# Herding Effect based Attention for Personalized Time-Sync Video Recommendation

## TL;DR

This paper introduces ITF-HEA, a novel model leveraging herding effect attention on time-sync comments and video images to improve personalized video recommendations, outperforming existing methods.

## Contribution

The paper proposes a new Image-Text Fusion model with Herding Effect Attention to incorporate real-time, context-dependent TSC data into video recommendation systems.

## Key findings

- ITF-HEA achieves 3.78% higher F1-score than state-of-the-art methods.
- The model effectively captures herding effects in TSC data.
- Incorporating time and semantic similarities improves recommendation accuracy.

## Abstract

Time-sync comment (TSC) is a new form of user-interaction review associated with real-time video contents, which contains a user's preferences for videos and therefore well suited as the data source for video recommendations. However, existing review-based recommendation methods ignore the context-dependent (generated by user-interaction), real-time, and time-sensitive properties of TSC data. To bridge the above gaps, in this paper, we use video images and users' TSCs to design an Image-Text Fusion model with a novel Herding Effect Attention mechanism (called ITF-HEA), which can predict users' favorite videos with model-based collaborative filtering. Specifically, in the HEA mechanism, we weight the context information based on the semantic similarities and time intervals between each TSC and its context, thereby considering influences of the herding effect in the model. Experiments show that ITF-HEA is on average 3.78\% higher than the state-of-the-art method upon F1-score in baselines.