John D. Hastings: Selected Publications

Journal Articles

• John Hastings, Karl Branting, and Jeffrey Lockwood, CARMA: A Case-Based Rangeland Management Adviser, AI Magazine, 23(2): 49-62 (2002).

  Abstract: CARMA is an advisory system for rangeland grasshopper infestations that demonstrates how AI technology can deliver expert advice to compensate for cutbacks in public services. CARMA uses two knowledge sources for the key task of predicting forage consumption by grasshoppers: (1) cases obtained by asking a group of experts to solve representative hypothetical problems and (2) a numerical model of rangeland ecosystems. These knowledge sources are integrated through the technique of model-based adaptation, in which case-based reasoning is used to find an approximate solution, and the model is used to adapt this approximate solution into a more precise solution. CARMA has been used in Wyoming counties since 1996. The combination of a simple interface, flexible control strategy, and integration of multiple knowledge sources makes CARMA accessible to inexperienced users and capable of producing advice comparable to that produced by human experts. Moreover, because CARMA embodies diverse forms of expertise, it has been used in ways that its developers did not anticipate, including pest management research, development of industry strategies, and in-state and federal pest-management policy decisions.

  PDF (1466K, 14 pages)

• L. Karl Branting, John D. Hastings, and Jeffrey A. Lockwood, Integrating Cases and Models for Prediction in Biological Systems, AI Applications, 11(1):29-48 (1997).

  Abstract: Many complex biological systems are characterized both by incomplete models and limited empirical data. Accurate prediction of the behavior of such systems requires exploitation of multiple, individually incomplete, knowledge sources. Model-based adaptation is a technique for integrating case-based reasoning with model-based reasoning to predict the behavior of biological systems. This approach is implemented in CARMA, a system for rangeland grasshopper management advising that implements a process model derived from protocol analysis of human expert problem-solving episodes. CARMA's ability to predict the forage consumption judgments of expert pest managers was empirically compared to that of case-based and model-based reasoning techniques in isolation. This evaluation provided initial confirmation for the hypothesis that an integration of model-based and case-based reasoning can lead to more accurate predictions than either technique individually.

  Postscript (255K, 32 pages) PDF (1.0M, 32 pages)

• John Hastings, Karl Branting, and Jeff Lockwood, A Multi-Paradigm Reasoning System for Rangeland Management. Computers and Electronics in Agriculture, 16(1):47-67 (1996).

  Abstract: Polycultural agroecosystems, such as rangelands, are too complex and poorly understood to permit precise numerical simulation. Management decisions that depend on predictions of the behavior of such systems therefore require a variety of knowledge sources and reasoning techniques. Our approach to designing a computer system to provide advice concerning such systems is to incorporate a variety of reasoning paradigms, permitting the computer system to apply whatever reasoning paradigm is most appropriate to each task as it arises in the process of giving advice. This approach is based on a process description of expert human problem solving that uses four different reasoning paradigms: model-based reasoning; case-based reasoning; rule-based reasoning; and probabilistic reasoning. The process description is implemented in CARMA, a computer system for advising ranchers about the best response to rangeland grasshopper infestations. CARMA reflects an approach that attempts to emulate the human ability to integrate multiple knowledge sources and reasoning techniques in a flexible and opportunistic fashion. The goal of this approach is to enable computer systems to optimize the use of the diverse and incomplete knowledge sources and to produce patterns of reasoning that resemble those of human decision makers.

  Postscript (897K, 33 pages) PDF (461K, 33 pages)

• Alexandre Latchininsky, John Hastings, and Scott Schell, Good CARMA for the High Plains, Proceedings of the 2007 Americas' Conference on Information Systems (AMCIS 2007), Keystone, Colorado, August 9-12, 2007.

