Chinese text classification for small sample set

Text classification is one of the most important topics in the fields of Internet information management and natural language processing. Machine learning based text classification methods are currently most popular ones with better performance than rule based ones. But they always need lots of training samples, which not only brings heavy work for previous manual classification, but also puts forward a higher request for storage and computing resources during the computer post-processing. Naïve Bayes algorithm is one of the most effective methods for text classification with the same problem. Only in the large training sample set can it get a more accurate result. This paper mainly studies Naïve Bayes classification algorithm for Chinese text based on Poisson distribution model and feature selection. The experimental results have shown that this method keeps high classification accuracy even in a small sample set.     

A Framework for Semantic Interpretation of Noun Compounds Using Tratz Model and Binary Features

Semantic interpretation of the relationship between noun compound (NC) elements has been a challenging issue due to the lack of contextual information, the unbounded number of combinations, and the absence of a universally accepted system for the categorization. The current models require a huge corpus of data to extract contextual information, which limits their usage in many situations. In this paper, a new semantic relations interpreter for NCs based on novel lightweight binary features is proposed. Some of the binary features used are novel. In addition, the interpreter uses a new feature selection method. By developing these new features and techniques, the proposed method removes the need for any huge corpuses. Implementing this method using a modular and plugin‐based framework, and by training it using the largest and the most current fine‐grained data set, shows that the accuracy is better than that of previously reported upon methods that utilize large corpuses. This improvement in accuracy and the provision of superior efficiency is achieved not only by improving the old features with such techniques as semantic scattering and sense collocation, but also by using various novel features and classifier max entropy. That the accuracy of the max entropy classifier is higher compared to that of other classifiers, such as a support vector machine, a Naïve Bayes, and a decision tree, is also shown.     

基于深度特征学习的网络流量异常检测方法

针对网络流量异常检测过程中提取的流量特征准确性低、鲁棒性差导致流量攻击检测率低、误报率高等问题,该文结合堆叠降噪自编码器(SDA)和softmax,提出一种基于深度特征学习的网络流量异常检测方法。首先基于粒子群优化算法设计SDA结构两阶段寻优算法:根据流量检测准确率依次对隐藏层层数及每层节点数进行寻优,确定搜索空间中的最优SDA结构,从而提高SDA提取特征的准确性。然后采用小批量梯度下降算法对优化的SDA进行训练,通过最小化含噪数据重构向量与原始输入向量间的差异,提取具有较强鲁棒性的流量特征。最后基于提取的流量特征对softmax进行训练构建异常检测分类器,从而实现对流量攻击的高性能检测。实验结果表明:该文所提方法可根据实验数据及其分类任务动态调整SDA结构,提取的流量特征具有更高的准确性和鲁棒性,流量攻击检测率高、误报率低。     

基于MapReduce技术的并行集成分类算法

由于计算机内存资源限制,分类器组合的有效性及最优性选择是机器学习领域的主要研究内容。经典的集成分类算法在处理小数据集时,拥有较高的分类准确性,但面对大量数据时,由于多基分类器学习、分类共用1台计算机资源,导致运算效率较低,这显然不适合处理当今的海量数据。针对已有集成分类算法只适合作用于小规模数据集的缺点,剖析了集成分类器的特性,采用基于聚合方式的集成分类器和云计算的MapReduce技术设计了并行集成分类算法(EMapReduce),达到并行处理大规模数据的目的。并在Amazon计算集群上模拟实验,实验结果表明该算法具有一定的高效性和可行性。     

基于特征选择的推荐系统托攻击检测算法

基于协同过滤的电子商务推荐系统极易受到托攻击,托攻击者注入伪造的用户模型增加或减少目标对象的推荐频率,如何检测托攻击是目前推荐系统领域的热点研究课题.分析五种类型托攻击对不同协同过滤算法产生的危害性,提出一种特征选择算法,为不同类型托攻击选取有效的检测指标.基于选择出的指标,提出两种基于监督学习的托攻击检测算法,第一种算法基于朴素贝叶斯分类;第二种算法基于k近邻分类.最后,通过实验验证了特征选择算法的有效性,及两种算法的灵敏性和特效性.     

Detection method of LDoS attacks based on combination of ANN ＆ KPCA

Low-rate denial-of-service (LDoS) attack is a new type of attack mode for TCP protocol.Characteristics of low average rate and strong concealment make it difficult for detection by traditional DoS detecting methods.According to characteristics of LDoS attacks,a new LDoS queue future was proposed from the router queue,the kernel principal component analysis (KPCA) method was combined with neural network,and a new method was present to detect LDoS attacks.The method reduced the dimensionality of queue feature via KPCA algorithm and made the reduced dimension data as the inputs of neural network.For the good sell-learning ability,BP neural network could generate a great LDoS attack classifier and this classifier was used to detect the attack.Experiment results show that the proposed approach has the characteristics of effectiveness and low algorithm complexity,which helps the design of high performance router.     

