Weightless Artificial Neural Networks have proved to be a promising paradigm for classification tasks. This work introduces the WANN-Tagger, which makes use of weightless artificial neural networks for labelling Portu...
详细信息
ISBN:
(纸本)9789898425324
Weightless Artificial Neural Networks have proved to be a promising paradigm for classification tasks. This work introduces the WANN-Tagger, which makes use of weightless artificial neural networks for labelling Portuguese sentences, tagging each of its terms with its respective part-of-speech. A first experimental evaluation using the CETENFolha corpus indicates the usefulness of this paradigm and shows that it outperforms traditional feedforward neural networks in both accuracy and training time, and also that it is competitive in accuracy with the Hidden Markov Model in some cases. Additionally, WANN-Tagger shows itself capable of incrementally learning new tagged sentences during runtime.
Short-term load forecasting has received a lot of attention from both researchers and practitioners. Many techniques, such as neural networks, fuzzy logic, and time series models, are developed to improve the modeling...
详细信息
Meta-heuristics are efficient techniques for solving large scale optimization problems in which traditional mathematical techniques are impractical or provide suboptimal solutions. The Shuffled Frog Leaping algorithm ...
详细信息
Meta-heuristics are efficient techniques for solving large scale optimization problems in which traditional mathematical techniques are impractical or provide suboptimal solutions. The Shuffled Frog Leaping algorithm (SFLA) is a stochastic iterative method, bio-inspired on the memetic evolution of a group of frogs when seeking for food, which combines the social behavior-based of the particle swarm optimization technique (PSO) and the global information exchange of memetic algorithms. However, the SFLA algorithm suffers on large execution times, being this problem clearly evident when solving complex optimization problems for embedded applications. This drawback can be overcome by exploiting the parallel capabilities of the SFLA. This paper proposes a hardware parallel implementation of the SFLA algorithm (HPSFLA) using FPGAs (Field programmable gate Arrays) and the efficient floating-point arithmetic. The proposed architecture allows the SFLA to improve the functionality of the algorithm as well as to decrease the execution times by implementing parallel frogs and parallel memeplexes. Three well-known benchmark problems have been used to validate the implemented algorithm and simulation results demonstrate that the HPSFLA speeds-up by factors of 362, 727 and 211 a C-code implementation using an embedded microprocessor for the Sphere, Rastrigin and Rosenbrock benchmarks problems, respectively. Synthesis, simulation and execution time results demonstrate the effectiveness of the proposed HPSFLA architecture for embedded optimization systems.
PatternLab for proteomics is a one-stop shop computational environment for analyzing shotgun proteomic data. Its modules provide means to pinpoint proteins/peptides that are differentially expressed and those that are...
详细信息
Graphs are among the most frequently used structures in computer Science. Some of the properties that must be checked in many applications are connectivity, acyclicity and the Eulerian and Hamiltonian properties. In t...
详细信息
The main knowledge management challenges are to capture, store and reuse contextual knowledge generated during interactions that occur daily in an organization. In this paper, we propose an activity context-aware arch...
详细信息
The main knowledge management challenges are to capture, store and reuse contextual knowledge generated during interactions that occur daily in an organization. In this paper, we propose an activity context-aware architecture to support knowledge management in working processes. The required features for this architecture are processing, reasoning and sharing contextual knowledge involving information about activities performed. We also present results from evaluation of our proposal — A-CoBrA — for a specific domain.
Speculative Multithreading (SpMT) increases the performance by means of executing multiple threads speculatively to exploit thread-level parallelism. By combining software and hardware approaches, we have improved the...
详细信息
We have proposed a checklist-based usability inspection technique (WDP - Web Design Perspectives-Based Usability Evaluation) specific for Web applications' usability evaluation and we are following an experimentat...
详细信息
The technical literature regarding Model-based Testing (MBT) has several techniques with different characteristics and goals available to be applied in software projects. Besides the lack of information regarding thes...
详细信息
ISBN:
(纸本)9781424437436
The technical literature regarding Model-based Testing (MBT) has several techniques with different characteristics and goals available to be applied in software projects. Besides the lack of information regarding these techniques, they could be applied together in a software project aiming at improving the testing coverage. However, this decision needs to be carefully analyzed to avoid loss of resources in a software project. Based on this scenario, this paper proposes an approach with the purpose of supporting the unique or combined selection of MBT techniques for a given software project considering two aspects: the adequacy level between MBT techniques and the software project characteristics and impact of more than one MBT technique in some testing process variables. At the end, preliminary results of an experimental evaluation are presented.
Real time performance of non-interactive rendering of three-dimensional scenes is usually unachievable. Ray tracing is one of the methods used for rendering such scenes. The performance achieved by a sequential softwa...
详细信息
暂无评论