基于Agent技术的人性化E-learning系统研究

设计了以Agent技术为核心的人性化的E-learning系统。以Agent技术和人工心理理论为基础,构建了ISM多层级结构化Learning-Map,在此基础上实现个性化的Learning-Map。在人性化研究方面,运用基于图像处理的人脸检测、表情识别技术实现情感的认知,并构建情感认知模型,通过智能Agent助手对认知的情感进行智能处理。     

基于OCC的Agent情感模型研究

Agent不仅要具有逻辑推理能力,还应当具有类似人类的情感能力。通过对情感理论的分析,提出了一个基于OCC的Agent情感模型,使Agent能够模拟像人类一样的认知能力和情感能力,从而行为决策更加智能。通过模型在虚拟环境角色中的应用结果,验证了此模型的合理性。     

基于模糊认知图的情感Agent模型研究

人工情感是人工心理的一个主要研究内容。从研究人工情感出发,提出一种基于模糊认知图的情感Agent建模的方法。模糊认知图模型通过在传统认知图模型中引入模糊测度来量化概念间因果关系的影响程度。Agent的知识由内部组元的状态以及组元之间的关系权值进行描述,用简单数值运算代替了复杂的符号逻辑来实现Agent的智能推理和决策。通过实验表明,该模型设计简单、易于扩展、适用性好。     

用Java实现基于CORBA的Agent应用系统开发

由于Agent在适应软件开发需求方面的有效性,利用Agent技术开发实际应用系统引起了人们的高度重视,但是基于Agent系统的开发仍然缺乏开发环境和编程工具的支持。笔者在构建远程智能教育通信平台的过程中,尝试借鉴Agent思想解决分布式智能应用问题,采用Java语言实现了基于CORBA的多Agent系统监控模型(CAMM)。文章描述了采用Java多线程机制实现Agent生命周期控制、采用对象序列化机制实现移动Agent的思想方法和技术手段。     

一个情感驱动的3D虚拟角色行为动画创作工具

情感作为人类具有智能的一个重要体现,是创建丰富细腻的虚拟智能体的不可或缺的环节。本文结合认知对情感的影响,提出一个改进的情感建模方法和一种新的行为建模方法,设计具体的情感情境,并通过行为树的行为组织形式对虚拟环境中的行为进行管理,以产生具体、真实的虚拟角色情感行为。提出＂个性-情感-情绪＂的三层情感层次结构,并最后完成了一个实验系统。实验结果证明,该情感模型和行为组织模型能有效地反映虚拟角色在虚拟环境中的智能情感行为。     

基于多Agent的工作流引擎的研究

将多Agent技术引入分布式工作流引擎中,构造设计了基于多Agent的工作流引擎模型,并分别解释了引擎模型中模板解析Agent、流程管理Agent、任务分配Agent、活动执行Agent、资源Agent和监控管理Agent的具体功能。该模型利用智能Agent对动态复杂活动进行解析、选取最佳执行路径,解决了工作流引擎运行的瓶颈问题,改善了工作流引擎的运行效率,提高了企业的生产力和处理复杂问题的能力。     

基于Agent技术的机群智能构造器的设计与实现

提出了一种基于Agent技术的机群智能构造器CIB。通过一种系统构造机制,CIB实现了对机群系统的自动配置、部署和引导,从而构造出用户定制的机群系统。同时,它提供了遵循用户心智模型的GUI,减轻了用户的认知负担。文章概述了CIB系统提出的背景,分析了同类机群管理软件的不足之处,介绍了采用Agent机制解决问题的方法,描述了CIB的设计与实现,并从易用性和效率两个方面对系统进行了评测。     

信息系统授权管理Agent的设计与实现

本文提出了一种基于角色的授权管理模型和授权管理智能代理系统的实现方法。文中提出的授权管理模型在授权主客体的安全级别、角色结构及授权继承性、权限类别及授权约束机制等方面对RBAC进行了扩充,使之更完善。文中还介绍了授权管理Agent的功能、工作流程、知识库和方法库。     

多Agent专家系统安全性设计与研究

针对多Agent专家系统面临的安全威胁,提出了单个Agent的安全体系结构及多Agent专家系统安全架构模型。利用容器砂箱模型,实现了多Agent专家系统基于角色的访问控制。讨论了PKI在多Agent专家系统中的实际应用,在此基础上,研究了多Agent专家系统身份验证、安全通信、安全管理、安全审计及安全恢复机制。应用开发实践表明,系统具有资源占用少、安全性高、稳定性好以及可伸缩性强等优点,可应用于企业网智能安全管理、电子商务等领域。     

智能居住环境的Multi—Agent体系构架

智能居住环境系统能够监测环境的参数、设备的状态,并根据居住者的偏好进行主动控制.在ZigBee无线传感器网络的基础上,提出了一种智能居住环境Multi-Agent(MAS)体系结构.该体系结构中ZigBee无线传感器网络的协调器对应MAS中的管理Agent、路由器对应功能Agent、终端设备对应现场Agent;同时建立了现场Agent、功能Agent、管理Agent这三者的内部结构模型,为智能居住环境Multi-Agent系统中各Agent之间的协同合作、冲突消解、组织联盟提供了基础构架.     

