Anti-germ Performance Prediction for Detergents Based on Elman Network on Small Data Sets
Anti-germ performance test is critical in the production of detergents. However, actual biochemical tests are often costly and time consuming. In this paper, we present a neural network based model to predict the performance. The model made it much faster and cost less than doing actual biochemical tests. We also present preprocessing methods that can reduce data conflicts while keeping the monotonicity on small data sets. This model performs well though the training data sets are small. Its input is the actual value of key ingredients, which is not widely used in solving biochemical problems. The results of experiments are generated on the base of two detergent products for two types of bacteria, and appear reasonable according to natural rules. The prediction results show a high precision and fitting with the monotonicity rule mostly. Experts in biochemical area also give good evaluations to the proposed model.
KeywordsAnti-germ performance prediction Artificial neural networks Monotonicity rule Pre-processing methods
Unable to display preview. Download preview PDF.
- 7.Petrocci, A.M., Clarke, P.: Proposed Test Method for Antimicrobial Laundry Additives. Journal of the Association of Official Analytical Chemists 52, 836–842 (1969)Google Scholar
- 9.Demuth, H., Beale, M., Hagan, M.: Neural Network Toolbox 5 User’s Guide. The MathWorks, Inc. (2007)Google Scholar