<![CDATA[Introduction – The trend of e-commerce and online shopping has offered customers more product choices, but it also resulted in information overload. Nowadays, users are equipped with technology that allows websites to automatically deliver products that they may be interested in so that they can easily locate their favorite items from enormous options. To automate the recommendation process, recommender systems are created and built. This research creates a musical instrument recommendation system based on user reviews. Methodology/Approach – In this paper, we design and implement a recommendation system that combines the k-Nearest Neighbor (kNN) algorithm with a collaborative filtering framework. Collaborative filtering is chosen in this case because of its capability of providing new information to users by collecting information that has been obtained from the other users. Furthermore, kNN is considered as a suitable combination in this case since this method is relatively simple and able to find the similarity of objects being compared. Findings – To evaluate this study, the recommendation results are evaluated using the Root Mean Square Error (RMSE) calculation method, and the RMSE result obtained is 0.8734 for schema that divides dataset into 70% data train and 30% dataset using KNNWith Means with pearson measurements, and the MAE (Mean Absolute Error) result obtained is 0.5998 with schema 60% data train and 40% data test using KNNBasic algorithm with cosine similarity. Originality/ Value/ Implication – We present experimental results of consolidating the kNN algorithm in the collaborative filtering framework using Amazon’s musical instrument dataset. Furthermore, we can see that kNN together with a collaborative filtering algorithm performs a satisfactory outcome.]]>
