Therefore, even though there is no explicit reference to the “IsA” context nodes from Zone MFrag in the individuals displayed in Figure 31, the object property hasOVariable [3] linking the Zone_MFrag with its respective ordinary variables implicitly conveys that subtyping restriction. As an example of the mapping between MEBN and PROWL depicted in Table 5., representing the context node “z=StarshipZone(st)” requires decomposing it into random variable terms “equal(z, StarshipZone(st))” and “StarshipZone(st)”. In the PROWL ontology these RV terms are respectively represented by Z_ZSZoneST_context and Z_ZSZoneST_inner_SZoneST, both individuals of class Context. The other properties depicted in Figure 31 are the object property hasSkolem [4], which links a quantifier MFrag with its respective Skolem constants, and the properties hasResidentNode [5], hasInputNode [6], and hasContextNode [7], all subproperties of hasNode. Figure 32 portrays the representation of node ZoneMD(z, t), from the Zone MFrag. Object property isNodeFrom [1] provides the link between the node and its MFrag. Further structural information is provided by the parent list formed with the object property hasParent [2], the object property isResidentNodeIn [6] and the properties that link ZoneMD(z, t) with its “copies” (instances in which the node is used as input and/or context node in other MFrags), which are hasContextInstanceIn [7] and hasInputInstanceIn (not visible in the picture). ZoneMD(z, t) is a resident node, so it has a probability distribution conditioned on its parents. The link between an individual of class Domain_res and the many possible representations of its probability distribution is provided by the object property hasProbDist [3]. Subsection 5.2.2 explains the different possibilities of representing probability distributions in PROWL. The list of possible states of ZoneMD(z, t) is made using the object property hasPossibleValues [4], while its arguments (ordinary variables z and t) are linked using the hasArgument object property [5]. Note that property hasArgument doesn’t actually point to an individual of class OVariable, which is the class that has the ordinary variables z and t represented (as Z_z and Z_t respectively). Instead, it points to individuals of class SimpleArgRelationship, a subclass of ArgRelationship. These two classes are reified relations specifying the many possibilities of arguments in a random variable. In this specific case (i.e. the ZoneMD(z, t) node), the two arguments are OVariables, so links to both are represented by individuals of the class SimpleArgRelationship, which works as a pointer to individuals of Class OVariable only. When a node has composite arguments, the parent class ArgRelationship should be used, since it works as a pointer to individuals of classes OVariable, Node, Entity, and Skolem. ZoneMD Resident Node Object properties isArgTermIn [8] and hasInnerTerm [9] provide further support to reified relations, by keeping track of the complex relationships in which each node is participating. The use of reification is also important for representing probability distributions in PROWL, which may be conveyed in different ways. 1.21.2Representing a Probability Distribution Representing probability distributions is a key issue in achieving a balance between interoperability and conciseness. Proprietary formats usually convey all the necessary information in a compact way, thus simply using that format in a xsd:string to convey that information is an attractive option. However, this option ties the ontology to a specific format that might not be universally known or might be inappropriate to a range of applications. Also, annotating probability distributions might reduce the ability to use that data in complex environments with many systems working with different formats, rules or requirements. PROWL is supposed to facilitate interoperability and thus should be as flexible as possible in terms of how to represent probability distributions. Therefore, it allows using multiple declarative distributions and/or a RDF table format to represent the probability distribution of a given RV. Each probability distribution can be expressed in different formats using PROWL’s declarative distributions represented via the DeclarativeDist class, which is depicted in Figure 33. Possible formats include Netica tables, Netica equations, Quiddity formulas, MEBN syntax, and others. However, the declaration itself is stored as a string so parsers should be compatible with the specific text format of each declaration. Every individual of class DeclarativeDist has an object property isProbDistOf [1] linking it with their respective resident node. A datatype property isRepresentedAs [2] defines how a given declarative probability distribution is expressed. A datatype property isDefault [3] flags it (or not) as a default distribution. finally, a datatype hasDeclaration [4] includes the probability distribution itself in the format previously defined. Declarative Distributions in PROWL PROWL tables have a different representational scheme. Each individual of class PROWLTable is actually a label that links the many components that collectively form a probability distribution of a resident node. As an example, the individual Z_ZoneFShips_table has three properties: isDefault, which states whether or not that individual represents a default probability distribution, isProbDistOf, which links the individual with the node it represents (Z_ZoneFShips in this case), and hasProbAssign, which links the individual with all the individuals of class ProbAssign that collectively form the probability distribution of node Z_ZoneFShips. One of those ProbAssign individuals is Z_ZoneFShips_table_2.3, which is depicted in Figure 34. A Probabilistic Assignment in a PROWL Table Z_ZoneFShips_table_2.3 corresponds to the probability assigned to the second state of node ZoneFShips (ZFS_1) given that its parent node has value ZN_PlanetarySystem (the third state of that parent). The probability itself (.20) is represented as a xsd:decimal that is linked to Z_ZoneFShips_table_2.3 via the datatype property hasStateProb [1]. The link between the ProbAssign individual and the state of ZoneFShips it refers to is made via the object property hasStateName [2], while property isProbAssignIn [3] links the probability assignment