In a higher education environment that carries out tridharma obligations, one of the roles of a lecturer is to be a researcher. Collaboration between researchers who have different expertise is important to provide solutions to problems and can also improve the quality of the research itself. This research is present to provide a research partner recommendation system by applying the Naïve Bayes machine learning algorithm and the Rapid Application Development (RAD) method. Software development is done with PHP programming language and MySQL database. Testing of the software developed using black box testing. The application provides recommendations for research partners based on keywords entered in the form of research topics. Research partner recommendations are given in the form of a weighted value of the percentage of suitability for the keywords entered. The algorithm has been successfully applied and testing has been successfully passed so that researchers can use this software to get the right research partner according to the research topic concerned.