Download | - View accepted manuscript: Algorithme d'apprentissage pour inférer les paramètres de proaftn (PDF, 548 KiB)
|
---|
Author | Search for: Benabbou, L.; Search for: Guitouni, A.; Search for: Belacel, N. |
---|
Format | Text, Article |
---|
Conference | Administrative Sciences Association of Canada Conference 2004 (ASAC 2004), June 5-8, 2004, Québec City, Québec, Canada |
---|
Abstract | PROAFTN is a multicriteria classification method that makes it possible to directly assign potential actions to different classes. In this work, we suggest a training algorithm that can be used to automatically adjust the parameters of the method. This algorithm utilizes local search and meta-heuristic tools to minimize the difference between the classification results obtained by the PROAFTN method and the a priori results in the training set. |
---|
Publication date | 2004 |
---|
In | |
---|
Language | French |
---|
NRC number | CNRC 47128 |
---|
NPARC number | 9146107 |
---|
Export citation | Export as RIS |
---|
Report a correction | Report a correction (opens in a new tab) |
---|
Record identifier | 4714c65a-9bfc-4a53-9055-7f14ec39cb0e |
---|
Record created | 2009-06-25 |
---|
Record modified | 2021-01-05 |
---|