Text mining for chemical information: the ChiKEL project
As scientific and patent literature expands, we need more efficient ways to find and extract information. Text mining is already being used successfully to analyse sets of documents after they are found by structure search, in a two‐step process. Integrating name‐to‐structure and structure search directly within an interactive text mining system enables structure search to be mixed with linguistic constraints for more precise filtering. This talk will describe work done in partnership between ChemAxon and Linguamatics in the EU funded project, ChiKEL, including improvements made to name‐to‐structure software, how we evaluated this, and the approach taken to integrating name to structure within the text mining platform, I2E.