Tree-Structured Parallel Regeneration for Multiple Data Losses in Distributed Storage Systems Based on Erasure Codes

To reduce the time required to complete the regeneration process of erasure codes, we propose a Tree-structured Parallel Regeneration (TPR) scheme for multiple data losses in distributed storage systems. Under the scheme, two algorithms are proposed for the construction of multiple regeneration trees, namely the edge-disjoint algorithm and edge-sharing algorithm. The edge-disjoint algorithm constructs multiple independent trees, and is simple and appropriate for environments where newcomers and their providers are distributed over a large area and have few intersections. The edge-sharing algorithm constructs multiple trees that compete to utilize the bandwidth, and make a better utilization of the bandwidth, although it needs to measure the available band-width and deal with the bandwidth changes; it is therefore difficult to implement in practical systems. The parallel regeneration for multiple data losses of TPR primarily includes two optimizations: firstly, transferring the data through the bandwidth optimized-paths in a pipe-line manner; secondly, executing data regeneration over multiple trees in parallel. To evaluate the proposal, we implement an event-based simulator and make a detailed comparison with some popular regeneration methods. The quantitative comparison results show that the use of TPR employing either the edge-disjoint algorithm or edge-sharing algorithm reduces the regeneration time significantly.

THE NEW METHOD OF GENERATION OF BINARY TREES FROM ZER0(0) AND ONE(1) CODES

A binary tree can be represented by a code reflecting the traversal of the corresponding regular binary tree in given monotonic order. A different coding scheme based on the branches of a regular binary tree with n-nodes is proposed. It differs from the coding scheme generally used and makes no distinction between internal nodes and terminal nodes. A code of a regular binary tree with nnodes is formed by labeling the left branches by O’s and the right branches by l’s and then traversing these branches in pre-order. Root is always assumed to be on a left branch.

A 4-Corner Codes Classifier Based on Decision Tree Inductive Learning for Handwritten Chinese Characters

The classification for handwritten Chinese character recognition can be viewed as a transformation in discrete vector space. In this paper, from the point of discrete vector space transformation, a new 4-corner codes classifier based on decision tree inductive learning algorithm ID3 for handwritten Chinese characters is presented. With a feature extraction controller, the classifier can reduce the number of extracted features and accelerate classification speed. Experimental results show that the 4-corner codes classifier performs well on both recognition accuracy and speed.

Matching Source Code Using Abstract Syntax Trees in Version Control Systems

Software projects are becoming larger and more complicated. Managing those projects is based on several software development methodologies. One of those methodologies is software version control, which is used in the majority of worldwide software projects. Although existing version control systems provide sufficient functionality in many situations, they are lacking in terms of semantics and structure for source code. It is commonly believed that improving software version control can contribute substantially to the development of software. We present a solution that considers a structural model for matching source code that can be used in version control.

Code Clone Detection Method Based on the Combination of Tree-Based and Token-Based Methods

This article proposes the high-speed and high-accuracy code clone detection method based on the combination of tree-based and token-based methods. Existence of duplicated program codes, called code clone, is one of the main factors that reduces the quality and maintainability of software. If one code fragment contains faults (bugs) and they are copied and modified to other locations, it is necessary to correct all of them. But it is not easy to find all code clones in large and complex software. Much research efforts have been done for code clone detection. There are mainly two methods for code clone detection. One is token-based and the other is tree-based method. Token-based method is fast and requires less resources. However it cannot detect all kinds of code clones. Tree-based method can detect all kinds of code clones, but it is slow and requires much computing resources. In this paper combination of these two methods was proposed to improve the efficiency and accuracy of detecting code clones. Firstly some candidates of code clones will be extracted by token-based method that is fast and lightweight. Then selected candidates will be checked more precisely by using tree-based method that can find all kinds of code clones. The prototype system was developed. This system accepts source code and tokenizes it in the first step. Then token-based method is applied to this token sequence to find candidates of code clones. After extracting several candidates, selected source codes will be converted into abstract syntax tree (AST) for applying tree-based method. Some sample source codes were used to evaluate the proposed method. This evaluation proved the improvement of efficiency and precision of code clones detecting.

An Optimal Algorithm for Prufer Codes

This paper studies the algorithms for coding and decoding Prufer codes of a labeled tree. The algorithms for coding and decoding Prufer codes of a labeled tree in the literatures require time usually. Although there exist linear time algorithms for Prufer-like codes [1,2,3], the algorithms utilize the integer sorting algorithms. The special range of the integers to be sorted is utilized to obtain a linear time integer sorting algorithm. The Prufer code problem is reduced to integer sorting. In this paper we consider the Prufer code problem in a different angle and a more direct manner. We start from a naïve algorithm, then improved it gradually and finally we obtain a very practical linear time algorithm. The techniques we used in this paper are of interest in their own right.