  Abstract: CARMA is a decision-support system for grasshopper infestations that has been successfully used since 1996. Rising treatment costs coupled with shrinking rangeland profit margins increasingly demand accurate selection of the most cost-effective responses to grasshopper infestations, and CARMA fills that need. In the process CARMA provides advice regarding grasshopper population management options in an environmentally and economically sound fashion, and is the only pest management software that includes the more environmentally-friendly Reduced Agent-Area Treatments (RAATs) as a treatment option and an open-ended capacity for user-based treatment updates. This paper describes the most recent changes to CARMA with particular attention to the new architecture which demonstrates an approach to integrating an artificially intelligent LISP reasoner with a Java graphical user interface (GUI) in a way which combines the strengths of the two languages (i.e., LISP for artificial intelligence and Java for graphical user interfaces) in order to provide a strong reasoner while at the same time producing an appealing user interface which is platform independent and web capable.

  PDF (1.4M, 9 pages)

• Jay H. Powell and John D. Hastings, An Empirical Evaluation of Automated Knowledge Discovery in a Complex Domain, Proceedings of the Workshop on Heuristic Search, Memory Based Heuristics and Their Applications and Proceedings of the Workshop on Learning for Search, Twenty-First National Conference on Artificial Intelligence (AAAI-06), Boston, MA, July 16-20, 2006.

  Abstract: Automatically acquiring knowledge in complex and possibly dynamic domains is an interesting, non-trivial problem. Case-based reasoning (CBR) systems are particularly well suited to the tasks of knowledge discovery and exploitation, and a rich set of methodologies and techniques exist to exploit the existing knowledge in a CBR system. However, the process of automatic knowledge discovery appears to be an area in which little research has been conducted within the CBR community. An approach to automatically acquiring knowledge in complex domains is automatic case elicitation (ACE), a learning technique whereby a CBR system automatically acquires knowledge in its domain through real-time exploration and interaction with its environment. The results of empirical testing in the domain of chess suggest that it is possible for a CBR system using ACE to successfully discover and exploit knowledge in an unsupervised manner. Results also indicate that the ability to explore is crucial for the success of an unsupervised CBR learner, and that exploration can lead to superior performance by discovering solutions to problems which would not otherwise be suggested or found by static or imperfect search mechanisms.

  PDF (310K, 6 pages)

• Siva N. Kommuri, Jay H. Powell, and John D. Hastings, On the Effectiveness of Automatic Case Elicitation in a More Complex Domain, Proceedings of the Workshop on Computer Gaming and Simulation Environments, Sixth International Conference on Case-based Reasoning (ICCBR-05), Chicago, Illinois, August 23-26, 2005.

  Abstract: Automatic case elicitation (ACE) is a learning technique in which a case-based reasoning system acquires knowledge automatically from scratch through repeated real-time trial and error interaction with its environment without dependence on pre-coded domain knowledge. ACE represents an alternative to manually constructed case bases and domain specific techniques, and is generally applicable to any domain for which knowledge can be obtained from a series of observations of an environment (e.g., checkers or massively multiplayer games). A priority is placed on maintaining the flexibility necessary to learn new domains with only negligible manual configuration. We found during testing that the current approach to ACE with a reliance on experience and exploration, while quite capable in the domain of checkers, did not perform adequately in the exponentially more complex domain of chess. Our results suggest that experience alone, without the ability to adapt for case differences between new and prior cases, is insufficient in more complex domains.

  PDF (88K, 8 pages)

• Jay H. Powell, Brandon M. Hauff, and John D. Hastings, Evaluating the Effectiveness of Exploration and Accumulated Experience in Automatic Case Elicitation, Proceedings of the Sixth International Conference on Case-based Reasoning (ICCBR-05), Chicago, Illinois, August 23-26, 2005.

  Abstract: Non-learning problem solvers have been applied to many interesting and complex domains. Experience-based learning techniques have been developed to augment the capabilities of certain non-learning problem solvers in order to improve overall performance. An alternative approach to enhancing pre-existing systems is automatic case elicitation, a learning technique in which a case-based reasoning system with no prior domain knowledge acquires knowledge automatically through real-time exploration and interaction with its environment. In empirical testing in the domain of checkers, results suggest not only that experience can substitute for the inclusion of pre-coded model-based knowledge, but also that the ability to explore is crucial to the performance of automatic case elicitation.

