Knowledge Extraction from the Neural ‘Black Box’ in Ecological Monitoring

Phytoplankton biomass within the Saginaw Bay ecosystem (Lake Huron, Michigan, USA) was characterized as a function of select physical/chemical indicators. The complexity and variability of ecological systems typically make it difficult to model the influences of anthropogenic stressors and/or natural disturbances. Here, Artificial Neural Networks (ANNs) were developed to model chlorophyll a concentrations, a measure for water-column phytoplankton biomass and a proxy for system-level health. ANNs act like “black boxes” in the sense that relationships are encoded as weight vectors within the trained network and as such, cannot easily support the generation of scientific hypotheses unless these relationships can be explained in a comprehensible form. Accordingly, the ‘knowledge’ and/or rule-based information embedded within ANNs needs to be extracted and expressed as a set of comprehensible ‘rules’. Such extracted information would enhance the delineation and understanding of ecological complexity and aid in developing usable prediction tools. Comparisons of various computational approaches (including TREPAN, an algorithm for constructing decision trees from neural networks) used in extracting rule-based information from trained Saginaw Bay ANNs are discussed.

[1]  Jie Cheng,et al.  Applying machine learning to semiconductor manufacturing , 1993, IEEE Expert.

[2]  Sreerama K. Murthy,et al.  Automatic Construction of Decision Trees from Data: A Multi-Disciplinary Survey , 1998, Data Mining and Knowledge Discovery.

[3]  Oren Etzioni,et al.  Representation design and brute-force induction in a Boeing manufacturing domain , 1994, Appl. Artif. Intell..

[4]  I. Aoki,et al.  Analysis and prediction of the fluctuation of sardine abundance using a neural network , 1997 .

[5]  Leech Wj,et al.  A rule-based process control method with feedback. , 1987 .

[6]  Uygar Özesmi,et al.  An artificial neural network approach to spatial habitat modelling with interspecific interaction , 1999 .

[7]  A. Famili,et al.  Use of decision-tree induction for process optimization and knowledge refinement of an industrial process , 1994, Artificial Intelligence for Engineering Design, Analysis and Manufacturing.

[8]  Gary Rawlinson,et al.  Applications of Expert Systems, Volume 2 , 1990 .

[9]  J. Strickler,et al.  Automatic classification of field-collected dinoflagellates by artificial neural network , 1996 .

[10]  Julian D. Olden,et al.  An artificial neural network approach for studying phytoplankton succession , 2000, Hydrobiologia.

[11]  Donald A. Jackson,et al.  Fish–Habitat Relationships in Lakes: Gaining Predictive and Explanatory Insight by Using Artificial Neural Networks , 2001 .

[12]  Kevin J. Dooley,et al.  Distinguishing between mean, variance and autocorrelation changes in statistical quality control , 1995 .

[13]  Yan Huang,et al.  Neural network modelling of coastal algal blooms , 2003 .

[14]  Friedrich Recknagel,et al.  Ecological Informatics: Understanding Ecology by Biologically-Inspired Computation , 2003 .

[15]  Sanchoy K. Das,et al.  A decision tree approach for selecting between demand based, reorder, and JIT/kanban methods for material procurement , 1994 .

[16]  LiMin Fu,et al.  Rule Learning by Searching on Adapted Nets , 1991, AAAI.

[17]  David Biggs,et al.  A method of choosing multiway partitions for classification and decision trees , 1991 .

[18]  Jude W. Shavlik,et al.  Extracting refined rules from knowledge-based neural networks , 2004, Machine Learning.

[19]  Jude W. Shavlik,et al.  in Advances in Neural Information Processing , 1996 .

[20]  Jude W. Shavlik,et al.  Using Sampling and Queries to Extract Rules from Trained Neural Networks , 1994, ICML.

[21]  Stephen I. Gallant,et al.  Connectionist expert systems , 1988, CACM.

[22]  Tharam S. Dillon,et al.  Automated knowledge acquisition , 1994, Prentice Hall International series in computer science and engineering.

[23]  P. Culverhouse,et al.  Automatic classification of field-collected dinoflagellates by artificial neural network , 1996 .

[24]  Ding-Geng Chen,et al.  A neural network model for forecasting fish stock recruitment , 1999 .

[25]  Sylvie Thiria,et al.  Applying artificial neural network methodology to ocean color remote sensing , 1999 .

[26]  Yagmur Denizhan,et al.  Classification trees prove useful in nondestructive testing of spot weld quality , 1993 .

[27]  G. David Garson,et al.  Interpreting neural-network connection weights , 1991 .

[28]  Chandrasekhar Ganduri,et al.  Rule Driven Job-Shop Scheduling Derived from Neural Networks through Extraction , 2004 .

[29]  Jude W. Shavlik,et al.  Extracting Refined Rules from Knowledge-Based Neural Networks , 1993, Machine Learning.

[30]  Sebastian Thrun,et al.  Extracting Rules from Artifical Neural Networks with Distributed Representations , 1994, NIPS.

[31]  Thomas G. Dietterich.  What is machine learning? , 2020, Archives of Disease in Childhood.

[32]  Douglas H. Fisher,et al.  Overcoming process delays with decision tree induction , 1994, IEEE Expert.

[33]  Michael J. Shaw,et al.  Learning-based scheduling in a flexible manufacturing flow line , 1994 .

[34]  Ingrid M. Schleiter,et al.  Modelling water quality, bioindication and population dynamics in lotic ecosystems using neural networks , 1999 .

[35]  Joachim Diederich,et al.  Survey and critique of techniques for extracting rules from trained artificial neural networks , 1995, Knowl. Based Syst..

[36]  Mark Craven,et al.  Extracting comprehensible models from trained neural networks , 1996 .

[37]  W. Leech.  A rule-based process control method with feedback. , 1987, ISA transactions.