Accurate decision trees for mining high-speed data streams

By João Gama, Ricardo Rocha, Pedro Medas. Knowledge Discovery and Data Mining, 2003

355 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Information Systems

View Paper: https://www.semanticscholar.org/paper/b635adfae0d08d00162c09c957931ded07495f20

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Information Systems

Citations in the corpus

Speeding-Up Hoeffding-Based Regression Trees With Options (2011, background): “In all these situations, the data ﬂows continuously and dy-Appearing Algorithms for incremental learning of classiﬁcation (Domingos & Hulten, 2000; Gama et al., 2003; 2004), regression and model trees (Potts & Sammut, 2005; Ikonomovska et al., 2010) represent a well established line of research on…”
Learning under Concept Drift: A Review (2019, background): “2 Covertype [108] 581012 54 7 [13], [23], [35], [36], [59], [64], [74], [83], [84], [85], [88], [89], [104], [117]”
Decision trees: a recent overview (2011, methodology): “To further improve VFDT, Gama et al. (2003) extended the VFDT system in two directions: the ability to deal with continuous data and the use of more powerful classification techniques at tree leaves.”
New Options for Hoeffding Trees (2007, methodology): “The sufficient statistics turn out to be beneficial for both tree growth and prediction as they can be used to build Naive Bayes models at the leaves of the tree that are more accurate than majority class estimates [ 7 ].”
Learning model trees from evolving data streams (2010, methodology): “Another representative algorithm is VFDTc (Gama et al. 2003), which also performs continuous monitoring of the differences in the class-distributions of the examples from two time-windows.”
Leveraging Bagging for Evolving Data Streams (2010, methodology): “It contains 581, 012 instances and 54 attributes, and it has been used in several papers on data stream classification [18, 25].”
Incremental on-line learning: A review and comparison of state of the art algorithms (2018, background): “It is often used as a benchmark for drift algorithms [82, 83, 84].”
Issues in evaluation of stream learning algorithms (2009, methodology): “Most recent learning algorithms [6, 1, 10, 20, 15, 12] maintain a decision model that continuously evolve over time, taking into account that the environment is non-stationary and computational resources are limited.”
A Survey of Classification Methods in Data Streams (2007, methodology): “While the selection of the techniques is based on the performance and quality of addressing the research challenges, there are a number of other methods [11, 8, 15, 22, 311 which we have not discussed in greater detail in this chapter.”
Online Machine Learning in Big Data Streams (2018, background): “s [13]. For example, concept drift adaptive online decision trees based on a statistical change detector that works on sliding windows are described in [26, 24]. More examples include Bayesian models [94, 19], neural networks [95, 153], and SVM [215, 137]. Concept drift adaptation methods exist for ”
Decision trees for mining data streams (2006, neutral): “We show that incremental tree induction methods that wait till enough statistical support for install a split-test will large beneﬁt of using more appropriate classiﬁcation strategies at tree leaves [12].”
On evaluating stream learning algorithms (2012, background): “…and O’Callaghan (2003); Rodrigues, Gama, and Pedroso (2008) for clustering, Domingos and Hulten (2000); Hulten, Spencer, and Domingos (2001); Gama, Rocha, and Medas (2003); Liang, Zhang, and Song (2010) for decision trees, Ferrer-Troyano, Aguilar-Ruiz, and Riquelme (2004); Gama and Kosina…”
VHT: Vertical hoeffding tree (2016, background): “This work has been the basis for a large number of improvements, such as dealing with concept drift [12] and handling continuous numeric attributes [11].”