Tracking recurring contexts using ensemble classifiers: an application to email filtering

229 citations recorded by the source corpus.

How researchers see this paper

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

Decentralized Online Learning: Take Benefits from Others’ Data without Sharing Your Own to Track Global Trend (2019, background): “Both usenet2 and spam are łconcept driftž [20] datasets, for which the optimal model changes over time.”
On evaluating stream learning algorithms (2012, background): “…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 (2011) for decision rules, Li, Wu, and Hu (2010); Katakis, Tsoumakas, and…”
Learning under Concept Drift: an Overview (2010, methodology): “[3 7, 74 , 80 , 16 2] ye s (i ns t.”
A Survey on Ensemble Learning for Data Stream Classification (2017, methodology): “2011b], recurrent concepts [Nishida et al. 2005; Ramamurthy and Bhatnagar 2007; Katakis et al. 2010; Jaber et al. 2013a; Gonçalves Jr and de Barros 2013], novel class detection [Masud et al.”
Learning under Concept Drift: A Review (2019, background): “11 Spam_data [112] 9324 499 2 [4], [5], [23], [36], [104], [118]”
Proximal Online Gradient Is Optimum for Dynamic Regret: A General Lower Bound (2021, background): “Both usenet2 and spam are “concept drift” datasets [52], for which the optimal model changes over time.”