<![CDATA[The problem of selecting laptops that match user needs often becomes a challenge for many users, especially for those who are less familiar with technical specifications. To address this problem, this research develops a Natural Language Processing (NLP)-based chatbot application capable of providing automatic laptop recommendations based on user needs. This application implements the TF-IDF algorithm to extract features from user input in natural language, then calculates cosine similarity with laptop specification datasets stored in a MySQL database to generate the most relevant recommendations. The results of black box testing show that the system is capable of providing recommendations with a precision rate of 87.5%, recall of 83.2%, and F1-score of 85.3% in understanding user preferences based on criteria such as price range, weight, and usage type. This research contributes to the development of NLP-based chatbot technology by integrating the TF-IDF approach for more accurate natural language understanding compared to conventional rule-based chatbots, as well as providing interactive solutions that facilitate ordinary users in obtaining laptop recommendations without requiring in-depth technical knowledge.]]>
