Application of machine learning to growth model in fisheries
Abstract
Traditional growth models, such as length-weight relationships (LWRs) and the von Bertalanffy (VB) growth function, have been widely used in fishery science. Their limitations in capturing nonlinear patterns necessitate alternative approaches. Machine learning (ML) techniques have recently gained attention as a powerful tool for enhancing predictive accuracy in biological studies. In this study, the growth parameter of Eastern mosquitofish, Gambusia holbrooki (135 females: 21–58.78 mm and 0.152–3.424 g; 59 males: 19.25–43.20 mm; 0.108–1.075 g), was determined with traditional LWRs, VB, and machine learning algorithms. The LWRs growth equations of female and male individuals were W=0.00002102 L2.8849 and W=0.00003064 L2.8212, respectively. The VB equations were determined Lt=80.990 [1–e–0.990(t+0.208)] for female and Lt=64.172 [1-e–0.610(t+0.271)] for male. In general, the performance of both methods (VB model and ML algorithms) in predicting lengths, as measured by Mean Absolute Percentage Error (MAPE), was satisfactory, with the VB model demonstrating slightly superior performance (2.734). In addition, the ML algorithm gives better results in length data prediction with multilayer perceptron and in weight data prediction with Sequential Minimum Optimization (SMO) algorithm when ML algorithms are examined. The diverse ML algorithms positively impacted the investigations addressing growth-related issues in fisheries.
References
Armstrong JS, Collopy F (1992) Error measures for generalizing about forecasting methods: empirical comparisons. International Journal of Forecasting 8(1): 69–80.
Beaudouin R, Ginot V, Monod G (2008) Growth characteristics of eastern mosquitofish Gambusia holbrooki in a northern habitat (Brittany, France). Journal of Fish Biology 73(10): 2468–2484.
Benzer R, Benzer S (2018) Alternative approaches for growth models: artificial neural networks (pp. 1–4). In: 2018 26th Signal Processing and Communications Applications Conference (SIU). IEEE, İzmir.
Benzer S, Benzer R (2020) Growth properties of Pseudorasbora parva in Süreyyabey Reservoir: traditional and artificial intelligent methods. Thalassas: An International Journal of Marine Sciences 36: 149–156.
Benzer S, Benzer R (2022) Morphometric analysis of Crayfish–traditional and artificial intelligent approach. Thalassas: An International Journal of Marine Sciences 38(2): 989–996.
Benzer S, Garabaghi FH, Benzer R, Mehr HD (2022) Investigation of some machine learning algorithms in fish age classification. Fisheries Research 245: 106151.
Beverton RJ, Holt SJ (2012) On the dynamics of exploited fish populations, Volume 11. Springer Science & Business Media, UK.
Bimba AT, Idris N, Al-Hunaiyyan A, Mahmud RB, Abdelaziz A, ... Chang V (2016) Towards knowledge modeling and manipulation technologies: a survey. International Journal of Information Management 36(6): 857–871.
Chai T, Draxler RR (2014) Root mean square error (RMSE) or mean absolute error (MAE)? – Arguments against avoiding RMSE in the literature. Geoscientific Model Development 7(3): 1247–1250.
Eagderi S, Radkhah A (2015) Length-weight relationship and condition factor of mosquitofish (Gambusia holbrooki) in three inland basins of Iran. Poeciliid Research 5(1): 39–43.
Erguden SA (2013) Age, growth, sex ratio and diet of eastern mosquitofish Gambusia holbrooki Girard, 1859 in Seyhan Dam Lake (Adana/Turkey). Iranian Journal of Fisheries Sciences 12(1): 204–218.
Flinn SA, Midway SR (2021) Trends in growth modeling in fisheries science. Fishes 6(1): 1.
Frank E, Hall M, Trigg L, Holmes G, Witten IH (2004) Data mining in bioinformatics using Weka. Bioinformatics 20(15): 2479–2481.
Friedrichs S, Igel C (2005) Evolutionary tuning of multiple SVM parameters. Neurocomputing 64: 107–117.
Froese R (2006) Cube law, condition factor and weight-length relationships: history, meta-analysis and recommendations. Journal of Applied Ichthyology 22: 241–253.
Goodsell JA, Kats LB (1999) Effect of introduced mosquitofish on Pacific treefrogs and the role of alternative prey. Conservation Biology 13: 921–924.
Hall M, Frank E, Holmes G, Pfahringer B, Reutemann P, Witten IH (2009) The WEKA data mining software: an update. ACM SIGKDD Explorations Newsletter, 11(1): 10–18.
Hecht-Nielsen R (1992) Theory of the backpropagation neural network. Neural networks for perception computation, learning, and architectures. Academic Press Inc, San Diego, CA. pp. 65–93.
Heinen-Kay JL, Morris KE, Ryan NA, Byerley SL, Venezia RE, ... Langerhans RB (2015) A trade-off between natural and sexual selection underlies diversification of a sexual signal. Behavioral Ecology 26(2): 533–542.
Huang L, Asteris PG, Koopialipoor M, Armaghani DJ, Tahir MM (2019) Invasive weed optimization technique-based ANN to the prediction of rock tensile strength. Applied Sciences 9(24): 5372.
