基于视觉感知特征的手机应用流量识别方法

由于大多数手机应用通过HTTP协议进行通信,传统的端口识别方法已经基本失效。另外,深度包检测和基于流统计特征的机器学习方法均存在手工设计特征和标记样本的困难。借鉴计算机视觉领域的优势,提出了一种基于视觉感知特征的手机应用流量识别方法。首先,将应用层载荷数据转换为视觉上有意义的图像,并从网络关口采集真实数据,建立了样本数据集IMTD17;然后,设计了具有视觉特征提取能力的卷积感知网络模型2D-CPN,利用卷积自编码实现了对大量无标记样本的学习,并通过多类型回归建立起从隐层特征到应用类型的映射。实验结果表明,该方法的流量识别准确率满足实际使用的需求。     

Personalized app recommendation based on app permissions

With the development of science and technology, the popularity of smart phones has made exponential growth in mobile phone application market. How to help users to select applications they prefer has become a hot topic in recommendation algorithm. As traditional recommendation algorithms are based on popularity and download, they inadvertently fail to recommend the desirable applications. At the same time, many users tend to pay more attention to permissions of those applications, because of some privacy and security reasons. There are few recommendation algorithms which take account of apps’ permissions, functionalities and users’ interests altogether. Some of them only consider permissions while neglecting the users’ interests, others just perform linear combination of apps’ permissions, functionalities and users’ interests to implement top-N recommendation. In this paper, we devise a recommendation method based on both permissions and functionalities. After demonstrating the correlation of apps’ permissions and users’ interests, we design an app risk score calculating method ARSM based on app-permission bipartite graph model. Furthermore, we propose a novel matrix factorization algorithm MFPF based on users’ interests, apps’ permissions and functionalities to handle personalized app recommendation. We compare our work with some of the state-of-the-art recommendation algorithms, and the results indicate that our work can improve the recommendation accuracy remarkably.     

Examining the impacts of fitness app features on user well-being

Drawing on the self-regulation theory, the current paper explores the impacts of two types of fitness app feature sets (i.e., personal-oriented and social-oriented features) on users’ health behavior and well-being. The results from fitness app users show that both personal-oriented features and social-oriented features of fitness apps can significantly improve exercise adherence and social engagement of users. Users’ exercise proficiency level negatively moderates the relationship between social-oriented features and (a) exercise adherence and (b) social engagement. High levels of social engagement promote users’ physical adherence to exercises. Exercise adherence and social engagement both enhance users’ subjective well-being, but their impacts on different dimensions of well-being vary. Furthermore, regardless of specific features, sufficient use of fitness apps, in general, can significantly help users lead more positive and healthier lives by maintaining exercise adherence, reducing emotional exhaustion, and improving their satisfaction with the overall quality of life. Our findings offer important insights into the underlying mechanisms that help explain fitness app features on users’ well-being, and on a practical level, provide suggestions for mobile app developers in designing better fitness app products and for exercisers in optimizing the benefits of fitness technology adoption.     

Usability evaluation of a virtual reality smartphone app for a living museum

This paper elaborates the empirical evidence of a usability evaluation of a VR and non-VR virtual tour application for a living museum. The System Usability Scale (SUS) was used in between participants experiments (Group 1: non-VR version and Group 2: VR version) with 40 participants. The results show that the mean scores of all components for the VR version are higher compared to the non-VR version, overall SUS score (72.10 vs 68.10), usability score (75.50 vs 71.70), and learnability (58.40 vs 57.00). Further analysis using a two-tailed independent t test showed no difference between the non-VR and VR versions. Additionally, no significant difference was observed between the groups in the context of gender, nationality, and prior experience (other VR tour applications) for overall SUS score, usability score, and learnability score. Α two-tailed independent t test indicated no significant difference in the usability score between participants with VR experience and no VR experience. However, a significant difference was found between participants with VR experience and no VR experience for both SUS score (t(38) = 2.17, p = 0.037) and learnability score (t(38) = 2.40, p = 0.021). The independent t test results indicated a significant difference between participant with and without previous visits to SCV for the usability score (t(38) = −2.31, p = 0.027), while there was no significant differences observed in other components. It can be concluded that both versions passed based on the SUS score. However, the sub-scale usability and learnability scores indicated some usability issue.