  PDF (459K, 11 pages)

• Jay H. Powell, Brandon M. Hauff, and John D. Hastings, Utilizing Case-Based Reasoning and Automatic Case Elicitation to Develop a Self-Taught Knowledgeable Agent, Proceedings of the Workshop on Challenges in Game AI, Nineteenth National Conference on Artificial Intelligence (AAAI-2004), San Jose, California, July 25-29, 2004.

  Abstract: Traditionally case-based reasoning (CBR) systems have relied on information manually provided by domain experts to form their knowledge bases. Additional domain knowledge is often used to improve performance of such systems. A less costly method of knowledge acquisition is automatic case elicitation, a learning technique in which a CBR system acquires knowledge automatically during real-time interaction with its environment with no prior domain knowledge (e.g., rules or cases). For problems that are observable, discrete and either deterministic or strategic in nature, automatic case elicitation can lead to the development of a self-taught knowledgeable agent. This paper describes the use of automatic case elicitation in CHEBR, a CHEckers case-Based Reasoner that employs self-taught knowledgeable agents. CHEBR was tested using model-based versus non-model-based matching to evaluate its ability to learn without predefined domain knowledge. The results suggest that additional experience can substitute for the inclusion of precoded model-based knowledge.

  PDF (162K, 5 pages)

• John Hastings, Karl Branting, Jeffrey Lockwood, and Scott Schell, CARMA+: A General Architecture for Pest Management, Proceedings of the Workshop on Environmental Decision Support Systems, Eighteenth International Joint Conference on Artificial Intelligence (IJCAI-2003), Acapulco, Mexico, August 9-15, 2003.

  Abstract: CARMA is a decision-support system for rangeland pest infestations that has been used successfully in Wyoming counties since 1996. CARMA is limited to the specific task for which it was designed: providing advice to ranchers concerning insect infestations on rangeland. This paper describes CARMA+, an architecture that permits CARMA's design to be applied to other pest-management tasks. A task analysis is described for a crop protection module for CARMA+ that is currently under development.

  PDF (363K, 4 pages)

• L. Karl Branting, John Hastings, and Jeffrey Lockwood, CARMA: A Case-Based Range Management Advisor, Proceedings of The Thirteenth Innovative Applications of Artificial Intelligence Conference (IAAI-2001), Seattle, Washington, August 7-9, 2001.

  Abstract: CARMA is an advisory system for rangeland grasshopper infestations that demonstrates how AI technology can deliver expert advice to compensate for cutbacks in public services. CARMA uses two knowledge sources for the key task of predicting forage consumption by grasshoppers: cases obtained by asking a group of experts to solve representative hypothetical problems; and a numerical model of rangeland ecosystems. These knowledge sources are integrated through the technique of model-based adaptation, in which CBR is used to find an approximate solution and the model is used to adapt this approximate solution into a more precise solution. CARMA has been used in Wyoming counties since 1996. The combination of a simple interface, flexible control strategy, and integration of multiple knowledge sources makes CARMA accessible to inexperienced users and capable of producing advice comparable to that produced by human experts. Moreover, because CARMA embodies diverse forms of expertise, it has been used in ways that its developers did not anticipate, including pest management research, development of industry strategies, and in state and federal pest management policy decisions.

  Postscript (1,563K, 8 pages) PDF (157K, 8 pages)

• John D. Hastings, L. Karl Branting, and Jeffrey A. Lockwood, Case Adaptation Using an Incomplete Causal Model Proceedings of the First International Conference on Case-Based Reasoning, Sesimbra, Portugal, October 23-26, 1995.

  Abstract: This paper describes a technique for integrating case-based reasoning with model-based reasoning to predict the behavior of biological systems characterized both by incomplete models and insufficient empirical data for accurate induction. This technique is implemented in CARMA, a system for rangeland pest management advising. CARMA's ability to predict the forage consumption judgments of 15 expert entomologists was empirically compared to that of CARMA's case-based and model-based components in isolation. This evaluation confirmed the hypothesis that integrating model-based and case-based reasoning through model-based adaptation can lead to more accurate predictions than the use of either technique individually.

