Search and recommendation are the two main means for users to obtain information, and the main differences between their evaluation systems are:
An important indicator for evaluating the quality of search results is to help users find the results they need as soon as possible and click away.
In designing a search ranking algorithm, we need to try our best to get the best results to the top. Often, the top three results of search engines gather the vast majority of user clicks. Simply put, a "good" search algorithm needs to allow users to obtain information more efficiently and stay shorter.
To measure whether the recommendation system is good enough, it is often based on whether it can allow users to stay for more time (such as buying more products, reading more news, etc.). The higher the success rate of the recommendation, the more willing users are to stay in the product.
Original link:
http://www.infoq.com/cn/articles/recommendation-and-searchengine