理性Agent的意图维护模型

在理性Agent的形式化中,通常认为Agent的思维状态包含信念、愿望帮意图这3个属性,它们之间的静态关系已被人们充分重视,但它们随时间的动态变化还未被很好研究,考察了意图维护直观语义中的一些问题,给出了随信念、愿望改变的意图维护的动态模型,与Rao和Georgeff的意图维护模型相比,适用性更广、更简单,不必限制可友爱世界的时序结构,不必另外引入信念=愿望、意图算子的on ly模态形式,并且对愿望和意图不存在逻辑全知问题和重言隐含下的副作用等问题。     

基于作者关联和概念网的科技论文搜索方法研究

针对论文检索的基本要求，详细分析了论文学科领域与论文作者信息间的关系，建立了论文作者关联概念网模型，提出了基于概念网（CN）和作者关联网络（ARN）结合的Internet科技论文索引机制，并提出了根据查询词集分析搜索者意图的信息提交层次策略，以及作者关联网络的自维护技术路线，从而提高搜索引擎的智能化水平和服务质量。     

虚拟维修仿真中手势识别的研究与应用

虚拟维修训练仿真要求良好的逼真性和沉浸感,对训练操作者的动作意图能高效地转化.将手势识别技术应用于虚拟维修仿真作为的获取可以有效满足这种要求.通过装备维修动作的建模研究分析定义了能够体现动作意图的必需手势;之后采用模糊神经网络技术实现对手势的识别并建立基于手势识别的虚拟手交互控制机制.还重点对神经网络的构建以及学习算法进行了介绍,并在现有适用度计算方法的基础上结合网络实际提出了适用度的平均求解法.最后将基于手势识别的虚拟手交互控制方法应用于具体维修训练仿真中并证明了方法的可行性和高效性.     

面向人机对话意图分类的混合神经网络模型

随着人机对话的不断发展,让计算机能够准确地理解用户查询意图,对整个人机对话领域都有着重要意义.意图分类的主要目标是在人机对话的过程中判断用户的意图,提升人机对话系统的准确度与自然度.首先分析多个分类模型在意图分类任务上的优缺点.在此基础上,提出一种混合神经网络模型,综合利用多个深度网络模型的多样性输出.在输入特征预处理上,采用语言模型词向量,将语言模型拥有的语义挖掘能力应用到混合网络中,可以进一步提升模型的表达能力.所提出的混合神经网络模型相对于最好的基准模型在两份数据集上分别取得了2.95%和3.85%的性能提升.新模型在该数据上取得了最优的性能.     

BP神经网络在维修保障系统能力评估中的应用

维修保障系统能力的高低已逐渐成为影响战争胜负的重要因素,对其能力的评估是实现维修保障目标、形成战斗力的重要和有效的辅助决策手段.该文首先介绍了BP神经网络的基本原理和方法,在系统分析维修保障系统能力各种影响因素的基础上,构建了维修保障系统能力评估指标体系.接着详细设计了用于评估维修保障系统能力的BP神经网络模型,并利用MATLAB仿真软件对结果进行了计算和分析.结果表明BP神经网络具有很强的解决复杂非线性关系问题的特点,适用于对维修保障系统等复杂系统的分析和评价.     

农产品网上交易系统的设计及应用

以互联网为核心的新技术革命正促使传统观念发生深刻的变革,农产品网上交易系统从农产品经销企业的实际出发,实现了信息发布、合同签订、仓库管理、财务管理及数据的查询与维护等功能,系统的应用将有利于加强产品信息交流、提高企业整体管理水平和核心竞争力。     

Does social network always promote entrepreneurial intentions? An empirical study in China

In order to verify how social network affects entrepreneurial intentions, a theoretical model is developed between network, entrepreneurial desirability, entrepreneurial feasibility and entrepreneurial intentions. As for empirical studies, a structural equation model is developed and several hypotheses are tested based on Chinese College-graduate Village Officials’ samples. Results showed that three social network dimensions, which are network size, network heterogeneity and properties of top node, affect entrepreneurial intentions significantly. Specifically, entrepreneurial intention is negatively correlated with properties of top node and the size of the social network; significant positive correlation is found between heterogeneity of social network and entrepreneurial intentions. Besides, entrepreneurial desirability and entrepreneurial feasibility are two mediators between network heterogeneity and entrepreneurial intension. This paper has also found that social networks sometimes may become an obstacle for College-graduate Village Official to start a business. The theoretical and practical implications for policymaker of the study’s findings are also discussed.     

A hybrid agent architecture integrating desire,intention and reinforcement learning

This paper presents a hybrid agent architecture that integrates the behaviours of BDI agents, specifically desire and intention, with a neural network based reinforcement learner known as Temporal Difference-Fusion Architecture for Learning and COgNition (TD-FALCON). With the explicit maintenance of goals, the agent performs reinforcement learning with the awareness of its objectives instead of relying on external reinforcement signals. More importantly, the intention module equips the hybrid architecture with deliberative planning capabilities, enabling the agent to purposefully maintain an agenda of actions to perform and reducing the need of constantly sensing the environment. Through reinforcement learning, plans can also be learned and evaluated without the rigidity of user-defined plans as used in traditional BDI systems. For intention and reinforcement learning to work cooperatively, two strategies are presented for combining the intention module and the reactive learning module for decision making in a real time environment. Our case study based on a minefield navigation domain investigates how the desire and intention modules may cooperatively enhance the capability of a pure reinforcement learner. The empirical results show that the hybrid architecture is able to learn plans efficiently and tap both intentional and reactive action execution to yield a robust performance.     

