Maximizing diversity by transformed ensemble learning

Mao, Shasha; Chen, Jiawei; Jiao, Licheng; Gou, Shuiping; Wang, Rongfang

doi:10.1016/j.asoc.2019.105580

Cited by 35 publications

(14 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…[16] The former is focused on a pair of classifiers [17], [18], [19], while the latter seeks diversity presented by a whole ensemble [20], [21]. Various attempts were made to obtain the best possible diversity [22] such as tuning learning algorithms [23], [24], sampling methods [25], [26], and feature selection [27], [28], [29], [30].…”

Section: Related Work a Static Ensemble Selectionmentioning

confidence: 99%

DDES: A Distribution-Based Dynamic Ensemble Selection Framework

Choi

Lim

2021

IEEE Access

View full text Add to dashboard Cite

Dynamic Ensemble Selection (DES) is a special type of ensemble modeling that selects different subsets of base classifiers for different test sample cases. In this process, multiple base classifiers are compared in terms of their competence as an ensemble to the best possible prediction for a given test sample case. Traditional DES methods rely on the Euclidean distancebased k-Nearest Neighbor (kNN) algorithm to identify the most relevant reference data points in such a way that the base classifiers correctly classifying them can be considered for an optimal ensemble. However, this approach is sensitive to the local structure of the data and the presence of noisy or irrelevant attributes. This study proposes a novel distribution-based DES (DDES) framework that takes the data structure into consideration when selecting reference data points. The experimental results for 30 classification problems indicate that the proposed approach yielded the best accuracy among the competing DES methods. Additionally, we discuss the correlation between the data complexity and the improvement in classification performance.

show abstract

Section: Related Work a Static Ensemble Selectionmentioning

confidence: 99%

DDES: A Distribution-Based Dynamic Ensemble Selection Framework

Choi

Lim

2021

IEEE Access

View full text Add to dashboard Cite

show abstract

“…They integrated unsupervised clustering with a fuzzy assignment process to make full use of data patterns to improve the ensemble performance. Mao et al [36] propose a transformation ensemble learning framework in which the combination of multiple base learners is converted into a linear transformation of all these base learners and which constructs an optimization objective function for balancing accuracy and diversity. The alternating direction multiplier method is used to solve this problem.…”

Section: Introductionmentioning

confidence: 99%

“…In previous studies [35,36], when the ensemble learning model was built, although the objective function includes accuracy and diversity factors, the diversity factor is used as a regularization term to avoid overfitting. Nevertheless, the diversity factor considers only the prediction results of the classifier.…”

Section: Introductionmentioning

confidence: 99%

A Weighted Ensemble Learning Algorithm Based on Diversity Using a Novel Particle Swarm Optimization Approach

et al. 2020

View full text Add to dashboard Cite

In ensemble learning, accuracy and diversity are the main factors affecting its performance. In previous studies, diversity was regarded only as a regularization term, which does not sufficiently indicate that diversity should implicitly be treated as an accuracy factor. In this study, a two-stage weighted ensemble learning method using the particle swarm optimization (PSO) algorithm is proposed to balance the diversity and accuracy in ensemble learning. The first stage is to enhance the diversity of the individual learner, which can be achieved by manipulating the datasets and the input features via a mixed-binary PSO algorithm to search for a set of individual learners with appropriate diversity. The purpose of the second stage is to improve the accuracy of the ensemble classifier using a weighted ensemble method that considers both diversity and accuracy. The set of weighted classifier ensembles is obtained by optimization via the PSO algorithm. The experimental results on 30 UCI datasets demonstrate that the proposed algorithm outperforms other state-of-the-art baselines.

show abstract

“…While literature [30] and [31] describe that the ideal ensemble is constructed using learners of small error and good diversity. However, rich diversity may cause the predicted value of meta-learners to deviate from the true values, and the improvement of individual accuracy often reduces the diversity of meta-learners, that is, accuracy and diversity are usually conflicting to each other.…”

Section: Introductionmentioning

confidence: 99%

An Optimal Stacking Ensemble for Remaining Useful Life Estimation of Systems Under Multi-Operating Conditions

Zhang

Chen

et al. 2020

IEEE Access

View full text Add to dashboard Cite

Remaining useful life (RUL) estimation is expected to provide appropriate maintenance for components or systems in industry to improve the reliability of the systems. Most data-based methods are limited to a single model, which is susceptible to various factors like environmental variability and diversity of operating conditions. In this paper, we propose an optimal stacking ensemble method combining different learning algorithms as meta-learners to mitigate the impact of multi-operating conditions. The selection of meta-learners follows a multi-objective evolutionary algorithm named non-dominated sorting genetic algorithms-II to balance the two conflicting objectives in terms of accuracy and diversity. Then the eventually evolved meta-learners are integrated by the meta-classifier for RUL estimation. In addition, a long-short-term feature extraction strategy is proposed to capture more degradation information from lifecycle data dynamically. Extensive experiments are performed on aero-engine dataset and battery dataset provided by NASA, which achieves the higher prognostic accuracy compared with the single models and existing methods. INDEX TERMS Remaining useful life (RUL), reliability, optimal stacking ensemble, meta-learners.

show abstract

Maximizing diversity by transformed ensemble learning

Cited by 35 publications

References 34 publications

DDES: A Distribution-Based Dynamic Ensemble Selection Framework

DDES: A Distribution-Based Dynamic Ensemble Selection Framework

A Weighted Ensemble Learning Algorithm Based on Diversity Using a Novel Particle Swarm Optimization Approach

An Optimal Stacking Ensemble for Remaining Useful Life Estimation of Systems Under Multi-Operating Conditions

Contact Info

Product

Resources

About