Abstract
Labor potential estimation is a complex analysis task of a large data sets of digital profiles of workers. Digital profile is a structure that contains heterogeneous data and has high complexity. The attribute space of digital profiles datasets has a large dimension. Individual features are distinguished by a large variability of types and include qualitative and quantitative characteristics that can be both continuous or discrete. This fact significantly complicates the use of traditional statistical methods and hinder manual selection of machine learning algorithms for labor potential estimation. Automated machine learning (AutoML) can help to deal with this problem. The results of AutoML can significantly depend on the used AutoML library. AutoML libraries differ in the set of algorithms among which they select the best one, in the constrains that are used to stop the search of the optimal solution, in the techniques used for hyperparameters optimization and in metrics that are calculated for the obtained results. In this work we develop an AutoML framework that unites several commonly used AutoML libraries and enables use of common constrains and metrics. Processing dataset about employees’ performance using the developed framework we show the applicability of the AutoML for the task of labor potential estimation.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Patutina, E.S.: Main approaches to the interpretation of the concepts of “labor potential” and “human potential” as the basis for the quality of the labor force in the works of domestic and foreign scientists. In: Science Research Practice. Collection of Selected Articles Based on the Materials of the International Scientific Conference, pp. 215–220 (2020)
Human Resources Management: Concepts, Methodologies, Tools, and Applications, p. 1513. IGI Global, Pennsylvania (2012)
Page, S.E.: What sociologists should know about complexity. Ann. Rev. Sociol. 41(1), 21–41 (2015). https://doi.org/10.1146/annurev-soc-073014-112230
Heravi, G., Eslamdoost, E.: Applying artificial neural networks for measuring and predicting construction-labor productivity. J. Constr. Eng. Manag. 141(10) (2016)
Ogiy, O.G., Osipov, V.Yu., Tristanov, A.B., Zhukova, N.A.: The process of managing labor potential of the fishery complex as an object of modeling using artificial neural networks. In: AIP Conference Proceedings, vol. 2661, p. 030001 (2022)
Thornton, C., Hutter, F., Hoos, H.H., Leyton-Brown, K.: Auto-WEKA: combined selection and hyperparameter optimization of classification algorithms. In: 19th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD 2013. ACM, New York (2013). https://doi.org/10.1145/2487575.2487629
Guo, X., van Stein, B., Bäck, T.: A new approach towards the combined algorithm selection and hyper-parameter optimization problem. In: 2019 IEEE Symposium Series on Computational Intelligence (SSCI), pp. 2042–2049. IEEE, Xiamen (2019). https://doi.org/10.1109/ssci44817.2019.9003174
Auto-WEKA. https://www.cs.ubc.ca/labs/algorithms/Projects/autoweka/. Accessed 07 Apr 2023
Hyperopt-sklearn. https://hyperopt.github.io/hyperopt-sklearn/. Accessed 10 Apr 2023
Feurer, et al.: Auto-Sklearn 2.0: Hands-Free AutoML via Meta-Learning. arXiv (2020)
Olson, R.S., Bartley, N., Urbanowicz, R.J., Moore, J.H.: Evaluation of a tree-based pipeline optimization tool for automating data science. In: Proceedings of GECCO 2016, pp. 485–492 (2016)
H2O.AI - The fastest, most accurate AI Cloud Platform. https://h2o.ai. Accessed 10 Apr 2023
Jin, H., Chollet, F., Song, Q., Hu, X.: AutoKeras: an AutoML library for deep learning. J. Mach. Learn. Res. 6, 1–6 (2023)
Scikit-Learn. Machine Learning in Python. https://scikit-learn.org/stable/. Accessed 07 Apr 2023
Weka 3: Machine Learning Software in Java. https://www.cs.waikato.ac.nz/ml/weka/. Accessed 07 Apr 2023
Hyperopt: Distributed Asynchronous Hyper-parameter Optimization. http://hyperopt.github.io/hyperopt/. Accessed 07 Apr 2023
OpenML. A worldwide machine learning lab. https://www.openml.org. Accessed 07 Apr 2023
TensorFlow. An end-to-end open source machine learning platform for everyone. https://www.tensorflow.org. Accessed 07 Apr 2023
AutoML Framework for AutoML libraries comparison. https://github.com/DarkEol/AutoML-Framework. Accessed 07 Apr 2023
Kaggle. https://www.kaggle.com. Accessed 07 Apr 2023
Employee Performance Prediction. https://www.kaggle.com/datasets/gauravduttakiit/employee-performance-prediction. Accessed 07 Apr 2023
Human Resources Data Set. https://www.kaggle.com/datasets/rhuebner/human-resources-data-set. Accessed 07 Apr 2023
IBM HR Analytics Employee Attrition & Performance. https://www.kaggle.com/datasets/pavansubhasht/ibm-hr-analytics-attrition-dataset. Accessed 07 Apr 2023
Employee Performance Analysis INX Future Inc. https://www.kaggle.com/datasets/eshwarganta/employee-performance-analysis-inx-future-inc/. Accessed 07 Apr 2023
Acknowledgments
This work was supported by a state grant (project No. FFZF-2022-0006) and as part of the agreement on the provision of subsidies from the federal budget for financial support for the implementation of the state task for 2023 (registration number of the EGISU NIOKTR card 1122110800010-7 28/11/2022).
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Kovalevsky, V., Stankova, E., Zhukova, N., Ogiy, O., Tristanov, A. (2023). AutoML Framework for Labor Potential Modeling. In: Gervasi, O., et al. Computational Science and Its Applications – ICCSA 2023. ICCSA 2023. Lecture Notes in Computer Science, vol 13957. Springer, Cham. https://doi.org/10.1007/978-3-031-36808-0_6
Download citation
DOI: https://doi.org/10.1007/978-3-031-36808-0_6
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-36807-3
Online ISBN: 978-3-031-36808-0
eBook Packages: Computer ScienceComputer Science (R0)