Extremely Fast Decision Tree
By Chaitanya Manapragada, Geoffrey I. Webb, Mahsa Salehi. 2018
179 citations recorded by the source corpus.
We introduce a novel incremental decision tree learning algorithm, Hoeffding Anytime Tree, that is statistically more efficient than the current state-of-the-art, Hoeffding Tree. We demonstrate that an implementation of Hoeffding Anytime Tree---"Extremely Fast Decision Tree'', a minor modification to the MOA implementation of Hoeffding Tree---obtains significantly superior prequential accuracy on most of the largest classification datasets from the UCI repository. Hoeffding Anytime Tree produces the asymptotic batch tree in the limit, is naturally resilient to concept drift, and can be used as a higher accuracy replacement for Hoeffding Tree in most scenarios, at a small additional computational cost.
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
- Shutao Mei
- Fuyi Li
- André Leier
- Tatiana T. Marquez-Lago
- K. Giam
- Nathan P. Croft
- Tatsuya Akutsu
- A. Ian Smith
Citations in the corpus
- Memory Efficient Experience Replay for Streaming Learning (2018, neutral): “[37], [38], [39], [56], [63] or ensemble methods [7], [9], [27], [44], [46], [51], [54], [61], [70], [72].”
- Deep Neural Networks and Tabular Data: A Survey (2021, unknown): “For this reason, despite the success of deep learning in other domains, less complex methods, such as incremental decision trees [207], [208], are often preferred in online learning applications.”
- Cost-Effective Incremental Deep Model: Matching Model Capacity With the Least Sampling (2023, unknown): “Furthermore, with the newly labeled data, we compare our model with several state-of-the-art incremental methods, i.e., DNN-SGD, Adwin-b [58], EFDT [59], and ODLD [27], note that EFDT and ODLD are both incremental ensemble methods.”
- A comprehensive review and performance evaluation of bioinformatics tools for HLA class I peptide-binding prediction (2020, methodology): “It is based on the conventional Hoeffding Tree [113] but with a minor modification that allows it to achieve a significantly improved predictive performance on most of the largest classification data sets [114].”