Worked on the EmilHvitfeldt/xgboost repository to enhance the clarity of feature importance visualizations by updating the default x-axis label in the plot_importance function from 'F score' to 'Importance score'. This change aimed to improve interpretability for analysts and stakeholders reviewing model outputs. The update was implemented in Python, leveraging skills in data visualization and machine learning to ensure the new label accurately reflected the underlying metric. Associated tests were revised to cover the updated label, providing regression protection and maintaining code reliability. The work focused on improving end-user understanding of model insights without introducing breaking changes.