The book presented an approach to leveraging cognitive features for NLP by harnessing eye-movement information from human readers and annotators. Eye-tracking technology is primarily used to record and analyze shallow cognitive information during text reading and annotation, to achieve the following goals: (a) better assessment of annotation effort to (a) increase annotation efficiency and (b) rationalize annotation pricing, and (b) augmenting text-based features with Cognition-Driven Features. The efficacy of the approaches has been exemplified by translation, sentiment analysis, and sarcasm detection tasks.
Sarcasm Detection Sentiment Analysis Gaze Data Learning Using Privileged Information (LuPI) Basic Convolutional Neural Networks Architecture
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.
This is a preview of subscription content, log in to check access.
Cortes, C., & Mohri, M. (2007). On transductive regression. In Advances in Neural Information Processing Systems (pp. 305–312).Google Scholar
Vapnik, V., & Vashist, A. (2009). A new learning paradigm: Learning using privileged information. Neural Networks, 22(5), 544–557.CrossRefGoogle Scholar
Zhou, Z.-H., & Li, M. (2005). Semi-supervised regression with co-training. In IJCAI (pp. 908–916).Google Scholar