Predicting Theropod Hunting Tactics using Machine Learning.

Abstract

The use of machine learning in different fields is becoming a more common practice thanks to Big Data and better granularity in data being collected. The application of machine learning to animal behavioral pattern analysis is becoming more popular due to the increase in size, types, and quality of data that can be gathered. Machine learning can even be used to predict the actual behavior of animals based off of certain features. This approach can also be used for predicting the behavior of extinct animals. This paper is the goal is to explore the possibility of using machine learning techniques to predict the hunting habits of dinosaurs based solely off of physical characteristic of the animal. By using the biomechanical features, a model can be created to aid in the classification of animals into either a scavenger or hunter roles. The results from the test show that there is a strong correlation between the physical characteristics and potential hunting habits. The models used here can then be used as a good baseline in predicting other theropods based solely on their bodies. The T-Rex was used as the test subject and was correctly classified as a primary hunter in most of the models.