Showing papers in "Computing in 2018"
TL;DR: In this paper, the authors present new methods for predicting the remaining time of running cases, assuming process stationarity, which achieves state-of-the-art performances and two other methods which are able to make predictions even with non-stationary processes.
Abstract: The ability to know in advance the trend of running process instances, with respect to different features, such as the expected completion time, would allow business managers to timely counteract to undesired situations, in order to prevent losses. Therefore, the ability to accurately predict future features of running business process instances would be a very helpful aid when managing processes, especially under service level agreement constraints. However, making such accurate forecasts is not easy: many factors may influence the predicted features. Many approaches have been proposed to cope with this problem but, generally, they assume that the underlying process is stationary. However, in real cases this assumption is not always true. In this work we present new methods for predicting the remaining time of running cases. In particular we propose a method, assuming process stationarity, which achieves state-of-the-art performances and two other methods which are able to make predictions even with non-stationary processes. We also describe an approach able to predict the full sequence of activities that a running case is going to take. All these methods are extensively evaluated on different real case studies.
81 citations
TL;DR: Using the mathematical model of BP neural networks, a prediction model based on satellite remote sensing data for the pollutant concentration in regional scale was explored, and the forecast for Fuling 3-h PM2.5 concentration was realized and the algorithm effectively establishes the correlation between AOD and PM 2.5.
Abstract: PM2.5 hadn’t received much attention until 2013 when people started to understand its dreadful impacts to human health. According to the meteorological monitoring data of PM2.5 from September 9, 2016 to September 9, 2017 in Fuling district, Chongqing, this paper analyzed the impact of temperature, humidity and the power of wind on PM2.5. Using the mathematical model of BP neural networks, a prediction model based on satellite remote sensing data for the pollutant concentration in regional scale was explored, and the forecast for Fuling 3-h PM2.5 concentration was realized. The algorithm effectively establishes the correlation between AOD and PM2.5 concentration, and it suppresses the overfitting phenomenon very well, as well as it makes up the limitation of machine learning for single site prediction.
56 citations
TL;DR: An ILP-based process discovery approach, based on region theory, that guarantees to discover relaxed sound workflow nets and devise a filtering algorithm that is able to cope with the presence of infrequent, exceptional behaviour.
Abstract: Process mining is concerned with the analysis, understanding and improvement of business processes. Process discovery, i.e. discovering a process model based on an event log, is considered the most challenging process mining task. State-of-the-art process discovery algorithms only discover local control flow patterns and are unable to discover complex, non-local patterns. Region theory based techniques, i.e. an established class of process discovery techniques, do allow for discovering such patterns. However, applying region theory directly results in complex, overfitting models, which is less desirable. Moreover, region theory does not cope with guarantees provided by state-of-the-art process discovery algorithms, both w.r.t. structural and behavioural properties of the discovered process models. In this paper we present an ILP-based process discovery approach, based on region theory, that guarantees to discover relaxed sound workflow nets. Moreover, we devise a filtering algorithm, based on the internal working of the ILP-formulation, that is able to cope with the presence of infrequent, exceptional behaviour. We have extensively evaluated the technique using different event logs with different levels of exceptional behaviour. Our experiments show that the presented approach allows us to leverage the inherent shortcomings of existing region-based approaches. The techniques presented are implemented and readily available in the HybridILPMiner package in the open-source process mining tool-kits ProM (
http://promtools.org
) and RapidProM (
http://rapidprom.org
).
56 citations
TL;DR: This paper classify, survey, model and compare the most relevant and recent QoS-based routing protocols proposed in the framework of WBAN, and provides a study of adaptability of the surveyed protocols related to the healthcare sector.
Abstract: Wireless Body Area Network (WBAN) constitutes a set of sensor nodes responsible for monitoring human physiological activities and actions. The increasing demand for real time applications in such networks stimulates many research activities in quality-of-service (QoS) based routing for data delivery. Designing such scheme of critical events while preserving the energy efficiency is a challenging task due to the dynamic of the network topology, severe constraints on power supply and limited in computation power and communication bandwidth. The design of QoS-based routing protocols becomes an essential part of WBANs and plays an important role in the communication stacks and has significant impact on the network performance. In this paper, we classify, survey, model and compare the most relevant and recent QoS-based routing protocols proposed in the framework of WBAN. A novel taxonomy of solutions is proposed, in which the comparison is performed with respect to relevant criteria. An analytical model is proposed in order to compare the performances of all the solutions. Furthermore, we provide a study of adaptability of the surveyed protocols related to the healthcare sector.
