Accuracy Updated Ensemble for Data Streams with Concept Drift
نویسندگان
چکیده
In this paper we study the problem of constructing accurate block-based ensemble classifiers from time evolving data streams. AWE is the best-known representative of these ensembles. We propose a new algorithm called Accuracy Updated Ensemble (AUE), which extends AWE by using online component classifiers and updating them according to the current distribution. Additional modifications of weighting functions solve problems with undesired classifier excluding seen in AWE. Experiments with several evolving data sets show that, while still requiring constant processing time and memory, AUE is more accurate than AWE.
منابع مشابه
Detecting Concept Drift in Data Stream Using Semi-Supervised Classification
Data stream is a sequence of data generated from various information sources at a high speed and high volume. Classifying data streams faces the three challenges of unlimited length, online processing, and concept drift. In related research, to meet the challenge of unlimited stream length, commonly the stream is divided into fixed size windows or gradual forgetting is used. Concept drift refer...
متن کاملHandling Gradual Concept Drift in Stream Data
Data streams are sequence of data examples that continuously arrive at time-varying and possibly unbound streams. These data streams are potentially huge in size and thus it is impossible to process many data mining techniques (e.g., sensor readings, call records, web page visits). Tachiniques for classification fail to successfully process data streams because of two factors: their overwhelmin...
متن کاملBatch Weighted Ensemble for Mining Data Streams with Concept Drift
This paper presents a new framework for dealing with two main types of concept drift (sudden and gradual) in labeled data with decision attribute. The learning examples are processed instance by instance. This new framework, called Online Batch Weighted Ensemble, introduces element of incremental processing into a block-based ensemble of classi ers. Its performance was evaluated experimentally ...
متن کاملConcept Drift Adaptation by Exploiting Historical Knowledge
Incremental learning with concept drift has often been tackled by ensemble methods, where models built in the past can be re-trained to attain new models for the current data. Two design questions need to be addressed in developing ensemble methods for incremental learning with concept drift, i.e., which historical (i.e., previously trained) models should be preserved and how to utilize them. A...
متن کاملAlgorithm to handle Concept Drifting in Data Stream Mining
Data Stream Mining is the evolving field of research. Mining continuous data streams brings unique opportunities but also new challenges. This paper will describe and evaluate the proposed classifier which uses ensemble classifier along with the boosting concept. Adaptive windowing is also used for handling the data stream. Empirical study will show that the proposed classifier takes less memor...
متن کامل