WHAT’S THIS SESSION ABOUT?
Responsible text mining is essential for ensuring that natural language processing systems are fair, transparent, and accountable. As large language models become increasingly important in research and decision-making, they can also reinforce societal biases, misclassify marginalized groups, and generate harmful content.
This Data Snack presents a practical approach to integrating ethical evaluation into text mining workflows. By using methods such as CheckList and HateCheck, researchers can assess how models perform in realistic scenarios, beyond simple accuracy metrics. These evaluations reveal key limitations, including misclassification of harmful language and unintended blocking of valid or supportive content.
The framework also highlights the importance of evaluating model performance across different groups, including women, people with disabilities, and racial and ethnic minorities. Even with balanced data, models often perform less effectively for these groups due to hidden biases. In addition, the high energy consumption required to train large models raises important environmental concerns. Overall, responsible text mining should be treated as a core competency in data science, supporting ethical reasoning and transparent research practices.
WHERE AND WHEN?
The session will take place from 1:00 to 1:30pm via Zoom. There will be a 15-20 minute presentation followed by an open forum for questions and discussion. The slides will be shared afterwards. We look forward to exciting discussions!
Zoom Link: https://uni-bremen.zoom-x.de/j/61666538039?pwd=JZTcc15FcsrZwedVYuMoVuhp3sPLCF.1
Download .ics calendar entry
ABOUT THE SPEAKER
Maryam Movahedifar holds a PhD in Statistics and has extensive experience in interpretable machine learning. With a strong foundation in statistical methods and practical experience in applying these techniques to real world problems, her work focuses on making complex models more transparent and understandable. In addition, she has expertise in text mining and the analysis of large scale textual data.