Automated training-set creation for software architecture traceability problem

Automated trace retrieval methods based on machine-learning algorithms can significantly reduce the cost and effort needed to create and maintain traceability links between requirements, architecture and source code. However, there is always an upfront cost to train such algorithms to detect relevant architectural information for each quality attribute in the code. In practice, training supervised or semi-supervised algorithms requires the expert to collect several files of architectural tactics that implement a quality requirement and train a learning method. Establishing such a training set can take weeks to months to complete. Furthermore, the effectiveness of this approach is largely dependent upon the knowledge of the expert. In this paper, we present three baseline approaches for the creation of training data. These approaches are (i) Manual Expert-Based, (ii) Automated Web-Mining, which generates training sets by automatically mining tactic’s APIs from technical programming websites, and lastly (iii) Automated Big-Data Analysis, which mines ultra-large scale code repositories to generate training sets. We compare the trace-link creation accuracy achieved using each of these three baseline approaches and discuss the costs and benefits associated with them. Additionally, in a separate study, we investigate the impact of training set size on the accuracy of recovering trace links. The results indicate that automated techniques can create a reliable training set for the problem of tracing architectural tactics.     

DCTracVis: a system retrieving and visualizing traceability links between source code and documentation

It is well recognized that traceability links between software artifacts provide crucial support in comprehension, efficient development, and effective management of a software system. However, automated traceability systems to date have been faced with two major open research challenges: how to extract traceability links with both high precision and high recall, and how to efficiently visualize links for complex systems because of scalability and visual clutter issues. To overcome the two challenges, we designed and developed a traceability system, DCTracVis. This system employs an approach that combines three supporting techniques, regular expressions, key phrases, and clustering, with information retrieval (IR) models to improve the performance of automated traceability recovery between documents and source code. This combination approach takes advantage of the strengths of the three techniques to ameliorate limitations of IR models. Our experimental results show that our approach improves the performance of IR models, increases the precision of retrieved links, and recovers more correct links than IR alone. After having retrieved high-quality traceability links, DCTracVis then utilizes a new approach that combines treemap and hierarchical tree techniques to reduce visual clutter and to allow the visualization of the global structure of traces and a detailed overview of each trace, while still being highly scalable and interactive. Usability evaluation results show that our approach can effectively and efficiently help software developers comprehend, browse, and maintain large numbers of links.     

Assessing traceability of software engineering artifacts

The generation of traceability links or traceability matrices is vital to many software engineering activities. It is also person-power intensive, time-consuming, error-prone, and lacks tool support. The activities that require traceability information include, but are not limited to, risk analysis, impact analysis, criticality assessment, test coverage analysis, and verification and validation of software systems. Information Retrieval (IR) techniques have been shown to assist with the automated generation of traceability links by reducing the time it takes to generate the traceability mapping. Researchers have applied techniques such as Latent Semantic Indexing (LSI), vector space retrieval, and probabilistic IR and have enjoyed some success. This paper concentrates on examining issues not previously widely studied in the context of traceability: the importance of the vocabulary base used for tracing and the evaluation and assessment of traceability mappings and methods using secondary measures. We examine these areas and perform empirical studies to understand the importance of each to the traceability of software engineering artifacts.     

基于深度学习的网络流时空特征自动提取方法

流量异常检测是网络入侵检测的主要途径之一,也是网络安全领域的一个热门研究方向。通过对网络流量进行实时监控,可及时有效地对网络异常进行预警。目前,网络流量异常检测方法主要分为基于规则和基于特征工程的方法,但现有方法需针对网络流量特征的变化需重新人工收集规则或 构造特征,工作量大且繁杂。为解决上述问题,该文提出一种基于卷积神经网络和循环神经网络的深度学习方法来自动提取网络流量的时空特征,可同时提取不同数据包之间的时序特征和同一数据包内字节流的空间特征,并减少了大量的人工工作。在 MAWILab 网络轨迹数据集上进行的验证分析结果表明,该文所提出的网络流时空特征提取方法优于已有的深度表示学习方法。     

