首頁電子資訊資訊軟體

Development Trend of a Smart Recommender: Advancing into an Individualized Full-Channel Model

免費
字數 1133
頁數 5
出版作者 Allison Tsui
出版單位 工研院IEK電子分項
出版日期 2015/11/18
出版類型 產業評析
所屬領域 資訊軟體
瀏覽次數 506
加入購物車 直接下載 直接下載 加入最愛
摘要

Amazon is definitely one of the world’s premier commercial recommender systems, with its precise recommendations still generating immense benefits for the website itself even today. Approximately 30%–40% of Amazon’s revenue is derived from product recommendations. Physical stores can provide only bestselling or newly launched products because of regional constraints, limited inventory space and cost. According to the 80/20 Rule, 80% of a vendor’s revenues are generated from only 20% of products. Most products cannot appear on the shelves in physical stores, or are ignored by customers. In the past decade, the rise of e-commerce has reduced the cost of displaying products for sale. The recommender system has engendered business opportunities for products that were previously unknown to consumers. Products that previously had failed to attract consumers’ attention could be recommended to them through this system, greatly increasing the probability that they would be purchased. According to statistics, 57% of Amazon’s sales and 20% of Netflix’s sales were derived from products that are not sold in physical retail stores.

內文標題/圖標題/表標題

I.E-Commerce Recommender System Initiating Product Long-Tail Effect

II.Evolution of Big Data Technological Platform, Optimization of Recommender System Performance

III.RichRelevance Offers 020 Comprehensive Shopping Integrated Service

IV.A Solution to a Full-Channel, Multi-Screen, Virtually Integrated Recommender System

Fig. 1 RichRelevance's 020 recommender system service model

上一篇人工智慧影響機器人產業能發展至...
下一篇Market Status o...
熱門點閱
推薦閱讀
推薦新聞

若有任何問題,可使用下方檢索互動介面找解答,或是寫信到客服信箱。

itismembers@iii.org.tw

星期一~五
9:00-12:30/13:30-18:00