News Popularity Prediction with Local-Global Long-Short-Term Embedding

“…The prevalence of online social networks has been profoundly changing our daily life. Social media users often share interesting content (e.g., literals (Fan et al, 2021), images (Wang et al, 2020), videos (Xie et al, 2020), etc.) with one another.…”

Section: Introductionmentioning

confidence: 99%

Explicit time embedding based cascade attention network for information popularity prediction

Sun

Zhou

Liu

et al. 2023

Information Processing & Management

View full text Add to dashboard Cite

“…As a result, news content feature modeling is more reliable in the early stages, while it fails to take advantage of the temporal evolution of popularity. Some works adopt multi-feature fusion methods to take advantage of time series and content features [12][13][14]. However, how to effectively represent and flexibly integrate the two types of features to exert their respective performance at different stages requires further research.…”

Section: Introductionmentioning

confidence: 99%

“…LSTM modeled the time propagation pattern, CNN learned semantic features and deep neural network (DNN) learned other auxiliary features in the model. Fan et al [14] proposed a neural model to predict news popularity by learning news embedding from global, local, long-term and short-term factors. Although the multi-feature fusion method uses the advantages of various features, how to represent the multiple features effectively and flexibly needs to be further studied.…”

Section: Introductionmentioning

confidence: 99%

Predicting the Popularity of Online News Based on the Dynamic Fusion of Multiple Features

Song,

Wang,

et al. 2023

Computers, Materials &Amp; Continua

View full text Add to dashboard Cite

Predicting the popularity of online news is essential for news providers and recommendation systems. Time series, content and meta-feature are important features in news popularity prediction. However, there is a lack of exploration of how to integrate them effectively into a deep learning model and how effective and valuable they are to the model's performance. This work proposes a novel deep learning model named Multiple Features Dynamic Fusion (MFDF) for news popularity prediction. For modeling time series, long short-term memory networks and attention-based convolution neural networks are used to capture long-term trends and short-term fluctuations of online news popularity. The typical convolution neural network gets headline semantic representation for modeling news headlines. In addition, a hierarchical attention network is exploited to extract news content semantic representation while using the latent Dirichlet allocation model to get the subject distribution of news as a semantic supplement. A factorization machine is employed to model the interaction relationship between metafeatures. Considering the role of these features at different stages, the proposed model exploits a time-based attention fusion layer to fuse multiple features dynamically. During the training phase, this work designs a loss function based on Newton's cooling law to train the model better. Extensive experiments on the real-world dataset from Toutiao confirm the effectiveness of the dynamic fusion of multiple features and demonstrate significant performance improvements over state-of-the-art news prediction techniques.

show abstract

News Popularity Prediction with Local-Global Long-Short-Term Embedding

Cited by 2 publications

References 23 publications

Explicit time embedding based cascade attention network for information popularity prediction

Explicit time embedding based cascade attention network for information popularity prediction

Predicting the Popularity of Online News Based on the Dynamic Fusion of Multiple Features

Contact Info

Product

Resources

About