隨堂測驗
透過集合功能整理內容
你可以依據偏好儲存及分類內容。
你為什麼要使用推薦系統?
提供建議引擎,讓瀏覽內容更輕鬆。
此外,優質的推薦系統也能協助使用者找到他們認為不在意尋找的內容。
你覺得所有水資源都很厲害。
看起來好像沒問題,但實際上,使用機器學習有更好的理由。
您想將使用者導向贊助商項目。
哎呀,這不是使用任何機器學習解決方案的理由。
推薦功能系統的主要元件為何?
候選人產生、評分和重新排名
非常好!以下為任何推薦系統的三個主要元件。
嵌入、相似度指標和放送次數
這些元素與推薦系統相關,但並非主要元件。
矩陣分解、DNN 和排名
儘管重新排名是元件,矩陣矩陣分解和 DNN 便是候選的產生器類型。
除非另有註明,否則本頁面中的內容是採用創用 CC 姓名標示 4.0 授權,程式碼範例則為阿帕契 2.0 授權。詳情請參閱《Google Developers 網站政策》。Java 是 Oracle 和/或其關聯企業的註冊商標。
上次更新時間:2024-07-26 (世界標準時間)。
[null,null,["上次更新時間:2024-07-26 (世界標準時間)。"],[[["\u003cp\u003eRecommendation systems enhance content browsing and help users discover new items.\u003c/p\u003e\n"],["\u003cp\u003eRecommender systems are built with three primary components: candidate generation, scoring, and re-ranking.\u003c/p\u003e\n"],["\u003cp\u003eWhile machine learning is often associated with recommendation systems, using it solely to promote sponsored items is not recommended.\u003c/p\u003e\n"]]],[],null,["# Check your understanding\n\n\u003cbr /\u003e\n\nWhy would you use recommendation systems? \nHaving a recommendation engine makes browsing content easier. \nPlus, a great recommendation system helps users find things they wouldn't have thought to look for on their own. \nYou think you have to sprinkle ML on everything. \nIt might seem that way, but in reality, there are much better reasons to use ML. \nYou want to direct users to sponsored items. \nYikes, this is not a great reason to use any ML solution. \nWhat are the primary components of a recommender system? \ncandidate generation, scoring, and re-ranking \nNicely done! These are the three primary components of any recommendation system. \nembedding, similarity metrics, and serving \nThese elements are related to recommendation systems, but they are not primary components. \nmatrix factorization, DNN, and reranking \nWhile re-ranking is a component, matrix factorization and DNN are types of candidate generators."]]