Summary of Suicide Risk Assessment on Social Media with Semi-supervised Learning, by Max Lovitt et al.
Suicide Risk Assessment on Social Media with Semi-Supervised Learning
by Max Lovitt, Haotian Ma, Song Wang, Yifan Peng
First submitted to arxiv on: 18 Nov 2024
Categories
- Main: Computation and Language (cs.CL)
- Secondary: Artificial Intelligence (cs.AI); Social and Information Networks (cs.SI)
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 semi-supervised framework is proposed for developing automated suicide risk assessment systems, which leverages both labeled and unlabeled data to address the class imbalance issue. The framework expands upon self-training with a pseudo-label acquisition process designed to handle imbalanced datasets. RoBERTa is identified as the best-performing model backbone, and by incorporating partially validated pseudo-labeled data alongside ground-truth labeled data, the model’s ability to assess suicide risk from social media posts is significantly improved. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary A team of researchers created a new system to help identify people who might be thinking about hurting themselves on social media. They developed a special way to use both labeled and unlabeled information to make this system better. This approach helps solve the problem of having more “not at risk” examples than “at risk” ones. The best-performing model was RoBERTa, and by combining this with some verified pseudo-labeled data, they were able to create a much better suicide risk assessment tool. |
Keywords
» Artificial intelligence » Self training » Semi supervised
