HAlf-MAsked Model for Named Entity Sentiment analysis

TL;DR

This paper introduces a novel half-masked approach for Named Entity Sentiment Analysis using transformers, improving accuracy and consistency by ensembling models and addressing overfitting issues.

Contribution

The paper proposes a new half-masked technique for entity-level sentiment analysis and demonstrates its effectiveness in ensemble models on the RuSentNE-23 dataset.

Findings

Achieved the best results on RuSentNE-23 evaluation data

Improved model consistency in entity-level sentiment analysis

Addressed overfitting challenges in transformer-based NESA models

Abstract

Named Entity Sentiment analysis (NESA) is one of the most actively developing application domains in Natural Language Processing (NLP). Social media NESA is a significant field of opinion analysis since detecting and tracking sentiment trends in the news flow is crucial for building various analytical systems and monitoring the media image of specific people or companies. In this paper, we study different transformers-based solutions NESA in RuSentNE-23 evaluation. Despite the effectiveness of the BERT-like models, they can still struggle with certain challenges, such as overfitting, which appeared to be the main obstacle in achieving high accuracy on the RuSentNE-23 data. We present several approaches to overcome this problem, among which there is a novel technique of additional pass over given data with masked entity before making the final prediction so that we can combine logits from the model when it knows the exact entity it predicts sentiment for and when it does not. Utilizing this technique, we ensemble multiple BERT- like models trained on different subsets of data to improve overall performance. Our proposed model achieves the best result on RuSentNE-23 evaluation data and demonstrates improved consistency in entity-level sentiment analysis.