Recommendation method of semi-supervised learning based on graph consistency model

Abstract

The invention discloses a recommendation method of the semi-supervised learning based on a graph consistency model. The recommendation method comprises the following steps: S1 obtaining a server data set, constructing the graph consistency model for the data set, and establishing data points and sides; S2 measuring the similarity of two data points through a mapping function, and giving a weighted value to the side which exists between the data points; S3 establishing and solving an objective function, and sorting solutions of the objective function; S4 performing convergence proof on the sorted solutions of the objective function, so as to obtain a recommendation list, and sending the recommendation list to a user terminal. According to the recommendation method, personalized recommendation of user behavior information and article content information can be realized.

Description

technical field [0001] The present invention relates to the field of computer data mining, in particular to a Figure 1 A recommended method for semi-supervised learning of consistent models. Background technique [0002] Traditional recommendation methods often adopt a hybrid recommendation method when using user behavior information and item content information to achieve personalized recommendations. The specific strategy is to use the collaborative filtering-based recommendation method to calculate the similarity of user behavior, use the content-based recommendation method to model the content information of the item, and then combine the two recommendation results according to certain principles to generate the final product. Recommended list. [0003] The selection of combination methods mainly includes: weighting, transformation, mixing, feature combination, cascading, feature expansion, and meta-level, but the formulation of combination strategies is a thorny issue...

AI Technical Summary

Problems solved by technology

For example: ①Using the weighted combination method, it is required to set different proportions for content-based recommendation results and collaborative filtering-based recommendation results, and then add up to get the final recommendation results, but there is no definite answer on how to determine the weighting factor; ②Using the transformation and combination method, the recommendation system changes different recommendation strategies according to the problem background and the actual situation. Since there are many recommendation technologies in the recommendation system, only one of the strategies can be adopted according to the specific environment each time. In the actual situation, it is recommended There are no exact criteria for the choice of strategy

Images