Survey on Ultra-Dense Networks (UDNs) and Applied Stochastic Geometry

The innovation of services offered by cellular networks gained the attention of researchers in the communication field. Thus, mobile industries deal with remarkable technological competition regarding service quality. The quality is determined by how superior, consistent, and quick a service is delivered to the user. Thus, mobility management is a basic factor as it deals with imperative information for managing user’s mobility. However, due to the expansion of connected devices, the users are set up densely which inspires the researcher for devising a novel mode switching model. This paper devises a novel mode switching model using the Naive Bayes classifier. Here, the switching of modes is based on certain quality parameters, like link utilization, bandwidth, delay, energy consumption, and signal strength. Whenever the network switches the communication link from cellular-mode to user-mode, it must maintain the quality parameters. For enhancing the performance of network mobility management, a mobility management model is devised in which user mobility is computed. Thus, the proposed method is essential for supporting improved user mobility during communication The proposed mode switching using Naïve Bayes classifier provides superior performance with a minimal delay of 0.164 s, maximal power of 58.786 bpm, maximal link utilization ratio of 0.727 and maximal throughput of 1,641,723 respectively.

     

一种基于树增强朴素贝叶斯的分类器学习方法

树增强朴素贝叶斯(TAN)结构强制每个属性结点必须拥有类别父结点和一个属性父结点,也没有考虑到各个属性与类别之间的相关性差异,导致分类准确率较差。为了改进TAN的分类准确率,该文首先扩展TAN结构,允许属性结点没有父结点或只有一个属性父结点;提出一种利用可分解的评分函数构建树形贝叶斯分类模型的学习方法,采用低阶条件独立性(CI)测试初步剔除无效属性,再结合改进的贝叶斯信息标准(BIC)评分函数利用贪婪搜索获得每个属性结点的父结点,从而建立分类模型。对比朴素贝叶斯(NB)和TAN,构建的分类器在多个分类指标上表现更好,说明该方法具有一定的优越性。     

基于组合式算法的Android恶意软件检测方法

为解决当前恶意软件静态检测方法中适用面较窄、实用性较低的问题,通过组合式算法筛选出最优分类器,并以此为基础实现了一个检测系统。首先使用逆向工程技术提取软件的特征库,并通过多段筛选得到分类器的初步结果。提出了一种基于最小风险贝叶斯的分类器评价标准,并以此为核心,通过对初步结果赋权值的方式得到最优分类器结果。最后以最优结果为核心实现了一个Android恶意软件检测系统原型。实验结果表明,该检测系统的分析精度为86.4%,并且不依赖于恶意代码的特征。     

基于改进的朴素贝叶斯的入侵检测方法

随着当前网络安全环境的日益严峻,针对网络入侵事件的检测至关重要。面对网络入侵检测中数据集合存在的冗余特征,提出一种改进后的朴素贝叶斯算法。该算法在原有朴素贝叶斯的基础上巧妙引入卡方检验,通过筛选数据集中占比重要的特征降低数据维度,提高了入侵检测的准确性。最后,结合实验结果证明,该方法有效提高了入侵检测的准确性。     

Clustering hypervisors to minimize failures in mobile cloud computing

Resource virtualization has become one of the key super‐power mobile computing architecture technologies. As mobile devices and multimedia traffic have increased dramatically, the load on mobile cloud computing systems has become heavier. Under such conditions, mobile cloud system reliability becomes a challenging task. In this paper, we propose a new model using a naive Bayes classifier for hypervisor failure prediction and prevention in mobile cloud computing. We exploit real‐time monitoring data in combination with historical maintenance data, which achieves higher accuracy in failure prediction and early failure‐risk detection. After detecting hypervisors at risk, we perform live migration of virtual servers within a cluster, which decreases the load and prevents failures in the cloud. We performed a simulation for verification. According to the experimental results, our proposed model shows good accuracy in failure prediction and the possibility of decreasing downtime in a hypervisor service. Copyright © 2017 John Wiley & Sons, Ltd.     

