Avian Hazard Mitigation: A Classification Approach to Bird Strike Analysis

Bird strikes during landing or takeoff can result in fatalities and catastrophic accidents, raising concerns about loss of life for the aviation sector. This study explored the potential of employing machine-learning algorithms to mitigate bird strikes in aviation. Bird strike data were collected, and the Boruta algorithm feature-selection technique was applied to identify relevant features for an experiment. Irrelevant features were omitted, and relevant features related to bird strikes were identified for analysis. The multinomial Naive Bayes algorithm is then used to segregate objects based on their features. The experimental results illustrate the power of using machine learning techniques to segregate and analyze bird strike data, thereby aiding in determining the causes and effects of damage.