基于记忆效应的协同过滤推荐算法

针对传统协同过滤算法无法及时反映用户兴趣变化的情况,将人脑的记忆和遗忘特性引入到个性化推荐中,提出基于记忆效应的协同过滤推荐算法。利用短时记忆体现用户近期兴趣变化,应用长时记忆强调用户早期兴趣的重要性,给出将短时记忆和长时记忆相结合的调和记忆,使推荐系统可以自适应地跟踪用户兴趣变化。实验结果表明,与CF算法、SCF算法和AUICF算法相比,该算法的推荐精度更高、收敛速度更快。     

集音节、梵文和词汇于一体的藏文智能输入系统*

研究和设计了以音节、词汇为输入单位的基于藏文拉丁转写的编码输入方案,设计了音节和词汇基字编码快速输入法,实现了集音节、梵文和词汇输入于一体的藏文智能输入系统,从而解决了藏文输入长期以来没有快速输入或智能输入法的问题。实验表明其新组词和智能词汇记忆使藏文输入更加快捷和高效。     

基于记忆的人工鱼认知模型

为丰富虚拟海洋环境中人工鱼的认知能力，以自然鱼的生物原理和记忆机制为理论依据，提出了一种基于记忆的人工鱼认知模型。信息在瞬时记忆、短时记忆和长时记忆3个阶段传输和存储，人工鱼通过聚焦器提取所关注的感知信息，通过决策器和短时记忆进行行为选择，将经验知识存储在长时记忆中，被短时记忆调用以优化当前决策。动画结果表明，采用此模型后人工鱼能产生记忆指导下的行为，表现出更真实更智能的生命特征。     

基于记忆原理的推荐系统托攻击检测模型

提出一种基于记忆原理的推荐系统托攻击检测模型。利用短时记忆元和长时记忆元所描述的记忆增强和衰减规律,以及这2种记忆元与综合记忆元的联系,对托攻击进行检测。该模型的特征记忆库可及时更新,由此节省系统开销。实验结果证明,基于该模型的推荐系统具有较高的托攻击检测正确率。     

基于NS-BML模型的记忆密度在交通信号灯控制系统中的研究

交通信号灯管理与控制直接影响着交通网络中的运行效率。NS-BML模型广泛应用于交通信号灯控制系统仿真,针对目前NS-BML模型中只考虑现在瞬时密度而忽略历史密度的问题,提出记忆密度策略,从长时记忆密度策略和短时记忆密度策略两个角度来分析该策略对曼哈顿式网络的影响,通过对时间离散化,求解短时记忆密度的最优比例因子。仿真结果表明,通过采用所提出的短时记忆密度策略可以有效提高系统的运行效率,同时保证计算机处理的速度,交通网络的平均速度和到达率分别同比增长8.51%和9.28%,说明了该策略的有效性。     

基于记忆原理的人工鱼智能认知模型

将生物学的记忆原理引入到计算机动画中,提出了一种基于记忆原理的人工鱼智能认知模型。在该模型中,利用短时记忆元和长时记忆衰减和更新的特点及其与综合记忆的联系,建立了相应的数学模型,综合记忆值的大小决定了人工鱼对该事物的记忆强度,特征记忆库的更新清理功能有效的节省了系统的开销。实验结果证明,该模型在一定程度上提高了人工鱼的认知能力,具有很好的应用价值。     

基于生物记忆原理的智能伸展树模型

针对普通伸展树节点移动方式单调、灵活性和实时性差的缺点,提出一种基于生物记忆原理的智能伸展树模型。该模型利用短时记忆元不断衰减和更新的特点及短时记忆元与长时记忆、综合记忆的关系,建立相应的数学模型,借助节点访问度反映节点向前移动的步幅大小。实验结果表明,智能伸展树具有良好的实时性和动态性,具有一定的应用价值。     

基于融合式神经网络的微生物生长环境关系抽取

为了构建完整的微生物生长环境关系数据库,提出基于卷积神经网络-长短时记忆(CNN-LSTM)的关系抽取系统.结合卷积神经网络(CNN)和长短时记忆(LSTM),实现对隐含特征的深度学习,提取分布式词向量特征和实体位置特征作为模型的特征输入.对比实验验证加入特征后CNN-LSTM模型的优势,并将CNN模型的特征输出作为LSTM模型的特征输入.在Bio-NLP 2016共享任务发布的BB-event语料集上得到目前最好的结果.     