Effective discovery of attacks using entropy of packet dynamics

Network-based attacks are so devastating that they have become major threats to network security. Early yet accurate warning of these attacks is critical for both operators and end users. However, neither speed nor accuracy is easy to achieve because both require effective extraction and interpretation of anomalous patterns from overwhelmingly massive, noisy network traffic. The intrusion detection system presented here is designed to assist in diagnosing and identifying network attacks. This IDS is based on the notion of packet dynamics, rather than packet content, as a way to cope with the increasing complexity of attacks. We employ a concept of entropy to measure time-variant packet dynamics and, further, to extrapolate this entropy to detect network attacks. The entropy of network traffic should vary abruptly once the distinct patterns of packet dynamics embedded in attacks appear. The proposed classifier is evaluated by comparing independent statistics derived from five well-known attacks. Our classifier detects those five attacks with high accuracy and does so in a timely manner.     

Sentiment analysis of extremism in social media from textual information

Uncertainty in political, religious, and social issues causes extremism among people that are depicted by their sentiments on social media. Although, English is the most common language used to share views on social media, however, other vicinity based languages are also used by locals. Thus, it is also required to incorporate the views in such languages along with widely used languages for revealing better insights from data. This research focuses on the sentimental analysis of social media multilingual textual data to discover the intensity of the sentiments of extremism. Our study classifies the incorporated textual views into any of four categories, including high extreme, low extreme, moderate, and neutral, based on their level of extremism. Initially, a multilingual lexicon with the intensity weights is created. This lexicon is validated from domain experts and it attains 88% accuracy for validation. Subsequently, Multinomial Naïve Bayes and Linear Support Vector Classifier algorithms are employed for classification purposes. Overall, on the underlying multilingual dataset, Linear Support Vector Classifier out-performs with an accuracy of 82%.     

Efficient cluster head selection using Naïve Bayes classifier for wireless sensor networks

Data mining and approaches based on it have always been of approaches that have been considered in solving problems in the field of computer, but on some issues, this approach has been neglected. The area of wireless sensor networks and specifically the issue of optimal determining of the cluster head node are of these issues. To solve the problem of optimal determining of the cluster head node, Naïve Bayes that is the subset of data mining techniques is used in this paper. The results obtained after simulation of the presented algorithm show that the efficiency of this algorithm is significantly higher compared with other approaches that have so far been used to solve this problem, and thus it can be said that using this algorithm will lead to improved outcomes of solving this problem.     

Internet of things intrusion detection model and algorithm based on cloud computing and multi-feature extraction extreme learning machine

With the rapid development of the Internet of Things (IoT), there are several challenges pertaining to security in IoT applications. Compared with the characteristics of the traditional Internet, the IoT has many problems, such as large assets, complex and diverse structures, and lack of computing resources. Traditional network intrusion detection systems cannot meet the security needs of IoT applications. In view of this situation, this study applies cloud computing and machine learning to the intrusion detection system of IoT to improve detection performance. Usually, traditional intrusion detection algorithms require considerable time for training, and these intrusion detection algorithms are not suitable for cloud computing due to the limited computing power and storage capacity of cloud nodes; therefore, it is necessary to study intrusion detection algorithms with low weights, short training time, and high detection accuracy for deployment and application on cloud nodes. An appropriate classification algorithm is a primary factor for deploying cloud computing intrusion prevention systems and a prerequisite for the system to respond to intrusion and reduce intrusion threats. This paper discusses the problems related to IoT intrusion prevention in cloud computing environments. Based on the analysis of cloud computing security threats, this study extensively explores IoT intrusion detection, cloud node monitoring, and intrusion response in cloud computing environments by using cloud computing, an improved extreme learning machine, and other methods. We use the Multi-Feature Extraction Extreme Learning Machine (MFE-ELM) algorithm for cloud computing, which adds a multi-feature extraction process to cloud servers, and use the deployed MFE-ELM algorithm on cloud nodes to detect and discover network intrusions to cloud nodes. In our simulation experiments, a classical dataset for intrusion detection is selected as a test, and test steps such as data preprocessing, feature engineering, model training, and result analysis are performed. The experimental results show that the proposed algorithm can effectively detect and identify most network data packets with good model performance and achieve efficient intrusion detection for heterogeneous data of the IoT from cloud nodes. Furthermore, it can enable the cloud server to discover nodes with serious security threats in the cloud cluster in real time, so that further security protection measures can be taken to obtain the optimal intrusion response strategy for the cloud cluster.     

基于HOG特征的织物瑕疵检测算法

传统织物瑕疵检测依赖人工,成本高且效率低.提出基于机器视觉和图像处理算法的织物瑕疵检测方法,采用方向梯度直方图描述织物纹理特征,通过计算局部区域的方向梯度直方图特征与无瑕疵织物图像特征的相似性来识别瑕疵,在织物图像数据集上的待测结果表明,此算法能够有效地提取出织物的瑕疵.     

