On the Problem of Local Minima in Backpropagation
IEEE Transactions on Pattern Analysis and Machine Intelligence
Some computer science issues in ubiquitous computing
Communications of the ACM - Special issue on computer augmented environments: back to the real world
Collaborative interface agents
AAAI '94 Proceedings of the twelfth national conference on Artificial intelligence (vol. 1)
Understanding and Using Context
Personal and Ubiquitous Computing
A Scalable Agent Location Mechanism
ATAL '99 6th International Workshop on Intelligent Agents VI, Agent Theories, Architectures, and Languages (ATAL),
LIME: A Middleware for Physical and Logical Mobility
ICDCS '01 Proceedings of the The 21st International Conference on Distributed Computing Systems
An ontology for context-aware pervasive computing environments
The Knowledge Engineering Review
| Hi-index | 0.00 |
How to choose an appropriate service from all the usable services regardless of user's location and heterogeneous architecture of underlying software and hardware infrastructure is the most important study content in ubiquitous computing domain. In order to overcome the shortcomings of blindness and randomicity in traditional and improved trust-mechanism-based service selection models, we propose a novel ANN-based (Artificial Neural Network) service selection model (called the ANNSS model). We adopt a novel method which according to the earlier information of the cooperation between the devices and the context information, an ANN-based evaluation standard for the service quality of service provider is given out so that user can acquire an effective guidance and choose the most appropriate service. At the same time, we improved the traditional BP algorithm based on three-term method (called the TTMBP algorithm) consisting of a learning rate (LR), a momentum factor (MF) and a proportional factor (PF) in order to satisfy the requirements of time issue in real-time system. The convergence speed and stability were enhanced by adding the proportional factor. The self-adjusting architecture method is adopted so that a moderate scale of neural network can be obtained. We have implemented the ANNSS algorithm in an actual power supply system for communication devices and fulfilled various simulations. The results of simulation show that the proposed service selection scheme is not only scalable but also efficient, and that the novel BP algorithm based on three-term has high convergence speed and good convergence stability. The novel service selection scheme superior to the traditional and improved trust-mechanism-based service selection scheme. The novel scheme can exactly choose a most appropriate service from many service providers and provide the most perfect service performance to users.