Using Inductive Logic Programming to Discover Knowledge Hidden in Chemical Data

This paper demonstrates how general purpose tools from the eld of Inductive Logic Programming ILP can be applied to analytical chemistry As far as these authors are aware this is the rst published work to describe the application of the ILP tool Golem to separation science An outline of the theory of ILP is given together with a description of Golem and previous applications of ILP The advantages of ILP over classical machine Correspondence to C H Bryant School of Computing and Mathematics The University of Hudder s eld HD DH United Kingdom induction techniques such as the Top Down Induction of Decision Tree family are explained A case study is then presented in which Golem is used to induce rules which predict with a high accuracy whether each of a series of attempted sepa rations succeed or fail The separation data was obtained from published work on the attempted separation of a series of substituted phthalide enantiomer pairs on R N dinitrobenzoyl phenylglycine