Design of electronic-commerce recommendation systems based on outlier mining

摘要: Prior studies mostly consider outliers as noise data and eliminate them, resulting in the loss of outlier knowledge. Based on existing technology recommendation systems detection, this research develops a new e-commerce recommended model from perspective knowledge management. Specifically, we apply mining integrate local coefficients into algorithm. The experimental results show that proposed extent performs better than traditional based collaborative filtering algorithm, which can effectively improve quality recommendation, enhance customer satisfaction loyalty, create potential benefits for business. Our study contributes to design recommending with some novel ideas provides useful guidelines developing extent.