to the table it belongs. Finally, each probability assigned to a state of a variable is conditioned to a combination of states of the parents of that variable. Object property hasConditionant [4] links a ProbAssign individual to the individuals of class CondRelationship that collectively form such a combination of parents. CondRelationship is a reified relation linking a parent with one of its possible states, and a set of CondRelationship individuals represents the combination of parents’ states to which a given probability assignment is conditioned. ZoneFShips has only one parent, so there is only one conditionant listed (Z_ZoneNature_cond_3.3), which is an individual of the reified class CondRelationship that links node ZoneNature with its third state (ZN_PlanetarySystem). If ZoneFShips had four parents, then four individuals of class CondRelationship (i.e. one for each parent) would have to be listed in order to represent the combination of parents under which that probability assignment is valid. One issue regarding the probability assignment is the use of xsd:decimal to convey a probability value, when the ideal situation would be to use a datatype that specifically covers the numerical range of probabilities (i.e. 0 to 1, including both extremities). However, at the time of this writing, OWL has no support for userdefined datatypes, so the closest datatype allowed by OWL is xsd:decimal. Although applications or plugins should be written to prevent invalid entries for probabilities, relying on external plugins to enforce this requirement is not an acceptable option. Therefore, a more robust solution must be sought. In the case of a future consideration of PROWL as a basis for a W3C Recomendation for representing uncertainty in the Semantic Web, a special datatype covering the numerical range of probabilities must be included. A very suitable name for such datatype is “prob” (prowl:prob), which has already been proposed by other researchers in this field (e.g., Ding & Peng, 2004). PROWL tables represent probabilities in a format that is highly interoperable, since each cell contains links to all the elements that are necessary for specifying the conditions in which the probability inside that cell applies. Also, those elements are available in a nonproprietary, syntaxindependent format, which makes it easier to be retrieved by diverse applications without the need for a format conversion. Yet, building PROWL tables the way it was done in this work is not a feasible option for a real life application or plugin. Fortunately, all the above steps can be avoided by developing automated tools. The next Section briefly covers such possibilities. 1.22A Proposed Operational Concept for Implementing PROWL In its current stage, PROWL contains only the basic representation elements that provide a means of representing any MEBNbased model. Such a representation could be used by a Bayesian tool (acting as a probabilistic ontology reasoner) to perform inferences to answer queries and/or to learn from newly incoming evidence via Bayesian learning. However, building MFrags and all their elements in a probabilistic ontology is a manual, error prone, and tedious process. Avoiding errors or inconsistencies requires very deep knowledge of the logic and of the data structure of PROWL. Without considering the future paths to be followed by research on PROWL (i.e. whether it will be kept as an upper ontology or transformed into an actual extension to the OWL language), the framework provided in this Dissertation makes it already possible to facilitate probabilistic ontology usage and editing by developing plugins to current OWL editors. Figure 35 illustrates an example of such a concept. In that figure, a possible plugin for the OWL Protégé editor (which is itself an OWL plugin) shows a graphical construction of an MFrag being performed in a very similar fashion as a BN is constructed in a graphical package such as Netica™. In this proposed scheme, in order to build an MFrag a user would only have to select the icon of the node he/she wants to create (e.g. resident, input, context, etc.), connect that node with its parents and children, and enter its basic characteristics (i.e. name, probability distribution, etc.) either by doubleclicking on it or via another GUIrelated facility. Snapshot of a Graphical PROWL Plugin The idea of such a plugin is to hide from users the complex constructs required to convey the many details of a probabilistic ontology, such as the reified relationships, composite RV term constructions (with or without quantifiers and Skolem constants), and others. In the figure, the Zone MFrag was selected from the combo box in the top of the viewing area, thus information about its nodes is displayed in a graphical format that allows the user to build more nodes, edit or view the existing ones. and then chose node ZoneEShips(z) so it appears highlighted (a red box around it) and all its data is shown in the lower square. Tedious tasks such as building a PROWL table with many cells could be carried out much more quickly and with fewer errors, thus providing a boost in productivity. In the probability table case, the user would only have to fill the probabilities in the correct cells of a CPT’s graphical display and the plugin would build their respective PR_OWL constructs. Another point of usage improvement is the intrinsic syntax check provided by a guided construction. As an example, when writing a composite RV term, the user would not have to actually write the complex reified relations (ArgRelationships, Skolem contants, OVariables, Inner terms, etc). Instead, a menu with the allowed connectives would be available so his/her task would be reduced to enter the arguments of the formula and embed the connectives the way he/she wants. The final result would be a valid formula that would then be transformed in PROWL syntax by the plugin. This brief idea of an operational concept barely scratches the surface of the many possibilities for the technology presented here, and its purpose is to point out one such possibility. As previously stated, the present dissertation is focused on defining a coherent, comprehensive probabilistic framework for the Semantic Web, in a way that any probabilistic system could be represented and made available to perform tasks such as plausible inference and