NLPBK at VLSP-2020 shared task: Compose transformer pretrained models for Reliable Intelligence Identification on Social network

TL;DR

This paper presents a transformer-based model combined with metadata features for reliable intelligence detection on Vietnamese social media, achieving top-tier ROC-AUC scores.

Contribution

It introduces a novel approach that integrates pretrained transformer models with metadata features for social media intelligence identification.

Findings

Achieved 0.9392 ROC-AUC on public test set

Final ROC-AUC of 0.9513 on private test set

Model outperforms baseline methods in the shared task

Abstract

This paper describes our method for tuning a transformer-based pretrained model, to adaptation with Reliable Intelligence Identification on Vietnamese SNSs problem. We also proposed a model that combines bert-base pretrained models with some metadata features, such as the number of comments, number of likes, images of SNS documents,... to improved results for VLSP shared task: Reliable Intelligence Identification on Vietnamese SNSs. With appropriate training techniques, our model is able to achieve 0.9392 ROC-AUC on public test set and the final version settles at top 2 ROC-AUC (0.9513) on private test set.