Designing and implementing a contradiction solver in an intelligent assistant decision system based on diversity of opinions

Number of pages: 105 File Format: word File Code: 31082
Year: 2011 University Degree: Master's degree Category: Computer Engineering
  • Part of the Content
  • Contents & Resources
  • Summary of Designing and implementing a contradiction solver in an intelligent assistant decision system based on diversity of opinions

    Master's thesis in the field of computer engineering (artificial intelligence)

    Abstract

    Designing and implementing a contradiction solver in an intelligent assistant decision system based on diversity of opinions

    Resolving inconsistency is an important procedure in many intelligent systems, including law-based systems. This procedure determines the order of implementation of the rules in situations where there is more than one rule to be implemented. There are different methods to fix the incompatibility. In this research, in order to resolve the inconsistency in a law-based system, it has been used to follow different points of view in separate inference paths. This system is an intelligent assistant decision system that, when inconsistencies occur, by considering separate lines of inference for each of the incompatible rules, it provides the possibility for the decision maker to know all the possible choices.

    In general, intelligent assistant decision systems have improved these systems by introducing various artificial intelligence techniques into the structure of decision assistant systems, in order to further support and improve decision making.

    An intelligent assistant decision system that In this research, it has been implemented in order to help a player in a real-time strategic game. Creating computer programs related to real-time strategic games is considered a new field in the field of artificial intelligence. These programs are not limited to creating intelligent opponents in order to entertain human players, and nowadays high-performance simulators for training military personnel are among the demands, and artificial intelligence research in the field of this type of games, in addition to the commercial producers of games, also enjoys the broad support of some defense institutions.

    Chapter One

    Introduction

    There are many problems that are not limited to a unique solution. In addition, some problems may have an unlimited number of similar answer paths. An inconsistency [1] occurs when various decisions are available, corresponding to distinct response paths.

    Generally, in a system that has a relatively large set of rules and facts, inserting a fact can lead to the correct value of several rules and, as a result, their activation. Any order of implementation of these rules can lead to different results, in which case this set of rules is called a set of incompatible rules. An inconsistency resolution strategy determines an order for the implementation of this set of rules. Intelligent systems, such as rule-based systems, planning tools, and knowledge-based structures, use different strategies to resolve inconsistency [2]. The proposed system is an intelligent decision support system designed and implemented to help a player in a real-time strategic game, and its structure and features will be described in the next chapters. Also, in this thesis, there are some materials related to intelligent assistant decision systems and various structures that researchers have considered for the implementation of these systems. One of the most important tasks performed by the inference engine is to resolve inconsistencies [47]. In general, inconsistency resolution is a strategy for choosing the order of rule execution when more than one rule can be executed.

    There are various methods for inconsistency resolution. The simplest solution is to randomly select rules. In some strategies, among the most important factors that are effective in the selection of rules is the priority value assigned to each rule by the system builder, in this method, another method should be used for rules with the same priority.More sophisticated methods use statistical information related to previous successes and failures while applying different rules in order to determine the probability of success. Also, some methods calculate the costs of the rules, which represent the efforts that the problem solver needs to perform the actions (such as time)[2].

    The method used in this research to resolve the inconsistency, by considering a separate line of inference for each of the inconsistent rules during the inference process, covers all possible situations for prioritization in the implementation of the rules.

     

    1-2- Assistant decision systems and intelligent assistant decision systems

    In general, decision-making is one of the most important and sensitive activities that takes place in any organization [48]. To support and support this complex process, a diverse group of independent information systems called decision support systems have emerged over the past two decades. These systems are computer-based tools that are created to support the complex process of decision-making and problem-solving, defined and designed to create an environment for problem analysis, building models and procedural simulation of decision-making and decision-making programs [49].

    These information systems, which are designed to interactively support all stages of a user's decision-making process, can include technologies derived from various scientific fields including accounting, cognitive science, computer science, economics, Engineering, management, statistics, etc. are and often consist of three components: data subsystem, model subsystem (which has a mechanism for data processing) and user communication subsystem] [19] Although, by using information sources and various analysis tools, decision support systems provide better and higher quality conditions for decision makers and having a supporting role instead of completely replacing people in the decision making process is one of their main goals [12], but they cannot be considered as Considered an intelligent assistant for decision makers. Intelligent decision support systems are useful and economically viable for general problems that require frequent decisions. These computer-based dialogue systems use a combination of data and specialized knowledge and models used to support decision makers in organizations with artificial intelligence techniques to solve semi-structured problems [50].

