应用最大熵马尔可夫模型标注阅读眼动序列

doi:10.3778/j.issn.1002-8331.2108-0090

摘要/Abstract

摘要： 人在阅读过程中的眼球运动具有一定规律，阅读眼动模型有助于人们更好地理解和认知这些规律。针对现有阅读眼动模型建模方法复杂的问题，突破传统阅读眼动模型注视粒度处理和回视处理模式，提出了基于单词的阅读眼动注视粒度处理模式和基于熟练读者的阅读眼动回视处理模式，利用阅读眼动注视序列标注与自然语言序列标注的高度相似性，形成了阅读眼动注视序列标注方法，从而把复杂的阅读眼动建模过程转化成了简单的语言序列标注过程，并使用最大熵马尔可夫模型实现了所提出的方法。实验结果表明，所提出的方法可以较好地描述阅读眼动任务，并且较易用机器学习模型进行实现。

关键词: 阅读眼动, 眼动模型, 眼动追踪, 序列标注, 最大熵马尔可夫

Abstract: The eye-movements during reading have certain rules, and the reading eye-movement models help researchers better understand and recognize these rules. However, the existing reading eye-movement models have disadvantages in complex modeling approaches. This paper proposes a word-based fixation granularity processing mode and a skilled reader-based regression processing mode for reading eye-movement, making a breakthrough in the fixation granularity processing and regression of the traditional reading eye-movement models. This paper makes use of the striking parallel between reading eye-movement fixation sequence labeling and natural language sequence labeling, transforming the complex reading eye-movement modeling process into a simple language sequence labeling process, and uses maximum entropy Markov model to implement the proposed method. The experimental results show that the proposed method can better describe the reading eye-movement task, and it is easy to implement with the machine learning model.

Key words: reading eye-movement, eye-movement model, eye-tracking, sequence labeling, maximum entropy Markov

王晓明, 王莹. 应用最大熵马尔可夫模型标注阅读眼动序列[J]. 计算机工程与应用, 2022, 58(24): 166-172.

WANG Xiaoming, WANG Ying. Applying Maximum Entropy Markov Model to Label Reading Eye-Movement Sequence[J]. Computer Engineering and Applications, 2022, 58(24): 166-172.

