Pub Date : 1994-03-01DOI: 10.1109/CAIA.1994.323642
G. Provan
The author describes a new approximation method which can significantly improve the computational efficiency of Bayesian networks. He applies this technique to the diagnosis of acute abdominal pain, with good results. This approach is based on using a reduced set of the model parameters for diagnostic reasoning. The tradeoffs in diagnostic accuracy required to obtain increased computational efficiency (due to the smaller models) are carefully specified using a variety of statistical metrics.<>
{"title":"Probabilistic diagnostic reasoning: towards improving diagnostic efficiency","authors":"G. Provan","doi":"10.1109/CAIA.1994.323642","DOIUrl":"https://doi.org/10.1109/CAIA.1994.323642","url":null,"abstract":"The author describes a new approximation method which can significantly improve the computational efficiency of Bayesian networks. He applies this technique to the diagnosis of acute abdominal pain, with good results. This approach is based on using a reduced set of the model parameters for diagnostic reasoning. The tradeoffs in diagnostic accuracy required to obtain increased computational efficiency (due to the smaller models) are carefully specified using a variety of statistical metrics.<<ETX>>","PeriodicalId":297396,"journal":{"name":"Proceedings of the Tenth Conference on Artificial Intelligence for Applications","volume":"4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131011937","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1994-03-01DOI: 10.1109/CAIA.1994.323667
S. Haller
The Interactive Discourse Planner (IDP) is designed to describe and justify domain plans interactively. The text plan that IDP incrementally formulates and executes is represented uniformly in the same knowledge base with the domain plans that are under discussion. In this way, the text plan and the domain plans are both accessible for analyzing the user's feedback. IDP can interpret vaguely articulated feedback, generate concise replies and metacomments, and detect when the user's feedback initiates a digression. As a testbed for my model, I am implementing IDP to interactively give driving directions and route advice.<>
{"title":"A model for cooperative interactive plan explanation","authors":"S. Haller","doi":"10.1109/CAIA.1994.323667","DOIUrl":"https://doi.org/10.1109/CAIA.1994.323667","url":null,"abstract":"The Interactive Discourse Planner (IDP) is designed to describe and justify domain plans interactively. The text plan that IDP incrementally formulates and executes is represented uniformly in the same knowledge base with the domain plans that are under discussion. In this way, the text plan and the domain plans are both accessible for analyzing the user's feedback. IDP can interpret vaguely articulated feedback, generate concise replies and metacomments, and detect when the user's feedback initiates a digression. As a testbed for my model, I am implementing IDP to interactively give driving directions and route advice.<<ETX>>","PeriodicalId":297396,"journal":{"name":"Proceedings of the Tenth Conference on Artificial Intelligence for Applications","volume":"19 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121577974","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1994-03-01DOI: 10.1109/CAIA.1994.323644
J. Barnett, T. Verma
An ideal automated reliability analysis system would take a CAD model as input, then identify critical components and output failure probabilities. Unfortunately, the practical systems in use today require a fault tree/spl minus/a kind of and/or graph/spl minus/as their input. Since the cognitive distance between the CAD model and its associated fault tree is large, the manual translation step is a source of many mistakes. Furthermore, the use of fault trees makes it difficult to handle the cyclical dependencies that naturally occur among system components. The functional dependency graph, a representation midway between CAD models and fault trees, is described and methodology that automates reliability and diagnostic analyses is developed. When the resultant technology is employed, the knowledge burden is more equally shared between the user and the computer. In addition, more complicated systems can be properly analyzed because cyclical dependencies can be represented.<>
{"title":"Intelligent reliability analysis","authors":"J. Barnett, T. Verma","doi":"10.1109/CAIA.1994.323644","DOIUrl":"https://doi.org/10.1109/CAIA.1994.323644","url":null,"abstract":"An ideal automated reliability analysis system would take a CAD model as input, then identify critical components and output failure probabilities. Unfortunately, the practical systems in use today require a fault tree/spl minus/a kind of and/or graph/spl minus/as their input. Since the cognitive distance between the CAD model and its associated fault tree is large, the manual translation step is a source of many mistakes. Furthermore, the use of fault trees makes it difficult to handle the cyclical dependencies that naturally occur among system components. The functional dependency graph, a representation midway between CAD models and fault trees, is described and methodology that automates reliability and diagnostic analyses is developed. When the resultant technology is employed, the knowledge burden is more equally shared between the user and the computer. In addition, more complicated systems can be properly analyzed because cyclical dependencies can be represented.<<ETX>>","PeriodicalId":297396,"journal":{"name":"Proceedings of the Tenth Conference on Artificial Intelligence for Applications","volume":"88 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122957363","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1994-03-01DOI: 10.1109/CAIA.1994.323623
