A Collaborative Filtering Recommendation Method Based on Decomposition Multi-objective Evolutionary Algorithm
A multi-objective evolution and collaborative filtering technology, applied in the field of recommendation systems, can solve the problem of high probability of being recommended, and achieve the effect of ensuring the accuracy and making the recommendation algorithm more user-friendly.
Active Publication Date: 2020-09-29
XIDIAN UNIV
View PDF3 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
But studies have shown that too much emphasis on recommendation accuracy, then the recommendation results will be very similar, and the probability of popular products being recommended will be greater
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0061] The present invention is described in further detail below in conjunction with accompanying drawing:
[0062] see Figure 1-3 , the present invention is based on the collaborative filtering recommendation method of decomposition multi-objective evolutionary algorithm, comprises the following steps:
[0063] Step 1: Input the existing user's evaluation information data on the item to get a rating matrix M u×i ;
[0064] Where u represents the number of users, i represents the number of items; each element of the scoring matrix represents the user's evaluation value of the item;
[0065] Step 2: Initialization: complete the input sparse scoring matrix,
[0066] (2a) According to the collaborative filtering algorithm based on the minimum k nearest neighbors, the similarity between items is obtained; here, the similarity between item i and item j is obtained by using Pearson correlation:
[0067]
[0068] Among them, U is a set composed of all users who have rated bo...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The invention discloses a decomposition-based multi-objective evolutionary algorithm-based collaborative filtering recommendation method, and mainly aims at solving the problem that the recommendation results on recommendation problems are convergent in the prior art. The method comprises the following realization steps of: (1) carrying out score estimation on articles by users by utilizing a collaborative filtering algorithm; (2) determining two objective functions; (3) constructing an initial population, and initializing individuals in the population by adoption of a real number encoding manner; (4) carrying out selection, cross and mutation operations on the individuals in the population to obtain a new filial generation; (5) updating the population by utilizing the filial generation; (6) judging whether to terminate the operation, when an iteration frequency achieves a set frequency, executing the step (7), and otherwise, returning to the step (4); (7) repeatedly executing the step (3) to each user; and (8) finally generating a series of non-dominant recommendation lists for each user.
Description
【Technical field】 [0001] The invention belongs to the field of recommendation systems, and relates to a collaborative filtering recommendation method based on a decomposition multi-objective evolutionary algorithm, in particular to a collaborative filtering recommendation algorithm based on a multi-objective evolutionary algorithm, which can be used in a variety of recommendation systems. 【Background technique】 [0002] The recommendation system is a very important information acquisition tool in a society with massive information. Facing the modern era of big data, the way people obtain information will change from traditional search engines to recommendation systems. For traditional search engines, each user will get the same results using the same keywords, and the server will present the results according to a certain sort. In the future, or in other words, the trend that is developing now will be to recommend information to users according to each user's individual sit...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More Patent Type & Authority Patents(China)
IPC IPC(8): G06Q30/06G06N3/12
CPCG06N3/12G06Q30/0631
Inventor 公茂果秦晓雷陈兰强王善峰
Owner XIDIAN UNIV