With the popularity of deep reinforcement learning(DRL), people have great interest in using deep reinforcement learning for application automated testing. However, most automated testing methods based on reinforcement learning ignore text information, use random sampling in experience replay and ignore the characteristics of Android automated testing. To solve above problem, this paper proposes ITPRTesting(Integrated Text feature information and Priority experience in Testing). It extracts the text information in the interface and uses the BERT algorithm to generate sentence vectors. It fuses the interactive control feature diagram(ICFD), which is mentioned in the previous work, and text information as the state required by reinforcement learning. And in reinforcement learning, the priority experience replay is combined, also the traditional priority experience replay is improved. This paper has carried out experiments on 10 open source applications. The experimental results show that ITPRTesting is superior to other methods in statement coverage and branch coverage.

With the improvements of Deep Reinforcement Learning (DRL), there have been tremendous interests in utilizing DRL for automated application testing. However, most automated testing methods based on reinforcement learning have the problem of interacting with invalid UI areas and invalid interactions with controls. To solve this problem, this paper extracts the page features, constructs the Interactive Control Feature Diagram(ICCD); improves the DDQN network structure, adds the residual network, makes the algorithm take the picture as the input, and splits the original single output action(n*w*h) into two successive outputs: the interaction(1,n) and the position(1,w*h); a new reward function which combines the interaction times and the image similarity of ICCD is proposed to explore different UIs and ensure that there will be more than one action will be executed under the same UI. Experiments are carried out on five open source applications. The experimental results show that the proposed method is superior to other methods in code coverage and branch coverage.

Android apps are popular in our daily life, while the testing and maintenance of them is still an open challenge. Random-based testing tools are time-consuming and model-based testing tools are unrealistic to construct all functional behaviors precisely. Existing testing tools based on reinforcement learning not only have difficulty in understanding the business logic of an application, but also face the problem of state explosion during testing. In this paper, we propose SQDroid, a semantic-driven approach for Android apps based Q-learning. SQDroid encourages the Q-learning agent to prefer the actions with functional semantics through a dynamic semantic reward function, which is beneficial to the understand business logic of an app. A state clustering module, employed to compress the state space in Q-table, utilizes the widget attributes on GUI hierarchy to abstract a state. It can reduce the number of states in Q-table and avoid the execution of repetitive actions. We evaluate SQDroid on 48 open-source Android apps. The results show SQDroid outperforms the state-of-the-art model-based/search-based technique/reinforcement learning-based Stoat, Sapienz and Q-testing in terms of code coverage and fault revelation.

combinatorial testing is an effective black box testing technique for the system with large numbers of parameters and their values. However, for significantly complex and key systems, combinatorial testing still owns high complexity. Testing completely the core part of these systems usually is a kind of solution and Variable strength combinatorial test data generation (VS-CTDG) emerges. In this paper, enhanced bird swarm algorithm (EBSA), a variant of Bird Swarm Algorithm (BSA), is employed into the problem. Moreover, a testing requirement reduction is proposed and makes EBSA more suitable into VS-CTDG than ever. Through benchmarks, EBSA is proved an effective approach.

The data quality directly determines the data further analysis and other applications. With the important of data quality, data is a major asset in software applications and information systems. Getting and ensuring high quality data is critical to the software and related business operations. At present, there are many methods and techniques of data quality assessment or evaluation. Based on the evaluation criteria of data quality, the paper mainly starts from the needs of application software and uses the requirements as a guide in the evaluation process. Define evaluation criteria based on user-defined requirements. In this paper, the method proposed in a practical case has been applied, and achieved good practical results.

Dynamic symbolic execution (DSE) is widely used to generate test cases for automated test. However, the practicability of DSE is greatly decreased by path explosion, which is an unavoidable challenge caused by that the number of execution paths increases exponentially with the increase in the size of program. To alleviate this problem, dynamic symbolic execution based on multiple test points (MPDSE) is proposed in this paper to improve the efficiency of DSE by reducing redundant paths. Through analyzing the dependences between test points and branches, the equivalent paths are figured out so that MPDSE can eliminate exploring redundant paths. Our experiments show that MPDSE has advantages in generating test cases compared to traditional DSE.

System and software quality evaluation is an important method of quality assurance, and its standardization provides quality requirements and evaluation supported by quality measurement. ISO/IEC organizations have also published a series of standards on system and software quality requirements and evaluation, which is significant for assuring the quality of system and software. This paper introduces quality evaluation based on SQuaRE series standards, and analyzes the relationships among quality model, measurement, requirements, evaluation. Besides, considerations and suggestions are given when applying the measurement function to quality evaluation. Finally, comparison with different methods of quantifying quality characteristics and sub-characteristics with the measured values is analyzed.