Bayesian learning. Therefore, implementing a plugin such as the one envisioned here is a development task that is outside the scope or this dissertation research. Nonetheless, it takes an important first step toward making probabilistic ontologies a reality. By opening the door to wide use of PROWL probabilistic ontologies, the present research makes a significant contribution to realizing the Semantic Web vision. Conclusion and Future Work 1.23Summary of Contributions The main objective of this research effort was to establish a framework that enables the use of Bayesian theory for representing and reasoning under uncertainty in the context of the Semantic Web. The key step for achieving such objective was the introduction of probabilistic ontologies, which were formally defined in Chapter 5. In order to provide the initial conditions for the future spread of probabilistic ontologies, we have developed a complete, modularized set of new definitions for the OWL language, which collectively form a coherent framework for building ontologies that are able to represent uncertainty from concepts of a given domain with full probabilistic firstorder logic expressiveness. Probabilistic ontologies written under this framework achieve a principled representation of uncertainty and allow for the use of different probabilistic reasoning systems as a means to perform plausible reasoning and learning from data on the MTheories represented in PROWL format. The contributions of this research effort also included the development and formalization of a typed version of MEBN logic. This extended version was needed as a means to achieve full compatibility with current Semantic Web languages, including OWL. A full implementation of MEBN logic and its typed extension does not yet exist. However, Quiddity*Suite is a powerful Bayesian probabilistic reasoning system that is capable of being applied as a PROWL reasoner. Therefore, we have also developed a set of rules for translating an MTheory written using the typed version of MEBN into a probabilistic model in IET’s Quiddity*Suite format. These rules were applied to the Starship MTheory specially developed for this research, and resulted in a running Quiddity*Suite model. The Starship MTheory includes some of the most complex aspects that can be expressed with MEBN logic, such as recursions, nodes with many uncertain parents, context constraints expressed as firstorder logic sentences with and without quantifiers, etc. Therefore, having achieved a Quiddity*Suite model capable of building any SSBN based on the original MTheory is a valid proof of concept of the feasibility of using Quiddity*Suite as a PROWL reasoner. The source code for the Quiddity Starship model is provided in the Appendix A of this dissertation. In order to demonstrate the feasibility of representing a complex MTheory using the concepts laid out in Chapter 5, the very same case study was used as a basis for writing a probabilistic ontology containing all the elements from the original model and exploring different possibilities for representing a probability distribution. The resulting PROWL ontology is logically equivalent to the original generative MTheory, and thus can be utilized as the basis for generating SSBNs to answer queries posed to the model. In addition, the representation of Finding MFrags was also covered, as a means to demonstrate how PROWL ontologies can incorporate new information, either via user insertion or by means of Bayesian learning from data. Therefore, the upper ontology presented here is capable of representing any MTheory, including both generative and MTheories with findings. In addition, it allows users to define probabilistic ontologies using a RDFbased syntax that is compatible with current OWL ontologies. Furthermore, translators could be written for thirdparty, oftheshelf probabilistic reasoners to make use of the ontology to perform Bayesian inference and learning. These capabilities were demonstrated by creating the case study ontology, translating its definitions into Quiddity*Suite and performing probabilistic inferences over it, a process that is documented in the appendices. 1.24A Long Road with Bright Signs Ahead The proposed framework can be understood as an initial solution situated in a middle ground between the extension approaches employed in OWLS and SRWL. In common with the first is the fact that no actual extension to OWL semantics and abstract syntax is performed at this time, since it is also an OWL upper ontology. Similarly to the latter, PROWL also has the need for specialized tools in order to realize its full potential, while also including concepts (e.g. the prob datatype, FOL connectives, quantifiers, etc.) that could greatly expand OWL expressiveness if adopted as a standard. Even though it is possible to represent a complex probabilistic system using PROWL definitions, performing plausible reasoning and learning from data requires an external tool (e.g. Quiddity*Suite). It is true that some preliminary consistency check and other OWLDL features are possible using PROWL (which is OWLDL compliant), and that any complex system can still be written in PROWL and be interpreted using different probabilistic reasoning systems, provided that PROWL plugins are written for capturing the data inside probabilistic ontologies in each package’s native format. Apart from the need for developing plugins for probabilistic packages so they can be used as plausible reasoners, a specific PROWL plugin for current OWL ontology editors remains a priority for future efforts. The process used here for writing probabilistic ontologies can be greatly improved via automation of most of the steps in the ontology building, mainly in the part of writing composite RV terms, but also for consistency checking, reified relations and other tasks that demand unnecessary awareness of the inner workings of the present solution. Once implemented, such plugin has the potential to make probabilistic ontologies a natural, powerful tool for helping to realize the Semantic Web vision. Furthermore, the technology has the potential to be used in important applications outside the Semantic Web, as we discuss in Appendix C In that discussion, our main point is that the proper use of probability information can help to establish reliable, more general semantic mapping schemas by