Article recommendation method and device and storage medium
A recommendation method and item technology, applied in the field of recommendation, can solve the problems of ignoring cold-start users or item learning, inaccurate recommendation results, emphasizing multi-modal attribute information, etc., and achieve the effect of improving accuracy
Pending Publication Date: 2022-03-25
HUNAN UNIV
View PDF0 Cites 1 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0004] However, in the methods based on graph neural network, the learning of cold-start users or items in graph neural network is often ignored, or the multi-modal attribute information is overemphasized, which leads to inaccurate recommendation results.
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
[0060] The following will clearly and completely describe the technical solutions in the embodiments of the application with reference to the drawings in the embodiments of the application. Apparently, the described embodiments are only some of the embodiments of the application, not all of them.
[0061] The algorithms recommended by cold start are divided into; only use attribute information algorithm, use traditional neural network model algorithm and migration learning algorithm. Use only attribute information, such as matrix factorization-based methods, feature-based factor models, probability-based models, and regression-based methods. A mixed-attribute-based matrix factorization method combines item attributes and interaction information for cold-start recommendation. The use of traditional neural network model algorithms includes solving the cold-start recommendation problem by using multi-layer neural networks, recurrent neural networks, and long-short-term memory neu...
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 provides an article recommendation method and related equipment. The accuracy of cold start can be improved. The method comprises the steps of determining multi-modal information of each cold start object in a cold start object set and multi-modal information of each warm object in a warm object set, wherein the multi-modal information comprises at least two of identification information, visual information features, audio information features and text information features; performing hierarchical clustering according to the multi-modal information of each cold start object and the multi-modal information of each warm object to obtain a plurality of clustering results; constructing a supplementary interaction graph between each cold start object and a warm object according to the plurality of clustering results, wherein the warm object corresponds to the warm object set; determining a first final representation of each cold start object and a second final representation of a warm object corresponding to each cold start object according to the supplementary interaction diagram; and recommending each cold start object according to the first final representation and the second final representation.
Description
【Technical field】 [0001] The present application belongs to the recommendation field, and in particular relates to an item recommendation method, device and storage medium. 【Background technique】 [0002] In the current era of information overload, people are increasingly inseparable from personalized recommendation systems, especially in the fields of e-commerce, social media and advertising. Current personalized recommendation systems highly rely on historical interactions between users and items, such as historical purchase records and click counts. However, they often fail to perform well when new users or new items appear, that is, the cold-start problem. How to recommend products to new users more accurately or recommend new items to users is an urgent task to be solved in the cold start problem. [0003] In order to alleviate this problem, common attribute information of users or items is used most, such as user’s age, gender and other attribute information as well ...
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 IPC IPC(8): G06F16/9535G06K9/62G06N3/04G06N3/08G06Q30/06
CPCG06F16/9535G06Q30/0631G06N3/08G06N3/045G06F18/2321G06F18/22
Inventor 曹达马守兴曾雅文陈诗雨陆绍飞陈浩
Owner HUNAN UNIV