55 citations
TL;DR: This paper presents the architectural design of WITS, the core algorithms, along with the solutions to the technical challenges in the system implementation, for convenient and efficient care delivery.
Abstract: Over the past few years, activity recognition techniques have attracted unprecedented attentions. Along with the recent prevalence of pervasive e-Health in various applications such as smart homes, automatic activity recognition is being implemented increasingly for rehabilitation systems, chronic disease management, and monitoring the elderly for their personal well-being. In this paper, we present WITS, an end-to-end web-based in-home monitoring system for convenient and efficient care delivery. The system unifies the data- and knowledge-driven techniques to enable a real-time multi-level activity monitoring in a personalized smart home. The core components consist of a novel shared-structure dictionary learning approach combined with rule-based reasoning for continuous daily activity tracking and abnormal activities detection. WITS also exploits an Internet of Things middleware for the scalable and seamless management and learning of the information produced by ambient sensors. We further develop a user-friendly interface, which runs on both iOS and Andriod, as well as in Chrome, for the efficient customization of WITS monitoring services without programming efforts. This paper presents the architectural design of WITS, the core algorithms, along with our solutions to the technical challenges in the system implementation.
53 citations
TL;DR: The main role of the model is to estimate the time needed to run a set of tasks in cloud and in turn reduces the processing cost, which demonstrates that the approach outperforms previous scheduling methods by a significant margin.
Abstract: We address in this paper the task-scheduling in cloud computing. This problem is known to be $${\mathcal {NP}}$$
-hard due to its combinatorial aspect. The main role of our model is to estimate the time needed to run a set of tasks in cloud and in turn reduces the processing cost. We propose a genetic approach for modelling and optimizing a task-scheduling problem in cloud computing. The experimental results demonstrate that our solution successfully competes with previous task-scheduling algorithms. For this, we develop a decision support system based on the core of CloudSim. In terms of processing cost, the obtained results show that our approach outperforms previous scheduling methods by a significant margin. In terms of makespan, the obtained schedules are also shorter than those of other algorithms.
47 citations
TL;DR: Experimental results show that the proposed Paragraph Vector method outperforms word-based embeddings and character n-gram based linear models, which are among the most effective approaches for identifying the writing style of an author.
Abstract: Recently, document embeddings methods have been proposed aiming at capturing hidden properties of the texts. These methods allow to represent documents in terms of fixed-length, continuous and dense feature vectors. In this paper, we propose to learn document vectors based on n-grams and not only on words. We use the recently proposed Paragraph Vector method. These n-grams include character n-grams, word n-grams and n-grams of POS tags (in all cases with n varying from 1 to 5). We considered the task of Cross-Topic Authorship Attribution and made experiments on The Guardian corpus. Experimental results show that our method outperforms word-based embeddings and character n-gram based linear models, which are among the most effective approaches for identifying the writing style of an author.
39 citations
TL;DR: A new diabetic diagnosis system which combines a newly proposed temporal feature selection and temporal fuzzy ant miner tree (TFAMT) classifier for effective decision making in type-2 diabetes analysis is introduced.
Abstract: Diabetic is becoming a very serious disease today for the most of people all over the world due to the unhealthy food habits. For predicting the diabetes, we introduce a new diabetic diagnosis system which combines a newly proposed temporal feature selection and temporal fuzzy ant miner tree (TFAMT) classifier for effective decision making in type-2 diabetes analysis. Moreover, a new temporal weighted genetic algorithm is proposed in this work for enhancing the detection accuracy by preprocessing the text and image data. Moreover, intelligent fuzzy rules are extracted from the weighted temporal capabilities with ant miner fuzzy decision tree classifier, and then fuzzy rule extractor is used to reduce the variety of functions in the extracted regulations. We empirically evaluated the effectiveness of the proposed TFAMT–TWGA model using the UCI Repository dataset and the collected retinopathy image dataset. The outcomes are analyzed and as compared with other exiting works. Furthermore, the detection accuracy is proven by way of using the ten-fold cross validation.
32 citations
TL;DR: A new paradigm of software service engineering, Requirement-Engineering Two-Phase of Service Engineering Paradigm (RE2SEP), which includes service oriented requirement engineering, domain oriented service Engineering, and the development approach of software services is presented.
