Comparing techniques for authorship attribution of source code

Attributing authorship of documents with unknown creators has been studied extensively for natural language text such as essays and literature, but less so for non‐natural languages such as computer source code. Previous attempts at attributing authorship of source code can be categorised by two attributes: the software features used for the classification, either strings of n tokens/bytes (n‐grams) or software metrics; and the classification technique that exploits those features, either information retrieval ranking or machine learning. The results of existing studies, however, are not directly comparable as all use different test beds and evaluation methodologies, making it difficult to assess which approach is superior. This paper summarises all previous techniques to source code authorship attribution, implements feature sets that are motivated by the literature, and applies information retrieval ranking methods or machine classifiers for each approach. Importantly, all approaches are tested on identical collections from varying programming languages and author types. Our conclusions are as follows: (i) ranking and machine classifier approaches are around 90% and 85% accurate, respectively, for a one‐in‐10 classification problem; (ii) the byte‐level n‐gram approach is best used with different parameters to those previously published; (iii) neural networks and support vector machines were found to be the most accurate machine classifiers of the eight evaluated; (iv) use of n‐gram features in combination with machine classifiers shows promise, but there are scalability problems that still must be overcome; and (v) approaches based on information retrieval techniques are currently more accurate than approaches based on machine learning. Copyright © 2012 John Wiley & Sons, Ltd.     

Multi-Wavelength vs. Binary Code Pattern Unwrapping in Fringe Projection Method

Phase extraction, phase unwrapping and phase to height conversion are consecutive steps within the fringe projection method for reconstruction of 3D objects. The procedure of determining discontinuities on the wrapped phase, resolving them and achieving the unwrapped phase is called phase unwrapping. A software system was developed that provides the ability to create many kinds of patterns with a verity of desirable properties suitable for different demands in fringe projection systems. In our previous work, an experimental study was reported based on a multi-wavelength unwrapping approach. In the following work, we implemented a new method called binary code pattern unwrapping that is based on time analysis. A comparison of the two unwrapping approaches in terms of resolution, quality, noise, depth accuracy, and computational cost is provided.     

口令重用行为与多维口令体系研究

互联网的迅速发展与网络服务的高度分散,促使广大网民不断注册更多的账户,并导致口令重用行为普遍化,使得用户信息面临泄露的风险。为此,基于2011年底互联网泄密门数据和大学生在线调查数据,分析了网民口令的结构特征和重用行为,并由此设计融入信息维度和分级管理思想的多维口令体系。该体系以根口令-重用码结构为基础,内容维包含多个独立的信息因子,构成口令的可记忆性主体;形式维负责形式变换,以提升口令的复杂性和安全性;时空维用于保障口令的时效性和重用性。对比量化分析结果表明,该口令体系具备良好的记忆性和便捷性,能有效抵御暴力攻击和熟人攻击。     

基于K-L散度的恶意代码模型聚类检测方法

在云计算应用环境下,由于服务系统越来越复杂,网络安全漏洞和被攻击情况急剧增加,传统的恶意代码检测技术和防护模式已无法适应云存储环境的需求。为此,通过引入高斯混合模型,建立恶意代码的分层检测机制,使用信息增益和文档频率等方法分析和提取样本数据特征值,结合K-L散度特性,提出基于K-L散度的恶意代码模型聚类检测方法。采用KDDCUP99数据集,使用Weka开源软件完成数据预处理和聚类分析。实验结果表明,在结合信息增益和文档频率进行特征分析的前提下,与贝叶斯算法相比,该方法在虚拟环境中恶意代码的平均检测时间降低16.6%,恶意代码的平均检测率提高1.05%。     

基于开放源代码的防火墙实验项目设计

当前计算机网络安全课程已经广泛地开设于计算机及信息安全等相关专业,而防火墙技术又是计算机网络安全课程的重要内容之一。用专用硬件防火墙作为实验设备具有价格昂贵、内部技术不透明的缺点,因此本文设计一系列基于Linux操作系统开放源代码的防火墙技术实验,这样不仅可以降低教学实验成本,也能让学生更清楚地了解各种防火墙技术及实现,增强学生的专业实践能力。     

Optimal binary codes and binary construction of quantum codes

This paper discusses optimal binary codes and pure binary quantum codes created using Steane construction. First, a local search algorithm for a special subclass of quasi-cyclic codes is proposed, then five binary quasi-cyclic codes are built. Second, three classical construction methods are generalized for new codes from old such that they are suitable for constructing binary self-orthogonal codes, and 62 binary codes and six subcode chains of obtained self-orthogonal codes are designed. Third, six pure binary quantum codes are constructed from the code pairs obtained through Steane construction. There are 66 good binary codes that include 12 optimal linear codes, 45 known optimal linear codes, and nine known optimal self-orthogonal codes. The six pure binary quantum codes all achieve the performance of their additive counterparts constructed by quaternary construction and thus are known optimal codes.     

基于安全协议代码的形式化辅助建模研究

随着安全协议形式化分析技术的不断发展,利用工具自动验证虽已得到实现,但建模环节仍需依赖专业人员手工建模,难度大且成本高,限制了此技术的进一步推广。为了提高建模的自动化程度,提出了依据安全协议代码进行形式化模型辅助生成的方案。首先,使用污点分析获取协议的通信流程,并且记录密码学原语操作;然后,根据通信流程之间的序列关系构建协议通信状态机;最终,根据目前主流的安全协议形式化分析工具Tamarin的模型语法生成形式化模型。实验结果表明,此方案可以生成形式化模型中的关键部分,提高了形式化建模的自动化程度,为形式化分析技术的推广作出贡献。     

一种使用帧间差值的图像传感器片上视频压缩算法

为了减少图像传感器视频数据的输出,提出了一种通过编码相邻两帧之间差值的无损视频压缩算法.算法首先将基于差分脉冲编码调制原理的差分操作在模拟域实现,减小了电路的复杂度.然后两帧之间的差值被基于块的无损压缩方案编码.实验结果证明,压缩后的图像数据可以被无损失的还原.通过对7个具有代表性的8位深度1280×720@60 fps的样本视频进行测试,在块大小为4×4和模式切换阈值为63时实现了最佳的压缩效果.在几乎没有光的条件下压缩率高达78.5％.在复杂运动场景下该算法压缩率为43.5％.提出的压缩算法更适用于长时间处于静止场景的视频录制.     

代码搜索方法研究进展

开源软件的成功推动了软件产业的蓬勃发展,大量代码资源为代码搜索创造了条件.如何通过代码搜索技术找到需求代码成为一个重要问题.为了更好地推进后续研究工作,首先对代码搜索相关概念及研究趋势进行介绍和说明;其次对使用不同技术的代码搜索研究工作进行综述,包括基于信息检索、查询和代码描述增强、程序特征分类以及深度学习等方面,并进一步总结归纳不同方法的优缺点;接下来针对代码搜索技术所应用的多个领域进行介绍,包括程序合成、代码推荐与补全和代码风格改善等方面;最后分析现阶段代码搜索面临的主要问题,为未来该方向的发展提供一定的参考与建议.