A Web-Based Laptop Purchase Recommendation Model Using Natural Language Processing (NLP) on Marketplace Reviews

This system will use the Natural Language Processing (NLP) method to analyze user reviews. In addition, the Naive Bayes classification algorithm will be used to provide recommendations based on the analysis. The methods used include collecting laptop user review data from the Shopee platform, Natural Language Processing (NLP) for text analysis, classification with the Naive Bayes algorithm, developing a recommendation system, and evaluating the system using relevant metrics. The results of the study show that this model achieves an accuracy of 0.86 with a precision of 0.93 for positive reviews and 0.69 for negative reviews. Of the total 42 reviews tested, the system provides a recall of 0.87 for positive reviews and 0.82 for negative reviews. The total reviews in the dataset consist of 96 positive reviews and 43 negative reviews. This study is expected to contribute to the development of review-based recommendation systems, so that users can make the right decisions.