Aiming at the challenges including long‑tailed salary distribution, difficulties in categorical‑feature processing, time‑series data leakage, and performance bottlenecks of single models in AI‑job salary prediction, this paper constructs a complete data preprocessing pipeline and proposes a LightGBM‑MLP weighted fusion prediction model. In the preprocessing stage, IQR‑based outlier elimination, logarithmic transformation for skewed‑distribution correction, categorical‑feature encoding, and time‑series dataset splitting are adopted to avoid temporal leakage. The LightGBM gradient‑boosting tree and regularized optimized MLP sub‑models are trained separately. Weight coefficients of 0.6 and 0.4 are assigned according to their performance on structured datasets for model fusion. Multiple baseline comparison experiments are carried out on the AI Job Market recruitment dataset. Experimental results show that the proposed fusion model achieves the highest coefficient of determination "R" ^"2"  "=0.4165"  and the lowest root‑mean‑square error "RMSE=23994.12" . Time‑series 5‑fold cross‑validation demonstrates low standard deviations for all metrics and good generalization stability, which can provide predictive references for AI‑job salaries.

Yuman Zou. LightGBM‑MLP Weighted Fusion Model for AI‑Job Salary Prediction. Academic Journal of Computing & Information Science (2026), Vol. 9, Issue 9: 8-14. https://doi.org/10.25236/AJCIS.2026.090902.

[1] Sofianopoulos, S., & Mavridis, P., “Comparative analysis of machine learning models for employee salary prediction,” Proceedings of the 12th International Conference on Information Systems and Technologies, Athens, Greece, 2021.

[2] Ali, M., & Gupta, S., “Machine learning approaches to salary prediction in human resource payroll systems,” Journal of Computing and Social Science, 5(2):45‑54, 2022.

[3] Li, Y., & Zhang, H., “A model for predicting salaries in big data roles: An integration of Random Forest and AdaBoost‑KNN models,” ACM Conference on Big Data and Artificial Intelligence, New York, USA, 189‑194, 2023.

[4] Novák, J., & Kučera, T., “Explainable AI and robust forecasting of global salary trends: Addressing data drift and unseen categories by tree‑based models,” Mathematical Modeling and Computing, 12(3):993‑1004, 2025.

[5] Shahin, N., “Job market cheat codes: Prototyping salary prediction and job grouping with synthetic job listings,” arXiv preprint arXiv:2506.15879, 2025.

[6] Sun, Y., Zhuang, F., Zhu, H., et al., “An interpretable salary prediction algorithm based on collective utility marginal contribution learning,” Journal of Computer Research and Development, 61(5):1276‑1289, 2024.

[7] Chen, L., & Zhang, Q., “Salary management data analysis and annual salary prediction based on artificial intelligence algorithms,” 2024 7th International Conference on Data‑Driven Intelligent Systems, New York, USA, 218‑223, 2024.

[8] Mobin, M., Zhang, T. Y., & Hu, B., “Interpretable salary prediction in the AI sector using ensemble learning with uncertainty quantification,” International Conference on Data Science and Human Resources, Piscataway, USA: IEEE, 2024.

[9] Wang, H., “Predicting AI position salaries: Model fusion via weighted averaging and feature correlation analysis,” ACM International Conference on Artificial Intelligence and Employment, New York, USA, 211‑216, 2024.

[10] Nidal, S., & Ahmad, M., “Ensemble regression strategies for job‑posting salary prediction,” 2025 International Conference on Applied AI and Data Analytics, Amman, Jordan: IEEE, 2025.