In order to improve the accuracy of sentiment classification of online product reviews, a model for sentiment analysis of unbalanced reviews is proposed. Firstly, the LDA model is used to balance the original review t...
详细信息
In this paper, LSOM (Load-balancing Self-Organizing Map), a neural network based on Kohonen39;s self-organizing map is proposed for the problem of mapping finite-element method (FEM) grids to distributed-memory para...
详细信息
Focal plane processingapplications present a growing computing need for portable telecomputing and videoputing systems. This paper demonstrates the integration of digital processing, analog interface circuitry, and t...
详细信息
Focal plane processingapplications present a growing computing need for portable telecomputing and videoputing systems. This paper demonstrates the integration of digital processing, analog interface circuitry, and thin film OE devices into a compact computing package. The SIMPil architecture provides a programmable, silicon efficient SIMD processor for effective execution of early image processingapplications such as edge detection, convolution, and compression. Results from a demonstration SIMPil node are presented including its microarchitecture, and performance on image processingapplications.
Advances in wireless communication, digital systems and micro-electronic-mechanical system technologies led to the development of wireless sensor networks (WSNs) which are used in various critical real-world applicati...
详细信息
ISBN:
(纸本)9781538621295
Advances in wireless communication, digital systems and micro-electronic-mechanical system technologies led to the development of wireless sensor networks (WSNs) which are used in various critical real-world applications. The fact that WSNs are low cost and eliminate the need for infrastructure led to their replacing traditional networks in area/event monitoring and tracking applications. WSNs consist of small and resource-limited sensor nodes, due to which several problems arise in the WSN development process. One of these problems is coverage. Providing the best coverage with a minimum number of sensor nodes is an NP-hard problem known as the maximum coverage sensor deployment problem (MCSDP). Genetic Algorithms (GAs) have been proved effective in solving optimization problems in many different disciplines (increasing coverage in WSNs, image processing, route planning, etc.). In this study, a GPU-based parallel GA solution for increasing the coverage of a given homogeneous WSN topology in a 2-D Euclidean area is proposed which is the first time this technique is used and parallelized on GPUs to the best of our knowledge. Finally, performance results of the proposed algorithm are compared to the previous work with the emphasis on the achieved performance improvement.
Knowledge that is needed during the design of a product is scattered among various systems containing diverse types of information. To integrate those information resources with the benefit of contextual augmentation ...
详细信息
ISBN:
(纸本)9781424450824
Knowledge that is needed during the design of a product is scattered among various systems containing diverse types of information. To integrate those information resources with the benefit of contextual augmentation and aggregation, meta-concepts are introduced that span across multiple information systems. A guiding instance is used to govern and validate the interactions between the information systems, which leads to virtualized information in a distributed environment. The necessary coverage of various implications for the content leads to a centralized module that controls the knowledge management across the distributed resources. Aspects as search and maintenance and the problem of outdated, contradicting and wrong information is addressed, while keeping the sovereignty of the individual systems intact. The presented approach relies on today's high performance computing environments that allows the alignment and strong comparison of multi parallel information resources.
The proceedings contain 142 papers. The topics discussed include: real-time middleware for the requirements of distributed process control;mobile information processing incorporating location-based services;architectu...
详细信息
ISBN:
(纸本)0780390946
The proceedings contain 142 papers. The topics discussed include: real-time middleware for the requirements of distributed process control;mobile information processing incorporating location-based services;architecture for automation system development based on IEC61499 standard;function Block applications based on an object oriented middleware;the JAKOBI architecture - A distributed dynamic execution environment in Java;development of a data pre-processing scheme and pluggable application modules for an intelligent equipment prognostics system;integrative approach of Web services and universal plug and play within an AV scenario;ontology-based context management components for service oriented architectures on wearable devices;and automated runtime management of embedded service systems based on design-time modeling and model transformation.
distributed query processing (DQP) has been widely used in data intensive applications where data of relevance to users are stored at multiple locations. This paper argues: (i) that DQP can be important in the Grid, a...
详细信息
distributed query processing (DQP) has been widely used in data intensive applications where data of relevance to users are stored at multiple locations. This paper argues: (i) that DQP can be important in the Grid, as a means of providing high-level, declarative languages for integrating data access and analysis;and (ii) that the Grid provides resource management facilities that are useful to developers of DQP systems. As well as discussing and illustrating how DQP technologies can be deployed within the Grid, the paper describes Polar*, a prototype implementation of a DQP system running over Globus. Polar* can handle complex data by adopting the ODMG object model and its query language OQL, which supports the invocation of user-defined operations. The Globus components are accessed through the MPICH-G interface rather than in a lower level way. A case study from bioinformatics is used throughout the paper, to show the benefits of the approach.
High-level data-structures are an important foundation for most applications. With the rise of multicores, there is a trend of supporting data-parallel collection operations in general purpose programming languages. H...
详细信息
ISBN:
(纸本)9781479984909
High-level data-structures are an important foundation for most applications. With the rise of multicores, there is a trend of supporting data-parallel collection operations in general purpose programming languages. However, these operations often incur high-level abstraction and scheduling penalties. We present a generic data-parallel collections design based on work-stealing for shared-memory architectures that overcomes abstraction penalties through callsite specialization of data-parallel operation instances. Moreover, we introduce work-stealing iterators that allow more fine-grained and efficient work-stealing. By eliminating abstraction penalties and making work-stealing data-structure-aware we achieve several dozen times better performance compared to existing JVM-based approaches.
We have developed an automatic camera control system for distance education. The system can calculate the 3D position and orientation of a marker by using the ARToolkit library, which is a software library for buildin...
详细信息
ISBN:
(纸本)0780389328
We have developed an automatic camera control system for distance education. The system can calculate the 3D position and orientation of a marker by using the ARToolkit library, which is a software library for building augmented reality applications. It can track a target indicated by the visual marker. According to the 3D position and orientation of the marker that is recognized, the system controls a pan/tilt/zoom video camera to track a target indicated by the 3D position and orientation of the recognized marker. Our system has functions such as marker prioritization, 3D marker offsets, and automatic position reset. These functions can alleviate the workload of lecturers who conduct lectures independently. Our system is based on distributedprocessing through IP multicasting, and flexibly supports various hardware configurations.
For the needs of real-time parallelprocessing of critical task and real-time state control in muti-node computing system, a lightweight distributedprocessing computer mechanism is designed for iterative computing ty...
详细信息
暂无评论