Recommended system classification
- Social recommendation
- Content-based filtering
- Recommendation based on collaborative filtering
- Item-based method
Ways to solve information overload
- Categories
- search engine
- Recommended system
Recommendation system evaluation
In order to fully evaluate the impact of the recommendation system on the interests of the three parties, this chapter will propose different indicators from different perspectives. These indicators include accuracy, coverage, novelty, surprise, trust, transparency, etc. Some of these indicators can be calculated offline, some can only be calculated online, and some can only be obtained through user questionnaires.
recording
- The main role of the personalized recommendation system in these websites is to improve the click-through rate and conversion rate of the website by analyzing a large number of user behavior logs and providing different users with different personalized page displays.
- Almost all recommendation system applications are composed of three parts: the front-end display page , the back-end log system, and the recommendation algorithm system .
- Amazon has two related product lists, one includes other products frequently purchased by users who have purchased this product (as shown in Figure 1-6), and the other includes other products frequently purchased by users who have viewed this product. . The difference between these two related recommendation lists is that different user behaviors are used to calculate the relevance of items. In addition, the most important application of related recommendation lists is cross selling. When you are buying an item, Amazon will tell you several other products that other users will buy while buying this product, and then let you choose whether you want to buy these products at the same time. If you click to buy at the same time, it will "package" these several items, sometimes offering a certain discount, and then sell it to you.
- Music recommendation
- Social network
- Location-based services
- Personalized mail
- Personalized advertising
problem
- What is the difference between social recommendation and collaborative filtering-based recommendation?