Abalone age prediction and analysis of influencing factors: combining machine learning, SHAP, and correlation coefficient methods

In biology and fishery research, accurate prediction of the age of abalone helps to deeply understand the life characteristics of abalone and provides a basis for scientific abalone culture. The traditional method of abalone age prediction is to infer the age by observing the number of annual rings on the shell of abalone, which is widely used but has great limitations in accuracy and operability due to the influence of individual differences in abalone and environmental factors. This study aims to provide a more accurate and efficient method for abalone age prediction using machine learning techniques combined with Shapley’s additive explanatory model(SHAP) and to analyze the key factors affecting the prediction results. By comparing six machine learning models using evaluation metrics such as MAE, RMSE, and MAPE, the experimental results showed that the XGBoost model outperformed the other models in terms of prediction performance, with its MAE value reduced by 28.5% compared to other methods. In addition, the SHAP model was used to further reveal the main features affecting abalone age prediction, such as abalone shell morphology and environmental factors, which provided a more interpretable analytical framework for abalone age prediction.