Use of Hybrid Methods in Making E-commerce Product Recommendation Systems to Overcome Cold Start Problems

The large number of users and the items offered in e-commerce make it difficult for buyers to choose right sellers offer their buyers. To overcome this problem, a system that can recommend goods automatically, namely recommendation is needed. One most popular methods used create collaborative filtering, recommendations are created based on similarities user behavior. Unfortunately, method has weakness, cold start, where will be inaccurate data lot new due minimal historical regarding This problem tried solved study using hybrid method, combines more than 1 list so cover shortcomings each method. uses Amazon's product transaction data. use start by switching mixed methods, not filtering model or who have little interaction. New receive combination popularity-based content-based models. seen from Mean Absolute Error (MAE) value model, MAE with minimum at least 3 times rating 0.566883, 7 times, smaller, 0.487553.