means of probabilistic ontologies, which can then be applied in applications spanning diverse domains, since it relies on a metaontology (i.e. a ontology about ontologies), carrying no domain information, which has the mappings between two or more ontologies as its instances. That is, PROWL has the potential for application in other semantic mapping solutions such as the DTB case study presented in Section C.1 of the appendices. It could also be applied to facilitate interoperability between systems as discussed in the Wise Pilot cases study, presented in Section C.2. The present work thus represents a step toward a generalpurpose solution for the semantic mapping problem. Finally, the most important requirement for adoption of a language is the standardization process. This process goes significantly beyond academic research and thus falls outside the scope of the present work. Nonetheless, we are confident of its feasibility, which we believe having demonstrated in this effort, and of its desirability, given its potential to help solve many of the obstacles that stand in the way of realizing the W3C’s vision for the Semantic Web.
Bibliography
Bibliography Adams, J. B. (1976). A Probability Model for Medical Reasoning and the MYCIN Model. Mathematical Biosciences, 32(1/2), 177186. Alberts, D. S., Garstka, J. J., & Stein, F. P. (1999). Network Centric Warfare: Developing And Leveraging Information Superiority. Washingtong, D.C., USA: National Defense University Press. Alghamdi, G., Laskey, K. B., Wang, X., Barbara, D., Shackleford, T., Wright, E. J., et al. (2004). Detecting Threatening Behavior Using Bayesian Networks, Conference on Behavioral Representation in Modeling and Simulation  BRIMS. Arlington, VA, USA. Alghamdi, G., Laskey, K. B., Wright, E. J., Barbará, D., & Chang, K.C. (2005). Modeling insider user behavior using MultiEntity Bayesian Networks. In Proceedings of the Tenth International Command and Control Research Technology Symposium (10^{th} ICCRTS). McLean, VA, USA: CCRP/DOD publications. Arpírez, J. C., Corcho, O., FernándezLópez, M., & GómezPérez, A. (2001). WebODE: A Scalable Workbench for Ontological Engineering, International Conference on Knowledge Capture (pp. 613). Victoria, British Columbia, Canada: ACM Press. Baader, F.; & Nutt, W. (2003). Basic Description Logics. Chapter in The Description Logics Handbook: Theory, Implementation and Applications. Baader, F.; Calvanese, D.; McGuiness, D.; Nardi, D.; PatelSchneider, P.; editors. 1^{st} edition, chapter 2, pages 47100. Cambridge, UK: Cambridge University Press. Ball, W. W. R. (2003). A Short Account of the History of Mathematics. New York, NY, USA: Main Street Books (originally published in 1908). Bangsø, O., & Wuillemin, P.H. (2000). Object Oriented Bayesian Networks: A Framework for Topdown Specification of Large Bayesian Networks and Repetitive Structures. Technical Report No. CIT87.200obphw1. Department of Computer Science, Aalborg University, Aalborg, Denmark. Baum, L. E., & Petrie, T. (1966). Statistical Inference for Probabilistic Functions of Finite State Markov Chains. Annals of Mathematical Statistics, 37, 15541563. BernersLee, T., & Fischetti, M. (2000). Weaving the Web: The Original Design and Ultimate Destiny of the World Wide Web by its Inventor . 1^{st} edition. New York, NY, USA: HarperCollins Publishers. BernersLee, T., Hendler, J., & Lassila, O. (2001). The Semantic Web, Scientific American Digital (pp. 7): Scientific American, Inc. Bobrow, D. G., & Winograd, T. (1977). An Overview of KRL, a Knowledge Representation Language. Cognitive Science, 1(1), 346. Boley, H., & Tabet, S. (2004). The Rule Markup Initiative. Retrieved May 29, 2005, from http://www.ruleml.org/ Booker, L. B., & Hota, N. (1986, August 810). Probabilistic Reasoning about Ship Images. Paper presented at the Second Annual Conference on Uncertainty in Artificial Intelligence, University of Pennsylvania, Philadelphia, PA. Bournez, C., & Hawke, S. (2004, November 23). Team Comment on the OWLS Submission. Retrieved June 29, 2005, from http://www.w3.org/Submission/2004/07/Comment Brachman, R. J. (1977). What's in a Concept: Structural Foundations for Semantic Networks. International Journal of ManMachine Studies, 9(2), 127152. Brachman, R. J., & Schmolze, J. G. (1985). An Overview of the KLONE Knowledge Representation System. Cognitive Science, 9(2), 171216. Buchanan, B. G., & Shortliffe, E. H. (1984). RuleBased Expert Systems: The MYCIN Experiments of the Stanford Heuristic Programming Project. Reading, MA, USA: AddisonWesley. Buntine, W. L. (1994a). Learning with Graphical Models. Technical Report No. FIA9403. NASA Ames Research Center, Artificial Intelligence Research Branch. Buntine, W. L. (1994b). Operations for Learning with Graphical Models. Journal of Artificial Intelligence Research, 2, 159225. Calvanese, D.; & De Giacomo, G. (2003). Expressive Description Logics. Chapter in The Description Logics Handbook: Theory, Implementation and Applications. Baader, F.; Calvanese, D.; McGuiness, D.; Nardi, D.; PatelSchneider, P.; editors. 1^{st} edition, chapter 5, pages 184225. Cambridge, UK: Cambridge University Press. Cardelli, L., & Wegner, P. (1985). On Understanding Types, Data Abstraction and Polymorphism. Computing Surveys, 17(4), 471522. Carnap, R. (1950). Logical Foundations of Probability. Chicago, IL, USA: University of Chicago Press. Charniak, E. (1991). Bayesian Networks without Tears. AI Magazine, 12, 5063. Charniak, E., & Goldman, R. P. (1989a). Plan Recognition in Stories and in Life. Paper presented at the Fifth Workshop on Uncertainty in Artificial Intelligence, Mountain View, California. Charniak, E., & Goldman, R. P. (1989b). A Semantics for Probabilistic QuantifierFree FirstOrder Languages with Particular Application to Story Understanding. Paper presented at the Eleventh International Joint Conference on Artificial Intelligence, August 1989, Detroit, Michigan, USA. Cheng, J., & Druzdzel, M. J. (2000). AISBN: An Adaptive Importance Sampling Algorithm for Evidential Reasoning in Large Bayesian Networks. Journal of Artificial Intelligence Research, 13, 155188. Clottes, J. E., Chauvet, J. M., BrunelDeschamps, E., Hillaire, C., Saugas, J. P., Evin, J., et al. (Eds.). (1995). Les Peintures Paléolithiques de la Grotte Chauvet Pontd'Arc à Vallon Pontd'Arc (Ardèche, France): Datations directes et indirectes par la méthode