Transition metals catalyzed reactions raised more attention in the development of organic synthesis. For example, nickel has been used to perform the cleavage of inert C-O bonds during coupling reaction. In recent decades, scientists have discovered varieties of catalytic cycle that are based on several readily available oxidation states that are commonly invoked by transition metals. The amount of chemical research using transition metals is increasing sharply every year, providing a great source of data to analyze using state-of-the-art machine learning based analysis techniques. In this project, we will conduct two-state research to study transition metals catalyzed reactions: first, build a high-quality database by extracting information from scientific publications; second, develop a machine learning model to analyze collected data.