form of linear functions in a high-dimensional feature space. Maximum Entropy is a probabilistic classification algorithm that belongs to the class of exponential models, which is based on the principle of Maximum Entropy. Maximum Entropy can be used to solve text classification problems such as Language detection, topic classification, and sentiment analysis. Sentiment analysis was tested using the Support Vector Machine (SVM) and Maximum Entropy methods to test the accuracy of each method in analyzing the sentiments of college alumni opinions. from the test results show Maximum Entropy has a better level of accuracy with the results of 95.45