du radiocarbone (Vol. 320). Paris, France: l'Académie de Sciences. CNN.com. (1998). Rationalizing Treason: An Interview with Aldrich Ames. Cold War Experience  Espionage Series Retrieved January 20, 2005, from http://www.cnn.com/SPECIALS/cold.war/experience/spies/interviews/ames. CNN.com. (2001). The Case Against Robert Hanssen: An FBI Insider and Admitted Spy. InDepth Special Series Retrieved January 20, 2005, from http://www.cnn.com/SPECIALS/2001/hanssen/ Codd, E. F. (1970). A Relational Model for Large Shared Data Banks. Communications of the ACM, 13(6), 377387. Cohen, L. J. (1989). An Introduction to the Philosophy of Induction and Probability. Oxford, UK: Clarenton Press. Connoly, D., Khare, R., & Rifkin, A. (1997). The Evolution of Web Documents: The Ascent of XML. Word Wide Web Journal (special issue on XML), 2(4), 119128. Cooper, G. F. (1987). Probabilistic Inference using Belief Networks is NpHard. Paper No. SMI870195. Knowledge Systems Laboratory, Stanford University. Stanford, CA, USA. Cooper, G. F., & Herskovits, E. (1992). A Bayesian Method for the Induction of Probabilistic Networks from Data. Machine Learning, 9, 309347. Costa, P. C. G. (1999). The Fighter Aircraft's Autodefense Management Problem: A Dynamic Decision Network Approach. Master of Science Thesis, School of Information Technology and Engineering, George Mason University. Fairfax, VA, USA. Costa, P. C. G., Laskey, K. B., Takikawa, M., Pool, M., Fung, F., & Wright, E. J. (2005). MEBN Logic: A Key Enabler for Network Centric Warfare. In Proceedings of the Tenth International Command and Control Research and Technology Symposium (10^{th} ICCRTS). Mclean, VA, USA: CCRP/DOD publications. Cox, R. T. (1946). Probability, Frequency and Reasonable Expectation. American Journal of Physics, 14, 113. Cristianini, N., ShaweTaylor, J., & Lodhi, H. (2001). Latent Semantic Kernels. Paper presented at the 18th International Conference on Machine Learning, San Francisco, CA, USA. DaConta, M. C., Obrst, L. J., & Smith, K. T. (2003). The Sematic Web: A Guide to the Future of Xml, Web Services, and Knowledge Management. Indianapolis, IN, USA: Wiley Publishing, Inc. Dagum, P., & Luby, M. (1993). Approximating Probabilistic Inference in Bayesian Belief Networks is NpHard. Artificial Intelligence, 60, 141153. Damas, L., & Milner, R. (1982). Principal Type Schemes for Functional Programs. Paper presented at the 9th ACM Symposium on Principles of Programming Languages. Albuquerque, NM, USA. de Finetti, B. (1990). Theory of Probability: A Critical Introductory Treatment. New York, NY, USA: John Wiley & Sons. Originally published in 1974. Dempster, A. P. (1967). Upper and Lower Probabilities Induced by a Multivalued Mapping. Annals of Mathematical Statistics, 38, 325339. Ding, Z., & Peng, Y. (2004). A Probabilistic Extension to Ontology Language OWL. Paper presented at the 37^{th} Annual Hawaii International Conference on System Sciences (HICSS'04). Jan, 58, 2004. Big Island, Hawaii, USA. Druzdzel, M. J., & Simon, H., A. (1993). Causality in Bayesian Belief Networks. Paper presented at the Ninth Annual Conference on Uncertainty in Artificial Intelligence (UAI93). San Francisco, CA, USA. Druzdzel, M. J., & Yuan, C. (2003). An Importance Sampling Algorithm Based on Evidence PrePropagation. Paper presented at the Nineteenth Annual Conference on Uncertainty in Artificial Intelligence. Acapulco, Mexico. Elliott, R. J., Aggoun, L., & Moore, J. B. (1995). Hidden Markov Models: Estimation and Control. New York, NY, USA: SpringerVerlag. Fellbaum, C. (Ed.). (1998). Wordnet, An Electronic Lexical Database. Cambridge, MA, USA: The MIT Press. Fikes, R. E., & Kehler, T. P. (1985). The Role Of FrameBased Representation in Knowledge Representation and Reasoning. Communications of the ACM, 28(9), 904920. Fine, T. L. (1973). Theories of Probability: An Examination of Foundations. New York, NY, USA: Academic Press. Frege, G. (1879) Begriffsschrift, Eine Der Arithmetischen Nachgebildete Formelsprache Des Reinen Denkens, Halle a. S.: Louis Nebert. Translated as Concept Script, A Formal Language of Pure Thought Modeled Upon that of Arithmetic, by S. BauerMengelberg in J. vanHeijenoort (ed.), From Frege to Gödel: A Source Book in Mathematical Logic, 18791931. Cambridge, MA, USA: Harvard University Press, 1967. Friedman, N., & Koller, D. (2000). Being Bayesian about Network Structure. Paper presented at the Sixteenth Conference on Uncertainty in Artificial Intelligence. San Mateo, California, USA. Fukushige, Y. (2004). Representing Probabilistic Knowledge in the Semantic Web, W3C Workshop on Semantic Web for Life Sciences. Cambridge, MA, USA. Fung, F., Laskey, K. B., Pool, M., Takikawa, M., & Wright, E. J. (2004). PLASMA: Combining Predicate Logic and Probability for Information Fusion and Decision Support. Paper presented at the AAAI Spring Symposium. Stanford, CA, USA. Fung, R., & Chang, K. C. (1989). Weighing and Integrating Evidence for Stochastic Simulation in Bayesian Networks. In M. Henrion, R. D. Shachter, L. N. Kanal & J. F. Lemmer (Eds.), Uncertainty in Artificial Intelligence 5 (pp. 209219). New York, NY, USA: Elsevier Science Publishing Company, Inc. Fung, R., & del Favero, B. (1994). Backward Simulation in Bayesian Networks. Paper presented at the Tenth Annual Conference on Uncertainty in Artificial Intelligence. San Francisco, CA, USA. Gelman, A. (2003). Bayesian Data Analysis. 2^{nd} edition. London, UK: Chapman and Hall. Getoor, L., Friedman, N., Koller, D., & Pfeffer, A. (2001). Learning Probabilistic Relational Models. New York, NY, USA: SpringerVerlag. Getoor, L., Koller, D., Taskar, B., & Friedman, N. (2000). Learning Probabilistic Relational Models with Structural Uncertainty. Paper presented at the ICML2000 Workshop on AttributeValue and Relational Learning:Crossing the Boundaries. Stanford, CA, USA. Gil, Y., & Ratnakar, V. (2004). Markup Languages: Comparison and Examples. Retrieved January 16, 2005, from http://trellis.semanticweb.org/expect/web/semanticweb/comparison.html Gilks, W., Thomas, A., & Spiegelhalter, D. J. (1994). A Language and Program for Complex Bayesian Modeling. The Statistician, 43, 169178. Giugno, R., & Lukasiewicz, T. (2002). PSHOQ(D): A Probabilistic Extension of SHOQ(D) for Probabilistic Ontologies in the Semantic Web. Paper presented at the European Conference on Logics in Artificial Intelligence (JELIA 2002), Sep 