Classification of Sentiment Analysis and Community Opinion Modeling Topics for Application of ICT in Government Operations

Utilizing information systems is very useful in the current era. Digitizing administration Village beneficial service process to public. This seen as a change that can make it easier or more difficult for people of Sanrobone take care at village office. study aims analyze public opinion regarding use e-government, predict and modeling topics related e-government. research applies text mining algorithm with sentiment analysis method see positive, negative, neutral perceptions also uses topic get most frequently appearing data. Stages this include Data Collection, Text Pre-processing, Sentiment Analysis, Topic Modelling, Classification, Evaluation. The results obtained are ten words appear often responses community: easy 122, help 96, 80, accessed letter permit 77, resident 73, manage 60, 52, person 52 words. 411 opinions, 37 negative 152 opinions. Finally, performance Nave Bayes predicting classification excellent, an accuracy rate 98 percent.