参考文献

[1] RAYNER K.Eye movements in reading and information processing：20 years of research[J].Psychological Bulletin，American Psychological Association Inc，1998，124（3）：372-422.
[2] 赵怡，高淑萍，何迪.基于深度学习的眼动跟踪数据融合算法[J].计算机工程与应用，2021，57（10）：211-217.
ZHAO Y，GAO S P，HE D.Eye movement and tracking data fusion algorithm based on deep learning[J].Computer Engineering and Applications，2021，57（10）：211-217.
[3] 王晓明，赵歆波.基于深度神经网络的个体阅读眼动预测[J].清华大学学报（自然科学版），2019，59（6）：468-475.
WANG X M，ZHAO X B.Eye movement prediction of individuals while reading based on deep neural networks[J].Journal of Tsinghua University（Science and Technology），2019，59（6）：468-475.
[4] MORRISON R E.Manipulation of stimulus onset delay in reading：evidence for parallel programming of saccades[J].Journal of Experimental Psychology：Human Perception and Performance，1984，10（5）：667-682.
[5] REICHLE E D，POLLATSEK A，FISHER D L，et al.Toward a model of eye movement control in reading[J].Psychological Review，1998，105（1）：125-157.
[6] REICHLE E D，RAYNER K，POLLATSEK A.The E-Z reader model of eye-movement control in reading：comparisons to other models[J].Behavioral and Brain Sciences，2003，26（4）：445-476.
[7] ENGBERT R，LONGTIN A，KLIEGL R.A dynamical model of saccade generation in reading based on spatially distributed lexical processing[J].Vision Research，2002，42（5）：621-636.
[8] ENGBERT R，NUTHMANN A，RICHTER E M，et al.Swift：a dynamical model of saccade generation during reading[J].Psychological Review，2005，112（4）：777-813.
[9] REILLY R G，RADACH R.Some empirical tests of an interactive activation model of eye movement control in reading[J].Cognitive Systems Research，2006，7（1）：34-55.
[10] SNELL J，VAN LEIPSIG S，GRAINGER J，et al.OB1-reader：a model of word recognition and eye movements in text reading[J].Psychological Review，2018，125（6）：969-984.
[11] TWEED D，VILIS T.A two dimensional model for saccade generation[J].Biological Cybernetics，1985，52（4）：219-227.
[12] DAI W，SELESNICK I，RIZZO J R，et al.A parametric model for saccadic eye movement[C]//2016 IEEE Signal Processing in Medicine and Biology Symposium，2016.
[13] KRUCHININA A P，YAKUSHEV A G.A mathematical model of optimal saccadic eye movement by a pair of muscles[J].Biophysics（Russian Federation），2018，63（2）：241-247.
[14] ISKANDER J，HOSSNY M，NAHAVANDI S，et al.An ocular biomechanic model for dynamic simulation of different eye movements[J].Journal of Biomechanics，2018，71：208-216.
[15] LEGGE G E，KLITZ T S，TJAN B S.Mr.Chips：an ideal-observer model of reading[J].Psychological Review，1997，104（3）：524-553.
[16] LEGGE G E，HOOVEN T A，KLITZ T S，et al.Mr.Chips 2002：new insights from an ideal-observer model of reading[J].Vision Research，2002，42（18）：2219-2234.
[17] REICHLE E D，LAURENT P A.Using reinforcement learning to understand the emergence of “intelligent” eye-movement behavior during reading[J].Psychological Review，2006，113（2）：390-408.
[18] BICKNELL K，LEVY R.A rational model of eye movement control in reading[C]//ACL 2010-48th Annual Meeting of the Association for Computational Linguistics，Proceedings of the Conference，2010：1168-1178.
[19] BICKNELL K.Eye movements in reading as rational behavior[J].Dissertation Abstracts International：Section B：The Sciences and Engineering，2012，72（11-B）：7071.
[20] RIMEY R D，BROWN C M.Controlling eye movements with hidden Markov models[J].International Journal of Computer Vision，1991，7（1）：47-65.
[21] NILSSON M，NIVRE J.Learning where to look：modeling eye movements in reading[C]//Proceedings of the Thirteenth Conference on Computational Natural Language Learning，2009：93-101.
[22] NILSSON M，NIVRE J.Entropy-driven evaluation of models of eye movement control in reading[C]//Human-Machine Interaction in Translation：Proceedings of the 8th International NLPCS Workshop，Samfundslitteratur，2011：201.
[23] HARA T，MOCHIHASHI D，KANO Y，et al.Predicting word fixations in text with a CRF model for capturing general reading strategies among readers[C]//Proceedings of the First Workshop on Eye-Tracking and Natural Language Processing，Mumbai，India，2012：55-70.
[24] MATTIES F，S?GAARD A.With blinkers on：robust prediction of eye movements across readers[C]//2013 Conference on Empirical Methods in Natural Language Processing，2013：803-807.
[25] KIT D，SULLIVAN B.Classifying mobile eye tracking data with hidden Markov models[C]//Proceedings of the 18th International Conference on Human-Computer Interaction with Mobile Devices and Services Adjunct，2016：1037-1040.
[26] HAHN M，KELLER F.Modeling human reading with neural attention[C]//Proceedings of the 2016 Conference on Empirical Methods in Natural Language Processing.Austin，Texas：Association for Computational Linguistics，2016：85-95.
[27] YU A W，LEE H，LE Q V.Learning to skim text[C]//55th Annual Meeting of the Association for Computational Linguistics，2017：1880-1890.
[28] DE VRIES W.Explaining lexical processing times with cognitively plausible computational models[C]//CEUR Workshop Proceedings，2019.
[29] ELSAYED A A H A.Probabilistic，deep，and metric learning for biometric identification from eye movements[D].Universit?t Potsdam，2019.
[30] WANG Y，WANG X M，WU Y W.A simple model of reading eye movement based on deep learning[J].IEEE Access，2020，8：193757-193767.
[31] JOTHI PRABHA A，BHARGAVI R.Predictive model for dyslexia from fixations and saccadic eye movement events[J].Computer Methods and Programs in Biomedicine，2020，195.
[32] KAZAMA J，MIYAO Y，TSUJII J.A maximum entropy tagging model with unsupervised hidden Markov models[J].Journal of Natural Language Processing，2004，11（4）：3-23.
[33] MEHDORN E，LAQUA H.Circulatory disorders of the retina and the optic nerve[C]//Fortschritte der Ophthalmologie，1990.
[34] LEE A G，MORGAN M L，PALAU A E B，et al.Anatomy of the optic nerve and visual pathway[C]//Nerves and Nerve Injuries，2015：277-303.
[35] SLATTERY T J，YATES M.Word skipping：effects of word length，predictability，spelling and reading skill[J].Quarterly Journal of Experimental Psychology，2018，71（1 Special）：250-259.
[36] SCHUSTER S，HAWELKA S，HUTZLER F，et al.Words in context：the effects of length，frequency，and predictability on brain responses during natural reading[J].Cerebral Cortex，2016，26（10）：3889-3904.
[37] LUKE S G，CHRISTIANSON K.Limits on lexical prediction during reading[J].Cognitive Psychology，2016，88：22-60.
[38] LIU Y，YU L，FU L，et al.The effects of parafoveal word frequency and segmentation on saccade targeting during Chinese reading[J].Psychonomic Bulletin and Review，2019，26（4）：1367-1376.
[39] LIU R，CHENG S，MA L，et al.Deep proximal unrolling：algorithmic framework，convergence analysis and applications[J].IEEE Transactions on Image Processing，2019，28（10）：5013-5026.
[40] LI X，LIU P，RAYNER K.Saccade target selection in Chinese reading[J].Psychonomic Bulletin and Review，2015，22（2）：524-530.
[41] KLAUDA S L，GUTHRIE J T.Relationships of three components of reading fluency to reading comprehension[J].Journal of Educational Psychology，2008，100（2）：310-321.
[42] LUKE S G，CHRISTIANSON K.The provo corpus：a large eye-tracking corpus with predictability norms[J].Behavior Research Methods，2018，50（2）：826-833.
[43] 王晓明，赵歆波.阅读眼动追踪语料库的构建与应用研究综述[J].计算机科学，2020，47（3）：133-140.
WANG X M，ZHAO X B.Survey of construction and application of reading eye-tracking corpus[J].Computer Science，2020，47（3）：133-140.
[44] ZHAO R，YAN R，CHEN Z，et al.Deep learning and its applications to machine health monitoring[J].Mechanical Systems and Signal Processing，2019，115：213-237.