YNU-HPCC at SemEval-2020 Task 8: Using a Parallel-Channel Model for Memotion Analysis

TL;DR

This paper presents a parallel-channel model combining text and visual cues for meme sentiment analysis, utilizing BERT and CNNs, achieving improved results in SemEval-2020 Task 8.

Contribution

It introduces a novel parallel-channel approach integrating textual and visual features with ensemble methods for meme sentiment classification.

Findings

System outperforms baseline algorithms

Achieved 19th place in sentiment classification subtask

Effective fusion of BERT and CNN features

Abstract

In recent years, the growing ubiquity of Internet memes on social media platforms, such as Facebook, Instagram, and Twitter, has become a topic of immense interest. However, the classification and recognition of memes is much more complicated than that of social text since it involves visual cues and language understanding. To address this issue, this paper proposed a parallel-channel model to process the textual and visual information in memes and then analyze the sentiment polarity of memes. In the shared task of identifying and categorizing memes, we preprocess the dataset according to the language behaviors on social media. Then, we adapt and fine-tune the Bidirectional Encoder Representations from Transformers (BERT), and two types of convolutional neural network models (CNNs) were used to extract the features from the pictures. We applied an ensemble model that combined the BiLSTM, BIGRU, and Attention models to perform cross domain suggestion mining. The officially released results show that our system performs better than the baseline algorithm. Our team won nineteenth place in subtask A (Sentiment Classification). The code of this paper is availabled at : https://github.com/YuanLi95/Semveal2020-Task8-emotion-analysis.