Proceedings of the 11th International Applied Business and Engineering Conference, ABEC 2023, September 21st, 2023, Bengkalis, Riau, Indonesia

Research Article

The Comparison of Ridge Regression Method and Lasso Regression Method to Predict The Graduation Time

Download40 downloads
  • @INPROCEEDINGS{10.4108/eai.21-9-2023.2342995,
        author={Humaira  Humaira and Nikita  Chairunnisa and Novi  Novi and Yulia Jihan Sy and Rika  Idmayanti},
        title={The Comparison of Ridge Regression Method and Lasso Regression Method to Predict The Graduation Time},
        proceedings={Proceedings of the 11th International Applied Business and Engineering Conference, ABEC 2023, September 21st, 2023, Bengkalis, Riau, Indonesia},
        publisher={EAI},
        proceedings_a={ABEC},
        year={2024},
        month={2},
        keywords={prediction graduation time ridge regression lasso regression mse},
        doi={10.4108/eai.21-9-2023.2342995}
    }
    
  • Humaira Humaira
    Nikita Chairunnisa
    Novi Novi
    Yulia Jihan Sy
    Rika Idmayanti
    Year: 2024
    The Comparison of Ridge Regression Method and Lasso Regression Method to Predict The Graduation Time
    ABEC
    EAI
    DOI: 10.4108/eai.21-9-2023.2342995
Humaira Humaira1,*, Nikita Chairunnisa1, Novi Novi1, Yulia Jihan Sy1, Rika Idmayanti1
  • 1: Politeknik Negeri Padang
*Contact email: humaira@pnp.ac.id

Abstract

Every year, graduates from the department of information technology are produced. There are not as many graduates as there are new students each year. This is as a result of the high rate of late graduations. The Department of Information Technology has a problem with this. Making an intelligent system is the answer to these issues. Intelligent system to estimate students' graduation times. Data from past years' graduations of students was gathered and trained. Utilizing regression to train on training data. The two types of regression are compared in this article: Ridge and Lasso. The prediction model's outputs had an accuracy of 92.22% and an MSE of 0.084, which is the best possible result. Ridge Regression, which produces the best prediction model, was used. Using its coefficients, Lasso Regression can identify the factors that have the greatest impact on the desired value.