Challenges in mining the literature for chemical information

Abstract

Chemical information extracted from the literature is of immense value for the pharmaceutical and chemical industries in many areas, including supporting drug discovery, manufacturing processes, or intellectual property protection. However, the exponential growth of the chemical literature has made it increasingly difficult for researchers to find the information they need within a reasonable time-frame. In order to address this issue, a large number of text mining approaches have been developed that can extract chemical information from different types of literature. But the lack of a single universal standard for chemical structure and nomenclature representation has posed significant challenges in mining the chemical information. Hence, a review on the current state of chemical text mining, problems confronted, solutions available, and future prospectus is presented.