Random walk is an effective network-based method for information mining. However, the computational complexity of random walk limits its application on large scale datasets. This study evaluating accuracy and performance of graphics processing units (GPUs) in accelerating random walk computation on heterogeneous networks. We first introduce a general heterogeneous network to describe user-item relevance model. We apply GPU-accelerated random walk method to this network and investigate the performance of the method. We demonstrate our method via large-scale experiments across MovieLens datasets and obtain a more than 12× speedup. The results shows the effectiveness and promise of the approach.

Dynamic symbolic execution (DSE) is a well-known program test technology which can generate test cases automatically through collecting constraints of the paths represented by symbolic value. Automated test based on DSE targets at covering all paths and branches in a limited time. Although DSE has been investigated for a few years, several challenges still exist which hinder the actual usage. Path explosion is one of the greatest challenges to solve, which decreases the test performance. Analyzing complexity of the execution paths of different test programs, and investigating the factors associated with the test efficiency become very important in the software testing life cycle.

This paper briefly analyzes the security mechanism of android platform and describes permission statement and request of android application. According to android permission mechanism and its shortcoming, the android privilege escalation detection technology and attacking principle is mainly analyzed, and privilege escalation attacking model and architecture of application is put forward. Penetration testing tool Drozer is used to detect application permission. Application package and Manifest file is analyzed to obtain privilege elevation vulnerability. At the same time, sensitive combination permission is distinguished to avoid and exclude the use of malicious code.

Virtualization technology not only is the basis of Cloud computing technology, but also plays an important role in Cloud Testing. Cloud Testing takes advantage of virtualization technology to generate VM (virtual machine) resources in the physical machine, and adopts the corresponding strategies to schedule the VM resources. VM scheduling strategies have a crucial impact on the overall performance of Cloud Testing. The paper first introduces the scheduling process of VM in Cloud Testing, and divides the common scheduling strategies into three categories: center on saving energy, center on load balancing and center on Qos performance. Then the common VM scheduling strategies in current Cloud Testing environment are analyzed. Finally, their advantages and disadvantages are also discussed.

The development and wide application of the internet technology produces a large amount of data, in order to storage and manage these massive data, NoSQL database technology comes into beings and develops rapidly, many manufacturers have introduced many different NoSQL storage solutions. In this context, the paper carries out the work about the testing method of the performance of NoSQL database. Firstly, the paper analyses the new features of the NoSQL database systems. Then, the paper proposes a general testing model for performance testing of the complicated system, which means that the architecture and its business should be considered when the testing work is carried out. Lastly, apply the model to perform the HBase's performance testing in the architecture level, mainly including some elements of the performance, such as the column-oriented data model, the spilt mechanism of the data table, the factor of the data replication. As to the every performance element, design the corresponding testing scenario and execute the testing procedure.

With the development and wide application of the computing technology, performance testing becomes more and more important. Real simulation of the user behavior becomes a concern of the performance testing. The paper introduces the potential model to establish the model of the visit amount which can be used in the performance testing. Firstly, the paper introduces the role of HBase in the search engine business and then illustrates the specific procedure about how to construct the visit amount model. Lastly, test the reading performance of HBase based on the model of the visit amount with the help of YCSB. At the same time, the paper simply introduces some applications of the potential cycle model in other contexts.

Lots of collaborative works are developed by workflow at present. The modeling of workflow is mainly by Petri Net. This paper studies the test coverage of collaborative workflow system to improve the system quality. By analyzing many Petri-Net models, the paper puts forward generic advance Petri Net model. And other workflow Petri Net models can be compatible with the generic Petri Net. Then test coverage strategy is considered and test case is designed according to the strategy. This method not only can meet the requirement of test coverage, but also can assist test case design and further execution of test case. Finally, a typical example is introduced to illustrate effectiveness of this method in collaborative workflow application.

This paper studies online quality measurement in cloud computing environment. The paper analyzes concentration measure evaluation method of the current software quality evaluation system. As is known to all, the existing evaluation technology uses concentrated evaluation before run-time. By testing and analyzing results, the quality is obtained. In this paper, by analyzing the features in cloud computing environment, we consider that service-oriented evaluation should be mainly based on runtime online measurement. The paper puts forward agent-based online measure infrastructure, which is evaluated by distributed assessment in service computing environment A quality model and corresponding online data collection strategy are described. In this approach, the previous quality evaluation, which is concentrated and simulating in a simulated environment, is substituted. Then, a novel measure method of quality data acquisition, which is based on distributed agent technology, is established. So, during online service operation, the obtained data can make measurement results to be accurate and credible.

Test cases are one of the most important assets in the testing process. This paper presents the testing ontology based SWEBOK and software quality model. The management and retrieval of test cases will play a vital role in test cases reuse. The keyword-based, as well as facet-based retrieval cannot meet user's flexible query requirement because of lack of semantic information. SWEBOK provides a broad agreement on the content of the software engineering discipline. At last this paper discusses the management and retrieval of test cases based on the semantic similarity of two test concepts in two ontologies according to difference sets of superconcept, subconcept, extension, and intension.