Abstract: In the big data era, servitization becomes one of the important development trends of the IT world. More and more software resources are developed and existed in the format as services on the Internet. These services from multi-domains and multi-networks are converged as a huge complicated service network or ecosystem, which can be called as Big Service. How to reuse the abundant open service resources to rapidly develop the new applications or comprehensive service solutions to meet massive individualized customer requirements is a key issue in the big data and big service ecosystem. Based on analyzing the ecosystem of big service, this paper presents a new paradigm of software service engineering, Requirement-Engineering Two-Phase of Service Engineering Paradigm (RE2SEP), which includes service oriented requirement engineering, domain oriented service engineering, and the development approach of software services. By means of the RE2SEP approach, the adaptive service solutions can be efficiently designed and implemented to match the requirement propositions of massive individualized customers in Big Service ecosystem. A case study of the RE2SEP applications, which is a project on citizens mobility service in smart city environment, is also given in this paper. The RE2SEP paradigm will change the way of traditional life-cycle oriented software engineering, and lead a new approach of software service engineering.
27 citations
TL;DR: An improved content-based model is proposed in this paper incorporating both semantics and context, and is evaluated using metrics and paralleled with the current methods grounded on the content.
Abstract: The existing content-based recommendation methods have two major limitations. First, due to the defects of the items and the user model matching algorithms, the recommendation results are very narrow. Second, scant attention is paid to the scenario, making the recommendation system not context-aware. It is essential to improve user satisfaction through high-quality recommendation. In this paper, two state-of-the-art methods are analyzed and extended to enhance recommendation performance. The first method is the context-aware recommender, which integrates context information into the recommendation process. The second method is the semantic analysis-based recommender, which incorporates domain semantics. Despite their compatibility, the challenge is to combine them in a way that will fully exploit their potential. An improved content-based model is proposed in this paper incorporating both semantics and context. Context-aware recommendation is performed to improve sensitivity to the context. Semantic relevance-based instance similarity is computed to address the problem of narrowness. The proposed recommendation system is evaluated using metrics (for instance, recall metric) and paralleled with the current methods grounded on the content. Results demonstrate the superiority of the proposed system in terms of accuracy.
26 citations
TL;DR: A methodology is proposed using RST to identify the efficient features for telecommunication customer churn prediction and it is identified that the proposed system designed with combining attribute selection with ensemble classification techniques works fine with classification accuracy of 95.13% compared to any single model.
Abstract: Rough set theory (RST) can be viewed as one of the classical set theory for handling with imprecision knowledge. The theory has discovered applications in numerous areas, for example, engineering, industries, environment and others. Churn in telecommunication sector, customer switching from one service provider to another. Predicting telecom customer churn is challenging due to the huge and inconsistent nature of the data. Churn prediction is crucial for telecommunication companies in order to build an efficient customer retention plan and apply successful marketing strategies. In this article, a methodology is proposed using RST to identify the efficient features for telecommunication customer churn prediction. Then the selected features are given to the ensemble-classification techniques such as Bagging, Boosting, Random Subspace. In this work the duke university-churn prediction data set is considered for performance evaluation and three sets of experiments are performed. Finally the performance of the proposed model is evaluated based on the following metrics such as true churn, false churn, specificity, precision and accuracy and it is identified that Proposed system designed with combining attribute selection with ensemble classification techniques works fine with classification accuracy of 95.13% compared to any single model.
TL;DR: The experimental results indicate that proposed method for segmentation of powdery mildew disease affected area from leaf image of cherry crops is convincing and computationally cheap.
Abstract: There are different reasons like pests, weeds, and diseases which are responsible for the loss of crop production. Identification and detection of different plant diseases is a difficult task in a large crop field and it also requires an expert manpower. In this paper, the proposed method uses adaptive intensity based thresholding for automatic segmentation of powdery mildew disease which makes this method invariant to image quality and noise. After the segmentation of powdery mildew disease from leaf images, the affected area is quantified which makes this method efficient for grading the level of disease infection. The proposed method is tested on the comprehensive dataset of leaf images of cherry crops, which achieved good accuracy of 99%. The experimental results indicate that proposed method for segmentation of powdery mildew disease affected area from leaf image of cherry crops is convincing and computationally cheap.
TL;DR: A low complexity heuristic scheduling algorithm called GeoDis is presented which allows data locality to cope with the data transfer requirement to achieve a greater performance on the makespan.
