Simple encoding of infrared spectra for pattern recognition Part 2. Neural network approach using back-propagation and associative Hopfield memory

Abstract By extending an adaptive momentum back-propagation two-layer network with a final associative Hopfield memory the network's total error convergence could be improved remarkably. This design enables simultaneous calculations of the network's weights and biases ( batch calculating network). Using only energy-orientated inputs of the mid-infrared spectra of 104 multi-functional carbonyl compounds, the networks were trained by 25 structutal features 104-fold for each of three input sets (19, 27 and 38 inputs). In a comprehensive statistical investigation the behavior was studied of the network's response to the increase of artificially produced noise to the inputs. Some of the chosen structural features to train the network remain reliable by increasing the disturbance of the input data and can be related to special regions of the original infrared spectra. Therefore the resulting network design could be suitable to verify the reliability of further structural features for classes of organic compounds other than carbonyl compounds.

[1]  Wendelin Feiten,et al.  Improving neural net training by mathematical optimization , 1992 .

[2]  Vladimir Nedeljkovic.  A novel multilayer neural networks training algorithm that minimizes the probability of classification error , 1993, IEEE Trans. Neural Networks.

[3]  Peter de B. Harrington,et al.  Optimal Associative Memory for Background Correction of Spectra , 1994 .

[4]  Wolfgang Porod,et al.  Analysis and synthesis of a class of neural networks: variable structure systems with infinite grain , 1989 .

[5]  Heskes,et al.  Learning in neural networks with local minima. , 1992, Physical review. A, Atomic, molecular, and optical physics.

[6]  Gerrit Kateman,et al.  Two-dimensional mapping of IR spectra using a parallel implemented self-organising feature map , 1993 .

[7]  Johann Gasteiger,et al.  Neural nets for mass and vibrational spectra , 1993 .

[8]  J.R.M. Smits,et al.  Practical implementation of neural networks for the interpretation of infrared spectra , 1993 .

[9]  John J. Hopfield,et al.  Simple 'neural' optimization networks: An A/D converter, signal decision circuit, and a linear programming circuit , 1986 .

[10]  Rainer Herges,et al.  Automatic interpretation of infrared spectra: recognition of aromatic substitution patterns using neural networks , 1992, J. Chem. Inf. Comput. Sci..

[11]  Dov Ingman,et al.  Local minimum escape using thermodynamic properties of neural networks , 1991, Neural Networks.

[12]  Dong-Jo Park,et al.  Novel fast training algorithm for multilayer feedforward neural network , 1992 .

[13]  R. Fessenden,et al.  Identifying functional groups in IR spectra using an artificial neural network , 1991 .

[14]  Henrik Schulz,et al.  Simple encoding of infrared spectra for pattern recognition. 1: Statistical examination of the effectiveness and information content by factor analysis , 1995 .

[15]  Tom Tollenaere,et al.  SuperSAB: Fast adaptive back propagation with good scaling properties , 1990, Neural Networks.

[16]  Geoffrey E. Hinton,et al.  Learning representations by back-propagating errors , 1986, Nature.

[17]  Astrom.  Computer Controlled Systems , 1990 .

[18]  Mark A. Kramer,et al.  Improvement of the backpropagation algorithm for training neural networks , 1990 .

[19]  Charles L. Wilkins,et al.  A novel algorithm for local minimum escape in back-propagation neural networks: application to the interpretation of matrix isolation infrared spectra , 1994, J. Chem. Inf. Comput. Sci..

[20]  Wolfgang Porod,et al.  Qualitative analysis and synthesis of a class of neural networks , 1988 .

[21]  M. Meyer,et al.  Neural networks for interpretation of infrared spectra using extremely reduced spectral data , 1993 .

[22]  Charles L. Wilkins,et al.  Evaluation and comparison of pattern classifiers for chemical applications , 1976 .

[23]  Kazutoshi Tanabe,et al.  Neural Network System for the Identification of Infrared Spectra , 1992 .

[24]  Claudio Nicolini,et al.  Modeling and analysis in biomedicine , 1984 .

[25]  A. Michel,et al.  Analysis and synthesis of a class of neural networks: linear systems operating on a closed hypercube , 1989 .

[26]  Daniel Cabrol-Bass,et al.  Neural network approach to structural feature recognition from infrared spectra , 1993, J. Chem. Inf. Comput. Sci..

[27]  J. Zupan,et al.  Neural networks: A new method for solving chemical problems or just a passing phase? , 1991 .

[28]  M. Meyer,et al.  Interpretation of infrared spectra by artificial neural networks , 1992 .

[29]  J J Hopfield,et al.  Neurons with graded response have collective computational properties like those of two-state neurons. , 1984, Proceedings of the National Academy of Sciences of the United States of America.

[30]  Morton E. Munk,et al.  A neural network approach to infrared spectrum interpretation , 1990 .

[31]  Leonardo Reyneri,et al.  Modified backpropagation algorithm for fast learning in neural networks , 1990 .

[32]  Bernard Widrow,et al.  Improving the learning speed of 2-layer neural networks by choosing initial values of the adaptive weights , 1990, 1990 IJCNN International Joint Conference on Neural Networks.

[33]  M. Munk,et al.  Neural network models for infrared spectrum interpretation , 1991 .

[34]  A. Savitzky,et al.  Smoothing and Differentiation of Data by Simplified Least Squares Procedures. , 1964 .

[35]  Gerrit Kateman,et al.  Interpretation of infrared spectra with modular neural-network systems , 1993 .

[36]  Charles J. Pouchert.  The Aldrich library of FT-IR spectra , 1985 .