E. H. Turner, T. L. Briggs
Agents must be able to react to the unpredictable elements of their world. At the same time, they can take advantage of reliable predictions that can be made about the world. In this paper, we present a method for taking advantage of such predictions to respond to unanticipated goals that become active after the agent has begun performing its task. The application we discuss is sequencing locations that an autonomous underwater vehicle (AUV) must visit to achieve its goals. Our method is embodied in NBA-PLANNER.<>
{"title":"Responding to unanticipated goals when planning travel for autonomous underwater vehicles","authors":"E. H. Turner, T. L. Briggs","doi":"10.1109/CAIA.1994.323623","DOIUrl":"https://doi.org/10.1109/CAIA.1994.323623","url":null,"abstract":"Agents must be able to react to the unpredictable elements of their world. At the same time, they can take advantage of reliable predictions that can be made about the world. In this paper, we present a method for taking advantage of such predictions to respond to unanticipated goals that become active after the agent has begun performing its task. The application we discuss is sequencing locations that an autonomous underwater vehicle (AUV) must visit to achieve its goals. Our method is embodied in NBA-PLANNER.<<ETX>>","PeriodicalId":297396,"journal":{"name":"Proceedings of the Tenth Conference on Artificial Intelligence for Applications","volume":"158 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132794908","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1994-03-01DOI: 10.1109/CAIA.1994.323684
M. Ishii, Y. Sasaki, S. Kaneda
This paper presents an interactive constraint satisfaction mechanism for building office systems. We noticed that regulations for business are primarily described in declarative constraints, and clerical work could be regarded as consistency maintenance governed by conformance to the regulations in terms of the data in business databases and application forms. From this view point, we model clerical work as a consistency maintenance model. On the basis of the model, we develop a constraint syntax and an interactive constraint satisfaction system appropriate to clerical work. In order to demonstrate the validity and effectiveness of our method, we rewrote the general affairs expert system KOA, an office system currently in operation, using constraints, and evaluated the resulting performance. Experimental results show that all the regulations can be easily described in constraints, the description size is reduced by 50%. The constraints have high modularity and cover rare cases exhaustively, the response time is short enough to be feasible, and the number of interactions is almost equivalent to that of the original system. The results confirm the important point that we can easily realize a maintainable office system with the interactive constraint satisfaction mechanism without increasing the number of interactions.<>
{"title":"Interactive constraint satisfaction for office systems","authors":"M. Ishii, Y. Sasaki, S. Kaneda","doi":"10.1109/CAIA.1994.323684","DOIUrl":"https://doi.org/10.1109/CAIA.1994.323684","url":null,"abstract":"This paper presents an interactive constraint satisfaction mechanism for building office systems. We noticed that regulations for business are primarily described in declarative constraints, and clerical work could be regarded as consistency maintenance governed by conformance to the regulations in terms of the data in business databases and application forms. From this view point, we model clerical work as a consistency maintenance model. On the basis of the model, we develop a constraint syntax and an interactive constraint satisfaction system appropriate to clerical work. In order to demonstrate the validity and effectiveness of our method, we rewrote the general affairs expert system KOA, an office system currently in operation, using constraints, and evaluated the resulting performance. Experimental results show that all the regulations can be easily described in constraints, the description size is reduced by 50%. The constraints have high modularity and cover rare cases exhaustively, the response time is short enough to be feasible, and the number of interactions is almost equivalent to that of the original system. The results confirm the important point that we can easily realize a maintainable office system with the interactive constraint satisfaction mechanism without increasing the number of interactions.<<ETX>>","PeriodicalId":297396,"journal":{"name":"Proceedings of the Tenth Conference on Artificial Intelligence for Applications","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132189693","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1994-03-01DOI: 10.1109/CAIA.1994.323668
A. Harjoko, B.J. Kurz
New results related to a novel approach in computer vision are reported in this paper. Rather than designing ever more complex algorithms to extract given objects from images, the design of simple targets for given fast and robust algorithms is promoted. The mathematical nature of this approach leads to optimized target structures. A unifying model for vision algorithms is introduced which combines target primitive extraction and target structure matching. The usefulness of this approach is demonstrated by a practical example of a vision system. The system employs a line cluster as the target, with the Hough transform as the extraction algorithm. In order to get the best results, the performance of the vision system is evaluated. The evaluation is attempted on two bases, by a mathematical model and by experimental results. The mathematical model is capable of predicting the effects of various target parameters as well as the effects of various distortions of the vision system. Target design can lead to simpler, faster and more robust computer vision systems for use in industry.<>