需求跟踪方法研究

对需求跟踪方法进行了分析、归纳,认为需求跟踪方法应易于掌握、能跟踪复杂系统、能自动生成跟踪链、跟踪链维护简单、有工具和文档支持等特性。     

基于最大最小距离的多中心数据综合增强方法

数据增强是解决数据集不平衡的有效方法，针对现有的数据增强方法存在生成样本越界和随机性差的问题，提出一种基于最大最小距离的多中心数据增强方法MCA。通过计算所有样本的加权密度，减少离群点对最终分类结果的影响，同时将抽样方法与最大最小距离算法相结合选择最优的数据，生成多中心点集，避免生成结果出现样本类别越界的情况，从而拓展样本数据的多样性，并且降低时间复杂度。在此基础上，根据样本的相似性构建权重函数，计算加权平均生成新的样本，解决原有数据集不平衡的问题。在SwedishLeaf数据集和实测数据集上进行实验，结果表明，相比SMOTE、Easy Ensemble、RR等方法，该方法的精确率和召回率均提高了1.17%以上，F1值提高了2%以上，能够有效提高泛化能力，在少数类和多数类样本不平衡率较高的情况下具有较优的分类性能。     

Hybrid evolutionary computation methods for quay crane scheduling problems

Quay crane scheduling is one of the most important operations in seaport terminals. The effectiveness of this operation can directly influence the overall performance as well as the competitive advantages of the terminal. This paper develops a new priority-based schedule construction procedure to generate quay crane schedules. From this procedure, two new hybrid evolutionary computation methods based on genetic algorithm (GA) and genetic programming (GP) are developed. The key difference between the two methods is their representations which decide how priorities of tasks are determined. While GA employs a permutation representation to decide the priorities of tasks, GP represents its individuals as a priority function which is used to calculate the priorities of tasks. A local search heuristic is also proposed to improve the quality of solutions obtained by GA and GP. The proposed hybrid evolutionary computation methods are tested on a large set of benchmark instances and the computational results show that they are competitive and efficient as compared to the existing methods. Many new best known solutions for the benchmark instances are discovered by using these methods. In addition, the proposed methods also show their flexibility when applied to generate robust solutions for quay crane scheduling problems under uncertainty. The results show that the obtained robust solutions are better than those obtained from the deterministic inputs.     

Towards automated traceability maintenance

Traceability relations support stakeholders in understanding the dependencies between artifacts created during the development of a software system and thus enable many development-related tasks. To ensure that the anticipated benefits of these tasks can be realized, it is necessary to have an up-to-date set of traceability relations between the established artifacts. This goal requires the creation of traceability relations during the initial development process. Furthermore, the goal also requires the maintenance of traceability relations over time as the software system evolves in order to prevent their decay. In this paper, an approach is discussed that supports the (semi-) automated update of traceability relations between requirements, analysis and design models of software systems expressed in the UML. This is made possible by analyzing change events that have been captured while working within a third-party UML modeling tool. Within the captured flow of events, development activities comprised of several events are recognized. These are matched with predefined rules that direct the update of impacted traceability relations. The overall approach is supported by a prototype tool and empirical results on the effectiveness of tool-supported traceability maintenance are provided.     

Inter-modelling with patterns

Inter-modelling is the activity of modelling relations between two or more modelling languages. The result of this activity is a model that describes the way in which model instances of these languages can be related. Many tasks in model-driven development can be classified as inter-modelling, for example designing model-to-model transformations, defining model matching and traceability relations, specifying model merging and model weaving, as well as describing mechanisms for inter-model consistency management and model synchronization. This paper presents our approach to inter-modelling in a declarative, relational, visual, and formal style. The approach relies on declarative patterns describing allowed or forbidden relations between two modelling languages. Such specification is then compiled into different operational mechanisms that are tailor-made for concrete inter-modelling scenarios. Up to now, we have used the approach to generate forward and backward transformations from a pattern specification. In this paper we demonstrate that the same specification can be used to derive mechanisms for other inter-modelling tasks, such as model matching and model traceability. In these scenarios the goals are generating the traces between two existing models, checking whether two models are correctly traced, and modifying the traces between two models if they are incorrect.     

基于反射特征的Android测试用例自动生成

随着Android应用软件数量的急速增加, Android应用质量的重要性越来越受到人们的重视.测试是高质量软件的重要保证,而测试用例生成技术是自动化测试的关键.数据显示,在Google Play中有将近88%的Android应用程序使用了反射.然而,现有的Android测试用例自动生成方法通常没有考虑反射技术的使用,无法检测出反射隐藏的恶意行为.为了进一步提高软件质量,本文提出一种新的Android测试用例生成方法,结合反射特征构造Android应用程序多粒度模型,同时对反射关系进行分析,生成能到达反射的调用路径,并利用自适应遗传算法生成覆盖反射路径的测试用例,对含反射特征的Android应用进行测试.为验证本文方法,分别从Android应用多粒度模型有效性及测试方法效率两方面对本文方法有效性进行评估.实验结果表明,本文提出的基于反射特征的Android测试用例自动生成方法对于反射的检测效果更好并且效率更高.     

