摘要 :
With approximately0.25ounce weight and about01(one)inch or25mm across size an eye is the fastest muscle in a human body,which is also second most complex part of the body after brain.To see the wonders of our universe,normal worki...
展开
With approximately0.25ounce weight and about01(one)inch or25mm across size an eye is the fastest muscle in a human body,which is also second most complex part of the body after brain.To see the wonders of our universe,normal working of each part of our eye is important and necessary.Studying eye movement trajectory through eye trackers can provide insights into classification mechanism from different aspect e.g.attention,mood,personalities,age grouping,eyes diseases,mental cognition and many more.But our work focus on experimental psychology for memory and reasoning classification for reading behavior.For the given text or images,memory and reasoning based observations are useful in multiple key areas of psychology,cognition,human computer interaction(HCI),gaming,virtual reality and related fields.Labeling and isolation of eye movement fixations and sequences-are considered an integral part of eye movement data analysis.The eye movement behaviors are generally divided into six types,but attention is often paid to fixation,saccade,and smooth pursuit.Therefore,it is essential to classify eye movement behaviors accurately using an excellent classification algorithm.The classification of eye movement should be a complete process,including three steps of pre-processing,classification and post-processing.However,it is very uncommon that all these steps are included in eye tracking literature where eye movement classification is discussed.For our project,screen-based EyeLink1000Plus is used to inspect and capture fixation,saccades,blinks etc.in eye tracking protocols of each participant(mature readers)during trials.Based on fixation sequence of each participant we trained deep recurrent neural network architecture–LSTM which classifies whether a participant is performing the given text-based task provided on monitor screen by memorizing it or by inference.To set our trial sentences,we follow the syllogism–A deductive type of reasoning in which conclusion is drawn from two given premises(propositions),each of which shares a term with the conclusion,and shares a common or middle term not present in the conclusion.A total of sixty(60)university students(all were mature readers and capable to understand given sentences)participated in memory-based reading trials and reasoning-based reading trials.Total number of participants were then divided into two equal groups and of which one group of thirty students was instructed to perform memory task first while the secondamp;nbsp;group of thirty students was instructed to perform reasoning based task first and then their order of performing the given tasks was changed.From sixty-one different sentences sixty(60)randomly selected sentences were presented to each participant.Eye movement fixations of all readers were captured with eye tracker and then s equential signals were generated from fixations of each participant,which were then processed to get hot vectors.Generated hot vectors were used as input to LSTM model.Our trained LSTM model successfully classify reading behavior of all participants.The obtained accuracy of our model for memory based reading behavior is95.39%and for reasoning based reading behavior it is94.5%.This high accuracy to classify the memory and the reasoning based reading behavior of participants ensures the significance of our w ork which will provide a solid base for future works on eye movements to build intelligent techniques for different areas of research especially in the field of A.I backed psychology,healthcare and neuro-marketing.Our trained model has the potential for(a)achieving very high accuracy for memory and reasoning classification,(b)for data learning,it saves enormous time,(c)Economical and comfortable for trials using EyeLink1000plus for data recording(d)and finally it can also provide a possible future work opportunity to classify human personalities and mood.Which might be future task as well.
收起