Forsense: Accelerating online research through sensemaking integration and machine research support.
summary
ForSense is a browser extension that integrates foraging and sensemaking by letting users clip paragraph-sized text, organize clips on a spatial canvas, and get relevance suggestions computed with BERT embeddings. A 10-participant study reports mean SUS 72.3, roughly five groups with ~three clips each, and participants accept imperfect suggestions as inspiration because suggestions are easy to undo. Key claims: paragraph-sized clips improve coherence, integrated UI reduces app switching, and lightweight corrective actions let users recover from imprecise machine support.
AI-written summary · about
chapters (10)
- 0:00Introduction to Forsense
- 1:06Problem and research goals
- 2:30Design principles for Forsense
- 3:54Forsense browser extension demo
- 5:37User study setup and findings
- 6:39Machine support effectiveness
- 7:21Integration benefits
- 8:02Handling imprecise machine support
- 9:25Summary and future work
- 10:07Acknowledgments and Q&A
related talks
13mSupercharging user experience with machine learning
19mSupporting Mobile Sensemaking Through Intentionally Uncertain Highlighting
26mAccelerating COVID-19 insights for clinical use with intelligent search
38mWhat the User (Doesn't) Say
36mUnlocking Smarter Typeahead Search
21mBring AI-Based Search to Your Web App
11mRethinking Dataset Discovery with DataScout
13mReMap: Lowering the Barrier to Help-Seeking with Multimodal Search