Tiktok Shop Seller Center is one of the leading e-commerce in Indonesia.
This app is designed to help sellers manage and increase sales. Companies or
organizations need to understand the response received from the public to the
products or services they offer. Opinions that arise from the public can have an
impact on the image of the company or organization. However, monitoring and
managing public opinion is not a simple task. The volume of incoming opinions is
too large to be processed manually. Therefore, special approaches or techniques
are needed that can automatically categorize those reviews as positive or negative.
One platform that provides reviews is Google Play. Data obtained from Google
Play is then labeled and analyzed using the Support Vector Machine method to
classify these reviews. From the results of the labeling that has been done,
visualizations will be seen in each sentiment class to find information that is
considered important and can be useful for decision making. Classification with the
SVM method obtained an accuracy rate of 91.59%. Furthermore, visualization
methods in the positive sentiment class include good, helpful, steady, easy, good,
bangat, please, application and sell. Meanwhile, the class of negative sentiment
that is often complained about includes applications, uploads, violations, blocks,
products, difficult, registers, complicated, fails and accounts.