河南理工大学学报（自然科学版）2025，Vol.44Issue(1)：145-153,9.DOI:10.16186/j.cnki.1673-9787.2024040012
基于生成式人工智能的眼动样本生成及识别
Generation and recognition of eye movement samples based on generative artificial intelligence
摘要
Abstract
Objectives Generative and traditional artificial intelligence models are pivotal tools in the infor-mation age.Leveraging these technologies,the generation and identification of eye movement samples have emerged as critical components,facilitating deeper explorations into cognitive mechanisms.Therefore,this study aims to promote the development of generative artificial intelligence in the field of eye tracking tech-nology,solve the problem of eye movement sample generation and the opacity and inexplicability caused by the increase in network depth,and deeply mine eye tracking data related to children's language develop-ment.Methods This study collected data on the eye movement process of 4～6 years old children's under-standing of different focus structures.Generative artificial intelligence model-variational autoencoder(VAE)and traditional models-multilayer perceptron(MLP)were used to identify the developmental differences in their eye movement patterns and attempt to generate new samples.Interpreting generative datasets based on grey relational analysis and confusion matrix.Results The results showed that:(1)the eye movement datasets generated by VAE for 4,5,and 6 years old children had higher accuracy than the MINIST dataset(mixed National Institute of Standards and Technology),and were consistent with the MLP analysis results,with accuracy,diversity,and certain interpretability;(2)The results of generative eye movement data and con-fusion matrix indicated that in unfocused structure,children's understanding level improved at the ages of 4～5 and 5～6,while the eye movement characteristics of object-focus structure and subject-focus structure changed less at the ages of 4～5 and more at the ages of 5～6,indicating that children's understanding of focus structure was a critical period at the age of 5,which was in line with the development law of children's un-derstanding of focus structure.Conclusions The artificial intelligence coupling analysis proposed in this ar-ticle could identify the development patterns of eye movement features and generate reliable new samples,providing new ideas for the combination of generative artificial intelligence and eye movement technology.关键词
生成式人工智能/变分自编码器/多层感知器/眼动Key words
generative artificial intelligence/variational autoencoder/multi-layer perceptron/eye movement分类
信息技术与安全科学引用本文复制引用
谭雪青,宋军,张慢慢,臧传丽..基于生成式人工智能的眼动样本生成及识别[J].河南理工大学学报（自然科学版）,2025,44(1):145-153,9.基金项目
国家自然科学基金资助项目(31800920) （31800920）
河南省教育科学规划重点课题资助项目(2025JKZD16) （2025JKZD16）
