The recommendation system is a dynamic information filtering system that is produced according to user interests or behavior. The recommendation system in the library can provide book references based on user interests or characteristics. UII Central Library has an information system in the form of a database of book lending transactions that can be used for a recommendation system. The method is a rough set and apriori algorithm. This study compares 2 methods to get the best method that can be applied in the recommendation system. The results obtained by the number of rough set rules as many as 14 rules with an average coverage of 0,01111 and average accuracy of 0,87416 and the number of apriori algorithm rules as many as 23 rules with an average support of 0,00276 and average confidence of 0,87458. Based on the number rules, the average value of accuracy or confidence, the apriori algorithm mehod is a method that can be used for the recommendation system in the UII Central Library.
                        
                        
                        
                        
                            
                                Copyrights © 2018