Bug localization based on syntactical and semantic information of source code

The existing software bug localization models treat the source file as natural language, which leads to the loss of syntactical and structure information of the source file. A bug localization model based on syntactical and semantic information of source code is proposed. Firstly, abstract syntax tree(AST) is divided based on node category to obtain statement sequence. The statement tree is encoded into vectors to capture lexical and syntactical knowledge at the statement level.Secondly, the source code is transformed into vector representation by the sequence naturalness of the statement. Therefore,the problem of gradient vanishing and explosion caused by a large AST size is obviated when using AST to the represent source code. Finally, the correlation between bug reports and source files are comprehensively analyzed from three aspects of syntax, semantics and text to locate the buggy code. Experiments show that compared with other standard models, the proposed model improves the performance of bug localization, and it has good advantages in mean reciprocal rank(MRR), mean average precision(MAP) and Top N Rank.

融合注意力机制和Child-Sum Tree-LSTM的二进制代码相似性检测

抽象语法树是一种代码的树型表示,它保留了代码中定义良好的语句组件、语句的显式顺序和执行逻辑。包含丰富语义信息的抽象语法树可以在二进制分析时通过反编译生成,并且已经作为代码特征应用于二进制代码相似度检测。抽象语法树中不同类别的节点承载着不同的语义信息,对整棵树的语义具有不同的贡献程度。然而现有的二进制代码相似度检测方法所用神经网络无法对抽象语法树节点进行重要性区分,影响了模型的训练效果。针对该问题,提出了一种融合注意力机制和Child-Sum Tree-LSTM神经网络的跨指令集、跨代码混淆二进制代码相似性检测方法。首先使用二进制分析工具IDA Pro对二进制代码反编译提取架构无关的抽象语法树特征,并利用随机采样构造训练样本对。然后使用抽象语法树训练样本对训练融合注意力机制和Child-Sum Tree-LSTM的神经网络模型。在公开数据集BINKIT上的实验表明,所提方法的AUC和Accuracy指标分别为94.1%、66.2%,优于Child-Sum Tree-LSTM算法。

基于大语言模型的个性化实验报告评语自动生成与应用

在计算机实验报告评阅过程中,不同的实验报告评价体系呈现出多样性和差异性,固化的实验评语模板缺乏个性化的内容,评价结果往往未给出可解释性的依据。针对以上问题,提出基于大语言模型的个性化实验报告评语自动生成框架。通过主题-评估决策-集成提示策略,从教师的实验需求、代码质量需求中抽取该实验特有的评价体系,形成评估决策树,构建计算机软件方向课程共享的评估决策树库。设计基于大语言模型和决策树的实验要求、代码质量主题评级方法,从评估决策树库检索匹配学生实验报告内容的评估决策树,结合实验报告和代码文本,自动生成实验主题、代码质量主题定量或定性的评级结果及对应的可解释性依据。在实验报告模板中融入学生已完成的实验任务、主题评级结果、评价依据等,生成个性化的实验评语。实验结果表明,基于主题-评估决策-集成提示策略的决策树生成结果明显优于未用提示的方法,该策略各部分具有一定的有效性和合理性,同时自动生成的评级结果和教师原先批阅的评阅结果对比,软件测试、面向对象程序设计、电商金融课程示例匹配正确率均达到90%以上。从任课教师对于自动生成的评语评分分析,评语在流畅性、相关性、合理性3个维度上达到了较高的质量水平。

带宽异构网络下的精确修复再生码数据修复方案

再生码技术以高容错性、低冗余开销等优点在数据存储领域得到了广泛应用,但基于再生码的冗余技术在修复失效数据时需从其他帮助节点下载多个编码块。考虑到节点间链路带宽的异构性,在实际网络中链路可用带宽容量变化很大,网络流量最小化并不一定意味着数据修复时间最小化,并且现有针对带宽异构网络下的再生码数据修复方案难以支持精确地修复再生码。由于精确修复再生码具有特定的数学结构,其并行修复难以实现,因此提出一个在带宽异构网络下实现精确修复再生码的数据修复方案ERC-TREE,此方案通过构建一棵最优树来有效利用帮助节点之间的可用带宽,从而实现失效节点数据的精确修复。仿真实验结果表明,在带宽异构网络下采用树型结构修复对精确修复再生码具有可行性。在带宽差异很大的情况下,ERC-TREE的数据修复时间相比星型结构的修复时间减少78%。

