Drug-drug interactions (DDIs) are a critical safety issue in clinical practice, as they can lead to severe and often unpredictable adverse effects. This risk becomes significantly higher in multi-drug therapies, which are increasingly used in the treatment of complex and chronic diseases such as cancer, cardiovascular disorders, and diabetes. However, identifying DDIs through in vivo studies is costly and time-consuming. In this study, a novel DDI prediction model, Mol2Image, has been proposed that utilizes chemical structure features derived from Simplified Molecular Input Line Entry System (SMILES) representations, including molecular property descriptors and structural fingerprints. The proposed model combines chemical structure information with automated feature learning. Molecular descriptors and structural fingerprints extracted from SMILES representations are converted into visual patterns that capture key chemical characteristics of each drug. These images are then processed by a Convolutional Neural Network (CNN) to learn high-level structural features associated with drug-drug interactions. The model is trained and evaluated using two benchmark DDI datasets: the Drugbank dataset, which consists of 443,046 interactions, and ChCh-Miner, which consists of 48,514 DDIs. Experimental results demonstrate that the proposed model (Mol2Image) achieves competitive performance compared with several state-of-the-art methods. Experimental results demonstrate that the proposed model consistently outperforms existing approaches, achieving accuracies of 0.9608 and 0.9683 using the Drugbank dataset and ChCh-Miner dataset, respectively. Ultimately, Mol2Image provides a highly scalable, strictly structure-centric framework that ensures superior predictive accuracy with minimal computational overhead, operating entirely independently of clinical data.

Recent advancements in single-cell transcriptomics have revolutionized data generation, resulting in the production of vast amounts of single-cell RNA sequencing (scRNA-seq) data. The intricate biology of tissues and organs may be better understood through the analysis of such data, which aims to not only identify known cell types but also uncover novel ones. In this paper, we outline the standard single-cell RNA-seq data analytic process with particular focus on the critical role of unsupervised clustering, since it has a significant influence over subsequent analyses and the resulting biological insights. However, navigating the computational landscape for clustering single-cell transcriptomic data poses significant challenges. Thus, in this article, we provide a simplified exploration of key computational challenges and considerations within the realm of scRNA-seq data clustering, emphasizing the application and relevance of selected unsupervised methods in current research, while acknowledging the potential for further investigation of the biological aspects in this field. Furthermore, we present recommendations on how such challenges may be addressed. This work not only sheds light on the complexities of scRNA-seq data cluster analysis, but also provides insights into how to effectively address and surmount these challenges.

Web services composition has drawn a great attention in computing industries to build complex and large systems. However, web services composition modeling has major challenges, including dependency determination, complexity, user requests dependency, handling cycles within a composition, service redundancy and scalability concerns. The service dependency graph (SDG) between services in a repository should be accurate to ensure the quality of composition and the associated user requests’ responses. Despite of the crucial importance of accurate dependencies for adequate web services compositions, current modeling approaches do not provide any metric to evaluate the dependency between services for quality estimation. In this paper, the Mutual Information-based Services Dependency (MISD) model is proposed as a graph-based modeling approach, independent to any given user request. It constructs services dependency graphs based on Web Services Mutual Information (WSMI), a proposed modified version of mutual information as the dependency metric, along with other criteria for an accurate, efficient dependency evaluation. It finds the optimum composition representing the structure of the web services composition in a repository rather than the path of given user requests. The experimental dependency analysis emphasizes the efficiency of the generated OC and accuracy of the constructed SDG to be 76% and 86% higher than the state-of-the-art models respectively. The time cost to build the SDG and to find the OC is reduced dramatically up to 99% for different public repositories compared to the state-of-the-art studies as the number of user requests increases.

Abstract Cloud testing is gaining much attention in both academia and industry as an emerging concept in the field of software testing. Cloud testing implies leveraging the resources of the cloud computing environment to overcome deficiencies of the traditional testing approaches. As a result, testing‐as‐a‐service (TaaS) is introduced as a service model that conducts all testing activities in a fully automated manner using cloud‐based resources. Performance testing is a type of software testing that validates the performance characteristics of the application under test (AUT) when subjected to different workloads during its operation. Performance characteristics include throughput, response time, and resource utilization of the AUT under a certain workload. This paper focuses on reviewing the literature related to the provision of performance testing as a service (P‐TaaS). In this study, we survey the previous work related to cloud‐based performance testing. We show the strengths and weaknesses of the current research. Besides, we compare the P‐TaaS with the traditional performance testing methodologies. A detailed discussion of the benefits and challenges of P‐TaaS is introduced along with identifying the research gaps and the future directions that can be adopted.