Influence of Expectation Confirmation,Network Externalities,and Flow on Use of Mobile Shopping Apps

ABSTRACT

The purpose of the current research is to examine the influence of expectation confirmation, network externalities, and flow on word-of-mouth intention and continued use of mobile shopping apps. A structured online survey questionnaire is used to collect data from 363 users of mobile shopping apps. Structural equation modeling is used to analyze the research model. The findings reveal that indirect network externalities, i.e., perceived complementarity, influence perceived usefulness of the mobile shopping app. Users’ confirmation of expectations significantly influences perceived usefulness, satisfaction, and continuance intention to use mobile apps. Satisfaction is found to be a significant predictor of continuance intention and word-of-mouth intention. Flow influences satisfaction of users, perceived usefulness, and continuance intention. Word-of-mouth intention is found to be an important post-adoption behavioral outcome. The results provide valuable theoretical insights for academics and managerial implications for providers of mobile shopping apps.     

Identification of human implicit visual search intention based on eye movement and pupillary analysis

We propose a novel approach for the identification of human implicit visual search intention based on eye movement patterns and pupillary analysis, in general, as well as pupil size, gradient of pupil size variation, fixation length and fixation count corresponding to areas of interest, and fixation count corresponding to non-areas of interest, in particular. The proposed model identifies human implicit visual search intention as task-free visual browsing or task-oriented visual search. Task-oriented visual search is further identified as task-oriented visual search intent generation, task-oriented visual search intent maintenance, or task-oriented visual search intent disappearance. During a visual search, measurement of the pupillary response is greatly influenced by external factors such the intensity and size of the visual stimulus. To alleviate the effects of external factors, we propose a robust baseline model that can accurately measure the pupillary response. Graphical representation of the measured parameter values shows significant differences among the different intent conditions, which can then be used as features for identification. By using the eye movement patterns and pupillary analysis, we can detect the transitions between different implicit intentions—task-free visual browsing intent to task-oriented visual search intent and task-oriented visual search intent maintenance to task-oriented visual search intent disappearance—using a hierarchical support vector machine. In the proposed model, the hierarchical support vector machine is able to identify the transitions between different intent conditions with greater than 90 % accuracy.     

集成状态监测和故障诊断的设备管理平台的建立

针对目前设备状态维修中故障诊断和维修管理相互脱节的问题，在状态维修思想的指导下，构建了实施状态维修的设备管理平台，给出了系统平台的集成架构。结合发电企业设备维修的实际，提出了状态维修实施的信息模型。运用IDEF方法对整个设备管理平台的总体功能进行了详细分析，建立了系统的功能模型。在构建系统模型的基础上，综合采用网络技术、计算机技术和数据库技术等开发出了集成状态监测和故障诊断的设备管理系统，为企业成功开展状态维修提供了有效的管理平台。     

改进高维数据相似度的目标意图识别方法

对于实际战场中目标属性要素呈现出的多样化,传统目标意图识别方法不能够较全面地建立属性之间的相似度模型.为了更好地阐述实际战场的复杂情况,提高目标意图识别的准确度,提出了一种利用改进的空间相似度与属性相似度融合的高维数据相似度模型,以全面地计算目标各种属性状态对态势意图的支持程度,再利用得到的高维数据相似度通过D-S证据理论对目标进行序贯识别.仿真实验表明:该方法具有有效性以及能够提高目标意图识别的准确度,为解决目标战术意图识别提供了新的方法.     

Factors influencing continuance intention to use social network sites: The Facebook case

Social networking sites (SNS) have become one of the most popular means of online social interaction. However, few studies have examined SNS from the context of post-adoption. This study focuses on continuance usage intention toward Facebook, a popular SNS, among university students in the United Arab Emirates. We have extended the expectation–confirmation model by investigating the influence of enjoyment and subjective norms as critical factors that directly influence continuance usage intention, and added habit as a mediator between satisfaction and continuance intention. We used structural equation modeling to validate the proposed model and hypotheses. We find that perceived usefulness, satisfaction, habit, enjoyment, and subjective norms explain 54.8% of the variance in continuance intention. Furthermore, our study shows that the effect of satisfaction on continuance intention is both direct and mediated by habit. Thus, practitioners should focus on all of these factors to encourage continuance usage intention in their site’s users.     