基于多特征融合的细胞特异性lncRNA的亚细胞定位预测

长链非编码RNA(long non-coding RNA,lncRNA)在细胞生物学过程和疾病发展中扮演着关键性角色。由于lncRNA的亚细胞定位和其生物学功能密切相关,因此确定lncRNA的亚细胞定位具有重要意义。目前已有一些基于机器学习的方法来识别lncRNA的亚细胞位置,但在识别人类lncRNA的细胞特异性定位方面的相关工作仍然有限。该模型对人类细胞系lncRNA亚细胞定位问题进行了研究,提取了k-mer、CKSNAP、SRS和TSS特征信息,并对各类特征信息进行了融合,基于XGBoost和LightGBM结合的算法来预测人类细胞系lncRNA的亚细胞位置,并通过10倍交叉检验对模型进行了评估。结果表明,该模型预测人类细胞系lncRNA亚细胞定位的方法与现有的预测方法相比,预测成功率均有一定改进,其基准数据集的AUROC值最高达到92.26%。

基于增强嵌入特征超图学习的恶意域名检测方法

攻击者利用域名灵活地实施各类网络攻击,诸多学者针对性地提出了一些基于统计特征和基于关联关系的恶意域名检测方法,但这2类方法在域名属性高阶关系表示方面存在不足,无法准确呈现域间全局高阶关系.针对这类问题,提出一种基于嵌入式特征超图学习的恶意域名检测方法:首先基于域名空间统计特征利用决策树构建域名超图结构,利用决策树倒数第2层节点的输出结果作为先验条件形成超边,快速将域名流量之间的多阶关联关系清晰地表示出来;其次基于超图结构特征对字符嵌入特征进行增强编码,基于域名空间统计特征和域名字符嵌入编码特征从域名数据中挖掘出字符间隐藏的高阶关系;最后结合中国科技网真实的域名系统(domain name system,DNS)流量,对有效性和可行性进行了分析与评估,能够快速高效地检测隐蔽的恶意域名.

一种基于编码单元快速划分的VVC帧内编码方法

相比于高效视频编码(high efficiency video coding,HEVC)标准,新一代编码标准多功能视频编码(versatile video coding,VVC)引入了很多新的技术,其中包括四叉树(quadtree,QT)和多类型树(multi-type tree,MTT)划分,MTT划分由HEVC中的QT划分延伸而来。新划分方法提高了压缩效率,但导致编码时间急剧增加。为了降低编码复杂度,提出了一种结合深度学习方法和MTT方向早期判决的快速帧内编码算法。首先使用轻量级的卷积神经网络(convolutional neural network,CNN)对QT和部分MTT进行预测划分,其余MTT则采用提前预测MTT划分方向的方法作进一步的优化。实验结果表明,所提方法能够大幅降低编码复杂度,相比于原始编码器的编码时间减少了74.3%,且只有3.3%的码率损失,性能优于对比的方法。

基于对比学习的跨语言代码克隆检测方法

代码克隆检测是提高软件开发效率、软件质量和可靠性的重要手段。基于抽象语法树(abstract syntax tree,AST)的单语言克隆检测已经取得了较为显著的效果,但跨语言代码的AST节点存在同义词、近义词且手工标注数据集成本高等问题,限制了现有克隆检测方法的有效性和实用性。针对上述问题,提出一种基于对比学习的树卷积神经网络(contrastive tree convolutional neuraln etwork,CTCNN)的跨语言代码克隆检测方法。该方法首先将不同编程语言的代码解析为AST,并对AST的节点类型和节点值作同义词转换处理,以降低不同编程语言AST之间的差异;同时,采用对比学习扩充负样本并对模型进行训练,使得在小样本数据集下能够最小化克隆对之间的距离,最大化非克隆对之间的距离。最后在公开数据集上进行了评测,精确度达到95.26%、召回率为99.98%、F_(1)为97.56%。结果表明,相较于现有的最好的CLCDSA和C4方法,该模型的检测精度分别提高了432%和3.73%,其F_(1)值分别提升了29.84%和6.29%,证明了所提模型是一种有效的跨语言代码克隆检测方法。

基于CB-Attention的JavaScript恶意混淆代码检测方法

当今JavaScript代码混淆方法日益多样,现有检测方法在对混淆代检测时会出现漏报和误报的情况,为解决该问题,提出一种基于CB-Attention的JavaScript恶意代码检测方法。由SDPCNN模型和BiLSTM+Attention模型构成,SDPCNN对短距离间的语义特征信息进行提取,BiLSTM+Attention获取JavaScript代码中长距离间的语义信息特征。为验证所提方法的有效性,将该方法与其它方法进行对比,对比结果表明,该方法具有较好的检测效果,F1-Score可达98.78%。

