Showing papers on "Abductive reasoning published in 1991"

On the Relationship Between Abduction and Deduction

Abstract: n a The aim of this paper is at analyzing from various points of view the relationships betwee bduction and deduction. In particular, we consider a meta-level definition of abduction in l d terms of deduction, similar to various definitions proposed in the literature, and an object-leve efinition in which abductive conclusions are expressed as a logical consequence of the obser. T vations and of a simple transformation of the domain theory based on predicate completion he equivalence between the two definitions is proved for domain theories of considerable s expressive power. The object-level characterization we propose uses very simple forms of rea oning and the equivalence result allows us to make explicit some of the assumptions underly-

A Prolog-like Inference System for Computing Minimum-Cost Abductive Explanations in Natural-Language Interpretation

Abstract: By determining what added assumptions would suffice to make the logical form of a sentence in natural language provable, abductive inference can be used in the interpretation of sentences to determine what information should be added to the listener's knowledge, i.e., what he should learn from the sentence. This is a comparatively new application of mechanized abduction. A new form of abduction — least specific abduction — is proposed as being more appropriate to the task of interpreting natural language than the forms that have been used in the traditional diagnostic and design-synthesis applications of abduction. The assignment of numerical costs to axioms and assumable literals permits specification of preferences on different abductive explanations. A new Prolog-like inference system that computes abductive explanations and their costs is given. To facilitate the computation of minimum-cost explanations, the inference system, unlike others such as Prolog, is designed to avoid the repeated use of the same instance of an axiom or assumption.

Extending abduction from propositional to first-order logic

Abstract: Abduction is often considered as inference to the best explanation. In this paper, we appeal to abduction as a way to generate all but only explanations that have “some reasonable prospect” of being valid.

Compiling a default reasoning system into Prolog

Abstract: Artificial intelligence researchers have been designing representation systems for default and abductive reasoning. Logic Programming researchers have been working on techniques to improve the efficiency of Horn clause deduction systems This paper describes how one such default and abductive reasoning system (namelyTheorist) can be translated into Horn clauses (with negation as failure), so that we can use the clarity of abductive reasoning systems and the efficiency of Horn clause deduction systems. We thus show how advances in expressive power that artificial intelligence workers are working on can directly utilise advances in efficiency that logic programming researchers are working on. Actual code from a running system is given.

Consequence-finding based on ordered linear resolution

Abstract: Since linear resolution with clause ordering is incomplete for consequence-finding, it has been used mainly for proof-finding. In this paper, we re-evaluate consequence-finding. Firstly, consequence-finding is generalized to the problem in which only interesting clauses having a certain property (called characteristic clauses) should be found. Then, we show how adding a skip rule to ordered linear resolution makes it complete for consequence-finding in this general sense. Compared with set-of-support resolution, the proposed method generates fewer clauses to find such a subset of consequences. In the propositional case, this is an elegant tool for computing the prime implicants/implicates. The importance of the results lies in their applicability to a wide class of AI problems including procedures for nonmonotonic and abductive reasoning and truth maintenance systems.

An inference rule for hypothesis generation

Abstract: There are many new application fields for automated deduction where we have to apply abductive reasoning. In these applications we have to generate consequences of a given theory having some appropriate properties. In particular we consider the case where we have to generate the clauses containing instances of a given literal L. The negation of the other literals in such clauses are hypothesis allowing to derive L. In this paper we present an inference rule, called L-inference, which was designed in order to derive those clauses, and a L-strategy. The L-inference rule is a sort of Input Hyper-resolution. The main result of the paper is the proof of the soundness and completeness of the L-inference rule. The L-strategy associated to the L-inference rule, is a saturation by level with deletion of the tautologies and of the subsumed clauses. We show that the L-strategy is also complete.

