Analysing Spectroscopic Data Using Hierarchical Cooperative Maximum Likelihood Hebbian Learning
A novel approach to feature selection is presented in this paper, in which the aim is to visualize and extract information from complex, high dimensional spectroscopic data. The model proposed is a mixture of factor analysis and exploratory projection pursuit based on a family of cost functions proposed by Fyfe and MacDonald  which maximizes the likelihood of identifying a specific distribution in the data while minimizing the effect of outliers [9,12]. It employs cooperative lateral connections derived from the Rectified Gaussian Distribution [8,14] to enforce a more sparse representation in each weight vector. We also demonstrate a hierarchical extension to this method which provides an interactive method for identifying possibly hidden structure in the dataset.
KeywordsCost Function Energy Function Spectroscopic Data Support Vector Regression Learning Rule
Unable to display preview. Download preview PDF.
- 3.Charles, D.: Unsupervised Artificial Neural Networks for the Identification of Multiple Causes in Data. PhD thesis, University of Paisley (1999)Google Scholar
- 4.Fyfe, C.: Negative Feedback as an Organising Principle for Artificial Neural Networks, PhD Thesis, Strathclyde University (1995)Google Scholar
- 6.Fyfe, C., Corchado, E.: Maximum Likelihood Hebbian Rules. In: European Symposium on Artificial Neural Networks (2002)Google Scholar
- 8.Seung, H.S., Socci, N.D., Lee, D.: The Rectified Gaussian Distribution. Advances in Neural Information Processing Systems 10, 350 (1998)Google Scholar
- 9.Smola, A.J., Scholkopf, B.: A Tutorial on Support Vector Regression. Technical Report NC2-TR-1998-030, NeuroCOLT2 Technical Report Series (1998)Google Scholar
- 12.Fyfe, C., MacDonald, D.: ε-Insensitive Hebbian learning. Neuro Computing (2002)Google Scholar
- 14.Corchado, E., Fyfe, C.: Orientation Selection Using Maximum Likelihood Hebbian Learning. International Journal of Knowledge-Based Intelligent Engineering Systems 7(2), Brighton, United Kingdom (April 2003) ISSN: 1327-2314Google Scholar
- 15.Bishop, C.M.: Neural Networks for Pattern Recognition, Oxford (1995)Google Scholar
- 16.Corchado, E., MacDonald, D., Fyfe, C.: Maximum and Minimum Likelihood Hebbian Learning for Exploratory Projection Pursuit. In: Data mining and Knowledge Discovery, Kluwer Academic Publishing, Dordrecht (in press)Google Scholar