Researchers have developed a novel approach to keyphrase extraction by incorporating human reading behavior, specifically eye-tracking data. Using a lightweight, webcam-based system and the open-source SearchGazer library, they created the Chinese LIS Eye-Tracking Corpus (CLIS-ET). This corpus includes features like first fixation duration, fixation number, and total fixation duration, which were integrated into an Att-BiLSTM+CRF model. The study found that these eye-tracking features significantly improved keyphrase extraction performance, with the combination of fixation number and total fixation duration yielding the best results. AI
IMPACT This research could lead to more accurate and context-aware keyphrase extraction tools by integrating human cognitive signals.
RANK_REASON The cluster describes an academic paper presenting a new methodology and dataset for keyphrase extraction.
- alphaXiv
- Att-BiLSTM+CRF
- CatalyzeX Code Finder for Papers
- CLIS-ET
- CORE Recommender
- DagsHub
- Gotit.pub
- Hugging Face
- Influence Flower
- library and information science
- ScienceCast
- SearchGazer
- CatalyzeX
AI-generated summary · Google Gemini · from 2 sources. How we write summaries →