ISSG (2000) 100 of the world’s worst invasive alien species: a selection from the global invasive species database. Invasive Species Specialist Group, University of Auckland, Auckland.
James G, Witten D, Hastie T, Tibshirani R (2013) An Introduction to statistical learning with applications in R. Springer, New York.
Jordan MI, Mitchell TM (2015) Machine learning: trends, perspectives, and prospects. Science 349(6245): 255–260.
Kurtul I, Sari HM (2020) Length-weight relationships of invasive mosquitofish (Gambusia holbrooki Girard, 1859) in 23 river basins of Turkey. Turkish Journal of Zoology 44(4): 324–334.
Lagler KF (1966) Freshwater fishery biology, second edition. WM. C. Brown Company. Dubuque. Iowa. 421 pp.
Leyse KE, Lawler SP, Strange T (2004) Effects of an alien fish, Gambusia affinis, on an endemic California fairy shrimp, Linderiella occidentalis: implications for conservation of diversity in fishless waters. Biological Conservation 118: 57–65.
Moon Y, Yao T (2011) A robust mean absolute deviation model for portfolio optimization. Computers & Operations Research 38(9): 1251–1258.
Osuna E, Freund R, Girosi F (1997) An improved training algorithm for support vector machines (pp. 276–285). In: Neural networks for signal processing VII. Proceedings of the 1997 IEEE signal processing society workshop. IEEE, New Jersey.
Ozcan EI (2024) Performance of artificial neural networks and traditional methods in determining selected growth parameters of Alburnus sellal Heckel, 1843. Oceanological and Hydrobiological Studies 53(2): 153–163.
Patimar R, Ghorbani M, Gol-Mohammadi A, Azimi-Glugahi H (2011) Life history pattern of mosquitofish Gambusia holbrooki (Girard, 1859) in the Tajan River (southern Caspian Sea to Iran). Chinese Journal of Oceanology and Limnology 29(1): 167–173.
Poggio T, Girosi F (1990) Regularization algorithms for learning that are equivalent to multilayer networks. Science 247: 978–982.
Pyke GH (2005) A review of the biology of Gambusia affinis and G. holbrooki. Reviews in Fish Biology and Fisheries 15(4): 339–365.
Rahman LF, Marufuzzaman M, Alam L, Bari MA, Sumaila UR, Sidek LM (2021) Developing an ensembled machine learning prediction model for marine fish and aquaculture production. Sustainability 13(16): 9124.
Rencher AC, Schaalje GB (2008) Linear models in statistics. John Wiley & Sons, Hoboken, NJ.
Ricker WE (1973) Linear regression in fisheries research. Journal of the Fisheries Board of Canada 30: 409–434.
Schnute JT, Richards LJ (1990) A unified approach to the analysis of fish growth, maturity, and survivorship data. Canadian Journal of Fisheries and Aquatic Sciences 47(1): 24–40.
Sellaoui N, Bounaceur F (2020) Growth and length-weight relationships of Gambusia affinis (Baird et Girard, 1853) population in Algeria (Cyprinodontiformes Poeciliidae). Biodivers J 11(4): 951–959.
Sparre P, Venema SC (1998) Introduction to tropical fish stock assessment. Part 1 manual. FAO Fisheries Technical Paper (306.1, Rev. 2), Food and Agriculture Organization of the United Nations, Rome. 407 pp.
Sun Z, Deng Z, Zhang Z (2016) Intelligent control theory and technology, 2nd edition. Tsinghua University Press, Beijing, China.
Suryanarayana I, Braibanti A, Rao RS, Ramam VA, Sudarsan D, Rao GN (2008) Neural networks in fisheries research. Fisheries Research 92(2–3): 115–139.
Tesch FW (1968) Age and growth. In: Ricker WE (Ed) Methods for assessment of fish production in freshwaters. Blackwell Scientific Publications, Oxford, UK. pp. 93–123.
Vapnik V (1995) The nature of statistical learning theory. Springer, New York.
Wang W, Xu Z (2004) A heuristic training for support vector regression. Neurocomputing 61: 259–275.
Willmott CJ, Matsuura K (2005) Advantages of the mean absolute error (MAE) over the root mean square error (RMSE) in assessing average model performance. Climate Research 30(1): 79–82.
Xiong W, Tao J, Zhang DC, Liu C, He DK, Chen YF (2015) Length-weight relationships for four small fish species caught in wetland of central Yangtze River, China. Journal of Applied Ichthyology 31: 219–220.
Xiong W, Zhu GP, Wang ZL, Ye N (2018) Length‐weight relationships of four fish species from mangrove of Zhanjiang, China. Journal of Applied Ichthyology 34(1): 167–168.
Zane L, Nelson WS, Jones AG, Avise JC (1999) Microsatellite assessment of multiple paternity in natural populations of a live-bearing fish, Gambusia holbrooki. Journal of Evolutionary Biology 12: 61–69.
Zhang Y, Zhang F, Cheng J, Zhao H (2021) Classification and recognition of fish farming by extraction new features to control the economic aquatic product. Complexity 2021: 5530453.
Zhao S, Zhang S, Liu J, Wang H, Zhu J, ... Zhao R (2021) Application of machine learning in intelligent fish aquaculture: a review. Aquaculture 540: 736724.

Copyright (c) 2025 The Author(s)

This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.