| Popis: |
Demand for coffee products continues to increase along with population growth. The demand for a product is seen from the public's perception of the product. Community perceptions can be found through tweets on social media Twitter. This perception can be used to analyze sentiment towards a product. This approach is called Sentiment Analysis and to analyze it requires Big Data Technology. This study aims to design a Sentiment Engine Analysis of the Level of Popularity of Coffee Based on Processing Using Big Data Technology to find out the sentiments of people around the world towards processed coffee drinks through Twitter social media. This study uses the process of collecting tweet data from Twitter using the Twitter API, data processing, classification into positive, negative, and neutral sentiments using the Naïve Bayes method, as well as data visualization in graphical form. The data generated in this study amounted to 1,253,971 data in 1 year with 1800 training data and 450 test data. The evaluation results with the Naïve Bayes classification are 92.67% accuracy, 98.52% precision, 88.67% recall, and 93.33% f-measure. |
