| Summary: | 碩士 === 國立臺北科技大學 === 商業自動化與管理研究所 === 93 === This research attempts to analyze several recommender systems based on the perspectives of sellers and buyers. From the seller’s perspective, recommendations are made based on the overall product purchase probability and the product profitability; from the buyer’s perspective, recommendations are made based on individual customer’s preferences.
Four recommender systems are compared in terms of recommendation accuracy and/or profit from cross-selling. Based on the experimental results, the following conclusions can be made:
1. As far as non-personalized recommendations are concerned, the recommendation accuracy will decrease when product profitability is additionally taken into consideration.
2. Compared with non-personalized recommendations, the recommendation accuracy will significantly increase when recommendations are made only from the buyer’s preference perspective.
3. As far as personalized recommendations are concerned, the profit from cross-selling significantly increases and the recommendation accuracy is only slightly different when product profitability is additionally taken into consideration.
4. The recommendation accuracy will significantly increase when recommendations are made in personalized approaches.
|
|---|
