Fixed-size ensemble classifier system evolutionarily adapted to a recurring context with an unlimited pool of classifiers

Jackowski, Konrad

doi:10.1007/s10044-013-0318-x

Cited by 22 publications

(11 citation statements)

References 41 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Although the NEVE algorithm have demonstrated satisfactory performance for the datasets used in the analysis of this study, it is strongly recommended to perform further tests -using different configurations, different datasets and performing different analysis -to confirm the results presented here. We also intend in the future to continue this work, analyzing other existing approaches, such as [17] and [18], and performing new experiments in comparison with these and other algorithms. We still need to investigate other factors related to QIEA-R fine tunning (genetic operators, population size, etc.…”

Section: Discussionmentioning

confidence: 99%

NEVE: A Neuro-Evolutionary Ensemble for Adaptive Learning

Escovedo

Cruz

Vellasco

et al. 2013

IFIP Advances in Information and Communication Technology

View full text Add to dashboard Cite

This work describes the use of a quantum-inspired evolutionary algorithm (QIEA-R) to construct a weighted ensemble of neural network classifiers for adaptive learning in concept drift problems. The proposed algorithm, named NEVE (meaning Neuro-EVolutionary Ensemble), uses the QIEA-R to train the neural networks and also to determine the best weights for each classifier belonging to the ensemble when a new block of data arrives. After running eight simulations using two different datasets and performing two different analysis of the results, we show that NEVE is able to learn the data set and to quickly respond to any drifts on the underlying data, indicating that our model can be a good alternative to address concept drift problems. We also compare the results reached by our model with an existing algorithm, Learn++.NSE, in two different nonstationary scenarios.

show abstract

Section: Discussionmentioning

confidence: 99%

NEVE: A Neuro-Evolutionary Ensemble for Adaptive Learning

Escovedo

Cruz

Vellasco

et al. 2013

IFIP Advances in Information and Communication Technology

View full text Add to dashboard Cite

show abstract

“…Apesar de o algoritmo NEVE ter demonstrado performance satisfatória para os conjuntos de dados utilizados nos experimentos deste trabalho, recomenda-se fortemente que sejam realizados novos testes, com diferentes configurações, a fim de que se possa confirmar ou não os resultados aqui apresentados. Também pretendemos no futuro dar continuidade neste trabalho, analisando outras abordagens existentes, como [5], [22] e [23], a fim de realizar novos experimentos de forma comparativa a estes algoritmos.…”

Section: Conclusões E Trabalhos Futurosunclassified

NEVE: Um Comitê de Classificadores Neuro-Evolucionário para Aprendizagem Adaptativa

Escovedo¹,

Cruz²,

Vellasco³

et al. 2016

Anais Do 11. Congresso Brasileiro De Inteligência Computacional

View full text Add to dashboard Cite

This work describes the use of a quantum-inspired evolutionary algorithm (QIEA-R) to construct a weighted ensemble of neural network classifiers for adaptive learning in concept drift problems. The proposed algorithm, named NEVE (meaning Neuro-EVolutionary Ensemble), uses the QIEA-R to train the neural networks and also to determine the best weights for each classifier belonging to the C when a new block of data arrives. After running eight simulations using two different datasets and performing two different analysis of the results, we show that NEVE is able to learn the data set and to quickly respond to any drifts on the underlying data, indicating that our model can be a good alternative to address concept drift problems.

show abstract

“…In such a case, OCC is the most proper solution. One may bring more examples of the potential use of OCC, such as image/video classification (where it is impossible to determine what will appear on the scene) [7], data stream analysis (where new, unknown classes may appear due to data shifts and drifts) [15,35], novelty detection [21] or bio-signal classification (where some pathologies may be dependent on the patient) [13].…”

Section: Classification In the Absence Of Counterexamplesmentioning

confidence: 99%

Selecting locally specialised classifiers for one-class classification ensembles

Krawczyk

Cyganek

2015

Pattern Anal Applic

View full text Add to dashboard Cite

One-class classification belongs to the one of the novel and very promising topics in contemporary machine learning. In recent years ensemble approaches have gained significant attention due to increasing robustness to unknown outliers and reducing the complexity of the learning process. In our previous works, we proposed a highly efficient one-class classifier ensemble, based on input data clustering and training weighted one-class classifiers on clustered subsets. However, the main drawback of this approach lied in difficult and time consuming selection of a number of competence areas which indirectly affects a number of members in the ensemble. In this paper, we investigate ten different methodologies for an automatic determination of the optimal number of competence areas for the proposed ensemble. They have roots in model selection for clustering, but can be also effectively applied to the classification task. In order to select the most useful technique, we investigate their performance in a number of one-class and multi-class problems. Numerous experimental results, backed-up with statistical testing, allows us to propose an efficient and fully automatic method for tuning the one-class clustering-based ensembles.

show abstract

Fixed-size ensemble classifier system evolutionarily adapted to a recurring context with an unlimited pool of classifiers

Cited by 22 publications

References 41 publications

NEVE: A Neuro-Evolutionary Ensemble for Adaptive Learning

NEVE: A Neuro-Evolutionary Ensemble for Adaptive Learning

NEVE: Um Comitê de Classificadores Neuro-Evolucionário para Aprendizagem Adaptativa

Selecting locally specialised classifiers for one-class classification ensembles

Contact Info

Product

Resources

About