基于依赖增强的分层抽象语法树的代码克隆检测

在软件工程领域,基于语义相似的代码克隆检测方法可以降低软件维护的成本并预防系统漏洞,抽象语法树(AST)作为典型的代码抽象表征形式,已成功应用于多种程序语言的代码克隆检测任务,然而现有工作主要利用原始AST提取代码的语义,没有深入挖掘AST中的深层语义和结构信息。针对上述问题,提出一种基于依赖增强的分层抽象语法树(DEHAST)的代码克隆检测方法。首先,对AST进行分层处理,将AST划分得到不同的语义层次;其次,为AST的不同层次添加相应的依赖增强边构建DEHAST,将简单的AST变成具有更丰富程序语义的异构图;最后,使用图匹配网络(GMN)模型检测异构图的相似性,实现代码克隆检测。在BigCloneBench和Google Code Jam两个数据集上的实验结果显示,DEHAST能够检测100%的Type-1和Type-2代码克隆、99%的Type-3代码克隆和97%的Type-4代码克隆;与基于树的方法ASTNN(AST-based Neural Network)相比,F1分数均提高了4个百分点,验证了DEHAST可以较好地完成代码语义克隆检测。

一种快速的多尺度多输入编码树单元互补分类网络

深度神经网络(DNN)已被广泛应用到高效视频编码(HEVC)编码树单元(CTU)的深度划分中,显著降低了编码复杂度。然而现有的基于DNN的CTU深度划分方法却忽略了不同尺度编码单元(CU)间的特征相关性和存在着分类错误累积等缺陷。基于此,该文提出一种多尺度多输入的互补分类网络(MCCN)来实现更高效且更准确的HEVC帧内CTU深度划分。首先,提出一种多尺度多输入的卷积神经网络(MMCNN),通过融合不同尺度CU的特征来建立CU间的关联,进一步提升网络的表达能力。然后,提出一种互补的分类策略(CCS),通过结合二分类和三分类,并采用投票机制来决定CTU中每个CU的最终深度值,有效避免了现有方法中存在的错误累积效应,实现了更准确的CTU深度划分。大量的实验结果表明,该文所提MCCN能够更大程度降低HEVC编码的复杂度,同时实现更准确的CTU深度划分:仅以增加3.18%的平均增量比特率(BD-BR)为代价,降低了71.49%的平均编码复杂度。同时,预测32×32 CU和16×16 CU的深度准确率分别提升了0.65%~0.93%和2.14%~9.27%。

一种基于文本和表达式的相似数学题推荐方法

目前数学习题的相似度计算方法主要利用的是文本信息,忽略了数学表达式信息。然而数学习题常常会包含大量的数学表达式,使得现有的相似度计算方法并不精确。为提高相似习题推荐的准确性,论文针对数学习题中文本信息计算相似度存在不足的问题,提出一种结合文本和数学表达式计算数学习题相似度的方法。该方法利用树形码计算数学表达式之间的相似度,结合文本信息和表达式信息计算得到数学习题的相似度,从而根据top-k进行习题推荐。比较论文的推荐方法和基于文本的推荐,实验结果显示该方法有效提高了推荐的准确率,是一种实用方法。

位置结构导向的多模态代码摘要生成方法

针对软件维护中的自动代码摘要任务,提出了一种创新的模型,旨在解决现有方法在保留源代码语义结构信息方面的不足。该模型采用图神经网络和Transformer技术,以更全面地捕捉代码的语义信息和结构信息。此外,采用字节对编码算法来处理未登录词问题,并通过四元组的形式保留抽象语法树的结构信息。这样的组合使得模型在处理源代码时不仅能够全面地捕捉代码的语义特征,还能够准确地学习到代码的语法结构。在真实Java数据集上的实验结果表明,该模型在BLEU、METEOR和ROUGE指标上均优于基线模型,从而验证了其在生成更准确代码摘要方面的有效性。

基于标准H.265的监控视频压缩编码方法

为了提升视频压缩编码效率,解决成本问题,基于标准H.265研究了一种新的监控视频压缩编码方法。通过H.265/HEVC选取色度帧内预测参数,确定参数格式,同时完成帧内和帧间预测。通过DM模式计算色度亮度参数的分量间相关性,得到五种候选模式,推导最优模式,根据推导结果得出分解结果。通过平滑处理确定像素精度,分别计算DCT2和DST7之间反复切换的参数,匹配监控视频的压缩编码速度与视频监控的播放采集速度,实现视频压缩编码。实验结果表明,标准H.265的压缩效率得到了大幅度提升,节省了大量成本,说明该方法能够提高编码性能,降低视频处理成本。