认知视角下的文本情感计算

尝试将认知语用学和情绪心理学相关知识引入到文本情感计算中,以拉扎勒斯的认知一评价理论和认知语用学的认知语境为理论背景,提出一个新的文本情感认知模型。它从情感的发生机制出发,以多种情感图式为基础,考虑否定状态下情感的极性关系等多方面因素,提高了文本情感识别的准确率。情感认知模型是从一个新的视角研究情感识别问题,拓宽了研究的维度和理论背景,提供了新的研究思路。实验证明该模型是有效的。     

Attention aware systems: Theories,applications, and research agenda

Human perceptual and cognitive abilities are limited resources. Attention is the mechanism used to allocate such resources in the most effective way. Current technologies, in addition to allowing fast access to information and people, should be designed to support human attentional processes on which they impose further strain. This paper analyses the issues related to the design of systems capable of such support: attention aware systems. We introduce the research aimed at understanding and modelling human attentional processes, including perceptual and cognitive processes as studied in cognitive psychology, as well as rhetorical, aesthetic, and social aspects related to attentional mechanisms. We analyse current approaches to the design of attention aware systems along three major features: detection of user’s current attentional state, detection and evaluation of possible alternative attentional states, strategies for focus switch or maintenance. Finally, we discuss the most promising research direction for the development of systems capable of supporting human attentional mechanisms.     

Guiding Attention by Cooperative Cues

A common assumption in visual attention is based on the rationale of ＂limited capacity of information processing＂. From this view point there is little consideration of how different information channels or modules are cooperating because cells in processing stages are forced to compete for the limited resource. To examine the mechanism behind the cooperative behavior of information channels, a computational model of selective attention is implemented based on two hypotheses. Unlike the traditional view of visual attention, the cooperative behavior is assumed to be a dynamic integration process between the bottom-up and top-down information. Furthermore, top-down information is assumed to provide a contextual cue during selection process and to guide the attentional allocation among many bottom-up candidates. The result from a series of simulation with still and video images showed some interesting properties that could not be explained by the competitive aspect of selective attention alone.     

EDITORIAL FOR SPECIAL ISSUE ON INTELLIGENT METHODS FOR PERFORMANCE ENHANCEMENT IN INDUSTRIAL SYSTEMS VOLUME II: INTELLIGENT SYSTEMS DEVELOPMENT - TOOLS AND METHODOLOGIES

We present a preliminary definition and theory of artificial emotion viewed as a sequential process comprising the appraisal of the agent global state, the generation of an emotion-signal, and an emotion-response. This theory distinguishes cognitive from affective appraisal on an architecture-grounded basis. Affective appraisal is performed by the affective component of the architecture; cognitive appraisal is performed by its cognitive component. A scheme for emotion classification with seven dimensions is presented. Among them, we emphasize the roles played by emotions and the way these roles are fulfilled. It is shown how emotions are generated, represented, and used in the Salt &; Pepper architecture for autonomous agents (Botelho, 1997). Salt &; Pepper is a specific architecture comprising an affective engine, a cognitive and behavioral engine, and an interruption manager. Most properties of the cognitive and behavioral engine rely upon a hybrid associative, schema-based long-term memory. In Salt &; Pepper, emotion-signals, represented by label, object of appraisal, urgency, and valence, are generated by the affective engine through the appraisal of the agent's global state. For each emotion-signal there are several nodes stored and interconnected in long-term memory. Each of these nodes contains an emotion response that may be executed when an emotion-signal is generated. Emotion intensity relates to the activation of the node. It is shown that the Salt &; Pepper architecture for autonomous agents exhibits several properties usually related to emotion: state and mood congruence, compound emotions, autonomic emotion-responses, and different emotion-responses to the same stimulus including the generation of different motives. The implementation of a concrete example is described.     