2326. Cosenza, Italy. Goldfarb, C. F. (1996). The Roots of SGML  A Personal Recollection. Retrieved September 30, 2004, from http://www.sgmlsource.com/history/roots.htm Greco, S., Leone, N., & Rullo, P. (1992). COMPLEX: An ObjectOriented Logic Programming System. IEEE Transactions on Knowledge and Data Engineering, 4(4), 344359. Grenander, U. (1995). Elements of Pattern Theory. Baltimore, MD, USA: Johns Hopkins University Press. Gruber, T. R. (1993). A Translation Approach to Portable Ontology Specifications. Knowledge Acquisition, 5(2), 199220. Gu, T., Keng, P. H., & Qing, Z. D. (2004). A Bayesian Approach for Dealing With Uncertainty Contexts. Paper presented at the Second International Conference on Pervarsive Computing, Apr 1823. Vienna, Austria. Haarslev, V., & Möller, R. (2001). RACER System Description. Paper presented at the International Joint Conference on Automated Reasoning (IJCAR'2001). Siena, Italy. Hacking, I. (1965). The Logic of Statistical Inference. Cambridge, MA, USA: Cambridge University Press. Hacking, I. (1975). The Emergence of Probability: A Philosophical Study of Early Ideas about Probability, Induction, and Statistical Inference. Cambridge, MA, USA: Cambridge University Press. Hansson, O., & Mayer, A. (1989). Heuristic Search as Evidential Reasoning. Paper presented at the Fifth Workshop on Uncertainty in Artificial Intelligence. Windsor, Ontario, Canada. Heckerman, D., Geiger, D., & Chickering, D. M. (1995a). Learning Bayesian Networks: The Combination of Knowledge and Statistical Data. Machine Learning (20), 197243. Heckerman, D., Mamdani, A., & Wellman, M. P. (1995b). RealWorld Applications of Bayesian Networks. Communications of the ACM, 38(3), 2468. Heckerman, D., Meek, C., & Koller, D. (2004). Probabilistic Models for Relational Data. Technical Report MSRTR200430, Microsoft Corporation, March 2004. Redmond, WA, USA. Heflin, J. (2004, February 10). OWL Web Ontology Language  Use Cases And Requirements. Retrieved December 04, 2005, from http://www.w3.org/TR/2004/RECwebontreq20040210/ Heinsohn, J. (1994). Probabilistic Description Logics. Paper presented at the Tenth Conference on Uncertainty in Artificial Intelligence (UAI94), Jul 2931.Seattle, WA, USA. Helsper, E. M., & van der Gaag, L. C. (2001). Ontologies for Probabilistic Networks: A Case Study in Oesophageal Cancer. Paper presented at the Thirteenth DutchBelgian Artificial Intelligence Conference. Amsterdam, The Netherlands. Hendler, J. (2004). Frequently Asked Questions on W3C's Web Ontology Language (OWL). Retrieved January 14, 2005, from http://www.w3.org/2003/08/owlfaq.html. Henrion, M. (1988). Propagation of Uncertainty by Probabilistic Logic Sampling in Bayes Networks. In J. F. Lemmer & L. N. Kanal (Eds.), Uncertainty in Artificial Intelligence 2 (pp. 149163). New York, NY, USA: Elsevier Science Publishing Company, Inc. Hillmann, D. (2001). Using Dublin Core. Retrieved June 28, 2005, from http://dublincore.org/documents/usageguide/. Hofmann, T. (1999). Probabilistic Latent Semantic Indexing. Paper presented at the 22^{nd} Annual ACM Conference on Research and Development in Information Retrieval, August 1519. Berkeley, CA, USA. Horrocks, I. (2002). DAML+OIL: A Reasonable Web Ontology Language. Keynote talk at the WES/CAiSE Conference. Toronto, Canada. Horrocks, I., PatelSchneider, P. F., Boley, H., Tabet, S., Grosof, B., & Dean, M. (2004). SWRL: A Semantic Web Rule Language Combining OWL and RuleML. W3C Member Submission. Retrieved May 29, 2005, from http://www.w3.org/Submission/2004/SUBMSWRL20040521/. Horrocks, I., & Sattler, U. (2001). Ontology Reasoning in the SHOQ(D) Description Logic. Paper presented at the Seventeenth International Joint Conference on Artificial Intelligence (IJCAI 2001), Aug 410. Seattle, WA, USA. Hotho, A., Staab, S., & Stumme, G. (2002). Text Clustering Based on Background Knowledge. Technical Report No. 425. Institute of Applied Informatics and Formal Description Methods AIFB, University of Karlsruhe. Karlsruhe, Germany. Hotho, A., Staab, S., & Stumme, G. (2003). Wordnet Improves Text Document Clustering, Semantic Web Workshop at SIGIR2003, 26^{th} Annual International ACM SIGIR Conference. Toronto, Canada. Jacobs, I., editor (2003, June 18). World Wide Web Consortium Process Document. Retrieved March 03, 2005, from http://www.w3.org/2003/06/Process20030618/cover.html Jaeger, M. (1994). Probabilistic Reasoning in Terminological Logics. Paper presented at the Fourth International Conference on Principles of Knowledge Representation and Reasoning (KR94), May 2427. Bonn, Germany. Jaeger, M. (1997). Relational Bayesian Networks. Paper presented at the 13th Annual Conference on Uncertainty in Artificial Intelligence (UAI97), August 13, Providence, RI, USA. Jensen, F. V. (1996). An Introduction to Bayesian Networks. New York, NY, USA: SpringerVerlag. Jensen, F. V. (2001). Bayesian Networks and Decision Graphs. New York, NY, USA: SpringerVerlag. Joachims, T. (1998). Text Categorization with Support Vector Machines: Learning With Many Relevant Features. Paper presented at the Tenth European Conference on Machine Learning, April 21 – 24. Chemnitz, Germany. Jordan, M. I., (Ed.). (1999). Learning in Graphical Models. Cambridge, MA, USA: MIT Press. Kadane, J. B., Schervish, M. J., & Seidenfeld, T. (1999). Rethinking the Foundations of Statistics. New York, NY, USA: Cambridge University Press. Kadane, J. B., & Schum, D. A. (1996). A Probabilistic Analysis of the Sacco and Vanzetti Evidence. New York, NY, USA: John Wiley & Sons. Kalyanpur, A., Sirin, E., Parsia, B., & Hendler, J. (2004). Hypermedia Inspired Ontology Engineering Environment: SWOOP. International Semantic Web Conference (ISWC2004). Hiroshima, Japan. Keynes, J. M. (2004). A Treatise on Probability. New York, NY, USA: Dover Publications. Originally published in 1921. Kifer, M., Lausen, G., & Wu, J. (1990). Logical Foundations of ObjectOriented and FrameBased Languages. Technical Report TR90003. University of Mannheim. Mannheim, BadenWürttemberg, Germany. Knublauch, H., Fergerson, R. W., Noy, N. F., & Musen, M. A. (2004). The Protégé OWL plugin: An Open Development Environment for Semantic Web Applications. Third International Semantic Web Conference (ISWC2004). Hiroshima, Japan. Kokkelink, S., & Schwänzl, R. (2001). Expressing Qualified Dublin Core in RDF / XML. Retrieved June 28, 2005, from http://dublincore.org/documents/dcqrdfxml/ . Koller, D., Levy, A. Y., & Pfeffer, A. (1997). PCLASSIC: A Tractable Probabilistic Description Logic. Paper presented at the Fourteenth National Conference on Artificial Intelligence (AAAI97), July 2731. Providence, RI, USA. Koller, D., & Pfeffer, A. (1997). ObjectOriented Bayesian Networks. Paper presented at the Thirteenth Conference on Uncertainty in Artificial Intelligence (UAI97). San Francisco, CA, USA. Kolmogorov, A. N. (1960). Foundations of the Theory of Probability. 