Cue-Based Segmentation of 4D Cardiac Image Sequences

Medical imaging scanners now exist that can generate 4D cardiac images. Since the heart moves, cardiac anatomy and physiology can be studied using 4D image sequences. Interactive manual 4D image analysis can be time-consuming and error-prone—automatic and semi-automatic methods have many advantages over manual segmentation. This paper describes a procedure for performing semi-automatic image segmentation on 4D image sequences. Our procedure is based on a small set of user-defined image-segmentation cues specified at certain time points in the sequence. These cues are then automatically interpolated or extrapolated for the remaining time points. The complete set of cues is interpreted and used to generate a sequence of image processing operations (such as operators for image enhancement, morphological processing, and region segmentation) that can subsequently segment the 4D image. This procedure permits 4D cardiac image segmentation with only a small amount of user interaction. The proposed approach compares favorably to results generated by defining cues on each individual volume and to results generated completely manually. The 4D approach also requires significantly less interaction time than pure manual analysis.     

Event-based traceability for managing evolutionary change

Although the benefits of requirements traceability are widely recognized, the actual practice of maintaining a traceability scheme is not always entirely successful. The traceability infrastructure underlying a software system tends to erode over its lifetime, as time-pressured practitioners fail to consistently maintain links and update impacted artifacts each time a change occurs, even with the support of automated systems. This paper proposes a new method of traceability based upon event-notification and is applicable even in a heterogeneous and globally distributed development environment. Traceable artifacts are no longer tightly coupled but are linked through an event service, which creates an environment in which change is handled more efficiently, and artifacts and their related links are maintained in a restorable state. The method also supports enhanced project management for the process of updating and maintaining the system artifacts.     

Goal-Centric Traceability: Using Virtual Plumblines to Maintain Critical Systemic Qualities

Successful software development involves the elicitation, implementation, and management of critical systemic requirements related to qualities such as security, usability, and performance. Unfortunately, even when such qualities are carefully incorporated into the initial design and implemented code, there are no guarantees that they will be consistently maintained throughout the lifetime of the software system. Even though it is well known that system qualities tend to erode as functional and environmental changes are introduced; existing regression testing techniques are primarily designed to test the impact of change upon system functionality rather than to evaluate how it might affect qualities such as performance, usability, or portability. The concept of using goal-centric traceability to establish relationships between a set of strategically placed assessment models and system goals is introduced. This paper describes the process, algorithms, and techniques for utilizing goal models to establish executable traces between goals and assessment models, detect change impact points through the use of automated traceability techniques, propagate impact events, and assess the impact of change upon systemic qualities. The approach is illustrated through two case studies respectively representing a small business application and a safety-critical real-time system.     

Tracing conceptual models' evolution in data warehouses by using the model driven architecture

Developing a data warehouse is an ongoing task where new requirements are constantly being added. A widely accepted approach for developing data warehouses is the hybrid approach, where requirements and data sources must be accommodated to a reconciliated data warehouse model. During this process, relationships between conceptual elements specified by user requirements and those supplied by the data sources are lost, since no traceability mechanisms are included. As a result, the designer wastes additional time and effort to update the data warehouse whenever user requirements or data sources change. In this paper, we propose an approach to preserve traceability at conceptual level for data warehouses. Our approach includes a set of traces and their formalization, in order to relate the multidimensional elements specified by user requirements with the concepts extracted from data sources. Therefore, we can easily identify how changes should be incorporated into the data warehouse, and derive it according to the new configuration. In order to minimize the effort required, we define a set of general Query/View/Transformation rules to automate the derivation of traces along with data warehouse elements. Finally, we describe a CASE tool that supports our approach and provide a detailed case study to show the applicability of the proposal.     

Human visual system-based image enhancement and logarithmic contrast measure.

