Weakly Supervised Part-of-speech Tagging Using Eye-tracking Data

Barrett, Maria; Bingel, Joachim; Keller, Frank; Søgaard, Anders

doi:10.18653/v1/p16-2094

Cited by 70 publications

(81 citation statements)

References 3 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Gaze information and other measures from psycholinguistics have been used in different ways in NLP. Some authors have used discretized, single features Goldwater, 2011, 2013;Plank, 2016;Klerke et al, 2016), whereas others have used multidimensional, continuous values (Barrett et al, 2016a;Bingel et al, 2016). We follow Gonzalez-Garduno and Søgaard (2018) in using a single, continuous feature.…”

Section: Discussion and Related Workmentioning

confidence: 99%

Sequence Classification with Human Attention

Barrett¹,

Bingel²,

Hollenstein³

et al. 2018

Proceedings of the 22nd Conference on Computational Natural Language Learning

Self Cite

View full text Add to dashboard Cite

Learning attention functions requires large volumes of data, but many NLP tasks simulate human behavior, and in this paper, we show that human attention really does provide a good inductive bias on many attention functions in NLP. Specifically, we use estimated human attention derived from eyetracking corpora to regularize attention functions in recurrent neural networks. We show substantial improvements across a range of tasks, including sentiment analysis, grammatical error detection, and detection of abusive language.

show abstract

Section: Discussion and Related Workmentioning

confidence: 99%

Sequence Classification with Human Attention

Barrett¹,

Bingel²,

Hollenstein³

et al. 2018

Proceedings of the 22nd Conference on Computational Natural Language Learning

Self Cite

View full text Add to dashboard Cite

show abstract

“…In previous studies the choice of number of gaze features used in the experiments has varied, seemingly, depending on the NLP task of interest. For instance, Barrett et al (2016) distinguish 31 features (where 22 are gaze features) for part-of-speech tagging while Hollenstein and Zhang (2019) use 17 features in the NER task. In another piece of work, 5 gaze features are used for relation classification and sentiment analysis .…”

Section: Gaze Informationmentioning

confidence: 99%

“…We have chosen 12 gaze features and based on the previous work (Barrett et al, 2016; we have divided them into 4 groups. In particular, we explore the informativeness of the basic gaze features: total fixation duration on a word w (total fix dur), mean fixation duration on w (mean fix dur, num-ber fixations on w (n fix) and fixation probability on w (fix prob).…”

Section: Gaze Informationmentioning

confidence: 99%

“…In this context, how to retrieve and leverage eye-tracking data has become an active area of research in different NLP fields. Previous studies (Barrett and Søgaard, 2015b;Barrett et al, 2016) suggest that real-time eye-tracking data can be collected at inference time, so that token-level gaze features are used during training but also at test time. However, even if in the near future every user has an eye tracker on top of their screen -a scenario which is far from guaranteed, and raises privacy concerns (Liebling and Preibusch, 2014)many running NLP applications that process data from various Internet sources will not expect to have any human being reading massive amounts of data.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Towards Making a Dependency Parser See

Strzyz¹,

Vilares²,

Gómez-Rodríguez³

2019

Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conferen

View full text Add to dashboard Cite

We explore whether it is possible to leverage eye-tracking data in an RNN dependency parser (for English) when such information is only available during training -i.e. no aggregated or token-level gaze features are used at inference time. To do so, we train a multitask learning model that parses sentences as sequence labeling and leverages gaze features as auxiliary tasks. Our method also learns to train from disjoint datasets, i.e. it can be used to test whether already collected gaze features are useful to improve the performance on new non-gazed annotated treebanks. Accuracy gains are modest but positive, showing the feasibility of the approach. It can serve as a first step towards architectures that can better leverage eye-tracking data or other complementary information available only for training sentences, possibly leading to improvements in syntactic parsing.

show abstract

“…This relationship between text and eye movements, has led to an influx of studies investigating the use of eye tracking data to improve and test computational models of language i.e. Barrett et al (2016); Demberg and Keller (2008); Klerke et al (2015). In this study we aim to incorporate eye movement data for the task of automatic readability assessment.…”

Section: Introductionmentioning

confidence: 99%

Proceedings of the 12th Workshop on Innovative Use of NLP for Building Educational Applications

Tetreault

Burstein

Leacock³

et al. 2017

View full text Add to dashboard Cite

Since the first workshop in 1997, BEA has become the leading venue for sharing and publishing innovative work that uses NLP to develop educational applications. The consistent interest and growth of the workshop has clear ties to challenges in education. The research presented at the workshop highlights advances in the technology and the maturity of the field of NLP in education. The capabilities serve as a response to educational challenges and are poised to support the needs of a variety of stakeholders, including educators, learners, parents, and administrators.NLP capabilities now support an array of learning domains, including writing, speaking, reading, and mathematics. In the writing and speech domains, automated writing evaluation (AWE) and speech assessment applications, respectively, are commercially deployed in high-stakes assessment and instructional settings, including Massive Open Online Courses (MOOCs). We also see widelyused commercial applications for plagiarism detection and peer review and explosive growth of mobile applications for game-based applications for instruction and assessment. The current educational and assessment landscape continues to foster a strong interest and high demand that pushes the state of the art in AWE capabilities to expand the analysis of written responses to writing genres other than those traditionally found in standardized assessments, especially writing tasks requiring use of sources and argumentative discourse.Steady growth in the development of NLP-based applications for education has prompted an increased number of workshops that typically focus on a single subfield. In BEA, we make an effort to have papers from many subfields, for example, tools for automated scoring, automated test-item generation, curriculum development, evaluation of text, dialogue, evaluation of genres beyond essays, feedback studies, and grammatical error correction.This year we received a record 62 submissions, and accepted 9 papers as oral presentations and 25 as poster presentation and/or demos, for an overall acceptance rate of 55 percent. Each paper was reviewed by three members of the Program Committee who were believed to be most appropriate for each paper. We continue to have a very strong policy to deal with conflicts of interest. First, we made a concerted effort to not assign papers to reviewers to evaluate if the paper had an author from their institution. Second, with respect to the organizing committee, authors of papers for which there was a conflict of interest recused themselves from the discussions.While the field is growing, we do recognize that there is a core group of institutions and researchers who work in this area. With a higher acceptance rate, we were able to include papers from a wider variety of topics and institutions. The papers accepted were selected on the basis of several factors, including the relevance to a core educational problem space, the novelty of the approach or domain, and the strength of the research. The accepted papers were highly diverse -...

show abstract

Weakly Supervised Part-of-speech Tagging Using Eye-tracking Data

Cited by 70 publications

References 3 publications

Sequence Classification with Human Attention

Sequence Classification with Human Attention

Towards Making a Dependency Parser See

Proceedings of the 12th Workshop on Innovative Use of NLP for Building Educational Applications

Contact Info

Product

Resources

About