1 / 11

컴퓨터공학과 98419-531 신수용

Application of Inductive Learning and Case-Based Reasoning for Troubleshooting Industrial Machines - Michel Manago and Eric Auriol. 컴퓨터공학과 98419-531 신수용. Inductive Learning (1/2). Abstract procedure 1. Creates a general description of past examples - create decision tree

cullen
Download Presentation

컴퓨터공학과 98419-531 신수용

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Application of Inductive Learning and Case-Based Reasoning for Troubleshooting Industrial Machines- Michel Manago and Eric Auriol 컴퓨터공학과 98419-531 신수용

  2. Inductive Learning (1/2) • Abstract procedure 1. Creates a general description of past examples - create decision tree 2. applies this description to new data • Inductive learning extracts relevant decision knowledge from case history

  3. Inductive Learning (2/2)

  4. Case-Based Reasoning (CBR) (1/2) • Abstract procedure 1. stores past examples - does not requires a tree structure 2. assigns decisions to new data by relating it to past cases • A case • (the description of a problem that has been successfully solved in the past, solutions) • When a new problem is encountered, CBR recalls similar cases and adapts the solutions that worked in the past for the current problem.

  5. CBR (2/2) • Application domain • poorly understood or where rules have many excepts • experience is as valuable as textbook knowledge • CBR makes direct use of past experience • historical cases are views as an asset that should be preserved and it is intuitively clear that remembering pat experience is useful • specialist talk about their domain by giving examples.

  6. Inductive learning vs. CBR • Help-desk areas; troubleshooting complex equipment • performance comparison • pure CBR retrieval is fast for DB with fewer than 10,000 cases

  7. Obtaining better feedback from experiences • CBR and inductive learning help to • improve after-sale support with help-desk software • develop diagnosis and fault analysis decision support system • regularly update troubleshooting manuals from observed faults • capture and reuse the experience of the most talented maintenance specialists • perform experience feedback to increase reliability and maintainability

  8. Applications (1/4) • Decision support system for the technical maintenance of the Cfm56-3 aircraft engines • Combination of inductive and CBR • gather the case data • fault trees have been generated by inductive learning

  9. Application (3/4) • LADI • troubleshoots axis positioning defects • SEPRO Robotique: AcknoSoft installed a CBR help-desk • performs a nearest-neighbor search on the relevant cases

More Related