Development of text-mining toolkit for extracting metal organic framework synthesis method

Abstract

As big data becomes more important, automated research on information extraction using text mining is being actively conducted. In this study, we develop an algorithm that automatically extracts synthesis information from papers of metal organic framework using text mining. Use bag of words model to classify synthesis paragraphs of the paper. Using mechanical learning, recognize materials from synthesis paragraphs and extract the names of metal organic frameworks and precursors. Using natural language processing, we detect the extensive property and condition of the synthesis paragraph, and match with each material. The extracted materials are divided into metal precursors, organic precursors, and solvents to store synthesis information.