Improving Adaptive Bagging Methods for Evolving Data Streams

Bifet, Albert; Holmes, Geoff; Pfahringer, Bernhard; Gavaldà, Ricard

doi:10.1007/978-3-642-05224-8_4

Cited by 50 publications

(26 citation statements)

References 16 publications

(23 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For bagging, using 10,000 instances for training in the classifier, the percentage correctly classified 8 For a more thorough analysis (more datasets), please see [20,174]. The main problem with this online method of boosting and bagging, is its inability to deal with drifting concepts since it assumes a stationary distribution [49].…”

Section: Overviewmentioning

confidence: 99%

“…close t−i − SMA(n 2 )) 2MovAvgVar(n 1 , n 2 ) = Moving Average of Variance(n 1 ) Moving Average of Variance(n 2 ) Demonstrating price with MovAvgVar(5,20) …”

mentioning

confidence: 99%

“…In our paper we use n = 5, 10, 20 for corrHighLow(n) and k = 1, 2, 3; n = 5, 10, 20 for corrClose(k, n). See Demonstrating price with corrClose(3,20) the following calculation:…”

mentioning

confidence: 99%

See 2 more Smart Citations

Machine-learning classification techniques for the analysis and prediction of high-frequency stock direction

Rechenthin¹

View full text Add to dashboard Cite

This thesis explores predictability in the market and then designs a decision support framework that can be used by traders to provide suggested indications of future stock price direction along with an associated probability of making that move.Markets do not remain stable and approaches that are highly predictive at one moment may cease to be so as more traders spot the patterns and adjust their trading techniques. Ideally, if these "concept drifts" could be anticipated, then the trader could store models to use with each specific market condition (or concept) and later apply those models to incoming data. The assumption however is that the future is uncertain, therefore future concepts are still undecided.Maintaining a model with only the most up-to-date price data is not necessarily the most ideal choice since the market may stabilize and old knowledge may become useful again. Additionally, decreasing training times to enable modified classifiers to work with streaming high-frequency stock data may result in decreases in performance (e.g. accuracy or AUC) due to insufficient learning times. Our framework takes a different approach to learning with drifting concepts, which is to assume that concept drift occurs and builds this into the model. The framework adapts to these market changes by building thousands of traditional base classifiers (SVMs, Decision Trees, and Neural Networks), using random subsets of past data, and covering similar (sector) stocks and heuristically combining the best of these base classifiers.This "ensemble", or pool of multiple models selected to achieve better predictive periii formance, is then used to predict future market direction. As the market moves, the base classifiers in the ensemble adapt to stay relevant and keep a high level of model performance. Our approach outperforms existing published algorithms.This thesis also addresses problems specific to learning with stock data streams, specifically class imbalance, attribute creation (e.g. technical and sentiment analysis), dimensionality reduction, and model performance due to release of news and time of

show abstract

Section: Overviewmentioning

confidence: 99%

“…close t−i − SMA(n 2 )) 2MovAvgVar(n 1 , n 2 ) = Moving Average of Variance(n 1 ) Moving Average of Variance(n 2 ) Demonstrating price with MovAvgVar(5,20) …”

mentioning

confidence: 99%

See 1 more Smart Citation

Machine-learning classification techniques for the analysis and prediction of high-frequency stock direction

Rechenthin¹

View full text Add to dashboard Cite

show abstract

“…Within the framework, it is possible to define the probability that instances of the stream belong to the new concept after the drift. It uses the sigmoid function, as an elegant and practical solution [15,21].…”

Section: Classification In Moamentioning

confidence: 99%

Streaming Data Mining with Massive Online Analytics (MOA)

Bifet¹,

Read²,

Holmes³

et al. 2018

Series in Machine Perception and Artificial Intelligence

Self Cite

View full text Add to dashboard Cite

Fast Big Data is being produced at high-velocity in real-time. To effectively deal with this type of streaming data produced in real time, we need to be able to adapt to changes on the distribution of the data being produced, and we need to do it using the minimum amount of time and memory. The Internet of Things (IoT) is a good example and motivation of this type of streaming data produced in real time.Massive Online Analytics (MOA) is a software environment for implementing algorithms and running experiments for online learning from evolving data streams. MOA is designed to deal with the challenging problem of scaling up the implementation of state of the art algorithms to real world dataset sizes. MOA includes classification and clustering methods. It contains collection of offline and online methods as well as tools for evaluation. MOA supports bi-directional interaction with WEKA, the Waikato Environment for Knowledge Analysis, and is released under the GNU GPL license.

show abstract

“…A VFDT like algorithm for learning regression and model trees appear in [39]. Bagging and boosting ensemble models, using VFDT like algorithms, appear in [14,13].…”

Section: Predictive Learning From Data Streamsmentioning

confidence: 99%

A survey on learning from data streams: current and future trends

Gama

2012

Prog Artif Intell

143

View full text Add to dashboard Cite

Nowadays, there are applications in which the data are modeled best not as persistent tables, but rather as transient data streams. In this article, we discuss the limitations of current machine learning and data mining algorithms. We discuss the fundamental issues in learning in dynamic environments like continuously maintain learning models that evolve over time, learning and forgetting, concept drift and change detection. Data streams produce a huge amount of data that introduce new constraints in the design of learning algorithms: limited computational resources in terms of memory, cpu power, and communication bandwidth. We present some illustrative algorithms, designed to taking these constrains into account, for decision-tree learning, hierarchical clustering and frequent pattern mining. We identify the main issues and current challenges that emerge in learning from data streams that open research lines for further developments.

show abstract

Improving Adaptive Bagging Methods for Evolving Data Streams

Cited by 50 publications

References 16 publications

Machine-learning classification techniques for the analysis and prediction of high-frequency stock direction

Machine-learning classification techniques for the analysis and prediction of high-frequency stock direction

Streaming Data Mining with Massive Online Analytics (MOA)

A survey on learning from data streams: current and future trends

Contact Info

Product

Resources

About