<![CDATA[Sentiment analysis is one branch of science from data mining that aims to analyze, understand, process, and extract textual data in the form of opinions on entities such as products, services, organizations, individuals, and certain topics. In determining positive, negative or neutral categories, a public response on twitter can be done manually by reading each tweet. This certainly requires a lot of time and takes a lot of energy. In this study using the Support Vector Machine classification algorithm to classify tweet data into positive, negative or neutral sentiments. Analysis is carried out based on a certain time span, because each time can have a different topic of discussion and from the results of these data can be seen the development of sentiment trends and can be seen how the public response to a particular topic. The tweet data is obtained by crawling periodically with the target keywords of the names of candidates and vice president in the 2019 election. The dataset used in this study uses 600 tweets. In testing the classification using k-fold cross validation by dividing into 10 data parts, average value of 66% accuracy, 67% precision and 66% recall.]]>
Copyrights © 2019
