An Information Theoretic Approach to Rule Induction from Databases
IEEE Transactions on Knowledge and Data Engineering
Experiments in Meta-level Learning with ILP
PKDD '99 Proceedings of the Third European Conference on Principles of Data Mining and Knowledge Discovery
An Apriori-Based Algorithm for Mining Frequent Substructures from Graph Data
PKDD '00 Proceedings of the 4th European Conference on Principles of Data Mining and Knowledge Discovery
Biochemical Knowledge Discovery Using Inductive Logic Programming
DS '98 Proceedings of the First International Conference on Discovery Science
DS '00 Proceedings of the Third International Conference on Discovery Science
| Hi-index | 0.00 |
The meta-analysis of the challenging data set on the mutagenicity of nitroaromatic compounds has been performed. There are two ways of structure coding: standard topological indexes or so-called fingerprint descriptors. In our previous work, a unique structure coding by fingerprint descriptors was used for the discovery of mutagenes with GUHA+/- software system. GUHA can process nominal variables, which are transformed to binary strings in the course of computation. Any structure coding can then be used for GUHA. The data encoded by topological indexes were processed by GUHA+/- software system as well. The hypotheses on the reasons for mutagenicity of nitroaromatic compounds were generated by GUHA+/- for Windows. Processing of data encoded by topological indexes was rather demanding because of the large number of structure descriptors. Meta-analysis by combining fingerprint descriptors for a posteriori structure templates resulting from previous analyses and more flexible topological indexes seems to be more appropriate.