Increased new students in a university making heaps of data more, therefore the search performed new knowledge with data mining. Grouping of data for prospective new students will be done by clustering method and using the K-means algorithm. In the data penmaru have 5 attributes used namely, hometown, gender, graduation status, category test, and majors. This analysis is done by using WEKA software and data retrieved from the data source admission of new students (penmaru) in the form of a data warehouse. Class of use this method is the major of attributes. Iteration is done 3 time iterations and the number of clusters on FKIK namely four clusters, namely FISIPOL 3 clusters. Clustering method can be applied to the determination of the major for prospective new students. Another that can be analyzed from the result of the grouping of data is the promotion strategy of each major to increase the quantity and quality.