An optimization method of a media personalized recommendation system
A recommendation system and optimization method technology, which is applied in the fields of instruments, computing, and electrical digital data processing, etc., can solve the problems of inaccurate recommendation results of personalized recommendation system and difficulty in project recommendation, etc.
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0126] The cold start problem of new projects is the main problem affecting the commercial value of collaborative filtering recommendation system. The cold start problem of new items means that when a new item is added to the recommendation system, the recommendation system cannot effectively filter target users for the new item due to the lack of rich user preference evaluation information or even no user preference evaluation information for the new item, resulting in When recommending new items to users, the target user's recommendation list hit rate is extremely low. Specifically, due to the lack of sufficient user preference evaluation information, it is difficult for the model-based collaborative filtering algorithm to effectively establish a user preference model for new items. best choice. Take Table 1 as an example:
[0127] Table 1 User-item rating matrix of a recommender system
[0128]
item 1
item 2
item 3
item 4
User A
5
4...
Embodiment 2
[0186] The new user cold start problem is an inherent problem in collaborative filtering recommender systems. The new user cold start problem means that when a new user joins the recommendation system, due to the lack of sufficient historical preference evaluation information for the new user, the collaborative filtering algorithm cannot perform efficient nearest neighbor search or preference modeling for the new user, resulting in the failure of the recommendation system. Make accurate item recommendations for new users. Take Table 3 as an example:
[0187] Table 3 User-item rating matrix of a recommender system
[0188]
item 1
item 2
item 3
item 4
User A
2
1
5
User B
3
5
User C
4
3
User D
4
[0189] Table 3 briefly shows a user-item rating matrix for a recommender system. Among them, user D is a new user of the recommendation system. Due to the sparseness of ...
Embodiment 3
[0224] The problem of data sparsity is one of the main research points of collaborative filtering recommender system. In an actual recommendation system, a large number of users and a large number of items lead to a huge dimension of the user-item rating matrix. A large number of ratings are missing from the rating matrix. When existing collaborative filtering algorithms deal with high-dimensional and extremely sparse user-item rating matrices, the item recommendation accuracy of the recommendation system drops severely, resulting in poor user experience and a large loss of users of the recommendation system. Take Table 4 as an example:
[0225] Table 4 User-item rating matrix of a recommender system
[0226]
item 1
item 2
item 3
item 4
...
User A
4
1
User B
2
User C
5
User D
3
...
[0227] Table 4 briefly shows a ...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More