脑视觉认知机制下意象-产品归类的集群用户偏好推理  

Cluster User Preference Inference for Image-product Categorization under Brain Visual Cognitive Mechanism

作　　者：任丽 林丽[1] 阳明庆[1] 郭主恩 REN Li;LIN Li;YANG Mingqing;GUO Zhuen(School of Mechanical Engineering,Guizhou University,Guiyang 550025)

机构地区：[1]贵州大学机械工程学院,贵阳550025

出　　处：《机械工程学报》2023年第15期282-292,共11页Journal of Mechanical Engineering

基　　金：国家自然科学基金(51865003);贵州省科技计划(黔科合基础-ZK[2021]重点055);黔科合平台人才计划([2018]5781);贵州大学培育(贵大培育[2019]06)资助项目。

摘　　要：精准获取用户客观需求是智能个性定制亟待解决的问题之一,传统方法明确用户认知偏好时存在多属性用户群体需求错位、对设计师经验依赖性强的问题,对此提出脑视觉认知机制下意象-产品归类的集群用户偏好推理方法。运用统计学方法甄选样本域及意象域,降低了人工干预;基于K-modes聚类用户认知特征元,构建了相似用户集群;采用K近邻算法智能化识别被试所属集群,并开展眼脑融合的主客观集成化认知实验,提取了集群中的关键主客观认知指标;最后运用逼近理想点排序法归类集群用户对应的意象-产品群,基于此归类推理出了新相似用户的认知偏好。以无人机用户认知偏好推理为例,建立了脑视觉认知机制下集群用户的意象-无人机归类,快捷推理新用户的需求案例及认知偏好,结果表明,运用现有集群用户的生理认知数据明确新用户的客观偏好的可行性和有效性,可提升生理实验的再利用率,为群体智能定制模式的个性设计提供理性客观视角。Accurate acquisition of users' objective needs is one of the urgent problems to be solved in intelligent personality customization.In traditional acquisition,there are problems such as misplacement of multi-attribute user groups' needs and strong dependence on experience designers.Therefore,a cluster user preference inference method based on image-product classification under the mechanism of brain visual cognition is proposed.Firstly,the sample domain and image domain are selected by statistical methods to reduce manual intervention;Then,similar user clusters are constructed based on k-modes clustering user cognitive feature elements;Intelligently identify the clusters of subjects through the K-nearest neighbor(KNN) algorithm,and carry out eye movements combined with electroencephalogram(EEG) experiments to extract key subjective and objective cognitive indicators in the clusters;Finally,technique for order preference by similarity to an ideal solution intelligence is used to classify the corresponding image-product group of cluster users and deduce the cognitive preferences of new similar users.Taking unmanned aerial vehicle(UAV) users as an example,the image-UAV classification of cluster users was established under the mechanism of brain visual cognition,and the demand preference of new users is quickly deduced.The results showed that it met the needs of users,and verified the effectiveness of the method.This study using the existing cluster the user's physiological and cognitive objective preference data to predict new users,improve the utilization rate of physical experiment,provide custom pattern swarm intelligence personality design rational objective perspective.

关 键 词：个性化需求 意象认知 集群理论 潜在偏好 

分 类 号：TH166[机械工程—机械制造及自动化]