Article recommendation method based on multi-attribute features

Abstract

The invention discloses an article recommendation method based on multi-attribute features, and belongs to the field of information processing. According to the method, more article features are extracted by using a multi-attribute article feature recommendation method, and the recommendation performance is improved: a struc2vec embedding vector based on an article quotation network, a metapath2vec embedding vector based on a heterogeneous network with article author and organization information, and an embedding vector of an article title and abstract content doc2vec are utilized, and on the basis of an original quotation network, through a graph reconstruction method, the embedding results of the isomorphic quotation network, the heterogeneous article network and the text information can be combined according to the weight. For a multi-attribute feature reconstruction network, graph embedding is carried out by using a method capable of combining structure information and homogeneous information, recommendation performance is improved, an embedding vector, containing the structure information and the homogeneous information, of an article node is obtained through a node2vec method, and finally recommendation is carried out through vector similarity.

Description

technical field [0001] The invention belongs to the field of information processing, in particular to an academic article recommendation method based on text features, citation network features and heterogeneous network features. Background technique [0002] Citation network: The citation network is a graph composed of academic articles as nodes. If there is a citation relationship between two academic articles, connect the two nodes; the citation network is G(E, V), v i ∈V means article v i In the data set, if e ij ∈E means article v i , v j There is a reference relationship; the reference relationship here is connected to the edge e ij is undirected, that is to say, it does not distinguish between articles v i By article v j citation or article v j By article v i quote. [0003] Heterogeneous network: For a graph G(E, V), if the points in the node set V are divided into different types, and there are edges between different types of points, then this graph is a h...

AI Technical Summary

Problems solved by technology

However, most graph-based recommendation algorithms ignore article text information.

[0021] Recommendation methods based on article text information, such as article information in the ACM classification tree, TF-IDF to extract the terminology of the article text for article recommendation, and doc2vec for text representation, etc., they cannot combine the citation relationship between articles, the author of the article, etc. Information such as the relationship between them is added to the recommendation results

Some graph embedding-based article recommendation methods that incorporate text features also have shortcomings

For example, paper2vec, although using the graph embedding method of the citation network, only uses the word2vec method to obtain the text features of the words, without considering the information brought by the order of the words in the text

Images