Soft Learning: A Conceptual Bridge between Data Mining and Machine Learning
| Autor: | Flaviu Adrian Mărginean |
|---|---|
| Rok vydání: | 2004 |
| Předmět: |
Association rule learning
Series (mathematics) Computer science business.industry Space (commercial competition) computer.software_genre Machine learning Rotation formalisms in three dimensions Task (project management) Computational learning theory Transversal (combinatorics) Data mining Artificial intelligence business computer Finite set |
| Zdroj: | Applications and Science in Soft Computing ISBN: 9783540408567 |
| DOI: | 10.1007/978-3-540-45240-9_33 |
| Popis: | It has been felt for some time that, despite employing different formalisms, being served by their own dedicated research communities and addressing distinct issues of practical interest, problems in Data Mining and Machine Learning connect through deep relationships. The paper [5] has taken a first step towards linking Data Mining and Machine Learning via Combinatorics by showing a correspondence between the problem of finding maximally specific sentences that are interesting in a database, the model of exact learning of monotone boolean functions in computational learning theory and the hyper-graph transversal problem in the combinatorics of finite sets. [5] summarises and concludes a series of valuable Data Mining research on fast discovery of association rules by the levelwise algorithm, series that includes [1, 4, 11]. Intuitively, a Data Mining task may consist of finding many weak predictors in a hypothesis space whereas in Machine Learning one strong predictor is sought. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|