     

Identification of malicious android app using manifest and opcode features

In this paper, we propose a statistical approach for smartphone malware detection. A set of features such as hardware, permission, application components, filtered intents, opcodes and strings are extracted from the samples to form a vector space model. Feature selection methods such as Entropy based Category Coverage Difference (ECCD) and Weighted Mutual Information (WI) are used to choose the prominent features. The performance of the system is analyzed using classifiers, Random Forest, Rotation Forest and Support Vector Machine (SVM). The system was evaluated on individual models as well as Meta feature space model for both malware and benign features. It was observed that the meta feature space model with malware features provide the best results for both feature selection. For ECCD, Random Forest classifier performs better [Dataset 1—0.972, Dataset 2—0.976 and Dataset 3—0.969] whereas in the case of WI, SVM gives highest F-measure [Dataset 1—0.993, Dataset 2—0.994 and Dataset 3—0.992]. From the overall analysis on the system, we conclude that the malware model outperforms it’s benign counterpart and also that WI is a better feature selection technique compared to ECCD.     

The differential impact of types of app innovation on customer evaluation

Digital innovation literature seldom considers the effects of innovation from app developers’ perspective. Even when it does, the findings on the value of app innovation are mixed. Based on technological innovation and early-mover advantages literature, we delineate types of app innovation and their impacts on customer evaluations. Our empirical results indicate that adding new business functions to provide new categories of product or service offerings increases customer evaluations, whereas, adding new supporting service functionalities (SSFs) decreases customer evaluations. For followers that “imitate” innovations, the effect of adding SSFs on customer evaluations is negatively contingent on the quality of early-mover apps.     

基于形象特征分析的垃圾邮件过滤系统的研究

在分析了大量包含图像的垃圾邮件后,提出基于形象特征分析的垃圾邮件过滤系统.它提取出有用的形象特征并使用一类支持向量机来过滤垃圾邮件.实验结果表明,该系统模型具有令人满意的过滤效果,并且有很高的探测率和较低的误判率.     

基于相容性分析的医疗诊断专家系统

医疗诊断专家系统作为人工智能重要应用领域之一,已被广泛应用于医疗诊断和网络医疗咨询等方面。其补充了人类专家的不足,并能有效地解决各种临床问题。然而,现有诊断推理模型在病例推理过程中过于依赖权值设定和医生经验值,诊断结果完全依赖于权值给定质量,并且目前尚未发现较实用的权值设定和自动赋值方法。为了解决权值设定和自动赋值的难题,提出了一种基于属性相容性分析的医疗诊断方法并进行了探讨。利用基于属性的相容性分析对医疗数据进行数字化处理,并根据属性之间的关联度构造属性相关矩阵和对属性进行剪枝处理,避免了无效属性的不必要运算;在相容性分析基础上提出了较为实用的权值设定数学模型;在进行病例相似度计算时,采用群体决策策略来完成诊断。针对所提方法进行了模拟。实验结果表明,该方法能有效解决权值自动设定问题和满足实际应用需求。     

Research on e-Government evaluation model based on the principal component analysis

Over the last few years, the area of electronic government (e-Government) has received increasing prominence and attention; people are interacting with e-Government systems to an ever greater extent. It is therefore important to measure the development of e-Government. Adopting principal component analysis (PCA), this study presents, validates and updates an evaluation model with 5 dimensions based on Socio-Technical model and Stakeholder Theory, which captures the multidimensional and interdependent nature of e-Government system. The validity of the model is empirically investigated using a sample of local e-Government of 18 cities in China, all of which have high Internet penetration and mature ICT use. The five dimensions of the evaluation model include project construction, information security management, special construction, transparency of government affairs and informationized ability. K-means clustering is applied in the subspace created by PCA to evaluate the local e-Government stages of growth of these 18 cities. The findings provide several important implications for e-Government research and practice.     

Facial beauty analysis based on features prediction and beautification models

This paper studies facial beauty analysis by using feature-based computer models in prediction and beautification. In order to assess the facial beauty index, a facial beauty prediction model is established, which can quickly and effectively estimate beauty indexes for new images. In the facial beautification model, we mainly achieve three functions: (1) propose an effective method to adjust the positions of the landmark points to beautify the facial geometry; (2) improve the multi-level median filtering method to perform facial skin beautification; and (3) design a method to achieve average facial beautification. Experimental results showed that the proposed models can significantly analyze the facial attractiveness of face images.     