Abstract The service level agreement (SLA) is an agreement between clients and the service provider, which defines the minimum performance and availability requirements for software. The service provider should have effective strategies for placement, migration, and replication of the tenants to reduce their operational costs, maximize the utilization of their hardware and software resources and accordingly meet the SLA requirements with slight SLA violations. In this research, a clustered‐based multi‐tenant database management system (CB‐MT DBMS) is proposed. Additionally, a dynamic proactive provisioning technique is built using three different prediction models: The Recursive Window Forecasting Autoregressive Integrated Moving Average (ARIMA) model, Exponential Moving Average (EMA) model, and the proposed Recurrent Neural Network (RNN) with Long Short‐Term Memory (LSTM) cells model. Various experimental scenarios with different datasets have been conducted to prove that the RNN model with LSTM cells is a promising solution in multi‐tenant environments, where the tenants have irregular workload patterns. Experimental results firstly show that the RNN model accuracy is superior to their counterparts (i.e., ARIMA and EMA models) when applied to multi‐tenant database workloads generated using TPC benchmarks, as it reduces the prediction error value which is computed using the Mean Absolute Percentage Error (MAPE) and the Root Mean Square Error (RMSE) metrics. Secondly, Experimental results prove that the RNN prediction model accuracy is superior to their counterparts for detecting SLA violation values and windows using different SLA values.

Summary The performance and efficiency of the distributed database system (DDBS) design are dependent on proper data fragmentation, reallocation, and replication of global relations. In this research, a Cluster Based Distributed and Parallel Database System (CB‐DDBS) architecture over a cloud environment is proposed. The proposed CB‐DDBS architecture processes the client's query requests, which access the clustered DDBS from anywhere. It also allows vertical and horizontal fragmentation, allocation, and replication decisions to be taken statically at the initial stage of the design. In addition, it allows migration and/or replication decisions to be taken by each cluster independently of other clusters using the proposed Optimal Fragment Reallocation and Replication (OFRAR) Algorithm. The proposed CB‐DDBS architecture and the proposed OFRAR Algorithm are tested in both Amazon cloud environment and a simulated environment. Experimental results show that the proposed OFRAR Algorithm efficiently reduces the communication costs for typical access patterns, the overloads of the sites, and the frequency and time spent on fragment migration over the network sites by imposing a stricter condition for fragment reallocation and replication, resulting in a great overall improvement in the DDBS performance. It also shows that the proposed CB‐DDBS architecture results in a significant reduction of the execution time needed for the transactions and for noticing an error.

Abstract Recently, testing mobile applications is gaining much attention due to the widespread of smartphones and the tremendous number of mobile applications development. It is essential to test mobile applications before being released for the public use. Graphical user interface (GUI) testing is a type of mobile applications testing conducted to ensure the proper functionality of the GUI components. Typically, GUI testing requires a lot of effort and time whether manual or automatic. Cloud computing is an emerging technology that can be used in the software engineering field to overcome the defects of the traditional testing approaches by using cloud computing resources. As a result, testing‐as‐a‐service is introduced as a service model that conducts all testing activities in a fully automated manner. In this paper, a system for mobile applications GUI testing based on testing‐as‐a‐service architecture is proposed. The proposed system performs all testing activities including automatic test case generation and simultaneous test execution on multiple virtual nodes for testing Android‐based applications. The proposed system reduces testing time and meets fast time‐to market constraint of mobile applications. Moreover, the proposed system architecture addresses many issues such as maximizing resource utilization, continuous monitoring to ensure system reliability, and applying fault‐tolerance approach to handle occurrence of any failure.

Summary With the proliferation of cloud storage services, data integrity verification becomes increasingly significant in order to guarantee the availability and correctness of the outsourced data. Recently, many auditing schemes have been proposed to verify data integrity without possessing or downloading the outsourced data files. However, such schemes in existence assume that there is a single data owner that can update the file and compute the integrity tags. They do not consider the efficiency of user revocation when auditing multi‐owner data in the cloud. In addition, existing schemes only support dynamic data operations over fixed‐size and single‐copy data blocks. Thus, every small update may need updating the tags for all file blocks in all replicas, which in turn causes higher storage and communication overheads. This paper proposes a public and dynamic auditing scheme that supports fully dynamic data operations over variable‐size data blocks for replicated and multi‐owner cloud storage. Moreover, the proposed scheme can also support efficient user revocation. By supporting batch auditing, the proposed scheme can handle multiple auditing tasks simultaneously and divide an auditing task into multiple sub‐tasks in order to increase the detection probability. Experimental results prove the effectiveness and efficiency of the proposed system. Copyright © 2016 John Wiley & Sons, Ltd.