2^{nd} edition. New York, NY, USA: Chelsea Publishing Co. Originally published in 1933. Korb, K. B., & Nicholson, A. E. (2003). Bayesian Artificial Intelligence. Boca Raton, FL, USA: Chapman & Hall/CRC Press. Langseth, H., & Nielsen, T. (2003). Fusion of Domain Knowledge with Data for Structured Learning in ObjectOriented Domains. Journal of Machine Learning Research, Special Issue on the Fusion of Domain Knowledge with Data for Decision Support, vol. 4, pp. 339368, July 2003. Laplace, P. S. (1996). A Philosophical Essay and Probabilities. New York, NY, USA: Dover Publications. Originally published in 1826. Laskey, K. B. (2005, March 15). FirstOrder Bayesian Logic. Draft paper, Department of Systems Engineering and Operation Research, School of Information Technology and Engineering, George Mason University. Retrieved July 3, 2005, from http://ite.gmu.edu/~klaskey/publications.html Laskey, K. B., & Lehner, P. E. (1994). Metareasoning and the Problem of Small Words. IEEE Transactions on Systems, Man and Cybernetics, 24(11), 16431652. Lauritzen, S., & Spiegelhalter, D. J. (1988). Local Computation and Probabilities on Graphical Structures and their Applications to Expert Systems. Journal of Royal Statistical Society, 50(2), 157224. Lee, P. M. (2004). Bayesian Statistics: An Introduction. 3^{rd} edition. London, UK: Edward Arnold Publishers. Lewis, D. (1980). A Subjectivist's Guide to Objective Chance. In Studies in Inductive Logic and Probability, Vol II. Berkeley, CA, USA: University of California Press. Littman, M. L., Letsche, T. A., Dumais, S. T., & Landauer, T. K. (1997, March 2426). Automatic CrossLinguistic Information Retrieval using Latent Semantic Indexing. Paper presented at the AAAI Spring Symposium on CrossLanguage Text and Speech Retrieval, Stanford University. Stanford, CA, USA. Lucas, P. J. F., van der Gaag, L. C., & AbuHanna, A. (2001, July 1^{st}). Workshop on Bayesian Models in Medicine. Paper presented at the European Conference on Artificial Intelligence in Medicine. Cascais, Portugal. Mahoney, S. M., & Laskey, K. B. (1998). Constructing Situation Specific Networks. In UAI’98: Proceedings of the Fourteenth Conference on Uncertainty in Artificial Intelligence, pages 370378. July 2426, University of Wisconsin Business School. Madison, WI, USA. Martin, D., Burstein, M., Hobbs, J., Lassila, O., McDermott, D., McIlraith, S., et al. (2004). OWLS: Semantic Markup for Web Services. Retrieved June 29, 2005, from http://www.daml.org/services/owls/1.1/overview/. Miller, E., & Hendler, J. (2004). Web Ontology Language (OWL). Retrieved January 14, 2004, from http://www.w3.org/2004/OWL/. Miller, E., & Hendler, J. (2005, February 04). Web Ontology Language (OWL). Retrieved June 28, 2005, from http://www.w3.org/2004/OWL/. Miller, G. A., Beckwith, R., Fellbaum, C., Gross, D., & Miller, K. (1990). Five Papers on WordNet. Technical Report No. 43, Cognitive Science Laboratory, Princenton University. Princenton, NJ, USA. Milner, R. (1978). A Theory of Type Polymorphism in Programming. Journal of Computer and System Sciences, 17, 348375. Minsky, M. L. (1975). Framework for Representing Knowledge. In The Psychology of Computer Vision. P. H. Winston (Eds.), pages 211277. New York, NY: McGrawHill. Mitchell, J. C. (1984). Type Inference and Type Containment. Paper presented at the International Symposium on Semantics of Data Types, June 2729. SophiaAntipolis, France. Murphy, K. (1998). Dynamic Bayesian Networks: Representation, Inference and Learning. Doctoral Dissertation, University of California. Berkeley, CA, USA. Neapolitan, R. E. (1990). Probabilistic Reasoning in Expert Systems: Theory and Algorithms. New York, NY, USA: John Wiley and Sons, Inc. Neapolitan, R. E. (2003). Learning Bayesian Networks. New York, NY, USA: Prentice Hall. Noy, N. F., Fergerson, R. W., & Musen, M. A. (2000). The Knowledge Model of Protégé2000: Combining Interoperability and Flexibility. Paper presented at the Twelfth International Conference on Knowledge Engineering and Knowledge Management (EKAW'2000). JuanlesPins, France. Noy, N. F., & Rector, A. (2004). Defining Nary Relations on the Semantic Web: Use with Individuals. W3C Working Draft Retrieved May 06, 2005, from http://www.w3.org/TR/2004/WDswbpnaryRelations20040721/. Noy, N. F., Sintek, M., Decker, S., Crubezy, M., Fergerson, R. W., & Musen, M. A. (2001). Creating Semantic Web Contents with Protégé2000. IEEE Intelligent Systems, 16(2), 6071. Oliver, R. M., & Smith, J. Q. (1990). Influence Diagrams, Belief Nets and Decision Analisys. 1^{st} edition. New York, NY, USA: John Willey & Sons Inc. Page, L. B. (1988). Probability for Engineering with Applications to Reliability. New York, NY, USA: Computer Science Press, Inc. PatelSchneider, P. F. (2005, April 11). A Proposal for a SWRL Extension Towards FirstOrder Logic. W3C Member Submission. Retrieved May 29, 2005, from http://www.w3.org/Submission/2005/SUBMSWRLFOL20050411/. PatelSchneider, P. F., Hayes, P., & Horrocks, I. (2004, February 10). OWL Web Ontology Language  Semantics and Abstract Syntax. W3C Recommendation. Retrieved December 05, 2004, from http://www.w3.org/TR/owlsemantics/. Pearl, J. (1988). Probabilistic Reasoning in Intelligent Systems: Networks of Plausible Inference. San Mateo, CA, USA: Morgan Kaufmann Publishers. Pearl, J. (2000). Causality: Models, Reasoning, and Inference. Cambridge, U.K.: Cambridge University Press. Peirce, C. S. (1885). On the Algebra of Logic. American Journal of Mathematics, 7, 180202. Pfeffer, A. (2001). IBAL: A Probabilistic Rational Programming Language International. In Proceedings of the Seventeenth International Joint Conference on Artificial Intelligence (IJCAI2001), August 410, vol. 1, pp. 733740. Seattle, WA, USA. Pfeffer, A., Koller, D., Milch, B., & Takusagawa, K. T. (1999). SPOOK: A System for Probabilistic ObjectOriented Knowledge Representation. In Proceedings