Abstract: Today, data-intensive applications rely on geographically distributed systems to leverage data collection, storing and processing. Data locality has been seen as a prominent technique to improve application performance and reduce the impact of network latency by scheduling jobs directly in the nodes hosting the data to be processed. MapReduce and Dryad are examples of frameworks which exploit locality by splitting jobs into multiple tasks that are dispatched to process portions of data locally. However, as the ecosystem of big data analysis has shifted from single clusters to span geo-distributed data centers, it is unavoidable that data may still be transferred through the network in order reduce the schedule length. Nevertheless, there is a lack of mechanism to efficiently blend data locality and inter-data center data transfer requirement in the existing scheduling techniques to address data-intensive processing across dispersed data centers. Therefore, the objective of this work is to propose and solve the makespan optimization problem for data-intensive job scheduling on geo-distributed data centers. To this end, we first formulate the task placement and the data access as a linear programming and use the GLPK solver to solve it. We then present a low complexity heuristic scheduling algorithm called GeoDis which allows data locality to cope with the data transfer requirement to achieve a greater performance on the makespan. The experiments with various realistic traces and synthetic generated workload show that GeoDis can reduce makespan of processing jobs by 44% as compared to the state-of-the-art algorithms and remain within $$91\%$$
closer to the optimal solution by the LP solver.
TL;DR: A hierarchical model-based strategy is proposed to evaluate distinct metrics by means of the composition of continuous-time Markov chains, reliability block diagrams and stochastic Petri nets, and the results obtained show that the proposed approach is indeed a good approximation to the measures obtained from the experiments conducted in a real cloud environment.
Abstract: Cloud computing brings new technologies and concepts that support communication services and data storage. Services like OneDrive, Google Drive and DropBox increase data availability and provide new features as synchronization and collaboration. These services require high availability and performance characteristics like high throughput and low probability that a timeout occurs, since it is fundamental to guarantee both business continuity and uninterrupted public services. In this research, we aim at evaluating availability and performance-related metrics for private cloud storage services. A hierarchical model-based strategy is proposed to evaluate distinct metrics by means of the composition of continuous-time Markov chains, reliability block diagrams and stochastic Petri nets. A case study is presented to illustrate the applicability of the proposed models through a cloud storage service hosted in the Eucalyptus platform. We also adopt availability importance index to identify the most critical components in relation to the system availability. Our numerical analyses indicate that, for instance, the adoption of redundant components reduces the probability that timeouts occur and the probability that users are attended due to failures. Furthermore, the results obtained from the stochastic models show that the proposed approach is indeed a good approximation to the measures obtained from the experiments conducted in a real cloud environment.
TL;DR: Its service oriented architecture ensures an easier modification for rapid updating and better performance and guarantees saving the search time and a better exploitation of the provider offerings thanks to a dedicated Cloud service description ontology.
Abstract: In this paper, we propose a Focused Crawler for Cloud service Discovery (FC4DC). Its service oriented architecture ensures an easier modification for rapid updating and better performance. Furthermore, the proposed crawler guarantees saving the search time and a better exploitation of the provider offerings thanks to a dedicated Cloud service description ontology. We finally present some experiments to evaluate the proposed crawler and demonstrate its effectiveness and efficiency.
TL;DR: The empirical results demonstrate that the proposed heuristics can alleviate the coincidental correctness problem and improve the accuracy of SBFL techniques.
Abstract: Despite the proven applicability of the spectrum-based fault localization (SBFL) methods, their effectiveness may be degraded due to the presence of coincidental correctness, which occurs when faults fail to propagate, ie, their execution does not result in failures This article aims at improving SBFL effectiveness by mitigating the effect of coincidentally correct test cases In this regard, given a test suite in which each test has been classified as failing or passing and each faulty program has a single-bug, we present a program slicing-based technique to identify a set of program entities that directly affect the program output when executed with failing test cases, called failure candidate causes (FCC) We then use FCC set to identify test cases that can be marked as being coincidentally correct These tests are identified based on two heuristics: the average suspiciousness score of the statements that directly affect the program output and the coverage ratio of those statements To evaluate our approach, we used several evaluation metrics and conducted extensive experiments on programs containing single and multiple bugs, including both real and seeded faults The empirical results demonstrate that the proposed heuristics can alleviate the coincidental correctness problem and improve the accuracy of SBFL techniques
TL;DR: A management strategy that focuses on the full life cycle management of geotechnical data together with the BIM model to improve the accuracy of decision making in the design, construction, operation and management stages of a construction project is put forward.
