Dynamic news recommendation method based on hierarchical attention network

Abstract

The invention provides a dynamic news recommendation method based on a hierarchical attention network. The adopted modules comprise the hierarchical attention network, a convolution layer and a full connection layer. The hierarchical attention network comprises a sentence level attention network and a news level attention network. A news sequence and candidate news of the user are subjected to thesentence level attention network to obtain attention weights of sentences in the news sequence; a weighted sum of the sentence content vectors is calculated to obtain a news content vector; the embedded representation of the news sequence is connected with the content vector to obtain the integral representation of the news, and the integral representation passes through the attention network ofthe news level to obtain the attention weight of the news in the news sequence and obtain the final representation of the news; in the convolutional layer, the final representations of the historicalnews are stacked in sequence to obtain a matrix, and the matrix is input into the convolutional layer to learn a user sequence reading mode to obtain a sequence preference vector; and in the full connection layer, the sequence preference vector, the candidate news overall representation and the user embedded representation are connected to obtain the probability that the user clicks the candidatenews.

Description

technical field [0001] The invention relates to a dynamic news recommendation method, in particular to a dynamic news recommendation method based on hierarchical attention networks. Background technique [0002] In recent years, with the rapid development of technologies such as cloud computing and big data, the emergence of various applications on the Internet has triggered an explosive growth of data scale. Big data contains rich value and great potential, which has brought transformative development to human society, but at the same time it has also brought about the problem of "information overload". How to quickly and effectively obtain valuable information from complex data has become a key problem in the development of big data. As an effective method to solve the "information overload" problem, recommender systems have become a hot spot in academia and industry and have been widely used. With the development of the World Wide Web, people's news reading habits have ...

AI Technical Summary

Problems solved by technology

Unable to find users with similar interests for new users with no history

More importantly, news is generated all the time, outdated news is quickly replaced by new news, and new news that has not been read cannot be recommended, so the collaborative filtering method is not suitable for the news field

[0006] The combination of the above two methods in the prior art three-hybrid method can alleviate their respective shortcomings, but the data often have complex characteristics such as multi-modality and heterogeneity, and the research of this method still faces severe challenges

[0007] The existing four deep learning-based methods all extract features from the news level, and few works consider fine-grained sentence-level distinctions, and few works consider the time decay characteristics of news reading and the sequence characteristics of news reading

Images