To predict first-year grades by including data on relationships with people inside and outside the university
DOI:
https://doi.org/10.52731/lir.v003.160Keywords:
Dropout, Predicting Performance, relation data, Random Forest, Logistic RegressionAbstract
In this study, in order to predict first-year credits, an important indicator for preventing students from dropping out of college, we created a model using three types of data: basic data before entering college, relationship data with non-students, and credit data, and conducted comparative verification for each combination of the three types of data. In the experiment, the accuracy of the model improved as the number of data was increased from basic data, relationship data, and unit data. In order to accurately predict students at high risk of dropping out, the number of credits for the spring semester of the first year should be available after the number of credits is available, but in order not to overlook high-risk students, we used a logistic regression model with an awareness of recall and showed that prediction is possible to some extent using basic data and relational data.
References
F. Del Bonifro, M. Gabbrielli, G. Lisanti, and S. P. Zingaro, “Student Dropout Prediction,” in Artificial Intelligence in Education, 2020, pp. 129–140.
武内清., “学生文化の実態と大学教育,” 高等教育研究, vol. 11, pp. 7–23, 2008.
河野銀子, “大学大衆化時代における’First-Generation'の位相,” 山形大学紀要 教育科学, vol. 13, no. 2, pp. 127–143, Jan. 2003.
A. Hellas et al., “Predicting academic performance: a systematic literature review,” in Proceedings Companion of the 23rd Annual ACM Conference on Innovation and Technology in Computer Science Education, Larnaca, Cyprus, Jul. 2018, pp. 175–199.
白鳥成彦, 大石哲也, 田尻慎太郎, 森雅生, and 室田真男, “中退確率の遷移を用いた中退学生の類型化,” 日本教育工学会論文誌, vol. 44, no. 1, pp. 11–22, 2020.
N. V. Chawla, K. W. Bowyer, L. O. Hall, and W. P. Kegelmeyer, “SMOTE: Synthetic Minority Over-sampling Technique,” J. Artif. Intell. Res., vol. 16, pp. 321–357, Jun. 200
