Resumen
The article focuses on scientific knowledge discovery using inductive logic programming (ILP). Scientific knowledge discovery tasks can be carried out using ILP. Other discovery areas of ILP include linguistic features in natural language data and patterns in highway traffic data. ILP can be more effective than neural nets for delving into biological function data for pharmaceutical engineers developing new drugs and their multiple varieties. There is an interactive cycle between human analysis and machine learning. Initially, traditional methods process the data and develop representations that characterize the system and rules describing the relationship between the components of the system. Next, machine learning uses these representations to identify new and hopefully more powerful and incisive, rules. Computer-based scientific discovery should support strong integration into the existing social environment of human scientific communities. The knowledge discovered should add to and builds on existing science. The ability to incorporate background knowledge and reuse learned knowledge, together with the coherence and lucidity of the hypotheses, have marked ILP as a particularly effective approach for scientific knowledge discovery.