Self-training for multi-target regression with tree ensembles

Levati, Jurica; Ceci, Michelangelo; Kocev, Dragi; Deroski, Sao

doi:10.1016/j.knosys.2017.02.014

Cited by 55 publications

(29 citation statements)

References 45 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…In our future work, we intend to pursue extensive empirical experiments to compare the proposed WvEnSL with other algorithms belonging to different SSL classes, and evaluate its performance using various component self-labeled algorithms and base learners. Furthermore, since our preliminary numerical experiments are quite encouraging, our next step is to explore the performance of the proposed algorithm on imbalanced datasets [39,40] and incorporate our proposed methodology for multi-target problems [41][42][43]. Additionally, another interesting aspect is the use of other component classifiers in the ensemble and enhance our proposed framework with more sophisticated and theoretically sound criteria for the development of an advanced weighted voting strategy.…”

Section: Discussionmentioning

confidence: 99%

A Weighted Voting Ensemble Self-Labeled Algorithm for the Detection of Lung Abnormalities from X-Rays

et al. 2019

View full text Add to dashboard Cite

During the last decades, intensive efforts have been devoted to the extraction of useful knowledge from large volumes of medical data employing advanced machine learning and data mining techniques. Advances in digital chest radiography have enabled research and medical centers to accumulate large repositories of classified (labeled) images and mostly of unclassified (unlabeled) images from human experts. Machine learning methods such as semi-supervised learning algorithms have been proposed as a new direction to address the problem of shortage of available labeled data, by exploiting the explicit classification information of labeled data with the information hidden in the unlabeled data. In the present work, we propose a new ensemble semi-supervised learning algorithm for the classification of lung abnormalities from chest X-rays based on a new weighted voting scheme. The proposed algorithm assigns a vector of weights on each component classifier of the ensemble based on its accuracy on each class. Our numerical experiments illustrate the efficiency of the proposed ensemble methodology against other state-of-the-art classification methods.

show abstract

Section: Discussionmentioning

confidence: 99%

A Weighted Voting Ensemble Self-Labeled Algorithm for the Detection of Lung Abnormalities from X-Rays

et al. 2019

View full text Add to dashboard Cite

show abstract

“…Applying unlabeled data in semi-supervised self-training is beneficial, but in some cases, it may degrade the classifier's performance if they are incorrectly labeled as an improper class by the initial classifier (Piroonsup et al, 2018;Levatić et al, 2017). Some studies try to avoid this issue mainly by post-processing, e.g., self-training with editing (Zhou et al, 2005), or applying a noise-filtering method to remove the mislabeled data (Triguero et al, 2014).…”

Section: Reliability Scorementioning

confidence: 99%

“…where ( ) is the prediction for sample returned by the jth estimator, and M(u) is the prediction for returned by the ensemble (i.e., the average of the predictions across all trees). This variance measure has been previously used in the context of bagging, where it performed the best among various approaches for estimating the reliability of regression predictions (Bosni et al, 2008;Levatić et al, 2017).…”

Section: Reliability Scorementioning

confidence: 99%

“…By contrast, when the reliability threshold is highly low, e.g., lower than 0.55 in experiments, the classifiers may be affected by increasingly more unreliable data, which will worsen the classification performance. In related work (Bosni et al, 2008;Levatić et al, 2017), they also argued that the disadvantage of self-learning is that if samples of wrong classification are added to the training set, the errors that they render in the subsequent training process will be increasingly deeper, and the other samples tend to render mistakes. Thus, the performance of self-training can deteriorate.…”

Section: Varying the Reliability Of Labelled Training Datamentioning

confidence: 99%

See 1 more Smart Citation

Opinion Spam Detection based on Annotation Extension and Neural Networks

Liu¹,

Pang²

2019

CIS

View full text Add to dashboard Cite

Online reviews play an increasingly important role in the purchase decisions of potential customers. Incidentally, driven by the desire to gain profit or publicity, spammers may be hired to write fake reviews and promote or demote the reputation of products or services. Correspondingly, opinion spam detection has attracted attention from both business and research communities in recent years. However, unlike other tasks such as news classification or blog classification, the existing review spam datasets are typically limited due to the expensiveness of human annotation, which may further affect detection performance even if excellent classifiers have been developed. We propose a novel approach in this paper to boost opinion spam detection performance by fully utilizing the existing labelled small-size dataset. We first design an annotation extension scheme that uses extra tree classifiers to train multiple estimators and then iteratively generate reliable labelled samples from unlabeled ones. Subsequently, we examine neural network scenarios on a newly extended dataset to learn the distributed representation. Experimental results suggest that the proposed approach has better generalization capability and improved performance than state-ofthe-art methods.

show abstract

“…Most recently, Levati et al (2017) proposed an algorithm to automatically identify an appropriate threshold from a candidate list for the reliability of predictions. The automatic selected threshold is used in the next iteration of self-training procedure.…”

Section: Related Workmentioning

confidence: 99%

Learning How to Self-Learn: Enhancing Self-Training Using Neural Reinforcement Learning

Chen

Zhang

Gao

2018

2018 International Conference on Asian Language Processing (IALP)

View full text Add to dashboard Cite

Self-training is a useful strategy for semisupervised learning, leveraging raw texts for enhancing model performances. Traditional self-training methods depend on heuristics such as model confidence for instance selection, the manual adjustment of which can be expensive. To address these challenges, we propose a deep reinforcement learning method to learn the self-training strategy automatically. Based on neural network representation of sentences, our model automatically learns an optimal policy for instance selection. Experimental results show that our approach outperforms the baseline solutions in terms of better tagging performances and stability.

show abstract

Self-training for multi-target regression with tree ensembles

Cited by 55 publications

References 45 publications

A Weighted Voting Ensemble Self-Labeled Algorithm for the Detection of Lung Abnormalities from X-Rays

A Weighted Voting Ensemble Self-Labeled Algorithm for the Detection of Lung Abnormalities from X-Rays

Opinion Spam Detection based on Annotation Extension and Neural Networks

Learning How to Self-Learn: Enhancing Self-Training Using Neural Reinforcement Learning

Contact Info

Product

Resources

About