BRAND REPUTATION MONITORING SYSTEM BASED ON SENTIMENT ANALYSIS USING THE K-NEAREST NEIGHBOR METHOD

Abstract

This research paper aims to describe the research and implementation of a brand reputation monitoring system. The system, called Monitool, is needed by the digital marketing division of PT Lingkar9 Titian Media to measure the brand reputation of its clients. Monitool focuses on monitoring and analyzing a brand on social media and news portals. With Monitool, the digital marketing division of PT Lingkar9 Titian Media can easily make digital marketing plans and can evaluate brand reputation efficiently. Currently the marketing division of PT Lingkar9 Titian Media measures brand reputation using paid tools to measure brand reputation, but it is quite costly, because the costs are not proportional to the profits, especially if the project being handled suffers a loss. Monitool uses the K-Nearest Neighbor (KNN) method to measure the reputation of a brand. Before measuring reputation, monitool will look for news or social media posts that offend the client's brand. After the data is collected, the data will be processed using the KNN method. After being processed, the system will issue a resume of sentiments consisting of negative, positive and neutral sentiments. The accuracy result of the current test data prediction is 81%. This prediction result will change when the training data is increased or decreased.