Abstract: For the convenience of collaborative design, virtual construction, construction process simulation and management, Building Information Modeling (BIM) is becoming an important tool in civil engineering. The current BIM team seems to neglect the geotechnical aspect of the model, which can result in costly mistakes, especially when the project is infrastructure based. The barrier between the BIM team and the geotechnical data provider is the difficulty in extracting and assimilating data from the archived geotechnical data, which is mainly in the form of geotechnical investigation reports and geological sections. Furthermore, the geotechnical data exposed from the construction can not be linked to the original data conveniently to correct the interpretated errors in the geotechnical data, and the monitoring data cannot be combined with the original geotechnical data to find the development trend of the monitoring variables. All of this indicates that current management strategy of geotechnical data should be improved upon. Therefore, a management strategy that focuses on the full life cycle management of geotechnical data together with the BIM model to improve the accuracy of decision making in the design, construction, operation and management stages of a construction project is particularly important. In this paper, we put forward a management strategy of geotechnical data that can help to integrate geotechnical information into the BIM of a construction project in order to realize the full life cycle management of geotechnical information. In this strategy, the geotechnical data from the geotechnical investigation is archived in the form of a centralized geotechnical database and an informative geotechnical model. The centralized geotechnical database is targeted to manage the factual data, base data, result data, and metadata. The informative geotechnical model can facilitate the geotechnical being used in the three-dimensional visualization environment where the collaborative design and virtual construction are accomplished. We discuss the specific workflow of building a centralized geotechnical database and an informative geotechnical model. Finally, we use the management of the geotechnical data in a hydropower station for experimental studies to verify the proposed management strategy. The result shows that it is advantageous to manage the geotechnical data in the proposed management strategy for the BIM of a construction project.
TL;DR: A comparative study of the various data gathering protocols that aim at balancing energy consumption using mixed/hybrid transmission schemes and highlights the aspects of the existing approaches that can be worked upon to achieve a more energy efficient and energy balanced data gathering approach in future.
Abstract: Wireless sensor networks (WSN) consists of small battery powered nodes. Energy efficiency and energy balancing are the most stringent needs of WSN for prolonging its lifetime. Due to many-to-one communication pattern in multi-hop communication, energy consumption is unbalanced in the network. Nodes which are closer to sink deplete their energy much faster than the nodes further away. This paper reviews and presents a comparative study of the various data gathering protocols that aim at balancing energy consumption using mixed/hybrid transmission schemes. It also highlights the aspects of the existing approaches that can be worked upon to achieve a more energy efficient and energy balanced data gathering approach in future.
TL;DR: A portable and automated sleep apnea detector that was designed and evaluated and achieved an average accuracy, sensitivity and specificity of 87.5, 79.5 and 90.8% respectively.
Abstract: Obstructive sleep apnea is a highly prevalent sleep related breathing disorder and polysomnography is the gold standard exam for diagnosis. Despite providing results with high accuracy this multi-parametric test is expensive, time consuming and does not fit with the new tendency in health care that is changing the focus to prevention and wellness. Home health care is seen as a possible way to address this problematic by using minimal invasive devices, providing low cost of diagnosis and higher accessibility. To address this, a portable and automated sleep apnea detector was designed and evaluated. The device uses one SpO2 sensor and the analysis is based on the connection between oxygen saturation and apnea events. The measured signals are received in a field-programmable gate array that checks for errors and implements the communication protocols of two wireless transmitters. Two solutions were implemented for processing the data: one based on a smartphone (due to availability and low cost) and another based on a personal computer (for a higher computation capability). The algorithms were implemented in Java, for the smartphone, and in Python, for the computer. Both implementations have a graphical user interface to simplify the device operation. The algorithms were tested using a database consisting of 70 patients with the SpO2 signal collected in a Hospital. The algorithm performance achieved an average accuracy, sensitivity and specificity of 87.5, 79.5 and 90.8% respectively.
TL;DR: The investment cost function is formulates which reflects the characteristics and impacts of online advertising spillover effect to enterprises and the improved Lanchester model is used based on the investment cost functions.
