- 期刊
Why 70/30 or 80/20 Relation Between Training and Testing Sets: A Pedagogical Explanation
摘要
When learning a dependence from data, to avoid overfitting, it is important to divide the data into the training set and the testing set. We first train our model on the training set, and then we use the data from the testing set to gauge the accuracy of the resulting model. Empirical studies show that the best results are obtained if we use 20-30% of the data for testing, and the remaining 70-80% of the data for training. In this paper, we provide a possible explanation for this empirical result.
延伸閱讀
- 彭喬渝（2019）。Analysis on the Pragmatic Strategies of Refusal Responses and Its Pedagogical Applications〔碩士論文，國立臺灣師範大學〕。華藝線上圖書館。https://doi.org/10.6345/NTNU201900922
- Chen, C. Y. (2020). Improving Meta-Learning by Regularized Pre-training [master's thesis, National Taiwan University]. Airiti Library. https://doi.org/10.6342/NTU202001626
- Chan, C. H. (1996). Examination performance, self-efficacy and attributional retraining [doctoral dissertation, The University of Hong Kong]. Airiti Library. https://www.airitilibrary.com/Article/Detail?DocID=U0029-1812201200007497
- Lin, H. T. (2015). Using Extended Classifier System for Pedagogical Tactics Optimization in Tutoring System [master's thesis, National Chiao Tung University]. Airiti Library. https://doi.org/10.6842/NCTU.2015.00104
- Shah, S. S. H., Fatima, M., Waris, S., Aziz, J., Jaffari, A. R., Ejaz, W., & Sherazi, S. K. (2012). The Causal Relationship of Training on Organizational Performance. Asian Journal of Business Management, 4(2), 111-113. https://www.airitilibrary.com/Article/Detail?DocID=20418752-201204-201508130021-201508130021-111-113