    There are different definitions of the differences between a decision support system and an intelligent decision support system, which is due to the existence of different types of intelligent decision support systems. In these systems, intelligent performance in decision-making has been made possible by improvements such as upgrading the model database management system or strengthening the user interface using various artificial intelligence techniques such as natural language processing or other similar techniques. Also, by supporting issues with uncertainty, this type of system provides the possibility of supporting a wider range of decisions and can control and manage territories in which the decision-making process is more complex and requires, in addition to skill and expertise, the evaluation of the effect of the proposed solution. Among the other advantages of intelligent decision support systems compared to decision support systems is the improvement of consistency in decisions, improvement of explanation and interpretation and justification of the suggestions provided by the system [19].

    Holsapple and Whinston were among the first researchers who designed and studied intelligent decision support systems [51]. They proposed the following characteristics for these systems:

    These systems include different types of knowledge that describe selected representations of the decision-making world.

    These systems have the ability to acquire and maintain descriptive knowledge[2] such as event maintenance[3] and other types of knowledge.

    These systems can produce and present knowledge in different ways.

    They can provide knowledge to provide or choose to acquire new knowledge.

    These systems can communicate directly (intelligently) with the decision-maker.

    Although these systems are intelligent and human-like supporters in the decision-making process, the decision-makers must make the final and critical decisions themselves.

  • Contents & References of Designing and implementing a contradiction solver in an intelligent assistant decision system based on diversity of opinions

    List:

    The first chapter. Introduction. 1

    1-1- Introduction 2

    1-2- Fixing the inconsistency. 3

    1-3- decision support systems and intelligent decision support systems. 4

    1-4- The purpose of this thesis. 6

    1-5- An overview of the thesis chapters. 6

    The second chapter. Methods of resolving incompatibility. 7

    2-1- Introduction. 8

    2-2- Some simple strategies to fix incompatibility. 9

    2-3- Fixing the inconsistency using a useful value. 13

    2-4- Fixing the inconsistency using random estimated costs. 15

    2-4-1- Estimation of mathematical hope of cost 17

    2-4-2- Recursive estimation 18

    2-4-3- Fixing inconsistencies 19

    2-5- Fixing inconsistencies using linear programming. 21

    2-6- Resolving inconsistencies using game theory. 22

    2-7- Fixing the inconsistency using the graph model. 23

    2-8- Fixing the inconsistency using the analytical hierarchical process and improving it. 25

    The third chapter. Intelligent decision support systems. 31

    3-1- Introduction. 32

    3-2- Features of intelligent assistant decision systems. 33

    3-3- Introducing several intelligent assistant decision systems with different structures. 36

    3-3-1- Using evolutionary algorithms in the IDSS structure 36

    3-3-2- Using intelligent agent in the IDSS structure 38

    3-3-3- Using data mining methods and artificial neural networks in the IDSS structure 40

    3-3-4- Using a decision-making method based on fuzzy logic in the IDSS structure 46

    3-3-5- The use of case-based inference in the structure of IDSS 51

    3-3-6- The use of law-based components in the structure of IDSS 55

    Chapter four. Real-time strategic computer games and intelligent systems related to them 57

    4-1- Introduction. 58

    4-2- Characteristics of real-time strategic games. 59

    4-3- An overview of intelligent systems related to real-time strategic games. 63

    The fifth chapter. The proposed system. 71

    5-1- Introduction. 72

    5-2- Introducing the proposed system. 73

    5-3- The main components of the proposed system. 74

    5-4- The method of solving the inconsistency used in the proposed system. 77

    Sixth chapter. Evaluation and results. 80

    Chapter Seven. Conclusion and future work. 89

    List of sources. 92

             

     

    Source:

    [1]        Giarratano, J. C. (2003). “CLIPS 6.21 User's Guide”, International Thompson Publishing.

    [2] Belavkin, R. V. (2003). "Conflict resolution by random estimated costs", Proceedings of the 17th European Simulation Multiconference (c) SCS Europe, BVBA.

    [3] Jones, G., Ritter, F. E., Wood, D. J. (2000). "Using a cognitive architecture to examine what develops", Psychological Science, 11(2), 93–100.

    [4] Belavkin, R. V., Ritter, F. E. (2003). "The use of entropy for analysis and control of cognitive models", Proceedings of the Fifth International Conference on Cognitive Modeling (pp. 21-26). Bamberg, Germany: Universit¨ats–Verlag Bamberg.

    [5]        Belavkin, R. V. (2002). "On emotion, learning and uncertainty: A cognitive modeling approach", PhD Thesis, The University of Nottingham, United Kingdom.

    [6] Lo C., Wu C. (2005). "The Survey and Expansion on Quantitative Methods to Conflict Resolution", Proceedings of the Third International Conference on Information Technology and Applications (ICITA'05), 268-271.

    [7] Fang L., Hipel K. W., Kilgour D. M. (1993). Interactive Decision Making: The Graph Model for Conflict Resolution. New York: Wiley.