Abstract: Online advertising has become an important marketing instrument for many enterprises, and the impact of enterprises’ online advertising has been increasing rapidly. Significant long-term enterprise profits are dynamically determined by the continuous online advertisement investment strategies implemented. This paper formulates the investment cost function which reflects the characteristics and impacts of online advertising spillover effect to enterprises. Then the improved Lanchester model is used based on the investment cost function. According to the model, our research calculates the Nash equilibrium and does the numerical analysis under open-loop strategy and closed-loop strategy. (1) With the change of time and under the condition of open-loop and closed-loop, when the spillover effect level is higher, the investment amount on fixed-position online advertisement by the enterprises becomes opposite. (2) When the level of spillover effect is strong, the change in the market share of competitive enterprises is related to the initial market share under the condition of open-loop. Under the condition of closed-loop, the market share does not change in accordance with the level of spillover effect. (3) No matter under the open-loop strategy or the closed-loop strategy, the competitive enterprises with lower initial market share should increase the investment of the online advertising in order to attract new customers as early as possible.
TL;DR: The validity of a convolutional neural network remote sensing classification model based on PTL-CFS has been proven and it is shown that the addition of PTL can accelerate the loss function of the convergence rate in CNN.
Abstract: Processing high-dimensional remote sensing images data with conventional convolutional neural networks raises certain issues such as prolonged model convergence time, vanishing gradient, convergence of the non-minimum values, etc. due to its high time-complexity and random initialization parameters nature. Aiming at those issues, this article proposes a convolutional neural network remote sensing classification model based on PTL-CFS. This model approach first utilizes parameter transfer learning algorithm to obtain the CNN initialization parameters of the target area, then it uses correlation-based feature selection algorithm to eliminate the redundant features and noises from the original feature set, finally, it classifies the remote sensing images using a conventional CNN model. This article has proven the validity of such network model when classifying remote sensing images in the Zha long wetland, Heilongjiang. Experiments show that the addition of PTL can accelerate the loss function of the convergence rate in CNN. The algorithm combined with CFS algorithm, compared with other algorithms to reduce the algorithm execution time and get better classification accuracy.
TL;DR: This study compared computation time between the computing methods with CPUs and GPUs in a simulation of neuronal models and found that the GPU-based computing system exhibits a higher computing performance than the CPU-based system, even if the GPU system includes data transfer from a graphics card to host memory.
Abstract: To understand the mechanism of information processing by a biological neural network, computer simulation of a large-scale spiking neural network is an important method However, because of a high computation cost of the simulation of a large-scale spiking neural network, the simulation requires high performance computing implemented by a supercomputer or a computer cluster Recently, hardware for parallel computing such as a multi-core CPU and a graphics card with a graphics processing unit (GPU) is built in a gaming computer and a workstation Thus, parallel computing using this hardware is becoming widespread, allowing us to obtain powerful computing power for simulation of a large-scale spiking neural network However, it is not clear how much increased performance the parallel computing method using a new GPU yields in the simulation of a large-scale spiking neural network In this study, we compared computation time between the computing methods with CPUs and GPUs in a simulation of neuronal models We developed computer programs of neuronal simulations for the computing systems that consist of a gaming graphics card with new architecture (the NVIDIA GTX 1080) and an accelerator board using a GPU (the NVIDIA Tesla K20C) Our results show that the computing systems can perform a simulation of a large number of neurons faster than CPU-based systems Furthermore, we investigated the accuracy of a simulation using single precision floating point We show that the simulation results of single precision were accurate enough compared with those of double precision, but chaotic neuronal response calculated by a GPU using single precision is prominently different from that calculated by a CPU using double precision Furthermore, the difference in chaotic dynamics appeared even if we used double precision of a GPU In conclusion, the GPU-based computing system exhibits a higher computing performance than the CPU-based system, even if the GPU system includes data transfer from a graphics card to host memory
TL;DR: A method for matrix representation of given query and its hierarchical ordering via calculus of applied lattice theory is introduced and the properties of Huffman coding are utilized to measure the changes in each query based on their Hamming distance.
Abstract: In the last decade, much attention has been paid towards connection among mobile and cloud devices for providing the optimum computational time to process any query of globally distributed users. This mathematics provides a large number of generated queries at a given phase of time. It creates a major problem in selecting some of the user required (or interested) queries and their changes to process the task within stimulated time. To elicit this problem the current paper introduces a method for matrix representation of given query and its hierarchical ordering via calculus of applied lattice theory. The importance of each query is decided through their entropy based computed weight and the level of granulation for their selection. The properties of Huffman coding are utilized to measure the changes in each query based on their Hamming distance. In addition, each of the proposed method are illustrated with an example.
