The Infona portal uses cookies, i.e. strings of text saved by a browser on the user's device. The portal can access those files and use them to remember the user's data, such as their chosen settings (screen view, interface language, etc.), or their login data. By using the Infona portal the user accepts automatic saving and using this information for portal operation purposes. More information on the subject can be found in the Privacy Policy and Terms of Service. By closing this window the user confirms that they have read the information on cookie usage, and they accept the privacy policy and the way cookies are used by the portal. You can change the cookie settings in your browser.
The purpose of this note is to present and evaluate the options available to a researcher wishing to use logic for representation, reasoning and computation in his application area. We examine two case studies, modal logic and resurce logics and show that in both cases neither classical logic nor its alternatives modal logic or substructural logic are completely satisfactory. We argue that there is...
The model elimination calculus is a linear, refulationally complete calculus for first order clause logic. We show how to extend this calculus with a framework for theory reasoning. Theory reasoning means to separate the knowledge of a given domain or theory and treat it by special purpose inference rules. We present two versions of theory model elimination: the one is called total theory...
We present a sound and complete calculus for an expressive sorted first-order logic. Sorts are extended to the semantic and pragmatic use of unary predicates. A sort may denote an empty set and the sort structure can be created by making use of the full first-order language. Technically spoken, we allow sort declarations to be used in the same way than ordinary atoms. Therefore we can compile every...
Along the line of the science of reasoning proposed by Alan Bundy [8], we present in this paper a computational theory accounting for human formal deductive competence. Our goal is primarily twofold. For one thing, it is aimed to establish an explanatory framework for human theorem proving. Devised as a computational theory, for another thing, it should also set up a theoretical foundation for deductive...
We present a new approach to Automated Deduction based on the concept of Shannon graphs, which are also known as Binary Decision Diagrams (BDDs). A Skoleinized formula is first transformed into a Shannon graph, then the latter is compiled into a set of Horn clauses. These can finally be run as a Prolog program trying to refute the initial formula. It is also possible to precompile axiomatizations...
The euphoric period in the history of expert systems has definitely come to an end. It is presently time to review the efforts which have been made in this field. Oriented to our conceptual framework of knowledge transformation, some trip wires which can seriously jeopardize the success of expert-system-developing projects are described in the following.
We view knowledge engineering as a constructive activity that encompasses both model building to make sense and model building to implement systems. We list four properties that we feel are important for environments that support this view on modeling and that exploit the symbiosis of both facets: epistemological modeling primitives, reusable templates, multifaceted modeling, and formal languages...
Modelling and formalizing knowledge are fundamental issues of knowledge acquisition. We argue that this process must be documented. The media break in the knowledge acquisition process—occurring in the transition from the informal knowledge to a formalized knowledge base—is one reason for lacking documentation. The system UFA supports and documents knowledge modelling and formalization of case-based...
Most problem solvers have a one-dimensional stop criterion: compute the correct and complete solution. Incremental algorithms can be interrupted at any time, returning a result that is more accurate the more time has been available. They allow the introduction of time as a new dimension into stop criteria. We can now define a system's utility in terms of the quality of its results and the time required...
In Computer Science, especially in AI, the treatment of temporal information is very important. For example, temporal restrictions for actions play a central role in planning. Mostly, qualitative constraints between actions, i.e. between time intervals assigned to actions, are represented in so called time interval networks. But humans involved often specify inconsistent networks. Thus, to support...
We shall consider an application in mechanical engineering, and shall show that the adequate modelling of the terminology of this problem domain in a conventional concept language poses two main representation problems. The first requires access to concrete domains, such as real numbers, while the second asks for a construct which can be used to represent sequences of varying length. As shown in recent...
Terminological reasoning systems directly support the abstraction mechanisms generalization and classification. But they do not bother about aggregation and have some problems with reasoning demands such as concrete domains, sequences of finite but unbounded size and derived attributes. The paper demonstrates the relevance of these issues in an analysis of a mechanical engineering application and...
Forward chaining of a logic program is specified by a meta-interpreter written in Prolog. Premises of forward rules are verified by Prolog's backward proof procedure. Since meta-interpreters in general suffer from the overhead of the additional meta-layer, a well-known technique, partial evaluation of the metainterpreter wrt the object-level program, is applied to reduce this overhead. Still several...
Neural networks are usually seen as obtaining all their knowledge through training on the basis of examples. In many AI applications appropriate for neural networks, however, symbolic knowledge does exist which describes a large number of cases relatively well, or at least contributes to partial solutions. From a practical point of view it appears to be a waste of resources to give up this knowledge...
The system LEFT is presented that learns most specific generalizations (MSGs) from structural descriptions. The new inductive multi-staged generalization algorithm is based on several new or enhanced concepts that improve the quality of generalization and make it applicable to real-world problems: LEFT evaluates the quality of each generated MSG using weighted predicates. The algorithm distinguishes...
ion has been identified as a powerful means to reduce the complexity of planning problems. In this paper, a formal model and a method are described for learning abstract plans from concrete plans. In this model, the problem of plan abstraction is decomposed into finding a state abstraction mapping and a sequence abstraction mapping. The definition of an abstract planning world and a generic state abstraction theory...
This paper considers the application of reinforcement learning to path finding tasks in continuous state space in the presence of obstacles. We show that cumulative evaluation functions (as Q-Functions [28] and V-Functions [4]) may be discontinuous if forbidden regions (as implied by obstacles) exist in state space. As the infinite number of states requires the use of function approximators such as...
In this paper we present an approach to teach classification knowledge. Our system TUDIS allows students to perform the complete process of solving a classification problem with the hypothesize-and-test strategy, that means interpreting the initially given data to data abstractions, generating and evaluating hypotheses and selecting additional tests to valuate the hypotheses. The student's actions...
In medicine large amounts of natural language documents have to be processed. Medical language is an interesting domain for the application of techniques developed in computational linguistics. Moreover, large scale applications of medical language processing raise the need to study the process of language engineering, which emphasizes some different problems than basic research. The texts found in...
Set the date range to filter the displayed results. You can set a starting date, ending date or both. You can enter the dates manually or choose them from the calendar.