Comparison of XGBoost and the Neural Network model on the class-balanced datasets
Authors:
- 12 November 2021
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 457-461
Abstract
The Extreme Gradient Boosting method and Deep learning methods are classical machine learning methods widely used in many fields. However, the advantages and disadvantages of these two methods have been a long-debated topic. This paper evaluates the classification performance of the XGBoost and Multiple-Layer Perceptron Neural Network on the structured data. We compare the classification performance of both methods, using the large scale, public datasets, and show the overall trend with the different percentages of datasets used. The experiment validated the higher accuracy in all datasets obtained through the XGBoost method. We concluded that the XGBoost method overcomes the complex data distribution with the feature space to better classify performance on the structured data.Keywords
This publication has 10 references indexed in Scilit:
- Machine Learning Interpretability: A Survey on Methods and MetricsElectronics, 2019
- DeepRED – Rule Extraction from Deep Neural Networks2016
- "Why Should I Trust You?"2016
- Principal component analysis: a review and recent developmentsPhilosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences, 2016
- An improved K-nearest-neighbor algorithm for text categorizationExpert Systems with Applications, 2012
- Principal component analysisWIREs Computational Statistics, 2010
- Rule extraction from neural networks via decision tree induction2002
- Extracting rules from trained neural networksIEEE Transactions on Neural Networks, 2000
- ANN-DT: an algorithm for extraction of decision trees from artificial neural networksIEEE Transactions on Neural Networks, 1999
- Connectionist expert systemsCommunications of the ACM, 1988