A proactive defense method for the stealthy EDoS attacks in a cloud environment

Cloud computing technology provides flexibility to Cloud Service Provider (CSP) for providing the cloud resources based on the users' requirements. In on‐demand pricing model, the attackers exploit this feature and cause unwanted scaling‐up of the cloud resources without any intent to pay for them. The associated cost for the unpaid malicious usage burdens the CSP, and over a long period, economic losses occur at the CSP end. Thus, the resources and services offered by the CSP become unsustainable, and the attack is termed as Economic Denial‐of‐Sustainability (EDoS) attack. The existing defense approaches for EDoS attacks are reactive. Thus, the associated attack detection/mitigation cost is high; consequently, the approaches are not suitable for the Small and Medium Enterprises (SMEs). The aim of this paper is to detect and mitigate, internal and external, stealthy EDoS attacks proactively. The attack is detected using average CPU utilization threshold and utility function (in terms of cost for the utilized cloud computing resources) and mitigated using virtual firewalls. Amazon Elastic Compute Cloud (Amazon EC2) is used to evaluate the performance of the proposed approach. The proposed approach accurately detects the EDoS attack and mitigates its effect as well with reduced cost. It is observed that the approach provides competitive response time, victim service downtime, and attack reporting time. Thus, the overall performance is improved.     

Error Rate Analysis of Uplink MC-CDMA Systems using Complementary Codes in Rayleigh Fading Channels

Cloud computing is a global technology for data storage and retrieving. Many organizations are switching their companies to cloud technology, so that they can lease cloud services for use on a membership or pay as you go basis rather than creating their own systems. Cloud service provider and the Cloud service accessibility are the two major problems in cloud computing. The Economic Denial of Sustainability (EDoS) attack is an important attack towards the cloud service providers. The attackers may send continuous requests to the cloud in a particular second. Hence the legitimate user cannot access the data due to heavy cloud traffic. Hence the paid user cannot access the data. However, this problem makes an economical issue to the users. So this paper presented a new technique as, ADS-PAYG (Attack Defense Shell- Pay As You Go) approach using Trust Factor method against the EDoS attack is proposed to improve more number of authenticated users by fixing a threshold value. The algorithm produced an effective result based on response time, accuracy and CPU utilization. The ADS-PAYG solution is applied using MATLAB, which outperforms other Trust factor estimation methods and effectively distinguishes attackers from legitimate users. The detection accuracy is 83.43% for the given dataset and it is high when compared to the existing algorithms.

     

基于多维熵序列分类的骨干网上流量异常检测

Detecting traffic anomalies is essential for diagnosing attacks. High-Speed Backbone Networks (HSBN) require Traffic Anomaly Detection Systems (TADS) which are accurate (high detection and low false positive rates) and efficient. The proposed approach utilizes entropy as traffic distributions metric over some traffic dimensions. An efficient algorithm, having low computational and space complexity, is used to estimate entropy. Entropy values over all dimensions are collected to form a detection vector for every sliding window. One class support vector machine classifies all detection vectors into one of two groups: abnormal vectors and normal vectors. A Multi-Windows Correlation Algorithm (MWCA) calculates comprehensive anomaly scores observed in a sequence of windows in order to reduce false positive rates and obtain high detection rates. Some real-world traffic traces have been used to validate and evaluate the efficiency and accuracy of this system through three experiments. In Experiment 1, the estimating algorithm of entropy which costs less memory and runs faster than traditional algorithms is more suitable for detection anomalies. In Experiment 2, the classification and correlation algorithms can improve the detection accuracy significantly. Experiment 3 compares the subject system and three well-known systems. Ours system is the most accurate one. Those results have indicated that the proposed system significantly improves the accuracy and efficiency.     

An energy‐saving algorithm for cloud resource management using a Kalman filter

Cloud computing has emerged as a promising technique to provide storage and computing component on‐demand services over a network. In this paper, we present an energy‐saving algorithm using the Kalman filter for cloud resource management to predict the workload and to further achieve high resource availability with low service level agreement. Using the proposed algorithm, one can estimate the potential future workload trend then predict the computing component workload utilizations and further retrench energy consumption and achieve load balancing in a cloud system. Experimental results show that the proposed algorithm achieves more than 92.22% accuracy in the computing component workload prediction, improves 55.11% energy in energy consumption, and has 3.71% in power prediction error rate, respectively. Copyright © 2013 John Wiley & Sons, Ltd.