Research on Training Performance Prediction Model Based on LightGBM

Authors

  • Jingzhe Peng
  • Feiyang Pan
  • Jiaxing Li
  • Yixin Duan

DOI:

https://doi.org/10.62051/84817739

Keywords:

LightGBM Model; Training Performance Prediction; Particle Swarm Algorithm; Gradient Boosting Decision Tree.

Abstract

This study aims to build an accurate performance prediction model to assist the training and development of skilled workers. This article uses the advanced LightGBM algorithm and combines it with rich technical worker training data in a certain industry to predict trainees' training results. In the data processing stage, this article conducted detailed data cleaning to eliminate outliers and ensure the accuracy and reliability of the data. At the same time, by introducing new variables, this paper further enhances the feature expression ability of the model, thereby improving the accuracy of prediction. To further optimize the model performance, this article innovatively uses the particle swarm algorithm to fine-tune the parameters of LightGBM. Experimental results show that the optimized model performs well in predicting students' training performance, with not only high prediction accuracy but also good stability. This model is expected to become a powerful tool for the training and development of skilled workers, helping various industries to evaluate and improve training effects more efficiently, thereby better meeting the needs of modern production and promoting the improvement of the overall quality of the skilled worker team.

Downloads

Download data is not yet available.

References

[1] BREJC M, ŠIROK K, KOREN A. Training as strategy for school self-evaluation capacity building [J/OL]. Management, 2019, 24: 73-87. http://dx.doi.org/10.30924/ mjcmi.24.si.5. DOI: 10.30924/mjcmi. 24. si. 5. DOI: https://doi.org/10.30924/mjcmi.24.si.5

[2] ZHANG Y, AN R, CUI J, et al. Undergraduate Grade Prediction in Chinese Higher Education Using Convolutional Neural Networks[C/OL]//LAK21: 11th International Learning Analytics and Knowledge Conference. 2021. http://dx.doi.org/10.1145/3448139.3448184. DOI: 10.1145/3448139. 3448184. DOI: https://doi.org/10.1145/3448139.3448184

[3] MAURYA L S, HUSSAIN M S, SINGH S. Developing Classifiers through Machine Learning Algorithms for Student Placement Prediction Based on Academic Performance[J/OL]. Applied Artificial Intelligence, 2021, 35 (6): 403 - 420. http://dx.doi.org/10.1080/08839514.2021.1901032. DOI:10.1080/08839514.2021.1901032. DOI: https://doi.org/10.1080/08839514.2021.1901032

[4] BAAK M A, KOOPMAN R F, SNOEK H, et al. A new correlation coefficient between categorical, ordinal and interval variables with Pearson characteristics[J]. Cornell University - arXiv, 2018.

[5] CHANDRASHEKAR G, SAHIN F. A survey on feature selection methods[J/OL]. Computers & Electrical Engineering, 2014: 16-28. http://dx.doi.org/10.1016/j.compeleceng.2013.11.024. DOI: 10.1016/j.compeleceng.2013. 11. 024. DOI: https://doi.org/10.1016/j.compeleceng.2013.11.024

[6] LIANG Y, WU J, WANG W, et al. Product marketing prediction based on XGboost and LightGBM algorithm[C/OL]//Proceedings of the 2nd International Conference on Artificial Intelligence and Pattern Recognition. 2019. http://dx.doi.org/10.1145/3357254.3357290. DOI: 10. 1145/3357254. 3357290. DOI: https://doi.org/10.1145/3357254.3357290

[7] BENTÉJAC C, CSÖRGŐ A, MARTÍNEZ-MUÑOZ G. A comparative analysis of gradient boosting algorithms[J/OL]. Artificial Intelligence Review, 2021: 1937 - 1967. http://dx.doi.org/10.1007/s10462-020-09896-5. DOI: 10.1007/s10462 - 020 - 09896 - 5. DOI: https://doi.org/10.1007/s10462-020-09896-5

[8] SUN X, LIU M, SIMA Z. A novel cryptocurrency price trend forecasting model based on Light [J/OL]. Finance Research Letters, 2020: 101084. http: //dx.doi.org/10.1016/j.frl.2018. 12. 032. DOI: 10.1016/j.frl.2018. 12. 032. DOI: https://doi.org/10.1016/j.frl.2018.12.032

[9] WANG D, TAN D, LIU L. Particle swarm optimization algorithm: an overview[J/OL]. Soft Computing, 2018: 387 - 408. http: //dx.doi.org/10.1007/s00500-016-2474-6. DOI: 10.1007/s00500-016-2474 - 6.

[10] ASTERIS P, JAIN M, SAIHJPAL V, et al. An Overview of Variants and Advancements of PSO Algorithm [J]. APPLIED SCIENCES-BASEL, 2022.

[11] DEGENHARDT F, SEIFERT S, SZYMCZAK S. Evaluation of variable selection methods for random forests and omics data sets[J/OL]. Briefings in Bioinformatics, 2019: 492 - 503. http: //dx.doi.org/10.1093/bib/bbx124. DOI:10.1093/bib/bbx124. DOI: https://doi.org/10.1093/bib/bbx124

Downloads

Published

12-08-2024

How to Cite

Peng, J. (2024) “Research on Training Performance Prediction Model Based on LightGBM”, Transactions on Computer Science and Intelligent Systems Research, 5, pp. 1470–1475. doi:10.62051/84817739.