    [8] Obeidi A., Hipel K.W., Kilgour D.M. (2005). "The Role of Emotions in Envisioning Outcomes in Conflict Analysis", Journal of Group Decision and Negotiation, Vol. 14, No. 6: 481-500.

    [9]        Kilgour D. M., Hipel K. W. (2005). "The Graph Model for Conflict Resolution: Past, present, and future", Group Decision and Negotiation, Vol. 14, No. 6: 441-460.

    [10] Obeidi A., Hipel K. W., Kilgour D. M. (2005). "Perception and Emotion in the Graph Model for Conflict"Perception and Emotion in the Graph Model for Conflict Resolution", International Conference on Systems, Man and Cybermetics, ISIC. [11] Saaty T. L. (2001). The Analytic Network Process: Decision Making with Dependence and Feedback. 2nd ed., Pittsburgh: RWS Publications.

    [12] Gupta J., et al. (2006). Intelligent Decision-making Support Systems, Foundations, Applications and Challenges. Springer, 1st Edition, XXIV, 503p.105 illus.

     [13]     Li Y, Hipel K.W. (2007). "The Integration of a Multiple Criteria Preference Ranking Approach for Conflict Resolution", International Conference on Systems, Man and Cybernetics, ISIC. [14] Phillips-Wren G., Forgionne G. (2002). "Evaluating web-based and real-time decision support systems", Proceedings of the DSIAge Conference, 166-175. [15] Han J. (1998). "Data Mining: An Overview from Databases Perspective", Tutorial on the Pacific-Asia Conference in Knowledge Discovery and Data Mining (PKDD-98).

    [16] Forgionne G., Mora M., Cervantes F., Gelman O. (2002). "I-DMSS: A conceptual architecture for the next generation of decision-making support systems in the internet age", Proceedings of the DSIAge Conference, 154-165.

    [17] Michalewicz Z., et al. (2005). "Case study: An intelligent decision-support system", IEEE Intelligent Systems: putting AI into practice, 20 (4):44-49.

    [18] Edwards JS., Duan Y., Robins PC. (2000). "An analysis of expert systems for business decision-making at different levels and in different roles", European Journal of Information Systems, 9:36-46. [19] Foster D., et al. (2005). "A Survey of Agent-Based Intelligent Decision Support Systems to Support Clinical Management and Research", In: 1st Intl. Workshop on Multi-Agent Systems for Medicine, Computational Biology, and Bioinformatics. [20] Turban E., Aronson J., Liang T. (2005). Decision Support Systems and Intelligent Systems. 7th Edn., New Jersey: Pearson Prentice-Hall.

    [21] Hayes-Roth F., Jacobstein N. (1994). "The state of knowledge-based systems", Communications of the ACM, 37 (3); 27-39.

    [22] Tecuci G., Kodratoff Y. (1995). Machine Learning and Knowledge Acquisition: Integrated Approaches. London UK: Academic Press.

    [23] Azvine B., Azarmi N., Nauck D. (2000). Intelligent Systems and Soft Computing: Prospects, Tools and Applications. (Eds. Carbonell, J. and Siekmann, J.), Lecture Notes in Artificial Intelligence. Adastral Park, UK: Springer-Verlag.

    [24] Sun R. (2001). "Artificial intelligence: connectionist and symbolic approaches", In (Eds. Smelser, N.J. and Baltes, P.B.), International Encyclopedia of the Social and Behavioral Sciences, 783-789.

    [25] Viademonte S. (2004). "A Hybrid Model for Intelligent Decision Support: Combining Data Mining and Artificial Neural Networks", PhD. Thesis, Faculty of Information Technology, Monash University, Australia.

    [26] Agrawal R., Imielinsk T., Swami A. (1993). "Mining association rules between sets of items in large databases", ACM SIGMOD Conference on Management of Data (SIGMOD'93), New York, NY: ACM Press: 207-216.

    [27] Machado RJ., Rocha AF. (1990). "The combinatorial neural network: a connectionist model for knowledge based systems", Uncertainty in knowledge bases. Berlin, Germany: Springer Verlag.

    [28] Leao BF., Rocha A. (1990). "Proposed methodology for knowledge acquisition: a study on congenital heart disease diagnosis", Methods of Information in Medicine, 29: 30-40.

    [29] El-Gayar OF., Leung PS. (2001). "A multiple criteria decision-making framework for regional aquaculture development", European Journal of Operational Research, 133: 462-482 [30] Kirby N. (2010). Introduction to Game AI. Course Technology PTR.

    [31]       Proctor M., et al. (2008). “Drools Documentation”, JBoss Community.

    [32] Buro M., Furtak T.M.

Designing and implementing a contradiction solver in an intelligent assistant decision system based on diversity of opinions