A biclustering approach for classification with mislabeled data

França, Fabrício Olivetti de; Coelho, André L. V.

doi:10.1016/j.eswa.2015.02.045

Cited by 10 publications

(6 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Recently, coherence biclustering has been also adopted as a source of accuracy improvement in the supervised classification of objects. For example, de França et al [26,12] have proposed novel algorithms for coping with multilabel classification and classification with noisy labels, respectively, by replacing or augmenting the feature space through the elicitation of good discriminative biclusters. By this means, novel binary features are extracted, each representing a discriminative bicluster between two classes of instances, and novel instances can be classified according to the way they match to these local patterns.…”

Section: Coherent and Discriminative Biclusteringmentioning

confidence: 99%

“…A possible extension to the present work is to adapt BicNeuron to handle multiclass and multilabel data sets [26] in a more straightforward manner. Moreover, we shall analyze how tolerant are the BicNeuron classifiers to noisy data [12].…”

Section: Final Remarksmentioning

confidence: 99%

“…Such definition generalizes the well-known concept of a cluster [19] in the sense that now instances can be grouped together considering only subsets of the features, and vice-versa, allowing for the elicitation of more contextualized data models [22]. Since different notions of similarity and bicluster representation can be adopted, different bicluster models have been investigated in the preceding years, giving birth to a number of biclustering algorithms deployed in several application domains [7,8,16,32,17,18,13,12].…”

Section: Introductionmentioning

confidence: 99%

“…Conventional bicluster analyses are unsupervised in nature, meaning that the class labels possibly associated with the instances (or features) are not taken into account. More recently, another class of algorithms, referred to as discriminative (or differential, supervised) biclustering algorithms, has gained increased attention [24,11,31,12]. These methods somehow incorporate class label details into the process of eliciting biclusters so as to produce better predictive models.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Enhancing perceptrons with contrastive biclusters

Coelho

França

2016

Electron. lett.

View full text Add to dashboard Cite

Perceptrons are neuronal devices capable of fully discriminating linearly separable classes. Although straightforward to implement and train, their applicability is usually hindered by non-trivial requirements imposed by real-world classification problems. Therefore, several approaches, such as kernel perceptrons, have been conceived to counteract such difficulties. In this paper, we investigate an enhanced perceptron model based on the notion of contrastive biclusters. From this perspective, a good discriminative bicluster comprises a subset of data instances belonging to one class that show high coherence across a subset of features and high differentiation from nearest instances of the other class under the same features (referred to as its contrastive bicluster). Upon each local subspace associated with a pair of contrastive biclusters a perceptron is trained and the model with highest area under the receiver operating characteristic curve (AUC) value is selected as the final classifier. Experiments conducted on a range of data sets, including those related to a difficult biosignal classification problem, show that the proposed variant can be indeed very useful, prevailing in most of the cases upon standard and kernel perceptrons in terms of accuracy and AUC measures.

show abstract

Section: Coherent and Discriminative Biclusteringmentioning

confidence: 99%

Section: Final Remarksmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Enhancing perceptrons with contrastive biclusters

Coelho

França

2016

Electron. lett.

View full text Add to dashboard Cite

show abstract

“…Indeed, true labels corresponding to the state of system are usually unavailable. Mislabeling may occur for several reasons including expert errors, lack of information or data labeling by nonexperts (de França & Coelho, 2015). Label uncertainty is an important issue in classification, because most classifiers are built on the hypothesis of a perfectly labeled training set.…”

Section: Introductionmentioning

confidence: 99%

Data-based fault detection in chemical processes: Managing records with operator intervention and uncertain labels

Askarian

Benítez

Graells

et al. 2016

Expert Systems with Applications

View full text Add to dashboard Cite

Developing data-driven fault detection systems for chemical plants requires managing uncertain data labels and dynamic attributes due to operator-process interactions.Mislabeled data is a known problem in computer science that has received scarce attention from the process systems community. This work introduces and examines the effects of operator actions in records and labels, and the consequences in the development of detection models. Using a state space model, this work proposes an iterative relabeling scheme for retraining classifiers that continuously refines dynamic attributes and labels.Three case studies are presented: a reactor as a motivating example, flooding in a simulated de-Butanizer column, as a complex case, and foaming in an absorber as an industrial challenge. For the first case, detection accuracy is shown to increase by 14% while operating costs are reduced by 20%. Moreover, regarding the de-Butanizer column, the performance of the proposed strategy is shown to be 10% higher than the filtering strategy. Promising results are finally reported in regard of efficient strategies to deal with the presented problem.

show abstract

A Survey of Methods for Detection and Correction of Noisy Labels in Time Series Data

Atkinson

Metsis

2021

IFIP Advances in Information and Communication Technology

View full text Add to dashboard Cite

A biclustering approach for classification with mislabeled data

Cited by 10 publications

References 38 publications

Enhancing perceptrons with contrastive biclusters

Enhancing perceptrons with contrastive biclusters

Data-based fault detection in chemical processes: Managing records with operator intervention and uncertain labels

A Survey of Methods for Detection and Correction of Noisy Labels in Time Series Data

Contact Info

Product

Resources

About