The Extraction of Information and Knowledge from Trained Neural Networks
In the past, neural networks were viewed as classification and regression systems whose internal representations were incomprehensible. It is now becoming apparent that algorithms can be designed that extract comprehensible representations from trained neural networks, enabling them to be used for data mining and knowledge discovery, that is, the discovery and explanation of previously unknown relationships present in data. This chapter reviews existing algorithms for extracting comprehensible representations from neural networks and outlines research to generalize and extend the capabilities of one of these algorithms, TREPAN. This algorithm has been generalized for application to bioinformatics data sets, including the prediction of splice site junctions in human DNA sequences, and cheminformatics. The results generated on these data sets are compared with those generated by a conventional data mining technique (C5) and appropriate conclusions are drawn.
- The Activity-Based Anorexia Mouse Model
- Molecular Probes for PNS Neurotoxicity, Degeneration, and Regeneration
- Y4 Receptor in Different Species: Functional Expression and Binding
- Hybrid Neuronal Network Studies Under Dynamic Clamp
- Determination of Histamine in Microdialysis Samples from the Rodent Brain by Column Liquid Chromatography
- Detection of Receptor Heteromers Involving Dopamine Receptors by the Sequential BRET-FRET Technology
- Intrathecal Delivery of Stem Cells to the Spinal Cord
- Cerebral Vascular Function in Genetically Altered Mice
- Determination of Adrenergic Receptor mRNAs by Quantitative Reverse Transcriptase-Polymerase Chain Reactions
- Fluorescent Angiogenesis Models Using Gelfoam Implanted in Transgenic Mice Expressing Fluorescent Proteins