Experiments with Domain Knowledge in Unsupervised Learning: Using and Revising Theories
Experiments with Domain Knowledge in Unsupervised Learning: Using and Revising Theories
| dc.contributor | es-ES | |
| dc.contributor | en-US | |
| dc.creator | CORTÉS, ULISES | |
| dc.creator | BÉJAR, JAVIER | |
| dc.date | 2009-10-05 | |
| dc.date.accessioned | 2018-03-16T14:22:11Z | |
| dc.date.available | 2018-03-16T14:22:11Z | |
| dc.identifier | http://ojs.unam.mx/index.php/cys/article/view/2454 | |
| dc.identifier.uri | http://repositorioslatinoamericanos.uchile.cl/handle/2250/1190326 | |
| dc.description | USING DOMIAN KNOWLEDGE IN USURPEVISED LEARNING HAS SHOWN TO BE USEFUL STRATEGY WHE THE SET OF EXAMPLES OF A GIVEN DOMAIN HAS NOT AN EVIDENT STRUCTURE OR PRESENTS SOME LEVEL OF NOISE. THIS BACKROUND KNOWLWDGE CAN BE EXPRESSED AS A SET OF CLASSIFICATION RULES AND INTRODUCED AS A SEMANTIC BIAS DURING THE LEARNING PROCESS. IN THS WORK WE PRESENT SOME EXPERIMENTS IN THE USE OF PORTIAL DOMAIN KNWLEDGE WHIT TOOL LINNEO, A CONCEPTUAL CLSTERING ALGORITHM. THE DOMAIN KNOWLEDGE HAS NEITHER TO BE COMPLETE NOR CONSISTENT. THIS BIAS WILL INCREASE THE QUALITY OF THE FINAL GROUOPS AND REDUCE THE EFFECT OF THE ORDER OF THE EXAMPLES. SOME MEASURES OF STABILITY OF CLASSIFICATION ARE USED. THE IMPROVEMENT OF THE CONCEPTS CAN BE USED TO ENHACE AND CORRECT THE DOMAIN KNOLEDGE. A SET OF HEURISTICS TO REVISE THE ORIGINAL DOMAIN THEORY HAS BEEN EXPERIMENTED, YIELDING TO SOME INTERESTING RESULTS. | es-ES |
| dc.description | USING DOMIAN KNOWLEDGE IN USURPEVISED LEARNING HAS SHOWN TO BE USEFUL STRATEGY WHE THE SET OF EXAMPLES OF A GIVEN DOMAIN HAS NOT AN EVIDENT STRUCTURE OR PRESENTS SOME LEVEL OF NOISE. THIS BACKROUND KNOWLWDGE CAN BE EXPRESSED AS A SET OF CLASSIFICATION RULES AND INTRODUCED AS A SEMANTIC BIAS DURING THE LEARNING PROCESS. IN THS WORK WE PRESENT SOME EXPERIMENTS IN THE USE OF PORTIAL DOMAIN KNWLEDGE WHIT TOOL LINNEO, A CONCEPTUAL CLSTERING ALGORITHM. THE DOMAIN KNOWLEDGE HAS NEITHER TO BE COMPLETE NOR CONSISTENT. THIS BIAS WILL INCREASE THE QUALITY OF THE FINAL GROUOPS AND REDUCE THE EFFECT OF THE ORDER OF THE EXAMPLES. SOME MEASURES OF STABILITY OF CLASSIFICATION ARE USED. THE IMPROVEMENT OF THE CONCEPTS CAN BE USED TO ENHACE AND CORRECT THE DOMAIN KNOLEDGE. A SET OF HEURISTICS TO REVISE THE ORIGINAL DOMAIN THEORY HAS BEEN EXPERIMENTED, YIELDING TO SOME INTERESTING RESULTS. | en-US |
| dc.format | application/pdf | |
| dc.language | spa | |
| dc.publisher | Computación y Sistemas | es-ES |
| dc.relation | http://ojs.unam.mx/index.php/cys/article/view/2454/2016 | |
| dc.source | Computación y Sistemas; Vol 1, No 003 (1998) | es-ES |
| dc.source | 1405-5546 | |
| dc.subject | es-ES | |
| dc.subject | KNOWLEDGE ACQUISITION; DOMAIN THEORY; ILL - STRUCTURED DOMAINS; CLUSTERING METHODS | en-US |
| dc.title | Experiments with Domain Knowledge in Unsupervised Learning: Using and Revising Theories | es-ES |
| dc.title | Experiments with Domain Knowledge in Unsupervised Learning: Using and Revising Theories | en-US |
| dc.type | Artículos de revistas | |
| dc.type | Artículos de revistas |