{"title":"Target design: a method for an accurate pose determination","authors":"A. Harjoko, B.J. Kurz","doi":"10.1109/CAIA.1994.323668","DOIUrl":"https://doi.org/10.1109/CAIA.1994.323668","url":null,"abstract":"New results related to a novel approach in computer vision are reported in this paper. Rather than designing ever more complex algorithms to extract given objects from images, the design of simple targets for given fast and robust algorithms is promoted. The mathematical nature of this approach leads to optimized target structures. A unifying model for vision algorithms is introduced which combines target primitive extraction and target structure matching. The usefulness of this approach is demonstrated by a practical example of a vision system. The system employs a line cluster as the target, with the Hough transform as the extraction algorithm. In order to get the best results, the performance of the vision system is evaluated. The evaluation is attempted on two bases, by a mathematical model and by experimental results. The mathematical model is capable of predicting the effects of various target parameters as well as the effects of various distortions of the vision system. Target design can lead to simpler, faster and more robust computer vision systems for use in industry.<<ETX>>","PeriodicalId":297396,"journal":{"name":"Proceedings of the Tenth Conference on Artificial Intelligence for Applications","volume":"33 1-2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134155992","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1994-03-01DOI: 10.1109/CAIA.1994.323690
I.F.C. Smith, B.V. Faltings
Reuse of designs is a traditional strategy that has become a new technique for knowledge-based design systems. This paper shows how the approach can support spatial adaptation of previous designs and improve integration of design decisions emanating from different views. Run-time parametrization of cases according to characteristics of new contexts helps support adaptation. Cases of previous designs of complex artifacts implicitly provide solutions for integrating diverse design criteria. We describe a prototype design system, CADRE, which had successfully applied the case-based paradigm to several examples of building design. The unique aspects of CADRE are not dependent upon the building design domain and therefore, our system has potential for supporting it range of spatial design tasks.<>
{"title":"Spatial design of complex artifacts using cases","authors":"I.F.C. Smith, B.V. Faltings","doi":"10.1109/CAIA.1994.323690","DOIUrl":"https://doi.org/10.1109/CAIA.1994.323690","url":null,"abstract":"Reuse of designs is a traditional strategy that has become a new technique for knowledge-based design systems. This paper shows how the approach can support spatial adaptation of previous designs and improve integration of design decisions emanating from different views. Run-time parametrization of cases according to characteristics of new contexts helps support adaptation. Cases of previous designs of complex artifacts implicitly provide solutions for integrating diverse design criteria. We describe a prototype design system, CADRE, which had successfully applied the case-based paradigm to several examples of building design. The unique aspects of CADRE are not dependent upon the building design domain and therefore, our system has potential for supporting it range of spatial design tasks.<<ETX>>","PeriodicalId":297396,"journal":{"name":"Proceedings of the Tenth Conference on Artificial Intelligence for Applications","volume":"44 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116586138","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1994-03-01DOI: 10.1109/CAIA.1994.323695
K. Miyashita, K. Sycara
We have developed an integrated framework of iterative revision and knowledge acquisition for schedule optimization, and implemented it in the CABINS system. The ill-structuredness of both the solution space and the desired objectives make scheduling problems difficult to formalize and costly to solve. In CABINS, situation-dependent user's preferences that guide schedule revision are captured in cases together with contextual information. During iterative repair, cases are exploited for multiple purposes, such as (1) repair action selection, (2) repair result evaluation and (3) recovery from revision failures. The contributions of the work lie in experimentally demonstrating in a domain where neither the human expert nor the program possess causal knowledge that search control knowledge can be acquired through past repair cases to improve the efficiency of rather intractable iterative repair process. The experiments in this paper were performed in the context of job-shop scheduling problems.<>
