| Download | - View accepted manuscript: Exploiting Context When Learning to Classify (PDF, 183 KiB)
|
|---|
| Author | Search for: Turney, Peter1 |
|---|
| Affiliation | - National Research Council of Canada. NRC Institute for Information Technology
|
|---|
| Format | Text, Article |
|---|
| Conference | ECML-93: European Conference on Machine Learning, April 5-7, 1993., Vienna, Austria |
|---|
| Subject | context; robust classification; context-sensitive features; machine learning; robust learning; classification robuste; caractéristiques sensibles au contexte; apprentissage machine; apprentissage robuste |
|---|
| Abstract | This paper addresses the problem of classifying observations when features are context-sensitive, specifically when the testing set involves a context that is different from the training set. The paper begins with a precise definition of the problem,then general strategies are presented for enhancing the performance of classification algorithms on this type of problem.These strategies are tested on two domains. The first domain is the diagnosis of gasturbine engines. The problem is to diagnose a faulty engine in one context, such as warm weather, when the fault has previously been seen only in another context, such as cold weather. These conddomain is speech recognition. The problem is to recognize words spoken by a new speaker, not represented in the training set. For both domains, exploiting context results in substantially more accurate classification. |
|---|
| Publication date | 1993 |
|---|
| In | |
|---|
| Language | English |
|---|
| NRC number | NRCC 35058 |
|---|
| NPARC number | 8913502 |
|---|
| Export citation | Export as RIS |
|---|
| Report a correction | Report a correction (opens in a new tab) |
|---|
| Record identifier | 294523ef-a629-46bd-b5cb-c65cb8857d2e |
|---|
| Record created | 2009-04-22 |
|---|
| Record modified | 2020-04-24 |
|---|
