A Survey on Supervised Classification on Data Streams

Lemaire, Vincent; Salperwyck, Christophe; Bondu, Alexis

doi:10.1007/978-3-319-17551-5_4

Cited by 40 publications

(14 citation statements)

References 83 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Intermediate nodes in the tree contains a test on a particular attribute that distribute data points in the different sub-trees. The objective from the learning phase is to identify groups of examples as consistent as possible while in the test phase the new example is assigned to the majority class of the leaf based on a score that corresponds to the proportion of training examples in the leaf that belong to the same class [6].…”

Section: Decision Trees Induction C45mentioning

confidence: 99%

“…An SVM model is a representation of the examples as points in the space such as they are divided by a clear gap as wide as possible. As [6] explains, 'the main idea is to maximize the distance between the separating hyper-plane and the closest training example'. After that, new examples are mapped to the same space where it is predicted that they belong to the gap they fall in [11].…”

Section: Support Vector Machinesmentioning

confidence: 99%

“…This approach as other approaches has some advantages such as its low training and prediction time complexity in addition to its low variance. It shines in classification problems where only few training examples are available [6].…”

Section: Naïve Bayesmentioning

confidence: 99%

See 2 more Smart Citations

Comparison of Classification Techniques for Wall Following Robot Navigation and Improvements to the KNN Algorithm

Madi¹,

Baba-Ali²

2019

9th International Conference on Computer Science, Engineering and Applications (ICCSEA 2019)

View full text Add to dashboard Cite

Autonomous navigation is an important feature that allows the robot to move independently from a point to another without a tele-operator. This feature makes mobile robots useful in many tasks that require transportation, exploration, surveillance, guidance, inspection …etc. Furthermore, autonomous robots deal with real time environments that tend to be complex, nonlinear and partially observed. They also operate with limited memory resources and tight time constraints. In this paper, we present an investigation related to mobile robot navigation. We first compare a group of classification algorithms using real traces of wall following robot navigation. Then we focus on the k Nearest Neighbors (KNN) algorithm to improve it and help it be more applicable in autonomous robot navigation. We applied a Nearest Neighbor set reduction technique to help reduce the high running time of KNN. The results indicate that KNN is a competing algorithm especially after decreasing the running time significantly by a factor of 19 and combining that with the KNN's existing features. Results are further improved by applying an attribute selection method.

show abstract

Section: Decision Trees Induction C45mentioning

confidence: 99%

Section: Support Vector Machinesmentioning

confidence: 99%

See 1 more Smart Citation

Comparison of Classification Techniques for Wall Following Robot Navigation and Improvements to the KNN Algorithm

Madi¹,

Baba-Ali²

2019

9th International Conference on Computer Science, Engineering and Applications (ICCSEA 2019)

View full text Add to dashboard Cite

show abstract

“…If a batch of data is generated from a variety of environments at the same time, the algorithms are more difficult to handle them. Some of the batch algorithms have utilized the sliding window mechanism, such as the single classifier and passive batch algorithms "STAGGER and FLORA" [4,5]. These algorithms have set up the dynamic window adjustment mechanism.…”

Section: Relevant Algorithmsmentioning

confidence: 99%

A fast learn++.NSE classification algorithm based on weighted moving average

Shen¹,

Zhu²,

Du³

et al. 2018

Filomat

View full text Add to dashboard Cite

Current researches of incremental classification learning algorithms mainly focus on learning from data in a stationary environment. The incremental learning in a non-stationary environment (NSE), where the underlying data probability distribution changes over time, however, has received much less attentions despite the abundant real applications have generated the long-term and cumulative big data in NSE. Thus, the incremental learning in NSE has gradually received extensive attentions. Nevertheless, the popular incremental classification learning algorithms currently for NSE such as SEA and DWM generally place strict restrictions on the changes. These algorithms can only deal with gradual drift and noncyclical and no new category situations. Therefore, it is highly necessary to develop a novel efficient incremental classification learning algorithm for the gradually cumulative big data in complex NSE. The recently proposed Learn++.NSE algorithm is an important research achievement in this field. However, the vote weight of each base-classifier of the Learn++.NSE depends on its whole error rates in the environments experienced. Therefore, the classification learning efficiency of the Learn++.NSE should be further improved. A novel fast Learn++.NSE algorithm based on weighted moving average (WMA-Learn++.NSE) is presented in this paper, which computes the weighted average of error rates using the sliding window technology to optimize the weight calculation. By only using the recent classification error rates of each base-classifier inside the sliding window to calculate the vote weight, the WMA-Learn++.NSE accelerates the compute of vote weight and improves the efficiency of classification learning. The verification experiments and performance analyses on both synthetic and real data set are presented in this paper. The experimental results show that the WMA-Learn++.NSE can achieve a higher execution efficiency compared to the Learn++.NSE in getting the equivalent classification correct rate.

show abstract

“…There are several interesting books or surveys on the data stream analysis and classification, but most of them focus on general methods of data stream analysis, not dedicating too much space to ensemble approaches [43,60,64,114,131] , and some have been written several years ago [59,107,109] . Therefore, there is still a gap in this literature with respect to present the development in learning ensembles from data streams.…”

Section: Introductionmentioning

confidence: 99%

Ensemble learning for data stream analysis: A survey

et al. 2017

View full text Add to dashboard Cite

In many applications of information systems learning algorithms have to act in dynamic environments where data are collected in the form of transient data streams. Compared to static data mining, processing streams imposes new computational requirements for algorithms to incrementally process incoming examples while using limited memory and time. Furthermore, due to the non-stationary characteristics of streaming data, prediction models are often also required to adapt to concept drifts. Out of several new proposed stream algorithms, ensembles play an important role, in particular for non-stationary environments. This paper surveys research on ensembles for data stream classification as well as regression tasks. Besides presenting a comprehensive spectrum of ensemble approaches for data streams, we also discuss advanced learning concepts such as imbalanced data streams, novelty detection, active and semisupervised learning, complex data representations and structured outputs. The paper concludes with a discussion of open research problems and lines of future research.

show abstract

A Survey on Supervised Classification on Data Streams

Cited by 40 publications

References 83 publications

Comparison of Classification Techniques for Wall Following Robot Navigation and Improvements to the KNN Algorithm

Comparison of Classification Techniques for Wall Following Robot Navigation and Improvements to the KNN Algorithm

A fast learn++.NSE classification algorithm based on weighted moving average

Ensemble learning for data stream analysis: A survey

Contact Info

Product

Resources

About