[Retracted] Exploration on the Action-Oriented Teaching Mode of Higher Vocational Political Courses under the Background of Internet + Education
Table 4
Comparative analysis of algorithms.
User based
Content based
Field
Strong timeliness
Strong personalized demand
Real-time
Users have new behaviors, which may not necessarily cause the recommendation results to change immediately
New user behavior will lead to changes in recommendation results
Cold boot
The user similarity matrix is calculated offline every other period of time, and new items can be recommended to interested users after they are online for a period of time
As long as a new user is interested in an item, he can recommend similar items to him