  Postscript (230K, 12 pages) PDF (168K, 12 pages)

• John D. Hastings, L. Karl Branting, Global and Case-Specific Model-based Adaptation Proceedings of the AAAI 1995 Fall Symposium on Adaptation of Knowledge for Reuse Cambridge, Massachusetts, November 10-12, 1995.

  Abstract: CARMA (CAse-based Range Management Adviser) is a system that integrates case-based reasoning with model-based reasoning for rangeland pest management. CARMA's predictions of rangeland forage loss by grasshoppers were compared to predictions by 15 expert entomologists using either global or case-specific adaptation weights. Under both conditions, CARMA's predictions were more accurate than CARMA's case-based and model-based components in isolation. However, CARMA's case-specific adaptation weights were consistently more accurate than global adaptation weights. The experimental results suggest that case-specific adaptation weights are more appropriate in domains that are poorly approximated by a linear function.

  Postscript (261K, 7 pages) PDF (264K, 7 pages)

• L. Karl Branting and John D. Hastings, An Empirical Evaluation of Model-Based Case Matching and Adaptation, Proceedings of the Workshop on Case-Based Reasoning, Twelfth National Conference on Artificial Intelligence (AAAI-94), Seattle, Washington, July 31-August 4, 1994.

  Abstract: Rangeland ecosystems typify physical systems having an incomplete causal theory. This paper describes CARMA, a system for rangeland pest management advising that uses model-based matching and adaptation to integrate case-based reasoning with model-based reasoning for prediction in rangeland ecosystems. An ablation study showed that removing any part of the CARMA's model-based knowledge dramatically degraded CARMA's predictive accuracy. By contrast, any of several prototypical cases could be substituted for CARMA's full case library without significantly degrading performance. This indicates that the completeness of the model-based knowledge used for matching and adaptation is more important to CARMA's performance than the coverage of the case library.

  Postscript (175K, 7 pages) PDF (142K, 7 pages)

• John Douglas Hastings, A Mixed Paradigm Reasoning Approach to Problem-Solving in Incomplete Causal-Theory Domains, Ph.D. Dissertation, University of Wyoming, Department of Computer Science, December 1996.

  Abstract: Many complex physical systems such as biological, ecological, and other natural systems are characterized both by incomplete models and limited empirical data. Accurate prediction of the behavior of such systems requires exploitation of multiple, individually incomplete, knowledge sources. This dissertation describes model-based adaptation, a technique for integrating case-based reasoning with model-based reasoning to predict the behavior of biological systems characterized both by incomplete causal models and insufficient emprical data for accurate induction. This approach is implemented in CARMA, a system for rangeland grasshopper management advising. CARMA implements a process model derived from protocol analysis of human expert problem- solving episodes. CARMA's design attempts to emulate the speed, graceful degradation, opportunism, and explanatory ability of human experts. CARMA's ability to predict the forage consumption judgements of expert entomologists was empirically compared to that of case-based and model-based reasoning techniques in isolation. This evaluation confirmed the hypothesis that integrating model-based integrating model-based and case-based reasoning can lead to more accurate predictions than the use of either technique individually.

  Postscript (Chapters 1 & 2, 615K, 24 pages)
  Postscript (Chapter 3 - Part 1, 3091K, 16 pages)
  Postscript (Chapter 3 - Part 2, 3314K, 11 pages)
  Postscript (Chapters 4 & 5, 1459K, 35 pages)
  Postscript (Chapters 6 & 7, Appendices, 1795K, 25 pages)

• John Douglas Hastings, Design and Implementation of a Speech Recognition Database Query System, M.S. Thesis, University of Wyoming, Department of Computer Science, August 1991.

  Abstract: This thesis introduces CONQUEST, a constrained natural language speech recognition database query system. The objective was to improve on previous natural language database query systems by designing and implementing a more user-friendly query system through the integration of speech and nondeterministic syntactic processing. This paper will discuss the areas in which improvements were attempted, the components required along with a discussion of each, an illustration of system operation, and an evaluation of the final product.

  Postscript (Thesis, 806K, 48 pages)
  Postscript (Appendices, 551K, 68 pages)