Summary of Explainable Multimodal Sentiment Analysis on Bengali Memes, by Kazi Toufique Elahi et al.
Explainable Multimodal Sentiment Analysis on Bengali Memes
by Kazi Toufique Elahi, Tasnuva Binte Rahman, Shakil Shahriar, Samir Sarker, Sajib Kumar Saha Joy, Faisal Muhammad Shah
First submitted to arxiv on: 20 Dec 2023
Categories
- Main: Computer Vision and Pattern Recognition (cs.CV)
- Secondary: Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Machine Learning (cs.LG)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary A machine learning-based framework is developed to analyze sentiment in Bengali memes, which are a unique form of digital communication that can convey a wide range of emotions. The approach combines visual features from ResNet50 and linguistic features from BanglishBERT, achieving a weighted F1-score of 0.71. This performance surpasses unimodal approaches, demonstrating the effectiveness of multimodality in detecting sentiment. Additionally, explainable artificial intelligence (XAI) techniques are used to interpret the behaviors of the models, providing insights into their decision-making processes. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Bengali memes have become a popular way for people to express themselves online, but understanding what they mean can be tricky. Researchers have been working on developing ways to analyze these memes and figure out how people feel about them. So far, most of this work has focused on languages like English, leaving Bengali memes behind. In this study, scientists came up with a new way to look at both the pictures and text in Bengali memes together. This approach worked better than just looking at one or the other, showing that combining different types of information can be really helpful. |
Keywords
* Artificial intelligence * F1 score * Machine learning
