Warung Wareg is a restaurant that serves a home-cooked menu with a mainstay menu in the form of various processed fish-based ingredients. In its business activities, restaurants need to carry out customer relationship management, one of which is by managing feedback well. One way that can be done is to manage reviews from users. The number of reviews numbering in the thousands makes it difficult for Warung Wareg to manage customer reviews, therefore a sentiment analysis is necessary to find out what customers think about Warung Wareg's services and products. The source of review data comes from Google Reviews and Tripadvisor by utilizing selenium for web scraping. Sentiment analysis was performed using the Support Vector Machine (SVM) method with term frequency - inverse document frequency (TF-IDF) as word weighting. Random undersampling technique is used to handle imbalance dataset. Hyperparameter tuning technique is done to produce the best model. Testing the results using the confusion matrix produces an accuracy value of 94%. The dashboard page is used to visualize the classification results using the Google Data Studio platform. The negative review ranking process is carried out to find the most negative reviews given by customers. From the results of the classification, a root cause analysis is also carried out to find the root causes of the negative reviews to formulate business recommendations that can be taken to overcome these problems.
Copyrights © 2023