Social recommendation method based on heterogeneous information network

Abstract

The invention requests to protect a social recommendation method based on a heterogeneous information network, which comprises the following steps of: for the heterogeneous information network, normalizing each associated edge weight W in the network by using a mapping function to obtain a processed weight omega; and then searching all path instances under the corresponding meta-path by using a meta-path-based method, and calculating a similarity relationship s(x,y) between objects of the same type under each meta-path. In order to deeply mine the similarity of the objects, matrix decomposition is used for projecting the similarity relation of the objects to a low-dimensional feature space Feature, and each object can be subjected to characterization representation through a unique vectorin the space. After the feature information under all meta-paths is solved, all the feature data is input into the gradient boosting decision tree model, model training is carried out, the linear relation and the nonlinear relation between the features are learned, and therefore the accuracy of the recommendation model is improved.

Description

technical field [0001] The invention belongs to the technical field of data mining and information fusion of heterogeneous information, in particular to a social recommendation algorithm. Background technique [0002] Traditional recommendation algorithms are mainly composed of content-based recommendation, collaborative recommendation, hybrid recommendation and model-based recommendation. Each recommendation algorithm has its own advantages and disadvantages. Collaborative filtering algorithm is an algorithm based on explicit data. Although it alleviates the data cold start problem to some extent, it is still affected by data sparsity. The content-based recommendation algorithm does not need to display rating data, it recommends based on the user's historical behavior records, and usually recommends some items similar to the previous items for the user. This recommendation method easily leads to over-personalized recommendations. The hybrid recommendation algorithm combin...

AI Technical Summary

Problems solved by technology

[0004] The problem mainly solved by the present invention is that the recommendation accuracy is not high due to data sparsity

Images