To search, Click below search items. |
| All Published Papers Search Service |
|
|
|
Title |
Implementation of Hybrid Software Architecture for Artificial Intelligence System |
|
Author |
B.Vinayagasundaram, S.K.Srivatsa |
| Citation |
Vol. 7 No. 1 pp. 35-41 |
|
Abstract |
An intelligent system should have the ability to process complex information. This information processing is not possible as whole in principle. We need some form of partial computational strategy, which is flexible that can adapt to changes in the environment and requirements. This paper introduces software architecture for specification and verification of Artificial Intelligence system incorporating conceptual and formal framework. The focus is based on reusable components. The architecture consists of four components; Task specification layer, Problem solver layer, domain layer and an adapter layer. These four components are hierarchically organized in a layered fashion. In this architecture, the overall structure is decomposed into sub components, in a layered way such that adding new layer without changing the existing layers can change the behavior of the system. Hence, an AI system can be built in an evolutionary way by combining and adapting several reusable components with out changing the existing functionality of the system. The paper is organized in to two parts. The first part of the paper summarizes the various AI architectures proposed in the literature. The second part of the paper addresses the hybrid layered software architecture for Artificial Intelligence system. Based on the proposed layered architecture, various text categorization methods are implemented in the problem solver layer and the performance is also discussed. |
|
Keywords |
Artificial Intelligence, software architecture, Reusability, Software engineering. |
|
URL |
