A modified particle swarm optimization (PSO) algorithm is proposed. Linear constraints in the PSO are added to satisfy the normalization conditions for different problems. A hybrid algorithm based on the modified PSO ...
详细信息
A modified particle swarm optimization (PSO) algorithm is proposed. Linear constraints in the PSO are added to satisfy the normalization conditions for different problems. A hybrid algorithm based on the modified PSO and combining forecasting is presented. Combining forecasting can improve the forecasting accuracy through combining different forecasting methods. The effectiveness of the algorithm is demonstrated through the prediction on the sunspots and the stocks data. Simulated results show that the hybrid algorithm can improve the forecasting accuracy to a great extent.
One developing commercial vehicle was simulated on crashworthiness using the nonlinear finite element method. The deformation of the auto-body, the movement of the steering wheel and the dynamic responses of the occup...
详细信息
One developing commercial vehicle was simulated on crashworthiness using the nonlinear finite element method. The deformation of the auto-body, the movement of the steering wheel and the dynamic responses of the occupant at the initial velocity of 50 km/h were studied. The results show that the design of the vehicle could be improved on structure and material. The frontal longitudinal beam, the main energy-absorbing part of the auto-body, was optimized on structure. The data of the simulation predict that the hinge of the engine hood would fracture during the crash. The failure of the engine hood hinge would be a danger to both the driver and passengers. Then the problem was solved by changing the engine hood and hinge on structure and material. Simulation results also show that applying new material and new manufacture techniques could improve the crashworthiness of the vehicle greatly. These improvement methods are valuable to the virtual design of vehicles.
The gene section ordering on solving traveling salesman problems is analyzed by numerical experiments. Some improved crossover operations are presented. Several combinations of genetic operations are examined and the ...
详细信息
The gene section ordering on solving traveling salesman problems is analyzed by numerical experiments. Some improved crossover operations are presented. Several combinations of genetic operations are examined and the functions of these operations are analyzed. The essentiality of the ordering of the gene section and the significance of the evolutionary inversion operation are discussed. Some results and conclusions are obtained and given, which provide useful information for the implementation of the genetic operations for solving the traveling salesman problem.
Computational intelligence is the computational simulation of the bio-intelligence, which includes artificial neural networks, fuzzy systems and evolutionary computations. This article summarizes the state of the art ...
详细信息
Computational intelligence is the computational simulation of the bio-intelligence, which includes artificial neural networks, fuzzy systems and evolutionary computations. This article summarizes the state of the art in the field of simulated modeling of vibration systems using methods of computational intelligence, based on some relevant subjects and the authors' own research work. First, contributions to the applications of computational intelligence to the identification of nonlinear characteristics of packaging are reviewed. Subsequently, applications of the newly developed training algorithms for feedforward neural networks to the identification of restoring forces in multi-degree-of-freedom nonlinear systems are discussed. Finally, the neural-network-based method of model reduction for the dynamic simulation of microelectromechanical systems (MEMS) using generalized Hebbian algorithm (GHA) and robust GHA is outlined. The prospects of the simulated modeling of vibration systems using techniques of computational intelligence are also indicated.
This book constitutes the refereed proceedings of the Third International Frontiers of Algorithmics Workshop, FAW 2009, held in Hefei, Anhui, China, in June 2009. The 33 revised full papers presented together with the...
详细信息
ISBN:
(数字)9783642022708
ISBN:
(纸本)9783642022692
This book constitutes the refereed proceedings of the Third International Frontiers of Algorithmics Workshop, FAW 2009, held in Hefei, Anhui, China, in June 2009. The 33 revised full papers presented together with the abstracts of 3 invited talks were carefully reviewed and selected from 87 submissions. The papers are organized in topical sections on graph algorithms; game theory with applications; graph theory, computational geometry; machine learning; parameterized algorithms, heuristics and analysis; approximation algorithms; as well as pattern recognition algorithms, large scale data mining.
Named Entity Recognition (NER) is an important task in knowledge extraction, which targets extracting structural information from unstructured text. To fully employ the prior-knowledge of the pre-trained language mode...
详细信息
Named Entity Recognition (NER) is an important task in knowledge extraction, which targets extracting structural information from unstructured text. To fully employ the prior-knowledge of the pre-trained language models, some research works formulate the NER task into the machine reading comprehension form (MRC-form) to enhance their model generalization capability of commonsense knowledge. However, this transformation still faces the data-hungry issue with limited training data for the specific NER tasks. To address the low-resource issue in NER, we introduce a method named active multi-task-based NER (AMT-NER), which is a two-stage multi-task active learning training model. Specifically, A multi-task learning module is first introduced into AMT-NER to improve its representation capability in low-resource NER tasks. Then, a two-stage training strategy is proposed to optimize AMT-NER multi-task learning. An associated task of Natural Language Inference (NLI) is also employed to enhance its commonsense knowledge further. More importantly, AMT-NER introduces an active learning module, uncertainty selective, to actively filter training data to help the NER model learn efficiently. Besides, we also find different external supportive data under different pipelines improves model performance differently in the NER tasks. Extensive experiments are performed to show the superiority of our method, which also proves our findings that the introduction of external knowledge is significant and effective in the MRC-form NER tasks.
暂无评论