Overfitting cautious selection of classifier ensembles with genetic algorithms

Santos, Eulanda M. dos; Sabourin, Robert; Maupin, Patrick

doi:10.1016/j.inffus.2008.11.003

Cited by 82 publications

(34 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Another perspective is to stop MOCA-I earlier. The work of Dos Santos et al shows that stopping the search before reaching the local optima could reduce over fitting [50] thus increase the results obtained on the test data -while reducing the execution time.…”

Section: Discussionmentioning

confidence: 99%

Conception of a dominance-based multi-objective local search in the context of classification rule mining in large and imbalanced data sets

Jacques

Taillard²,

Delerue³

et al. 2015

Applied Soft Computing

View full text Add to dashboard Cite

Section: Discussionmentioning

confidence: 99%

Conception of a dominance-based multi-objective local search in the context of classification rule mining in large and imbalanced data sets

Jacques

Taillard²,

Delerue³

et al. 2015

Applied Soft Computing

View full text Add to dashboard Cite

“…Mainly artificial neural networks (ANN) were used [13]. Although the prediction ability of ANNs is comparatively higher there are reasonable limitations such as the need to have a great experience in order to select the control parameters properly and difficulties with building the model itself [14,15].…”

Section: Literature Reviewmentioning

confidence: 99%

Modelling bankruptcy prediction models in Slovak companies

Kováčová

Klieštiková

2017

SHS Web Conf.

View full text Add to dashboard Cite

Abstract. An intensive research from academics and practitioners has been provided regarding models for bankruptcy prediction and credit risk management. In spite of numerous researches focusing on forecasting bankruptcy using traditional statistics techniques (e.g. discriminant analysis and logistic regression) and early artificial intelligence models (e.g. artificial neural networks), there is a trend for transition to machine learning models (support vector machines, bagging, boosting, and random forest) to predict bankruptcy one year prior to the event. Comparing the performance of this with unconventional approach with results obtained by discriminant analysis, logistic regression, and neural networks application, it has been found that bagging, boosting, and random forest models outperform the others techniques, and that all prediction accuracy in the testing sample improves when the additional variables are included. On the other side the prediction accuracy of old and well known bankruptcy prediction models is quiet high. Therefore, we aim to analyse these in some way old models on the dataset of Slovak companies to validate their prediction ability in specific conditions. Furthermore, these models will be modelled according to new trends by calculating the influence of elimination of selected variables on the overall prediction ability of these models.

show abstract

“…As a result of this, the learning algorithm's performance drops when it is tested in an unknown dataset. The amount of data used for the learning process is fundamental in this context [10]. Small datasets are more prone to over fitting than large data sets [27].…”

Section: Over Fitting and Diversitymentioning

confidence: 99%

“…The principle is that, given the initial pool C, the best performing subset of classifiers in P(C) must be found, and this is the powerset of C defining the population of all possible candidate ensembles Cj [10].…”

Section: Introductionmentioning

confidence: 99%

Rule Based Ensembles Using Pair Wise Neural Network Classifiers

Jenghara¹,

Ebrahimpour-Komleh²

2015

IJISA

View full text Add to dashboard Cite

Abstract-In value estimation, the inexperienced people's estimation average is good approximation to true value, provided that the answer of these individual are independent. Classifier ensemble is the implementation of mentioned principle in classification tasks that are investigated in two aspects. In the first aspect, feature space is divided into several local regions and each region is assigned with a highly competent classifier and in the second, the base classifiers are applied in parallel and equally experienced in some ways to achieve a group consensus. In this paper combination of two methods are used. An important consideration in classifier combination is that much better results can be achieved if diverse classifiers, rather than similar classifiers, are combined. To achieve diversity in classifiers output, the symmetric pairwise weighted feature space is used and the outputs of trained classifiers over the weighted feature space are combined to inference final result. In this paper MLP classifiers are used as the base classifiers. The Experimental results show that the applied method is promising.

show abstract

Overfitting cautious selection of classifier ensembles with genetic algorithms

Cited by 82 publications

References 29 publications

Conception of a dominance-based multi-objective local search in the context of classification rule mining in large and imbalanced data sets

Conception of a dominance-based multi-objective local search in the context of classification rule mining in large and imbalanced data sets

Modelling bankruptcy prediction models in Slovak companies

Rule Based Ensembles Using Pair Wise Neural Network Classifiers

Contact Info

Product

Resources

About