基于义类的无导词义消歧方法的研究*

词义消歧在自然语言处理的许多应用领域都起着十分重要的作用。为了适用于大规模的词义消歧,提出了一种无导的学习方法。基于向量空间模型,结合机读词典和义类词典建立从义项到义类的映射关系,再利用义类知识在语料库中无导学习消歧特征,最后利用这些特征实现词义消歧。     

Unsupervised Electric Motor Fault Detection by Using Deep Autoencoders

Fault diagnosis of electric motors is a fundamental task for production line testing, and it is usually performed by experienced human operators. In the recent years, several methods have been proposed in the literature for detecting faults automatically. Deep neural networks have been successfully employed for this task, but, up to the authors' knowledge, they have never been used in an unsupervised scenario. This paper proposes an unsupervised method for diagnosing faults of electric motors by using a novelty detection approach based on deep autoencoders. In the proposed method, vibration signals are acquired by using accelerometers and processed to extract Log-Mel coefficients as features. Autoencoders are trained by using normal data only, i.e., data that do not contain faults. Three different autoencoders architectures have been evaluated: the multi-layer perceptron (MLP) autoencoder, the convolutional neural network autoencoder, and the recurrent autoencoder composed of long short-term memory (LSTM) units. The experiments have been conducted by using a dataset created by the authors, and the proposed approaches have been compared to the one-class support vector machine (OC-SVM) algorithm. The performance has been evaluated in terms area under curve (AUC) of the receiver operating characteristic curve, and the results showed that all the autoencoder-based approaches outperform the OC-SVM algorithm. Moreover, the MLP autoencoder is the most performing architecture, achieving an AUC equal to 99.11%      

Finding the Iris Using Convolutional Neural Networks

Journal of Computer and Systems Sciences International - Iris-based identification is a well-known and widely used technology. However, its use in mobile devices is limited due to the algorithmic...     

面向上下文图形可视化挖掘企业网络行为

在企业分布式网络系统中,精确、动态地发现用户的网络行为日益具有挑战性。目前网络管理系统主要依赖于对用户身份的推理来进行网络行为管理,此类方法由于所收集到的数据缩放比粗糙,从而在大规模的网络环境下不能精确地对网络行为进行挖掘、发现和管理。对主机、用户、应用程序等网络上下文内容进行可视化挖掘,为动态地进行网络管理提供了重要帮助。     

利用语义词典Web挖掘语言模型的无指导译文消歧

为了解决困扰词义及译文消歧的数据稀疏及知识获取问题,提出一种基于Web利用n-gram统计语言模型进行消歧的方法.在提出词汇语义与其n-gram语言模型存在对应关系假设的基础上,首先利用Hownet建立中文歧义词的英文译文与知网DEF的对应关系并得到该DEF下的词汇集合,然后通过搜索引擎在Web上搜索,并以此计算不同DEF中词汇n-gram出现的概率,然后进行消歧决策.在国际语义评测SemEval-2007中的Multilingual Chinese English Lexical Sample Task测试集上的测试表明,该方法的P_mar值为55.9%,比其上该任务参评最好的无指导系统性能高出12.8%.     

Word Sense Disambiguation Using the Classification Information Model

A Classification Information Model is a pattern classification model.The model decides the proper class of an input instance by integrating individual decisions, each of which is made with each feature in the pattern.Each individual decision is weighted according to the distributional property of the feature deriving the decision. An individual decision and its weight are represented as classification information which is extracted from the training instances.In the word sense disambiguation based on the model, the proper sense of an input instance is determined by the weighted sum of whole individual decisions derived from the features contained in the instance.     

利用时滞补偿器降低广义预测控制动态关联作用

本文提出了通过预加前、后纯时滞补偿器的方法降低多时滞多变量广义预测控制的闭环动态关联作用的思想,给出了设计方法,并讨论了设计参数的选择对系统的动态关联作用的影响,通过仿真证明了本文所提出的方法的有效性。     

计算机人脸照片识别技术研究

介绍了计算机人脸照片识别技术的一般方法,以及为了提高识别的精确度和识别率的改进算法。     

用改进的非监督学习方法加速前馈神经网络的训练

本语文对非学习方法和前馈神经网络的结构进行了一些简单的改进,用改进后的结构和学习方法寝化前馈神经网络的权值,使前馈神经网络的训练速度和逼近精度显著提高。最后本语文通过仿真实例验证了该方法。     

Finding Global Minima of Maximum Functions by Using Exclusion Functions without Derivatives

The first part of the paper introduces a new exclusion function without derivatives and describes its fundamental characteristics. The second part uses the new notion for finding global minimum of a real function of few variables g : [a, b] Rm R, g(x) = max{g1(x),...,gn(x)}, x [a, b], where g1(x),...,gn(x) are special multivariant real functions. The structure of the Maple program and some numerical examples are also presented in this part.     

运用动态规则集机制改善IDS性能的研究

针对高速网下IDS因性能而造成的高漏报、误报的问题,提出分析强度、分析命中率以及动态规则集的概念,建立动态规则集机制的模型,该模型可以根据网络流量和事件反馈实时动态地调整规则集的规模和分布,以提高IDS系统的整体性能,并介绍了实现中的关键技术。     

双端固定音叉作谐振敏感元件的电磁激励效率的理论与实验研究

为了能够增强谐振音叉的输出信号,进而提高谐振音叉的激励效率,测量、分析并比较了谐振音叉在不同激励信号下的激励效率.针对双端固定音叉,对不同激励信号从理论上进行分析.最后,利用谐振式加速度测试平台,以不同的激励信号对谐振音叉进行激励,测量输出信号.测试结果表明:激励信号的激励效率从大到小的排列顺序是:方波、正弦波、三角波、锯齿波.激励信号所加的直流偏置越大,输出的信号波形的幅值就越大.     

应用变形引擎实现的软件动态保护方法研究

通过对软件动态保护的研究,提出了应用变形引擎实现的软件动态保护方法.将待保护软件的重要代码片段从软件中删除,利用自修改变形技术构造这些重要代码片段的变形引擎,在软件执行过程中,将它们依次还原并执行;同时,根据一些代码片段的执行时间控制变形子引擎的调度;并且利用动态加解密技术对变形子引擎进行保护.这种方法大大增加了软件攻击者成功逆向该软件所需的时间和精力.最后通过实验说明软件安全强度的提高对性能造成了一定的影响.     

应用红外遥感寻找地下热水试验研究

利用陆地卫星红外遥感数据能反映地物辐射温度差异这一特点,通过对遥感影像进行波段差值运算、IHS（HLS）彩色变换和热红外波段密度分割3种方法的增强处理,对其进行综合探索研究。并结合地质构造展布规律、地面调查,最后圈定地热异常的分布范围,获得了一项适宜全省乃至更大范围的探寻地下热水资源的技术方法,值得大面积推广应用。     

基于词典属性特征的粗粒度词义消歧

本文依据《现代汉语语法信息词典》中对词语多义的属性特征描述,对《人民日报》语料中155 个词语共 4 996 个同形实例进行了粗粒度词义自动消歧实验,同时用贝叶斯算法进行了比较测试。基于词典属性特征的消歧方法在同形层面上准确率达到 90%, 但召回率偏低。其优点在于两个方面: 1) 不受词义标注语料库规模的影响;2) 对特定词语意义的消歧准确率可达到100%。本文也讨论了适用于不同词类的消歧特征。