Varying scene illumination poses many challenging problems for machine vision systems. One such issue is developing global enhancement methods that work effectively across the varying illumination. In this paper, we introduce two novel image enhancement algorithms: edge-preserving contrast enhancement, which is able to better preserve edge details while enhancing contrast in images with varying illumination, and a novel multihistogram equalization method which utilizes the human visual system (HVS) to segment the image, allowing a fast and efficient correction of nonuniform illumination. We then extend this HVS-based multihistogram equalization approach to create a general enhancement method that can utilize any combination of enhancement algorithms for an improved performance. Additionally, we propose new quantitative measures of image enhancement, called the logarithmic Michelson contrast measure (AME) and the logarithmic AME by entropy. Many image enhancement methods require selection of operating parameters, which are typically chosen using subjective methods, but these new measures allow for automated selection. We present experimental results for these methods and make a comparison against other leading algorithms.     

A new similarity measure for collaborative filtering to alleviate the new user cold-starting problem

Collaborative filtering is one of the most successful and widely used methods of automated product recommendation in online stores. The most critical component of the method is the mechanism of finding similarities among users using product ratings data so that products can be recommended based on the similarities. The calculation of similarities has relied on traditional distance and vector similarity measures such as Pearson’s correlation and cosine which, however, have been seldom questioned in terms of their effectiveness in the recommendation problem domain. This paper presents a new heuristic similarity measure that focuses on improving recommendation performance under cold-start conditions where only a small number of ratings are available for similarity calculation for each user. Experiments using three different datasets show the superiority of the measure in new user cold-start conditions.     

未知协议的逆向分析与自动化测试

在工业控制、军事通信、金融信息等创新型网络中,大量未知(私有或半私有)协议被广泛采用.对通信协议及其实现进行严格的测试是确保网络系统安全性的重要手段,现有测试手段与方法大多只能针对已知协议进行,未知协议的广泛采用对协议测试提出了挑战.本文提出了针对未知协议的逆向分析与自动化测试方法,其基本思想是基于对协议流量的逆向分析,识别出协议特征,动态生成多维测试数据,自动监控被测系统的运行状态,获得准确的测试结果,为系统安全可靠运行提供依据.具体贡献包括:(1)自动化模糊测试框架;(2)基于协议特征库的逆向分析方法;(3)基于多维变异的测试数据生成方法;(4)基于主动探测的测试执行与异常定位方法.本文设计实现了自动化测试工具UPAFuzz,试验结果表明,UPAFuzz能够基于网络流量实现协议特征的自动识别,并自动生成海量模糊测试数据,对被测系统进行测试;在生成的测试数据量达到千万级时,UPAFuzz的内存占用率为现有模糊测试工具Boofuzz的50%,且其耗时仅为Boofuzz的10%,大大提升了测试执行效率.     

Learning Strategies for Mechanised Building of Decision Procedures

In this paper we present an investigation into whether and how decision procedures can be learnt and built automatically. Our approach consists of two stages. First, a refined brute-force search procedure applies exhaustively a set of given elementary methods to try to solve a corpus of conjectures generated by a stochastic context-free grammar. The successful proof traces are saved. In the second stage, a learning algorithm (by Jamnik et al.) tries to extract a required supermethod (i.e., decision procedure) from the given traces. In the paper, this technique is applied to elementary methods that encode the operations of the Fourier-Motzkin's decision procedure for Presburger arithmetic on rational numbers. The results of our experiment are encouraging.     

Best Practices for Automated Traceability

Automated traceability applies information-retrieval techniques to generate candidate links, sharply reducing the effort of manual approaches to build and maintain a requirements trace matrix as well as providing after-the-fact traceability in legacy documents.The authors describe nine best practices for implementing effective automated traceability.     

一种融合D_BBAS方法的重复缺陷报告检测

重复缺陷报告的自动化检测可以减少开发冗余和维护成本,最近重复缺陷报告的检测倾向于利用深度神经网络,并考虑结构化和非结构化信息来生成混合表示特征。为了更有效获得缺陷报告的非结构化信息的特征,提出一种D_BBAS（Doc2vec and BERT BiLSTM-attention similarity）方法,它基于大规模缺陷报告库训练特征提取模型,生成能反映深层次语义信息的缺陷摘要文本表示集和缺陷描述文本表示集;利用这两个分布式的表示集计算出缺陷报告对的相似度,从而得到两个新的相似度特征;这两个新特征将与基于结构化信息生成的传统特征结合后参与重复缺陷报告的检测。在著名开源项目Eclipse、NetBeans 和Open Office的缺陷报告库上验证了D_BBAS方法的有效性,其中包含超过50万个缺陷报告。实验结果表明,相比于代表性方法,该方法的F1值平均提升了1.7%,证明了D_BBAS方法的有效性。