EURO-Online login
- New to EURO? Create an account
- I forgot my username and/or my password.
- Help with cookies
(important for IE8 users)
3262. Preference Learning Approaches in Interactive Multiple Criteria Decision Making Using Augmented Training Sets
Invited abstract in session MB-44: Preference Learning 2, stream Multiple Criteria Decision Analysis.
Monday, 10:30-12:00Room: 20 (building: 324)
Authors (first author is the speaker)
| 1. | Srinivas Prasad
|
| Decision Sciences, The George Washington University | |
| 2. | Atilla Ay
|
| James Madison University | |
| 3. | Luis Novoa
|
| James Madison University |
Abstract
We investigate machine learning based preference elicitation strategies in the context of Interactive Multiple Criteria Decision Making. Specifically, such methods employ popular machine learning techniques such as neural networks among others. We study how to make the training phase efficient in such an approach by employing certain preference representation structures that provide for augmentation of the training set. Such augmentations help accelerate the preference learning phase by making maximum use of the elicited preferences in eventually estimating the decision maker's preference profile based on partial preference information. We study different elicitation strategies in combination with different levels of augmentation and provide some empirical results illustrating their effectiveness.
Keywords
- Multi-Objective Decision Making
- Machine Learning
- Artificial Intelligence
Status: accepted
Back to the list of papers