基于生物记忆原理的入侵检测模型

利用生物记忆原理中的记忆存储、更新与遗忘原理,建立一种基于生物记忆原理的入侵检测模型。在本模型中,利用瞬时记忆衰减更新速度快且对外界信息反应敏感的特点,对异常数据进行及时检测,尽可能较早地阻止入侵的发生;短时记忆和长时记忆可以随着异常访问频度调整记忆强度并进行相互转化,以达到最佳的记忆效果。短时记忆容量限制和记忆库自动更新清理机制能有效地节省入侵检测系统资源消耗,将更多空间用于存储重要信息。实践检验发现,该模型能实时追踪最新动态,借助记忆库对旧信息进行选择性更新和遗忘,并对未知入侵行为做出及时、高效、准确的判断。     

基于IRT的英语辅助学习系统设计与实现

为开发支持有效英语学习的现代辅助学习形式或工具提出了一种基于IRT和词汇学习记忆周期的个性化移动英语词汇学习系统,系统包括远程管理服务器、客户端移动学习系统和数据同步智能体等3部分.该系统基于个体的词汇能力和学习记忆周期,根据每个单词的难度和个体学习者的词汇能力,自适应地为个体学习者安排词汇学习.通过仿真分析,所提出的...     

基于生物记忆原理的个性化搜索引擎模型

提出一种基于记忆原理的个性化搜索引擎模型,根据记忆原理的基本模型与记忆元的转化关系生成相应的记忆模块,将该模块应用到搜索引擎模型中。建立了记忆模块页面记忆的数学更新模型并进行编程实现。鉴于系统存储容量的限制及一些访问过的页面利用率低的情况,对记忆存储库进行实时的更新。跳跃步数的设定使得该模型更具有灵活性,可根据用户需求设置跳跃步数。实验证明该模型能满足用户个性化的需求,提高搜索效率。     

SELF-STABILIZING DISTRIBUTED SORTING IN TREE NETWORKS

This paper presents a self-stabilizing distributed sorting algorithm for tree networks. The distributed sorting problem can be informally described as follows: Nodes cooperate to reach a global configuration where every node, depending on its identifier, is assigned a specific final value taken from a set of input values distributed across all nodes. The input values may change in time. In our solution, the system reaches its final configuration in a finite time after the input values are stable and the faults cease. The fault-tolerance and the adaptivity to changing input is achieved using Dijkstra's paradigm of self-stabilization. A self-stabilizing algorithm, regardless of the initial system state, will converge in finite time to a set of legitimate states without the need for explicit exception handlers or backward recovery. Our solution is based on a continuous broadcast with acknowledgment along the tree edges to achieve the synchronization among processes in the system. It has 0(n ×h) time complexity and only 0(log(n) × ) memory requirement where h is the degree of the tree and h is the height of the tree.     

Memory protection in embedded systems

With reference to an embedded system featuring no support for memory management, we present a model of a protection system based on passwords. At the hardware level, our model takes advantage of a memory protection unit (MPU) interposed between the processor and the complex of the main memory and the input-output devices. The MPU supports both concepts of a protection context and a protection domain. A protection context is a set of access rights for the memory pages; a protection domain is a set of one or more protection contexts. Passwords are associated with protection domains. A process that holds a given password can take advantage of this password to activate the corresponding domain. A small set of protection primitives makes it possible to modify the composition of the domains in a strictly controlled fashion.The proposed protection model is evaluated from a number of important viewpoints, which include password distribution, review and revocation, the memory requirements for storage of the information concerning protection, and the time necessary for password validation.     

Optimal scheduling of homogeneous job systems

Homogeneous job systems are systems in which all of a finite set of jobs to be processed by the system have exactly the same processing requirements. This paper assumes that each job first executes an input task requiring an input unit (channel or controller) for some amount of time T_c along with a memory unit. Then it executes a computational task requiring a processing unit and the memory for some amount of time T_p. Under these assumptions, it is possible to derive some inequalities concerning the relative number of memory, input, and processor units which can be efficiently used by the system as a function of T_c and T_p. The scheduling problem is to order tasks and assign resources to them in such a way as to minimize some cost function. The cost functions considered in this paper are job set finishing time and dwell time. Some theorems are stated and proved which yield closed form expressions for the minimum finishing time in batch and in time-shared systems as a function of the number of jobs, memories, processors, input units, and T_c and T_p. The purpose of this study is to derive some general results which aid in the efficient utilization of multiprocessor computer systems. Although this study is directed toward a specific type of homogeneous system, it is shown that the results are applicable to other systems (e.g., systems with output).     