TL;DR: It is shown how HADAS helps web server providers to make a trade-off between energy consumption and execution time, allowing them to sell different server configurations with different costs without modifying the hardware.
Abstract: The impact of energy consumption on the environment and the economy is raising awareness of “green” software engineering. HADAS is an eco-assistant that makes developers aware of the influence of their designs and implementations on the energy consumption and performance of the final product. In this paper, we extend HADAS to better support the requirements of users: researchers, automatically dumping the energy-consumption of different software solutions; and developers, who want to perform a sustainability analysis of different software solutions. This analysis has been extended by adding Pearson’s chi-squared differentials and Bootstrapping statistics, to automatically check the significance of correlations of the energy consumption, or the execution time, with any other variable (e.g., the number of users) that can influence the selection of a particular eco-efficient configuration. We have evaluated our approach by performing a sustainability analysis of the most common web servers (i.e. PHP servers) using the time and energy data measured with the Watts Up? Pro tool previously dumped in HADAS. We show how HADAS helps web server providers to make a trade-off between energy consumption and execution time, allowing them to sell different server configurations with different costs without modifying the hardware.
TL;DR: This research proposes particle swarm optimization (PSO) based algorithm, Integer PSO (IPSO) for design space exploration of reconfigurable computer architectures to have better energy and throughput balance.
Abstract: Most of recent research in multicore processor architectures has been shifted towards reconfigurable architectures due to increasing complexity of computing systems. These systems provide better application-specific energy and throughput balance with their reconfigurable behavior. They perform automatic run time resource allocation for an application as per its needs. But in terms of performance, current methodologies produce some unpredictable results because of the actual variety of the workloads. Therefore, we need optimization of the system resources usage by employing some optimization algorithms. Early research in the field of reconfigurable architecture using optimization algorithms has produced efficient results for energy consumption with the reconfiguration of cache sizes and associativity, number of cores and operating frequency. In this research, we propose particle swarm optimization (PSO) based algorithm, Integer PSO (IPSO) for design space exploration of reconfigurable computer architectures to have better energy and throughput balance. The results obtained by IPSO are evaluated by using various SPLASH-2 benchmark applications. Evaluation shows notable reduction in energy consumption without major effect on throughput. Simulation results also support the use of IPSO in design space exploration of multicore reconfigurable processor architectures.
TL;DR: The LeTS heuristic is a work-conserving algorithm that takes into account both locality and load balancing in order to reduce the execution time of target applications and outperforms state-of-the-art algorithms in amortizing inter-task communication cost.
Abstract: In systems with complex many-core cache hierarchy, exploiting data locality can significantly reduce execution time and energy consumption of parallel applications. Locality can be exploited at various hardware and software layers. For instance, by implementing private and shared caches in a multi-level fashion, recent hardware designs are already optimised for locality. However, this would all be useless if the software scheduling does not cast the execution in a manner that promotes locality available in the programs themselves. Since programs for parallel systems consist of tasks executed simultaneously, task scheduling becomes crucial for the performance in multi-level cache architectures. This paper presents a heuristic algorithm for homogeneous multi-core systems called locality-aware task scheduling (LeTS). The LeTS heuristic is a work-conserving algorithm that takes into account both locality and load balancing in order to reduce the execution time of target applications. The working principle of LeTS is based on two distinctive phases, namely; working task group formation phase (WTG-FP) and working task group ordering phase (WTG-OP). The WTG-FP forms groups of tasks in order to capture data reuse across tasks while the WTG-OP determines an optimal order of execution for task groups that minimizes the reuse distance of shared data between tasks. We have performed experiments using randomly generated task graphs by varying three major performance parameters, namely: (1) communication to computation ratio (CCR) between 0.1 and 1.0, (2) application size, i.e., task graphs comprising of 50-, 100-, and 300-tasks per graph, and (3) number of cores with 2-, 4-, 8-, and 16-cores execution scenarios. We have also performed experiments using selected real-world applications. The LeTS heuristic reduces overall execution time of applications by exploiting inter-task data locality. Results show that LeTS outperforms state-of-the-art algorithms in amortizing inter-task communication cost.
TL;DR: This paper proposes an approach for selecting a subset of the passed test suite when a failure revealed by a failed test case is revealed, and shows that the fault localization effectiveness can be significantly improved with less than 5% passed test cases.
