Relational learning algorithms mine complex databases for interesting patterns. Usually, the search space of patterns grows very quickly with the increase in data size, making it impractical to solve important problems. In this work we present the design of a relational learning system, that takes advantage of graphics processing units (GPUs) to perform the most time consuming function of the learner, rule coverage. To evaluate performance, we use four applications: a widely used relational learning benchmark for predicting carcinogenesis in rodents, an application in chemo-informatics, an application in opinion mining, and an application in mining health record data. We compare results using a single and multiple CPUs in a multicore host and using the GPU version. Results show that the GPU version of the learner is up to eight times faster than the best CPU version. 相似文献
Cloud computing systems handle large volumes of data by using almost unlimited computational resources, while spatial data warehouses (SDWs) are multidimensional databases that store huge volumes of both spatial data and conventional data. Cloud computing environments have been considered adequate to host voluminous databases, process analytical workloads and deliver database as a service, while spatial online analytical processing (spatial OLAP) queries issued over SDWs are intrinsically analytical. However, hosting a SDW in the cloud and processing spatial OLAP queries over such database impose novel obstacles. In this article, we introduce novel concepts as cloud SDW and spatial OLAP as a service, and afterwards detail the design of novel schemas for cloud SDW and spatial OLAP query processing over cloud SDW. Furthermore, we evaluate the performance to process spatial OLAP queries in cloud SDWs using our own query processor aided by a cloud spatial index. Moreover, we describe the cloud spatial bitmap index to improve the performance to process spatial OLAP queries in cloud SDWs, and assess it through an experimental evaluation. Results derived from our experiments revealed that such index was capable to reduce the query response time from 58.20 up to 98.89 %. 相似文献
Real-time interactive multimedia communications are becoming increasingly useful for education, business, e-commerce and e-government, providing an enriched user experience in teleconferencing, e-meetings, distance training and product demonstrations. Large corporations are usually located at several sites, so real-time multipoint sessions within corporations are especially difficult. IP multicast is available or feasible within each site of an organization. Thus, corporate networks can be considered as various multicast-capable networks interconnected through a wide area network without multicast connectivity. This paper proposes a resilient self-managed overlay network to support real-time multipoint interactive sessions within corporate networks. The proposed overlay takes advantage of the configuration of corporate networks to self-organize and provide an efficient media delivery service, making use of multicast communications wherever available. Various self-healing techniques are implemented allowing for the continuity of ongoing sessions in spite of network disruptions and entity failures. Extensive simulations and tests have been carried out to assess the performance and resilience of the overlay facing several types of disruptions. 相似文献
In this paper, we propose the problem of online cost-sensitive classifier adaptation and the first algorithm to solve it. We assume that we have a base classifier for a cost-sensitive classification problem, but it is trained with respect to a cost setting different to the desired one. Moreover, we also have some training data samples streaming to the algorithm one by one. The problem is to adapt the given base classifier to the desired cost setting using the steaming training samples online. To solve this problem, we propose to learn a new classifier by adding an adaptation function to the base classifier, and update the adaptation function parameter according to the streaming data samples. Given an input data sample and the cost of misclassifying it, we update the adaptation function parameter by minimizing cost-weighted hinge loss and respecting previous learned parameter simultaneously. The proposed algorithm is compared to both online and off-line cost-sensitive algorithms on two cost-sensitive classification problems, and the experiments show that it not only outperforms them on classification performances, but also requires significantly less running time.
Virtual organizations (VOs) are formed by an alliance of organizations linked by a partnership for dealing with emerging challenges. Information and communication technologies play a fundamental role facilitating cooperation, communication and collaboration among the VO members. The formal identification and representation of Requirements Engineering (RE) for one organization have been researched to a large extent along with several elicitation techniques. However, these techniques are not adequate for covering the major challenges of RE for VO. We explore the work done in the management and RE fields to propose a model-based framework for eliciting VOs’ requirements. The goal is to analyze the VO from two points of view: border (intra-organizational, inter-organizational and extra-organizational) and abstract (intentional, organizational and operational). This article includes the framework validation with a case study and a transformation process to develop partial Business Process Diagram from the intentional models. 相似文献
The problem of district design for the implementation of arc routing activities is addressed. The aim is to partition a road network into a given number of sectors to facilitate the organization of the operations to be implemented within the region. This problem arises in numerous applications such as postal delivery, meter readings, winter gritting, road maintenance, and municipal solid waste collection. An integer linear programming model is proposed where a novel set of node parity constraints to favor Eulerian districts is introduced. Series of instances were solved to assess the impact of these parity constraints on the objective function and deadhead distance. Networks with up to 401 nodes and 764 edges were successfully solved. The model is useful at a tactical level as it can be used to promote workload balance, compactness, deadhead distance reduction and parity in districts. 相似文献
Online bin stretching is a semi-online variant of bin packing in which the algorithm has to use the same number of bins as an optimal packing, but is allowed to slightly overpack the bins. The goal is to minimize the amount of overpacking, i.e., the maximum size packed into any bin. We give an algorithm for online bin stretching with a stretching factor of \(11/8 = 1.375\) for three bins. Additionally, we present a lower bound of \(45/33 = 1.\overline{36}\) for online bin stretching on three bins and a lower bound of 19/14 for four and five bins that were discovered using a computer search. 相似文献
The performance of state-of-the-art speaker verification in uncontrolled environment is affected by different variabilities. Short duration variability is very common in these scenarios and causes the speaker verification performance to decrease quickly while the duration of verification utterances decreases. Linear discriminant analysis (LDA) is the most common session variability compensation algorithm, nevertheless it presents some shortcomings when trained with insufficient data. In this paper we introduce two methods for session variability compensation to deal with short-length utterances on i-vector space. The first method proposes to incorporate the short duration variability information in the within-class variance estimation process. The second proposes to compensate the session and short duration variabilities in two different spaces with LDA algorithms (2S-LDA). First, we analyzed the behavior of the within and between class scatters in the first proposed method. Then, both proposed methods are evaluated on telephone session from NIST SRE-08 for different duration of the evaluation utterances: full (average 2.5 min), 20, 15, 10 and 5 s. The 2S-LDA method obtains good results on different short-length utterances conditions in the evaluations, with a EER relative average improvement of 1.58%, compared to the best baseline (WCCN[LDA]). Finally, we applied the 2S-LDA method in speaker verification under reverberant environment, using different reverberant conditions from Reverb challenge 2013, obtaining an improvement of 8.96 and 23% under matched and mismatched reverberant conditions, respectively. 相似文献