of the Fifteenth Conference on Uncertainty in Artificial Intelligence, pp. 541550, July 30 – August 1. Stockholm, Sweden. Polleres, A.; Bussler, C.; Travers, G.; Domingue, J. B.; & Burdett, D. (2005, April 4). Web Service Modeling Ontology (WSMO) Submission. W3C Member Submission. Retrieved June 29, 2005, from http://www.w3.org/Submission/2005/06/. Pool, M., & Aikin, J. (2004). KEEPER and Protégé: An Elicitation Environment for Bayesian Inference Tools. Paper presented at the Workshop on Protégé and Reasoning held at the Seventh International Protégé Conference, July 6 – 9. Bethesda, MD, USA. Pool, M. (2004). An OWL Based Implementation of Quiddity*Modeler. Technical Report. Information Extraction and Transport, Inc., July 2004. Rosslyn, VA, USA. Popper, K. R. (1957). The Propensity Interpretation of the Calculus of Probability and the Quantum Theory. In Stefan Körner, ed. Observation and Interpretation: A Symposium of Philosophers and Physicists. Proceedings of the Ninth Symposium of the Colston Research Society held in the University of Bristol, April 1st–April 4th, 1957. London: Butterworth Scientific Publications, 1957, 65–70. Popper, K. R. (1959). The Propensity Interpretation of Probability. British Journal for the Philosophy of Science 10 (1959), 25–42. Porter, M. F. (1980). An Algorithm for Suffix Stripping. Program  Electronic Library and Information Systems, 14(3), p.130137. Powers, S. (2003). Practical RDF. 1^{st} edition. Sebastopol, CA, USA: O'Reilly & Associates, Inc. Press, S. J. (1989). Bayesian Statistics: Principles, Models, and Applications. New York, NY, USA: John Wiley & Sons. Rabiner, L. R. (1989). A Tutorial on Hidden Markov Models and Selected Applications In Speech Recognition. Proceedings of the IEEE, 77(2), p. 257285. Ramsey, F. P. (1931). The Foundations of Mathematics and other Logical Essays. London, UK: Kegan Paul, Trench, Trubner & Co. Savage, L. J. (1972). The Foundations of Statistics. New York, NY, USA: Dover Publications (originally published in 1954). SchmidtSchauß, M., & Smolka, G. (1991). Attributive Concept Descriptions with Complements. Artificial Intelligence, 48(1), 126. Schum, D. A. (1994). Evidential Foundations of Probabilistic Reasoning. New York, NY, USA: Wiley. Sebastiani, F. (2002). Machine Learning in Automated Text Categorization. ACM Computing Surveys, 34(1), 147. Shachter, R. D., & Peot, M. A. (1990). Simulation Approaches to General Probabilistic Inference on Bayesian Networks. In M. Henrion, R. D. Shachter, L. N. Kanal & J. F. Lemmer (Eds.), Uncertainty in Artificial Intelligence 5. New York, NY: Elsevier Science Publishing Company, Inc. Shafer, G. (1976). A Mathematical Theory of Evidence. Princeton, NJ, USA: Princeton University Press. Shafer, G. (1986). The Construction of Probability Arguments. Boston University Law Review, 66(34), 799816. Shenoy, P. P., & Demirer, R. (2001). Sequential Valuation Networks for Asymmetric Decision Problems. European Journal of Operations Research: School of Business, University of Kansas. Lawrence, KS, USA. Shortliffe, E. H., Rhame, F. S., Axline, S. G., Cohen, S. N., Buchanan, B. G., Davis, R., et al. (1975, August). MYCIN: A Computer Program Providing Antimicrobial Therapy Recommendations. Clinical Medicine, 34. Siolas, G., & d'AlchéBuc, F. (2000, July 2427). Support Vector Machines Based on a Semantic Kernel for Text Categorization. Paper presented at the International Joint Conference on Neural Networks. Como, Italy. Smith, B. (2004). Ontology: Philosophical and Computational. Retrieved January 15, 2005, from http://ontology.buffalo.edu/smith//articles/ontologies.htm. Sowa, J. F. (2000). Knowledge Representation: Logical, Philosophical, and Computational Foundations. Pacific Grove, CA, USA: Brooks/Cole. Spiegelhalter, D. J., Franklin, R., & Bull, K. (1989). Assessment, Criticism, and Improvement of Imprecise Probabilities for a Medical Expert System. In Proceedings of the Fifth Conference on Uncertainty in Artificial Intelligence, pages 285294. Mountain View, CA. Spiegelhalter, D. J., Thomas, A., & Best, N. (1996). Computation on Graphical Models. Bayesian Statistics, 5, 407425. Spyns, P., Meersman, R., & Jarrar, M. (2002). Data Modeling versus Ontology Engineering. SIGMOD Record, 31(4), 1217. Stone, L. D., Barlow, C. A., & Corwin, T. L. (1999). Bayesian Multiple Target Tracking. Boston, MA, USA: Artech House. Takikawa, M., d’Ambrosio, B., & Wright, E. (2002). RealTime Inference with LargeScale Temporal Bayes Nets. In Proceedings of the Eighteenth Conference on Uncertainty in Artificial Intelligence (UAI2002), pages 477484, August 14. University of Alberta. Edmonton, Alberta, Canada. Toffler, A. (1980). The Third Wave, 1^{st} edition. New York, NY, USA: Morrow. Uschold, M., & Gruninger, M. (1996). Ontologies: Principles, Methods, and Applications. Knowledge Engineering Review, 11(2), 93155. von Mises, R. (1981). Probability, Statistics and Truth. 2^{nd} edition. New York, NY, USA: Dover Publications. Originally published in 1928. Watson, S. R., & Buede, D. M. (1987). Decision Synthesis: The Principles and Practice of Decision Analysis. Cambridge, UK: Cambridge University Press. Weatherford, R. (1982). Philosophical Foundations of Probability Theory. London, UK: Routledge & K. Paul.
Source Code for The Starship Model The source code listed below refers to the probabilistic model that was developed for this research effort. It corresponds to four separate files, namely: Starship_main.spi – It’s the execution manager for the Starship model Starship_framedefs.spi – Defines the model’s frame structure Starship _functions.spi – Defines the functions used in the model Starship_exec.spi – Create instances and built an SSBN The output of the models is a Netica file that will be saved in each model's respective folder and will be named as Starship_v00_SSBN_00t_00f_00e_00c.dne, where: v00  the model version 00t  number of time steps 00f  number of friend starships 00e  number of enemy starships 00c  number of enemy starships with cloak mode This source code was generated using the following configuration: Hardware: Apple PowerMac Dual G5 – 2.0 GHz – 1.5 GB RAM Software: Apple Mac OS X Panther (version 10.3.9) Java virtual machine (version 1.4.2_05) IET Quiddity*Suite (version 4.1.5– build Unix041217T1653) JEdit (version 4.2) Norsys Netica (version 2.17) running on top of MS Virtual PC (version 7.0)