Interplay between trust,information privacy concerns and behavioural intention of users on online social networks

Trust and privacy features of websites have evolved as an important concern for any businesses or interactions, particularly in online networks. The study investigates the relationship between trust, privacy concerns and behavioural intention of users on the social network. The behavioural intention of users on the online social network (OSN) is captured by intention to disclose information and intention to interact with others in OSN. The study was conducted on a sample of 457 active users from one of the major social networking website, Facebook. Partial least squares based structural equation modelling was used for analysing the results. The findings of the study reveal that intention to disclose information mediates the relationship between trust in the website and the intention to interact with others. Another important finding of the study indicates that prior positive experience with the website significantly impacts the trust in website, and the trust in website also plays a crucial role while determining the information privacy concerns in the OSN.     

基于Faster R-CNN的诱导维修自动交互设计

随着增强现实技术在机械领域的发展，已经有越来越多的例子证明了AR在工业维修方面提高操作效率的优越性。为了提高诱导维修操作过程的效率，针对传统的增强现实维修系统不能对维修状态进行感知和判断的问题，提出了一种基于快速区域卷积神经网络（Faster R-CNN）的进程识别自动交互方法。该方法基于Faster R-CNN建立零件识别的深度神经网络模型并利用反向传播进一步微调，通过对零件的识别输出零件的类型和编号，反馈给系统触发相应的操作步骤，无需用户进行另外的交互操作。实验结果表明，基于深度神经网络的维修零件识别率可达95%，平均识别速度为每帧300 ms，满足AR诱导维修系统的精度和交互性要求。     

基于对手思维建模的分布式入侵检测模型

研究网络入侵和入侵检测系统的现状和发展趋势,将对手思维建模和意图识别技术引入入侵检测系统,提出了一个基于对手思维建模的入侵检测模型(IRAIDS),为解决大规模、分布式、智能化入侵提供了解决方法.     

Degradation process prediction for rotational machinery based on hybrid intelligent model

Bearings are among the most critical and precise components in rotational machinery. The condition and health of bearings play an important role in the functionality and performance of rotational machinery. Since a neural network ensemble approach shows significantly improved generalization performance and outperforms those of a single neural network, one novel selective neural network ensemble model is developed for bearing degradation process prediction. An improved particle swarm optimization with simulated annealing is proposed to select the optimal subset formed by accurate and diverse networks and obtain a better ability to escape from the local optimum. An experimental setup to perform fatigue testing on ball bearings and several simulations are explored in order to validate the developed prediction model. Experimental results show that degradation process prediction based on the explored selective neural network ensemble model provides a means of enhancing the monitoring of ball bearings' condition, and the results of this model are superior in comparison with the results of a single neural network. This selective neural network ensemble model can be used as one excellent predictive maintenance tool in plants.     

面向配电网数据的命名实体识别

在电力系统中, 配电调度是一个复杂且统筹性较强的工作, 大多依赖于工作人员的经验和主观判断, 极易出现纰漏. 所以急需利用智能化手段来帮助检修计划的分析与生成. 命名实体识别是构建配电网知识图谱以及问答系统等任务的关键技术, 它能够将非结构化数据中的命名实体识别出来. 针对配电检修数据的复杂性及强关联性等特点, 本文采用BERT-IDCNN-BiLSTM-CRF深度学习模型. 该模型相较于传统的BERT-BiLSTM-CRF模型, 融入IDCNN神经网络模型, 更好地利用GPU的性能, 在保证识别准确率的前提下, 提高效率. 通过对标注好的检修计划数据进行训练, 并与其他常用模型对比, 在召回率、精确率以及F1值3个指标上, 本文提出的模型均达到最优的效果, F1值可以达到83.1%, 该模型在配电网数据识别任务上取得了很好的效果.     

基于泛终端边缘计算的智能图像电力监控系统网络安全防护方式的采用和研究

图像识别、边缘计算和AI技术的不断发展为电力监控方式方法提供了更便捷的手段和更多的迭代数据资料,使电力监控系统更便捷、更高效地迈向数字化发展。但日益增多的设备、计算方式的改变和更大的数据量带来了更多的网络安全漏洞节点和网络安全风险。本文就基于泛终端边缘计算的智能图像电力监控系统网络安全防护方式的采取做了具体的探讨,意图采用更贴近实际的防护手段,为电力智能图像监控系统提供安全稳定的运行保障。     

Modeling hedonic is continuance through the uses and gratifications theory: An empirical study in online games

It has long been a challenge for online game providers that online game players frequently switch to alternative games without much hesitation. Current IS continuance theories are mainly developed to interpret user continuance of general utilitarian IS and are ineffective in interpreting user continuance of hedonic IS. In this study based on the uses and gratifications theory, a hedonic IS continuance model is developed by incorporating three types of gratification: hedonic gratification (enjoyment, fantasy and escapism); social gratification (social interaction and social presence); and utilitarian gratification (achievement and self-presentation). Age and gender are the moderating factors in the model. The research model is empirically assessed based on 3919 validated responses from the users of a social network game in China. In this study we found that three types of gratification affect an individual’s continuance intention to use a social network game: hedonic gratification (enjoyment, fantasy and escapism), utilitarian gratification (achievement) and social gratification (social interaction and social presence). The results provide weak support for the moderating effect of gender on the relationship between each antecedent and continuance intention, but offer strong support for the moderating effect of age on the relationships.