Machine learning was used to predict biochar yield and carbon contents based on biomass and pyrolysis data. The study found that random forest accurately predicted these values using biomass characteristics and pyrolysis conditions. Pyrolysis conditions had a higher impact on biochar yield and carbon content than biomass characteristics. Structural information of biomass was crucial for predicting biochar yield, while element compositions were more important for carbon content. The analysis showed how different factors influenced biochar production and the interactions between them. This research provides new insights into the pyrolysis process of biomass and ways to improve biochar yield and carbon content.