Summary In this paper, we propose the Software‐as‐a‐Service (SaaS) Recommender (SaaSRec), a personalized reputation‐based QoS‐aware recommender system (RS) for SaaS services. SaaSRec semantically processes user requests in order to find business‐oriented matching services, which are then filtered to satisfy the user QoS requirements and service characteristics. Subsequently, hybrid filtering is utilized to validate the services set on the basis of services metadata, reputation and user interests. Finally, the recommended set of services is ranked using a unique combination of factors: Relevance to user profile, service reputations and service cost. Moreover, we propose a new method for calculating the service reputation from the objective time‐weighted user feedbacks. SaaSRec addresses many challenges faced by the generic RSs: User cold‐start problem, limited content analysis and low performance. In respect of service RSs, SaaSRec tackles the disregarding of relevant factors to services recommendation: Cloud service characteristics, user physical location, service reputation and user interests. Moreover, SaaSRec provides a hybrid justification for the recommended services to increase the user's acceptance. Experimental evaluation against a real‐world services dataset has been carried out, and the results show that the proposed recommendation approach surpasses other collaborative filtering‐based recommendation approaches in respect of both precision and recall. This performance improvement was verified using different matrix density levels and number of recommendations. Copyright © 2016 John Wiley & Sons, Ltd.

Summary Cloud users are overwhelmed with great numbers of cloud services. Service recommender systems evaluate the services that provide same functionalities according to the user requirements. A key enabler to accurate recommendation in recommender systems is the appropriate determination of similar users. This paper contributes to the personalized cloud services recommendation area. In specific, we introduce a user‐based similarity measure that integrates relevant similarity aspects: user demographic information, service ratings, and user interest. The proposed similarity measure is used in a hybrid collaborative filtering (CF) approach that leverages the advantages of both model‐ and memory‐based approaches to improve the recommendation process. Experimental evaluation on real‐world services data set shows that the proposed approach outperforms other CF approaches in respect of the prediction accuracy and recommendation time while maintaining better or same coverage.

Abstract Regression testing repeatedly executes test cases of previous builds to validate that the original features are not affected with any new changes. In recent years, regression testing has seen a remarkable progress with the increasing popularity of agile methods, which stress the central role of regression testing in maintaining software quality. The optimum case for regression testing in agile context is to run regression set at the end of each sprint and release, which requires a lot of cost and time. In this paper, we present an automated agile regression testing approach on both the sprints and release levels. The proposed approach addresses both weighted sprint test cases prioritization technique, which prioritizes test cases based on several parameters having real practical weight for testers, and Cluster‐based Release Test cases Selection technique that clusters user stories based on the similarity of covered modules to solve the scalability issue. Test cases are then selected based on issues logged for failed test cases using text mining techniques. The proposed approach achieves enhancement for both the prioritization and selection of test cases for agile regression testing. Copyright © 2016 John Wiley & Sons, Ltd.

Summary User authentication is a crucial requirement for cloud service providers to prove that the outsourced data and services are safe from imposters. Keystroke dynamics is a promising behavioral biometrics for strengthening user authentication, however, current keystroke based solutions designed for certain datasets, for example, a fixed length text typed on a traditional personal computer keyboard and their authentication performances were not acceptable for other input devices nor free length text. Moreover, they suffer from a high dimensional feature space that degrades the authentication accuracy and performance. In this paper, a keystroke dynamics based authentication system is proposed for cloud environments that is applicable to fixed and free text typed on traditional and touch screen keyboards. The proposed system utilizes different feature extraction methods, as a preprocessing step, to minimize the feature space dimensionality. Moreover, different fusion rules are evaluated to combine the different feature extraction methods so that a set of the most relevant features is chosen. Because of the huge number of users' samples, a clustering method is applied to the users' profile templates to reduce the verification time. The proposed system is applied to three different benchmark datasets using three different classifiers. Experimental results demonstrate the effectiveness and efficiency of the proposed system. Copyright © 2015 John Wiley & Sons, Ltd.

The Multi-agents computing paradigm offers support for large scale, widely distributed, high-performance computational systems. Several of such architectures and frameworks have been developed aimed at primarily computations in support of scientific, engineering calculations and managements. On the other hand agent software provides a number of issues including; autonomous, integrity, flexibility, ease of use and playing a number of different roles within a web scripting languages as an essential component of interactive web content to bridge the gap in these technologies. In this paper we address the architecture that support the publishing, description, managing, and communication between the agents of the business environment based on the feature that exists in the multi-agents and the agents' behaviours. This research based on multi agents approach and its associated agent description languages are used to facilitate the construction and management of ad-hoc federated software services.