Reacting to Different Types of Concept Drift: The Accuracy Updated Ensemble Algorithm

Brzeziński, Dariusz; Stefanowski, Jerzy

doi:10.1109/tnnls.2013.2251352

Cited by 376 publications

(241 citation statements)

References 28 publications

Supporting

Mentioning

230

Contrasting

Unclassified

Order By: Relevance

“…Optimizing Deep Neural Network model also plays a vital role in improving the results. It is observed in general that Deep Neural Network model gives accurate results than individual machine learning model, even whatever method we use to combine different machine learning models [12]- [14]. The Deep Neural Network method is the new method in the predictive modeling world.…”

Section: Discussionmentioning

confidence: 99%

Optimizing a Deep Learning Model in Order to Have a Robust Neural Network Topology

Khan¹,

Kumar²

2018

IJMO

View full text Add to dashboard Cite

Abstract-In this study, a method based on different feature engineering / feature extraction / feature derivation is proposed for improving air passenger forecasting by machine learning existing libraries. In this kind of formulation, we kept focus on creating different kinds of datasets that differ one from another by methodology so we extracted new features and compared new feature space with original feature space in terms of variable importance. We conducted experiments to improve the variance by aggregating all the features in final feature space. Finally, we optimized a deep learning model to have a Robust Neural Network Topology.Index Terms-Deep learning, logistic regression, feature extraction, neural network.

show abstract

Section: Discussionmentioning

confidence: 99%

Optimizing a Deep Learning Model in Order to Have a Robust Neural Network Topology

Khan¹,

Kumar²

2018

IJMO

View full text Add to dashboard Cite

show abstract

“…2. The AUE ensemble method (AUE method has been proved to deal well with recurrent concept drifts, although it does not provide a reduction of the training instances needed) presented in Brzezinski and Stefanowski (2013), using 10 Hoeffding Tree classifiers on it. 3.…”

Section: Methodsmentioning

confidence: 99%

“…In this way, most existing proposals do not exploit this aspect and have to learn new concepts from scratch even if they are recurrent as it is the case of Brzezinski and Stefanowski (2011) and Brzezinski and Stefanowski (2013) where an ensemble mechanism is used to deal with concept drift. Similarly, in Katakis et al (2010) an ensemble is also used, but incremental clustering is performed to maintain information on historical concepts.…”

Section: Data Stream Classificationmentioning

confidence: 99%

Predicting recurring concepts on data-streams by means of a meta-model and a fuzzy similarity function

Ángel

Bártolo²,

Ernestina

2016

Expert Systems with Applications

View full text Add to dashboard Cite

Stream-mining approach is defined as a set of cutting-edge techniques designed to process streams of data in real time, in order to extract knowledge. In the particular case of classification, stream-mining has to adapt its behavior to the volatile underlying data distributions, what has been called concept drift. It is important to note that concept drift may lead to situations where predictive models become invalid and have therefore to be updated to represent the actual concepts that data poses. In this context, there is a specific type of concept drift, known as recurrent concept drift, where the concepts represented by data have already appeared in the past. In those cases the learning process could be saved or at least minimized by applying a previously trained model. To deal with the aforementioned scenario, meta-models can be used in the process of enhancing the drift detection mechanisms used by data stream algorithms, by representing and predicting when the change will occur. There are some real-world situations where a concept reappears, as in the case of intrusion detection systems (IDS), where the same incidents or an adaptation of them usually reappear over time. In these environments the early prediction of drift by means of a better knowledge of past models can help to anticipate to the change, thus improving efficiency of the model regarding the training instances needed. Furthermore, as a complement of meta-models, a mechanism to assess the similarity between classifica-tion models is also needed when dealing with recurrent concepts. In this context, when reusing a previously trained model a rough comparison between concepts is usually made, applying boolean logic. The intro-duction of fuzzy logic comparisons between models could lead to a better efficient reuse of previously seen concepts, by applying not just equal models, but also similar ones. This work faces the aforementioned open issues by means of the MM-PRec system, that integrates a meta-model mechanism and a fuzzy similarity function. The theoretical proposal of MM-PRec is also validated in this paper by means of different experiments using both synthetic and real datasets.

show abstract

“…Brezinski et al [22] methods including the Naïve Bayes algorithm in reference to memory costs, processing time, and classification accuracy using various drift scenarios. The experiments were implemented on the MOA framework using real-world and synthetic datasets representing concept drifts occurring at various rates such as gradual, recurrent, and sudden.…”

Section: The Accuracy Updated Ensemble (Aue2)mentioning

confidence: 99%

Hoeffding Tree Algorithms for Anomaly Detection in Streaming Datasets: A Survey

Muallem¹,

Shetty²,

Pan³

et al. 2017

JIS

View full text Add to dashboard Cite

This survey aims to deliver an extensive and well-constructed overview of using machine learning for the problem of detecting anomalies in streaming datasets. The objective is to provide the effectiveness of using Hoeffding Trees as a machine learning algorithm solution for the problem of detecting anomalies in streaming cyber datasets. In this survey we categorize the existing research works of Hoeffding Trees which can be feasible for this type of study into the following: surveying distributed Hoeffding Trees, surveying ensembles of Hoeffding Trees and surveying existing techniques using Hoeffding Trees for anomaly detection. These categories are referred to as compositions within this paper and were selected based on their relation to streaming data and the flexibility of their techniques for use within different domains of streaming data. We discuss the relevance of how combining the techniques of the proposed research works within these compositions can be used to address the anomaly detection problem in streaming cyber datasets. The goal is to show how a combination of techniques from different compositions can solve a prominent problem, anomaly detection.

show abstract

Reacting to Different Types of Concept Drift: The Accuracy Updated Ensemble Algorithm

Cited by 376 publications

References 28 publications

Optimizing a Deep Learning Model in Order to Have a Robust Neural Network Topology

Optimizing a Deep Learning Model in Order to Have a Robust Neural Network Topology

Predicting recurring concepts on data-streams by means of a meta-model and a fuzzy similarity function

Hoeffding Tree Algorithms for Anomaly Detection in Streaming Datasets: A Survey

Contact Info

Product

Resources

About