Download presentation
Presentation is loading. Please wait.
Published byShannon Fitzgerald Modified over 5 years ago
1
PURE Learning Plan Richard Lee, James Chen,
2
Project Overview Mentored by PhD candidate Jason Cho
Advice from Professor Eric Meyer of the University of Illinois Department of Journalism Project in automatic bias-detection in newspaper articles Split into two parts: Similar article topic recognition Easier, but less interesting Automatic bias detection Harder, but more interesting
3
What Have We Learned? This is a very, very difficult problem.
Project involves a variety of fields, including computer science, English, linguistics, mathematics, journalism, etc. Relatively unique problem, no thorough solution has been attempted/researched Similar research done, such as sarcasm detection, bias research (linguistics), etc.
4
Getting Up to Speed Tools Papers OpenNLP, Python NLTK
Stanford NLP Toolkit Papers Shedding (a Thousand Points of) Light on Biased Language Recognizing stances in online debates Extracting opinion targets in a single-and cross-domain setting with conditional random fields
5
Ideas Investigating words that have a high correlation with bias, and seeing if the presence of them indicate bias Take presumably biased articles, use named- entity recognition to parse relevant sentences (eg, all sentences with "Obama" present) Use a combination of word-counting, part-of- speech tagging, and word sentiment determination to aggregate overall bias.
Similar presentations
© 2025 SlidePlayer.com. Inc.
All rights reserved.