| Presentation | 1998/12/1 Intelligent Agents for Probabilistic Diagnosis-Restoration Planning in WAN Hajji Hassan, Far B.H., Zenya Koono, |
|---|---|
| PDF Download Page | PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | As networks increase in size, heterogeneity and complexity, the need to maintain their availability and reliability grows in both importance and difficulty. In this paper, we propose a probabilistic distributed approach to automatic network fault diagnosis and correction in Wide Area Networks (WANs) using Bayesian networks. Each sub-domain of a WAN is assigned to a single agent that models its own partial knowledge as a Bayesian sub-net. When the agent is notified of an anomaly in one of its managed objects, it computes an initial optimal observation plan given reported abnormal observations. The plan specifies the order in which the objects are to be observed. Then, information gathering objects are sequentially evaluated using the myopic value of information, and the agent decides upon the object to observe based on its potential to reveal useful information about which component is faulty, and associated cost. To derive the globally optimal restoration plan, agents maintain an interface set of objects through which they coordinate observation and restoration actions, and keep track of relevant information in other domains. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | Fault management / wide area networks / Bayesian networks / decision-theoretic diagnosis-restoration planning |
| Paper # | AI98-56 |
| Date of Issue |
| Conference Information | |
| Committee | AI |
|---|---|
| Conference Date | 1998/12/1(1days) |
| Place (in Japanese) | (See Japanese page) |
| Place (in English) | |
| Topics (in Japanese) | (See Japanese page) |
| Topics (in English) | |
| Chair | |
| Vice Chair | |
| Secretary | |
| Assistant | |
| Paper Information | |
| Registration To | Artificial Intelligence and Knowledge-Based Processing (AI) |
|---|---|
| Language | ENG |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | Intelligent Agents for Probabilistic Diagnosis-Restoration Planning in WAN |
| Sub Title (in English) | |
| Keyword(1) | Fault management |
| Keyword(2) | wide area networks |
| Keyword(3) | Bayesian networks |
| Keyword(4) | decision-theoretic diagnosis-restoration planning |
| 1st Author's Name | Hajji Hassan |
| 1st Author's Affiliation | Department of Information and Computer Science, Saitama University() |
| 2nd Author's Name | Far B.H. |
| 2nd Author's Affiliation | Department of Information and Computer Science, Saitama University |
| 3rd Author's Name | Zenya Koono |
| 3rd Author's Affiliation | Department of Information and Computer Science, Saitama University |
| Date | 1998/12/1 |
| Paper # | AI98-56 |
| Volume (vol) | vol.98 |
| Number (no) | 437 |
| Page | pp.pp.- |
| #Pages | 8 |
| Date of Issue | |