人工智能情感认知推理的计算模型分析  

The Analysis of Computational Model for Emotion Cognitive Reasoning

作　　者：樊岳红[1] FAN Yuehong

机构地区：[1]山西大学哲学社会学学院,山西太原030006

出　　处：《上海师范大学学报（哲学社会科学版）》2020年第2期94-103,共10页Journal of Shanghai Normal University(Philosophy & Social Sciences Edition)

基　　金：国家社科后期资助项目“维特根斯坦教育哲学思想研究”(19FZXB037);山西省软科学项目(2018041050-5)。

摘　　要：对于观察者如何理解和推理他人的心理状态,当前研究范式比较少地使用计算模型。现今认知科学已经着手使用计算建模的方法来对他人的情感状态进行解释和推论。以此为背景,文章基于情感直观理论来研究人工智能情感认知的计算模型,并使用形式化的贝叶斯规则来分析情感认知的推理分类。作者从“情感认知”“第三人称评价”“情感原因的推断”“情感线索的整合”“反向评估”“对假设推理的预测”“反事实推理及解释”7个维度具体阐明了人工智能情感认知的推理分类,并进行计算建模,从而为人工智能情感认知提供一种统一的解释模型。The current research paradigm makes relatively little use of computational models regarding the observers’understanding and reasoning about the mental state of others.Research on cognitive science has fruitfully applied computational modeling approaches to explain how observers understand and reason about others’mental states.This paper explores emotion cognition based on the framework of emotional intuitive theory,and discusses the emotional cognitive reasoning’s classification by the method of formal Bayesian rule.This paper also classifies emotional recognition and builds the computation model by 7 dimensions,namely,emotion recognition,third-person appraisals,inferring causes of emotions,emotional cue integration,reverse appraisal,predictions for hypothetical reasoning,counterfactual reasoning and explanations.Accordingly,the study can provide a uniform calculation explanative model for emotional cognition.

关 键 词：人工智能 情感计算 计算模型 情感的直观理论 贝叶斯规则 心灵理论 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]