Developing Industrial Case-Based Reasoning Applications: The Inreca Methodology
Developing Industrial Case-Based Reasoning Applications: The Inreca Methodology
Conversational Case-Based Reasoning
Applied Intelligence
Exploiting Taxonomic and Causal Relations in Conversational Case Retrieval
ECCBR '02 Proceedings of the 6th European Conference on Advances in Case-Based Reasoning
minimizing dialog length in interactive case-based reasoning
IJCAI'01 Proceedings of the 17th international joint conference on Artificial intelligence - Volume 2
| Hi-index | 0.00 |
This research presents a conversational case-based reasoning (CBR) help desk utility: a system that provides service support to a company's products. The goal is to utilize an existing database of customer calls as a case base, and achieve high recall and precision. In large case bases with inputs that may include irrelevant information, a conversation between the system and the user is required to gain more information about the problem. The suggested help desk utility addresses these concerns by using a 2-steps method. The first step is to create a category tree to enable quick and direct access to the problem's general subject and characteristics. The second step is an interactive dialog with the user by asking relevant questions until a solution is found. The incremental clustering algorithm COBWEB was used to build the category tree and an information gain ratio was used to order potential questions. The system was evaluated by the technical support team at a CAD company and showed a practical potential.