Abstract: As spectra-based fault localization techniques report suspicious statements by analyzing the coverage of test cases, the effectiveness of the results is highly dependent on the composition of test suites. This paper proposes an approach for selecting a subset of the passed test suite when a failure revealed by a failed test case. The goal is to obtain a more effective fault localization using a minimal number of test cases than using the originally given large number of test cases. A novelty is that a prioritization criterion and a selection criterion are defined. Different from previous studies, the failed trace is fully considered. The prioritization criterion partitions statements in the failed trace into more suspicious and less suspicious, and then ranks passed test cases by their ability in distinguishing the more suspicious statements from the less suspicious ones. The selection criterion selects the minimal passed test suite which can maximize the number of coverage equivalent classes in the failed trace, so as to distinguish the suspicious statements and meanwhile reduce the size of the test suite. Another novelty is that our approach turns the test case selection into a multi-criteria optimization to make the prioritization and the selection criteria complement each other. This approach was evaluated with 5 fault localization techniques, 8 subject programs and 35,392 test cases. The results show that the fault localization effectiveness can be significantly improved with less than 5% passed test cases. Our approach has advantages over the statement- based and vector-based test suite reduction approaches in both fault localization effectiveness and test suite reduction rate.
TL;DR: This review, also characterized as an exploratory search, provides an overview of the techniques in the area that tries to look beyond accuracy in recommender systems, and finds the existence and characteristics of such approaches.
Abstract: Recommender systems were first conceived to provide suggestions of interesting items to users. The evolution of such systems provided an understanding that a recommender system is currently used to diverse objectives. One of the current challenges in the field is to have approaches of recommendation that go beyond accuracy metrics. Since it is a very recent interest of the community, this review, also characterized as an exploratory search, provides an overview of the techniques in the area that tries to look beyond accuracy. More specifically, one of the characteristics that would provide such evolution to these systems is the adaptation. This review is then performed to find the existence and characteristics of such approaches. Of the total 438 papers returned in the submission of the search string, 57 papers were analyzed after two filtering processes. The papers have shown that the area is little explored and one of the reasons is the challenge to validate non-accuracy characteristics in such approaches.
TL;DR: A generic approximate solution approach is adapted and developed for performability modelling which considers performance and availability issues of large number of nodes with multi-repairmen and is able to incorporate availability Issues of the system.
Abstract: The large-scale communication systems and computer networks provide flexible, efficient, and highly available services to their users. However, the practical large-scale systems result in unpredictable, fault-tolerant, often detrimental outcomes. This leads to developing and designing analytical models to understand and predict of complex system behaviour in order to ensure availability of large-scale systems. In this paper, analytical modelling and optimization analysis are presented for large-scale systems. The key contribution of this paper is twofold. First, a generic approximate solution approach is adapted and developed for performability modelling which considers performance and availability issues of large number of nodes with multi-repairmen. The analytical model and solution presented here are capable of considering large number of nodes up to thousands and able to incorporate availability issues of the system. Second and foremost, the relationship between the number of nodes and the number of repairmen is presented with an optimization analysis for large-scale systems. In order to show the efficacy and the accuracy of the proposed approach, the results obtained from the analytical model is validated with the results obtained from the simulations.
TL;DR: This paper proposes to introduce a network allocation vector (NAV), to reduce energy consumption and collisions in IEEE 802.15.4 networks and shows that using the NAV allows reducing significantly the energy consumption when applying the fragmentation technique in slotted CSMA/CA under saturated traffic conditions.
Abstract: Transmission delay, throughput and energy are important criterions to consider in wireless sensor networks (WSN). In this way, IEEE 802.15.4 standard was conceived with the objective to reduce resource’s consumption in both WSN and wireless personal area networks. In such networks, the slotted CSMA/CA still occupies a prominent place as a channel control access mechanism with its inherent simplicity and reduced complexity. In this paper, we propose to introduce a network allocation vector (NAV), to reduce energy consumption and collisions in IEEE 802.15.4 networks. A Markov chain-based analytical model of the fragmentation mechanism, in a saturated traffic, is given as well as a model of the energy consumption using the NAV mechanism. The obtained results show that the fragmentation technique improves at the same time the throughput, the access delay and the bandwidth occupation. They, also, show that using the NAV allows reducing significantly the energy consumption when applying the fragmentation technique in slotted CSMA/CA under saturated traffic conditions.