ABSTRACT
Computer-assisted extraction or mining of chemical structural information from the literature requires special tools that address the various ways of encoding structures. Traditionally, in the literature, chemical structures are identifi ed by textual names or images of structures. Chemical images of structures are in general very explicit and can convey a great deal of information to a chemist, but they cannot be read by computers (Ibison et al. 1993). To make these images machine-readable would involve a chemical image recognition capability as described in Chapter 4. This is a challenging area, and Chapter 4 provides the reader with an excellent background on what to expect when using these capabilities.
