Summary of Active Learning For Identifying Disaster-related Tweets: a Comparison with Keyword Filtering and Generic Fine-tuning, by David Hanny et al.
Active Learning for Identifying Disaster-Related Tweets: A Comparison with Keyword Filtering and Generic Fine-Tuning
by David Hanny, Sebastian Schmidt, Bernd Resch
First submitted to arxiv on: 19 Aug 2024
Categories
- Main: Computation and Language (cs.CL)
- Secondary: 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 novel Machine Learning (ML) approach is proposed to identify disaster-related posts on social media platforms, leveraging the power of Active Learning (AL). By fine-tuning RoBERTa models with CrisisLex data and incorporating AL techniques, a robust classification model is developed to outperform traditional methods. The study compares various approaches, including keyword filtering, topic modeling, and classification-based techniques, demonstrating the potential of AL in text classification tasks. With only minimal labelling effort required, this model can be applied to a wide range of use cases beyond disaster response, offering valuable insights for social media analysis. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Social media can provide vital information during natural disasters. But finding the important posts among all the data is hard. This study uses a new way called Active Learning (AL) to identify disaster-related tweets on social media. They compare different methods and find that using AL and fine-tuning models with some generic data works best. This means you can train a model to find disaster-related posts with very little extra work. It’s not just useful for disasters, but also for understanding how people use social media. |
Keywords
» Artificial intelligence » Active learning » Classification » Fine tuning » Machine learning » Text classification
