传统基于空中目标特征状态推理作战意图的方法,需要大量的领域专家知识对特征状态的权重、先验概率等进行量化,明确特征状态与意图之间的对应关系,而神经网络可以在领域专家知识不足条件下,通过自身训练得到特征状态与意图之间的规则。...传统基于空中目标特征状态推理作战意图的方法,需要大量的领域专家知识对特征状态的权重、先验概率等进行量化,明确特征状态与意图之间的对应关系,而神经网络可以在领域专家知识不足条件下,通过自身训练得到特征状态与意图之间的规则。针对反向传播(BP)算法在更新网络节点权值时收敛速度慢、容易陷入局部最优的问题,通过引入ReLU(Rectified Linear Unit)激活函数和自适应矩估计(Adam)优化算法,设计了基于深度神经网络的作战意图识别模型,提高了模型收敛速度,有效地防止陷入局部最优。仿真结果表明,所提方法能够有效识别空中目标作战意图,获得更高的识别率。展开更多
Purpose:A social question & answer(SQA) community's long-term sustainability depends on its members' willingness to stay and contribute their knowledge continuously in the community.This research aims to i...Purpose:A social question & answer(SQA) community's long-term sustainability depends on its members' willingness to stay and contribute their knowledge continuously in the community.This research aims to investigate the critical factors which influence users' intention to continue contributing knowledge in the SQA community.Design/methodology/approach:Grounded on information systems(IS) continuance theory,this study put forward a model of the factors that influence SQA community members' intention to continue contributing knowledge.Survey was conducted to gather data from knowledge contributors of four major Chinese SQA communities(Baidu Knows,Sina iAsk,Soso Ask and Yahoo! Knowledge).By using the partial least squares(PLS) technique,research hypotheses derived from the proposed model were empirically validated.Findings:Except enjoyment in helping others and knowledge self-efficacy,all other factors including extrinsic reward,reputation enhancement,realization of self-worth,perceived usefulness,attitude towards knowledge contribution,and satisfaction exert significant impacts on users' continuance intentions in an SQA community.Research limitations:First,important factors such as the ease of use of information systems which may influence users' continuance intentions were not investigated in the study.Second,the study sample needs to be enlarged,and users of smaller SQA communities should also be included,to make the results more representative.Practical implications:This study will help SQA community designers and managers develop or improve incentive mechanisms to attract more people to contribute their knowledge and promote the development of the SQA community.Originality/value:This study improves the previous research models and puts forward a model of user continuance intention to contribute knowledge in an SQA community.It will extend the understanding of SQA community users' intention to continue contributing knowledge by distinguishing these users' different roles and focusing only on knowledge contributor展开更多
文摘传统基于空中目标特征状态推理作战意图的方法,需要大量的领域专家知识对特征状态的权重、先验概率等进行量化,明确特征状态与意图之间的对应关系,而神经网络可以在领域专家知识不足条件下,通过自身训练得到特征状态与意图之间的规则。针对反向传播(BP)算法在更新网络节点权值时收敛速度慢、容易陷入局部最优的问题,通过引入ReLU(Rectified Linear Unit)激活函数和自适应矩估计(Adam)优化算法,设计了基于深度神经网络的作战意图识别模型,提高了模型收敛速度,有效地防止陷入局部最优。仿真结果表明,所提方法能够有效识别空中目标作战意图,获得更高的识别率。
基金supported by Wuhan University Development Program for Researchers Born after the 1970s
文摘Purpose:A social question & answer(SQA) community's long-term sustainability depends on its members' willingness to stay and contribute their knowledge continuously in the community.This research aims to investigate the critical factors which influence users' intention to continue contributing knowledge in the SQA community.Design/methodology/approach:Grounded on information systems(IS) continuance theory,this study put forward a model of the factors that influence SQA community members' intention to continue contributing knowledge.Survey was conducted to gather data from knowledge contributors of four major Chinese SQA communities(Baidu Knows,Sina iAsk,Soso Ask and Yahoo! Knowledge).By using the partial least squares(PLS) technique,research hypotheses derived from the proposed model were empirically validated.Findings:Except enjoyment in helping others and knowledge self-efficacy,all other factors including extrinsic reward,reputation enhancement,realization of self-worth,perceived usefulness,attitude towards knowledge contribution,and satisfaction exert significant impacts on users' continuance intentions in an SQA community.Research limitations:First,important factors such as the ease of use of information systems which may influence users' continuance intentions were not investigated in the study.Second,the study sample needs to be enlarged,and users of smaller SQA communities should also be included,to make the results more representative.Practical implications:This study will help SQA community designers and managers develop or improve incentive mechanisms to attract more people to contribute their knowledge and promote the development of the SQA community.Originality/value:This study improves the previous research models and puts forward a model of user continuance intention to contribute knowledge in an SQA community.It will extend the understanding of SQA community users' intention to continue contributing knowledge by distinguishing these users' different roles and focusing only on knowledge contributor