基于统计分析的作战行动效能评估方法

由于作战过程中不确定因素多,作战行动效果数据表现出显著的随机性。为了探索效果数据背后隐藏的作战规律,基于统计分析的方法研究作战行动效能的评估问题。分析了作战行动及其效能的基本概念,针对增强最简半自治适应性作战神经网络工具箱(EINSTein)产生的仿真数据采用单次、单组以及多组实验分析的手段,研究进攻行动效果数据的统计特征,发现对于具有作战优势的一方,相比通过增加作战人数而言,提高火力半径能够取得更好的作战效果。在此基础上,提出一种作战行动效能的评估方法,并结合仿真数据进行了验证,从而为基于实际训练效果数据的效能评估提供可行的解决方案。     

基于混沌理论的需求分析质量评价模型

在基于混沌理论和软件需求复杂性理论研究的基础上,提出一种理论模型.该模型是在已描述的非线性方程下得到的一个包含三段式的轨迹图,它可以对软件需求的演化过程进行指导,同时也可以评价需求过程的质量.初步研究表明,只要需求演化过程的变化图和模型的轨迹图一致,那么需求的演化过程就可以看成是正常的.最后通过对具体实例的分析,验证了该模型在评价需求演化过程上是有效可行的.     

基于改进的GLCM甲状腺纹理特征提取与分析

为了提高病变和正常的甲状腺核磁共振图像（MR）的分类准确率,提出了改进的窗口自适应灰度共生矩阵（Gray Level Co-occurrence Matrix,GLCM）纹理特征提取算法。采用基于统计的纹理特征GLCM算法提取感兴趣区域（ROIs）的纹理特征,由HOG特征启示,研究基于梯度信息的GLCM窗口自适应算法,考虑了梯度信息对GLCM中滑动窗口大小设置的影响,克服了传统方法的固定窗口对图像细节保留的影响,同时为了消除各向异性,取四个方向的共生矩阵的均值作为最终的共生矩阵,最后计算GLCM的相关、能量、对比、逆差矩和熵的均值和方差。对94幅甲状腺图像采用逻辑回归模型分析来预测分类准确度,结果显示,该方法优于其他的方法,对甲状腺图像诊断性能更好,预分类准确率达到96.8%,灵敏度97.90%,特异度95.7%,ROC曲线下面积（Area Under the Curve, AUC）为0.968。实验结果表明改进的GLCM能够有效辅助医生对甲状腺MR图像做出正确诊断。     

Government website accessibility: a cross-country analysis of G7 and BRICS countries

Universal Access in the Information Society - The Internet has revolutionized how users are able to work and gather information. In modern world, information is available on a mouse click. This is...     

Modeling,analysis, and evaluation of a carbon tax policy based on the emission factor

This paper studies a tax policy based on the emission factor, which is used as an intensity measure. Specifically, the paper models a situation where policy makers set a limit on the production emission factor of a regulated industry and require firms to pay tax if they exceed that target. The efficiency of the policy is evaluated against other existing environmental policies using a social welfare mathematical programming model. A case study is built within the context of the cement industry and is used to carry out the analysis. The results show that the intensity-based carbon tax achieves the highest consumers’ surplus and production quantities and the lowest prices. The policy is found to be effective in reducing CO₂ emissions with little impact on social welfare.     

One-dimensional V-Scope analysis of habituation to simulated cross-country skiing

Abstract

Responses to simulated cross-country skiing were measured using the V-Scope, a new telemetric ultrasound motion monitor. Ten young male adults performed a total of 45 minutes of distributed practice on a Nordic-Track^R ski simulator. Over a period of three 15-minute sessions cadence and velocity were unchanged. Step and stride lengths decreased significantly (p < 0-05) after the first 15-minute session and then remained unchanged. There were no left-right limb differences across all sessions indicating a normal gait. Response variability in velocity, step lengths and stride length was dramatically reduced after the first exposure period. This study demonstrates that the V-Scope system is a useful motion analysing device and, on the basis of the data presented in this preliminary investigation, at least two 15-minute habituation sessions are required for initial habituation to simulated cross-country skiing.     

Understanding the Windows EAL4 evaluation

On29 October 2002, Microsoft and Science Applications International Corporation issued press releases announcing that Windows 2000 with Service Pack 3 had received Evaluation Assurance Level 4 certification under the Common Criteria evaluation process. The Microsoft certification is sure to come up in many purchasing discussions over the next few years, so the question arises: What does this mean in plain English?.