Entering Aladdin's cave: Developing an app for children with Down syndrome

Tablets have much to offer children with learning difficulties, but evidence of their effectiveness to teach academic skills is limited and cannot be easily separated from the quality of the software. This paper analyses data from 3 iterative cycles of designing an app for children with Down syndrome to support their awareness of quantity through an inclusive game. Research with neurotypical children suggests that representation of quantity (or magnitude) is an area with considerable potential in supporting the foundations for children's mathematical learning. It has received little attention as an aspect for intervention for children with Down syndrome. Data collected in this study illustrate the need to carefully align the game mechanic to the target skills, strengthen levels of access, and introduce gradations of attentional demand. They also signal the interrelationship between children's cognitive and affective responses to the game, making it essential to find the optimal level of challenge. Children's strategies in response to mistakes indicate the importance of creating an agile responsive system. The data also suggest that developers routinely extend the number of features that are optional, enabling a greater level of personalization and a more inclusive game.     

Emotion recognition from physiological signals using wireless sensors for presence technologies

In this article we describe a new approach to enhance presence technologies. First, we discuss the strong relationship between cognitive processes and emotions and how human physiology is uniquely affected when experiencing each emotion. Secondly, we introduce our prototype multimodal affective user interface. In the remainder of the paper we describe the emotion elicitation experiment we designed and conducted and the algorithms we implemented to analyse the physiological signals associated with emotions. These algorithms can then be used to recognise the affective states of users from physiological data collected via non-invasive technologies. The affective intelligent user interfaces we plan to create will adapt to user affect dynamically in the current context, thus providing enhanced social presence.     

改进CNN的多通道语义合成情感分类模型研究

为了解决传统神经网络由于上下文语序变化而导致的情感歧义问题,提出一种多通道语义合成卷积神经网络（SFCNN）。使用改进的情感倾向注意力机制对词向量进行情感加权操作;将情感倾向词向量进行多通道语义合成,生成带有文本上下文语义信息的深度语义向量,构建情感分类模型;使用自适应学习率的梯度下降算法对模型参数进行优化,完成行情感分类任务。为了验证改进算法的有效性,使用多种微博数据样本集在提出的模型上进行对比实验。实验结果表明,改进的情感倾向注意力机制结合多通道语义合成卷积神经网络具有较好的情感分类能力,并且自适应学习率的梯度下降算法可以更快地完成模型收敛工作。     

Using eye movement data to infer human behavioral intentions

Behavior-directed intentions can be revealed by certain biological signals that precede behaviors. This study used eye movement data to infer human behavioral intentions. Participants were asked to view pictures while operating under different intentions, which necessitated cognitive search and affective appraisal. Intentions regarding the pictures were non-specific or specific, specific intentions were cognitive or affective, and affective intentions were to evaluate either the positive or negative emotions expressed by the individuals depicted. The affective task group made more fixations and had a larger average pupil size than the cognitive task group. The positive appreciation group made more and shorter fixations, on average, than the negative appreciation group. However, support vector machine algorithms revealed low classification accuracy. This was due to large inter-individual variance and psychological factors underlying intentions. We demonstrated improvement in classification accuracy using individual repeated measures data, which helped infer participants’ self-selected intentions.     

基于上下文感知的方面类别情感分类

由于一个评论往往会涉及多种方面类别及情感倾向,而传统注意力机制难以区分方面词和情感词的对应关系,从而影响评论同时存在多种方面类别时的情感极性分析.为了解决上述问题,提出了一种基于上下文感知的方面类别情感分类模型(MA-DSA).该模型通过重构方面向量捕获句子中更多样且有效的语义特征,并将其融入上下文向量,然后将上下文向量通过DiSA模块进一步捕捉句子内部情感特征,确定方面词与情感词的关系,进而对指定方面类别进行情感分类.在SemEval的三个数据集上的实验结果表明,MA-DSA模型在Restaurant-2014数据集上的三个指标值均优于基准模型,证明了该模型的有效性.     

基于ConvLSTM网络的维度情感识别模型研究

学业情绪能够影响和调节学习者的注意、记忆、思维等认知活动,情绪自动识别是智慧学习环境中情感交互和教学决策的基础。目前情绪识别研究主要集中在离散情绪的识别,其在时间轴上是非连续的,无法精准刻画学生学业情绪演变过程,为解决这个问题,基于众包方法建立真实在线学习情境中的中学生学习维度情感数据集,设计基于连续维度情感预测的深度学习分析模型。实验中根据学生学习风格确定触发学生学业情绪的学习材料,并招募32位实验人员进行自主在线学习,实时采集被试面部图像,获取157个学生学业情绪视频;对每个视频进行情感Arousal和Valence二维化,建立包含2 178张学生面部表情的维度数据库;建立基于ConvLSTM网络的维度情感模型,并在面向中学生的维度情感数据库上进行实验,得到一致性相关系数（Concordance Correlation Coefficient,CCC）均值为0.581,同时在Aff-Wild公开数据集上进行实验,得到的一致相关系数均值为0.222。实验表明,提出的基于维度情感模型在Aff-Wild公开数据集维度情绪识别中CCC相关度系数指标提升了7.6%~43.0%。