Using Abalone’s Physical Features to Predict its Age

Abstract

Abalone is one of the most delicious and highly-prized seafood around the world, its deliciousness also makes the abalone industry a non-negligible part of the global economic circle, a lot of people and countries rely on abalone for their lives and economy. Therefore, it means a lot for us to study about abalone and it’s population. However, as a necessary step when studying about abalone, getting the age of abalone is a very complicated and time-consuming task. That’s why we need a model to help us predict the age of abalone according to it’s physical measurements which are easy to acquire. The project considered three models, linear regression model, polynomial regression model and Random Forest. 10-fold cross validation is used to compute the mean square error, multiple R square is also considered when evaluating the models. In the results, the polynomial model is the best model among three models, with lowest mean square error and largest R-square. The research provides us with a model to get the age of abalone in an easy and convenient way, which makes the study about abalone more convenient and thus be beneficial for the development of abalone industry.