ReINTEL Challenge 2020: A Multimodal Ensemble Model for Detecting Unreliable Information on Vietnamese SNS

TL;DR

This paper introduces a multimodal ensemble approach combining text, images, and metadata to effectively identify unreliable information on Vietnamese social media, achieving high accuracy in a competitive challenge.

Contribution

The paper presents a novel multimodal ensemble model that integrates multiple neural networks for improved reliability classification on social media data.

Findings

Achieved 0.9445 ROC AUC score on the challenge test set.

Multimodal ensemble outperforms individual models in reliability detection.

Demonstrated effectiveness of combining text, images, and metadata features.

Abstract

In this paper, we present our methods for unrealiable information identification task at VLSP 2020 ReINTEL Challenge. The task is to classify a piece of information into reliable or unreliable category. We propose a novel multimodal ensemble model which combines two multimodal models to solve the task. In each multimodal model, we combined feature representations acquired from three different data types: texts, images, and metadata. Multimodal features are derived from three neural networks and fused for classification. Experimental results showed that our proposed multimodal ensemble model improved against single models in term of ROC AUC score. We obtained 0.9445 AUC score on the private test of the challenge.