{"title":"Learning control knowledge through cases in schedule optimization problems","authors":"K. Miyashita, K. Sycara","doi":"10.1109/CAIA.1994.323695","DOIUrl":"https://doi.org/10.1109/CAIA.1994.323695","url":null,"abstract":"We have developed an integrated framework of iterative revision and knowledge acquisition for schedule optimization, and implemented it in the CABINS system. The ill-structuredness of both the solution space and the desired objectives make scheduling problems difficult to formalize and costly to solve. In CABINS, situation-dependent user's preferences that guide schedule revision are captured in cases together with contextual information. During iterative repair, cases are exploited for multiple purposes, such as (1) repair action selection, (2) repair result evaluation and (3) recovery from revision failures. The contributions of the work lie in experimentally demonstrating in a domain where neither the human expert nor the program possess causal knowledge that search control knowledge can be acquired through past repair cases to improve the efficiency of rather intractable iterative repair process. The experiments in this paper were performed in the context of job-shop scheduling problems.<<ETX>>","PeriodicalId":297396,"journal":{"name":"Proceedings of the Tenth Conference on Artificial Intelligence for Applications","volume":"94 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116417007","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1994-03-01DOI: 10.1109/CAIA.1994.323657
Chih-Hung Wu, Shie-Jue Lee, Hung-Sen Chou
Keeping knowledge consistent is an important topic in the life cycle of developing expert systems. In this paper, we focus on some major problems in knowledge validation: redundancy, subsumption, cycles, conflict, and unnecessary conditions, and describe how these problems are solved in rule-based expert systems using dependency analysis. A rule-dependency graph is developed to describe the dependency relationship among the rules contained in a knowledge base. Since each type of inconsistent knowledge presents a specific topology in the rule-dependency graph, knowledge validation can be done by examining the structure of the graph. With the aid of the rule-dependency graph, we have developed a token-flow paradigm that identifies the inconsistent structure in the rule base. The idea is effective and can be easily implemented. Properties of our method are explored. Some practical examples are also presented.<>
{"title":"Dependency analysis for knowledge validation in rule-based expert systems","authors":"Chih-Hung Wu, Shie-Jue Lee, Hung-Sen Chou","doi":"10.1109/CAIA.1994.323657","DOIUrl":"https://doi.org/10.1109/CAIA.1994.323657","url":null,"abstract":"Keeping knowledge consistent is an important topic in the life cycle of developing expert systems. In this paper, we focus on some major problems in knowledge validation: redundancy, subsumption, cycles, conflict, and unnecessary conditions, and describe how these problems are solved in rule-based expert systems using dependency analysis. A rule-dependency graph is developed to describe the dependency relationship among the rules contained in a knowledge base. Since each type of inconsistent knowledge presents a specific topology in the rule-dependency graph, knowledge validation can be done by examining the structure of the graph. With the aid of the rule-dependency graph, we have developed a token-flow paradigm that identifies the inconsistent structure in the rule base. The idea is effective and can be easily implemented. Properties of our method are explored. Some practical examples are also presented.<<ETX>>","PeriodicalId":297396,"journal":{"name":"Proceedings of the Tenth Conference on Artificial Intelligence for Applications","volume":"34 16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116567869","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1994-03-01DOI: 10.1109/CAIA.1994.323625
V. Prabhakar, S. Sheppard
Finite element analysis (FEA) pre-processing is considered to be an art-form requiring many years of experience before the necessary expertise can be obtained by human analysts to efficiently and reliable develop a numerical model. In the "real" world, design models rarely match the simplistic approach adopted in an academic environment to FEA. Complex 3D parts with intricate boundary and loading conditions are commonly dealt with. Model idealization is an important pre-processing phase of FEA, intended to create a simplified analysis geometry while accurately representing the design form, function and intent. A rule-based expert system named DESIDE-X has been developed which integrates object-oriented programming, a CSG feature-based design representation system and an idealization rule-system that uses heuristic and procedural methods to remove or simplify geometric features that are insignificant from the FEA perspective.<>
{"title":"A knowledge-based approach to model idealization in FEM","authors":"V. Prabhakar, S. Sheppard","doi":"10.1109/CAIA.1994.323625","DOIUrl":"https://doi.org/10.1109/CAIA.1994.323625","url":null,"abstract":"Finite element analysis (FEA) pre-processing is considered to be an art-form requiring many years of experience before the necessary expertise can be obtained by human analysts to efficiently and reliable develop a numerical model. In the \"real\" world, design models rarely match the simplistic approach adopted in an academic environment to FEA. Complex 3D parts with intricate boundary and loading conditions are commonly dealt with. Model idealization is an important pre-processing phase of FEA, intended to create a simplified analysis geometry while accurately representing the design form, function and intent. A rule-based expert system named DESIDE-X has been developed which integrates object-oriented programming, a CSG feature-based design representation system and an idealization rule-system that uses heuristic and procedural methods to remove or simplify geometric features that are insignificant from the FEA perspective.<<ETX>>","PeriodicalId":297396,"journal":{"name":"Proceedings of the Tenth Conference on Artificial Intelligence for Applications","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128781616","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：