单元自动化测试中类的抽象内存模型研究

由于面向对象程序具有多态性等复杂特性,在软件单元测试中仅凭静态分析难以判断指针和引用指向对象的具体类型,为了解决这一问题,对类的抽象内存模型进行研究,并提出类的操作语义模拟算法;在路径分析时,通过构建和更新抽象内存模型,从而对变量所属类的范围进行限定;对于单元测试,对基于输入域的随机测试进行优化,提出基于路径的随机测试方法,得到输入变量的类型集合;实验表明,类的抽象内存模型结合操作语义模拟算法能够有效提取出路径中类相关的约束,基于路径的随机测试方法比起基于输入域的随机测试方法能够明显提高测试效率。     

非线性状态空间模型遗忘性研究

动态系统的遗忘性是描述系统行为主要取决于其近期输入的概念，与系统的稳定性、鲁棒性等密切相关。对于非线性状态空间模型，遗忘性在很弱的条件下与整体一致非常吸引等价。系统在凸集或不变区集上的遗忘性可由一类特殊Ｌｙａｐｕｎｏｖ函数的存在性保证。遗忘区域的大小可利用矩阵不等式来估计。     

Scene analysis by integrating primitive segmentation andassociative memory

Scene analysis is a major aspect of perception and continues to challenge machine perception. This paper addresses the scene-analysis problem by integrating a primitive segmentation stage with a model of associative memory. The model is a multistage system that consists of an initial primitive segmentation stage, a multimodule associative memory, and a short-term memory (STM) layer. Primitive segmentation is performed by a locally excitatory globally inhibitory oscillator network (LEGION), which segments the input scene into multiple parts that correspond to groups of synchronous oscillations. Each segment triggers memory recall and multiple recalled patterns then interact with one another in the STM layer. The STM layer projects to the LEGION network, giving rise to memory-based grouping and segmentation. The system achieves scene analysis entirely in phase space, which provides a unifying mechanism for both bottom-up analysis and top-down analysis. The model is evaluated with a systematic set of three-dimensional (3-D) line drawing objects, which are arranged in an arbitrary fashion to compose input scenes that allow object occlusion. Memory-based organization is responsible for a significant improvement in performance. A number of issues are discussed, including input-anchored alignment, top-down organization, and the role of STM in producing context sensitivity of memory recall.     

基于粗糙集与BP神经网络的多因素预测模型

运用粗糙集方法和信息熵概念，在不改变训练样本分类质量的条件下，按照输入影响因素相对于输出的重要度的大小，对输入参数集进行约简，确定神经网络输入层变量和神经元个数。通过对典型样本的学习，建立粗糙集BP神经网络多因素预测模型，将其用于导弹系统研制费用预测。结果表明，该方法减少了网络的训练时间，改善了学习效率，具有较高的预测精度，是可行的、有效的。     

改进果蝇算法优化CIAO-LSTM网络的时序预测模型

为了提高时间序列的预测精度,提出了一种基于改进果蝇算法优化直连长短期记忆网络的时间序列预测方法。将长短期记忆网络的多个时间步输入与输出进行全连接(CIAO-LSTM,直连长短期记忆网络),增强了对目标系统中线性成分的表征。提出了一种改进的果蝇优化算法(IFOA),通过动态改变果蝇的搜索半径和对适应度函数增加逃脱系数,提高了果蝇优化算法的全局寻优能力和局部收敛速度。使用IFOA优化CIAO-LSTM网络参数并构建预测模型(IFOA_CIAO-LSTM)。实验结果表明,优化后的时序预测方法相比传统的长短期记忆网络泛化能力更强、预测精度更高,对于波动较大的数据可以实现更好的拟合。     

Model and controller selection policies based on output predictionerrors

Based on observations of the past inputs and outputs of an unknown system Σ, a countable set of predictors, O_p, p∈P, is used to predict the system output sequence. Using performance measures derived from the resultant prediction errors, a decision rule is to be designed to select a p∈P at each time κ. We study the structure and memory requirements of decision rules that converge to some q∈P such that the qth prediction error sequence has desirable properties. In a very general setting we give a positive result that there exist stationary derision rules with countable memory that converge to a “good” predictor. These decision rules are robust in a sense made precise in the paper. In addition, we demonstrate that there does not exist a decision rule with finite memory that has this property. Based on the decision rule's selection at time κ, a controller for the system Σ is chosen from a family Γ_p ∈P of predesigned control systems. We show that for certain multi-input/multi-output linear systems the resultant closed-loop controlled system is stable and can asymptotically track an exogenous reference input