Recommendations of Thesis Supervisor using the Cosine Similarity Method

During the thesis writing process, role of supervisor is needed so that completion timely quality maintained. One problems in determining for undergraduate computer science at Bumigora University subjectivity head study program supervisor. Not only that, selection process supervisors done manually it can take a long time and slow down student work. Students' work will be late not on if topic accordance with competence lecturer. This aims to apply cosine similarity method recommendation University. The stages this research consist collecting documents, pre-processing text (Case Folding, Tokenization, Filtering, Stemming), word weighting TF-IDF, implementation method, accuracy testing. data used are 113 documents which divided as training many 90 testing 23 documents. Based test, correctly recommend 21 an 91.3%. Thus, applied case selecting because has very good accuracy.