Implementasi Chatbot Sebagai Virtual Assistant di Universitas Panca Marga Probolinggo menggunakan Metode TF-IDF

Abstrak

This research is to implement a chatbot as a virtual assistant that can be used as an academic information service for the general public as well as for the academic community of the Panca Marga University Probolinggo campus. The stages of developing this chatbot using the waterfall method include analysis, design, code, testing and maintenance. The method used for learning chatbots uses Tf-Idf and VSM for word weighting in documents and queries and Cosine similarity to calculate similarity (similarity) between documents and queries. The final result of this research is a chatbot application that can be used as a virtual assistant as customer service in serving and providing information about academics at Panca Marga University Probolinggo. Based on the results of the accuracy and UAT testing, the accuracy rate obtained by the chatbot reached 85.7% and the UAT test in the first test reached 84.1% with a total of 30 respondents, in the second test it reached 82.1% with a total of 92 respondents.