Passive concept drift handling via variations of learning vector quantization
Autor: | Frank-Michael Schleif, Moritz Heusinger, Christoph Raab |
---|---|
Rok vydání: | 2020 |
Předmět: |
0209 industrial biotechnology
Learning vector quantization Concept drift Computer science Work (physics) 02 engineering and technology Field (computer science) 020901 industrial engineering & automation Stochastic gradient descent Distribution (mathematics) Artificial Intelligence 0202 electrical engineering electronic engineering information engineering Benchmark (computing) 020201 artificial intelligence & image processing Algorithm Software |
Zdroj: | Neural Computing and Applications. 34:89-100 |
ISSN: | 1433-3058 0941-0643 |
Popis: | Concept drift is a change of the underlying data distribution which occurs especially with streaming data. Besides other challenges in the field of streaming data classification, concept drift has to be addressed to obtain reliable predictions. Robust Soft Learning Vector Quantization as well as Generalized Learning Vector Quantization has already shown good performance in traditional settings and is modified in this work to handle streaming data. Further, momentum-based stochastic gradient descent techniques are applied to tackle concept drift passively due to increased learning capabilities. The proposed work is tested against common benchmark algorithms and streaming data in the field and achieved promising results. |
Databáze: | OpenAIRE |
Externí odkaz: |