<![CDATA[One way to gauge users' thoughts and sentiments towards a particular product, service, or subject is by conducting sentiment analysis on reviews posted on the Google Playstore platform. Among the plethora of apps available on the Google Playstore is Shopee. Due to the vast and unstructured nature of user comments in the review section, it becomes challenging to quickly and accurately grasp the overall information. This research aims to classify sentiments as positive, negative, or neutral, with the hope that the Shopee app can improve. Hence, the K-Nearest Neighbor Algorithm is employed to analyze sentiments to ensure users' opinions regarding their interaction with the Shopee program. Sentiment analysis is utilized to categorize reviews into positive, neutral, and negative groups. A dataset of 2000 entries is used in this analysis, obtained through web scraping, with 70% as training data and 30% as test data. The results indicate that this data split scenario yields the best model, achieving an accuracy of 70%, precision of 50.5%, recall of 44.8%, and an F1-score of 48.3% overall. To optimize results further, the implementation of more optimal data sampling techniques is necessary to attain a more balanced class distribution in both training and test data.]]>
Copyrights © 2024
