||The volume of chemical documents has been rapidly increasing in the last years. To evaluate the newly discovered chemical reaction in those, process design is necessary to more accurately estimate CAPEX (Capital Expenditure) and OPEX (Operating Expenditure). However, there is no database containing sufficient data for process design such as conversion, composition, temperature, pressure. Therefore, process engineers spend a lot of time searching reaction data among the growing collection of papers and patents. With the advancement of Natural language processing, chemical text mining has been noticed as key solution for transforming unstructured data into a more structured data format easy to process for analysis. In this research, we investigated the effectiveness of the state-of-the-art Artificial Intelligence-based Natural Language Processing Model, BERT(Bidirectional Encoder Representations from Transformers), on extracting reaction information.