Abductive and approximate reasoning models for characterizing inference channels

Abstract: A serious problem in computer database and knowledge base security is detecting and eliminating so-called inference channels. The existence of such channels enables a user with access to information classified at a low level to infer information classified at a high level, and through the transformation of low level data to high level data may provide an unacceptable information flow. In order to estimate the presence of inference channels, determine the degree of risk which they present, and find ways to eliminate them, one needs a formal model to describe them. The authors introduce abductive reasoning. Abduction provides both the basis for a formal model for the inference problem and a computational mechanism for detecting inference channels. Abduction additionally provides a framework for reasoning with approximate and uncertain information, which enables them to extend the model for inference channels by taking into account the likelihood that a person might believe some statement of interest. >

Grammar-related transformations of logic programs

Abstract: We investigate a simple transformation of logic programs capable of inverting the order of computation. Several examples are given which illustrate how this transformation may serve such purposes as left-recursion elimination, loop-elimination, simulation of forward reasoning, isotopic modification of programs and simulation of abductive reasoning.

SRI International: description of the TACITUS system as used for MUC-3

Abstract: TACITUS is a system for interpreting natural language texts that has been under development since 1985. It has a preprocessor and postprocessor currently tailored to the MUC-3 application. It performs a syntactic analysis of the sentences in the text, using a fairly complete grammar of English, producing a logical form in first-order predicate calculus. Pragmatics problems are solved by abductive inference in a pragmatics, or interpretation, component.

The Elements of Reasoning

Abstract: Preface 1Reasoning: Are You For It or Against It? The Powers of Reasoning Elemental Questions Pluto and Plato Fine Language and Geometry Ethos, Logos, and Pathos The End of Reasoning Internal and External Reasoning: An Example Rhetorical Inventions Beyond "For or Against" Reasoning Practice 2Invention: Places, Paths, and Structures of Reasoning An Introduction to the Specific Elements Places of Reasoning: Topoi Paths of Reasoning: The Stases Structures of Reasoning From Invention to Judgment Stases and Time Reasoning Practice 3Conjectures: Places to Begin The Primary Stasis A Trove of Conjectural Claims How to Spot a Conjectural Claim Three Types of Conjectural Claims Reasoning Practice 4Definitions: They Can Change Everything Rhetoric and Definitions Dictionary Definitions Neologisms Stipulative Definitions Specific Means of Defining Reasoning Practice 5Causes and Consequences: A Sense of How the World Works How Could This Happen? Reasoning from Effect to Cause Reasoning from Cause to Effect Antecedence-Subsequence Post-Hoc, Ergo Propter Hoc Chance as a Causal Factor Chance and Causality, Myth and Cosmology Some Guideline for Causal Reasoning Causality and the Ends of Reasoning Reasoning Practice 6Values: Judgments Grounded in Nature and Consequences Criteria Supporting Value Claims: Nature and Consequences One Example of Claims about Value: Music Another Example: Family Farms Weighting Criteria Guidelines for Reasoning about Values Reasoning Practice 7Procedures and Proposals: Actualizing the Potential for Change Ready? "Houston: We Have a Problem" A Modest Proposal Feasibility, Plausibility, Credibility Guidelines for Reasoning about Procedures and Proposals Reasoning Practice 8Becoming a Citizen Critic: Where Rhetoric Meets the Road Diversions of Reasoning Spectator Culture, Consumer Culture, Democratic Culture Reasoning to Invoke Citizen Critics What Is a Citizen? And a Citizen of What? The Enthymemes of This Book Reasoning Practice Index

Localized abductive planning for robot assembly

Abstract: The application of localized abductive reasoning in assembly planning with multiple robots is explained. The assembly problem is described in the event calculus using Horn clause logic. The assembly problem domain is decomposed into regions of activity, and the actions, predicates, and constraints of the planning problem are distributed over these regions. The reasoning component of the planner is abduction; the planner makes abduction steps locally within the regions, and the search algorithm shifts control from one region to another to find an overall global plan. An example is shown of a flashlight assembly problem in a robot cell with two robots that share one feeder. >

An efficient hypothetical reasoning system for predicate-logic knowledge-base

Abstract: The methods of fast hypothetical reasoning systems developed for propositional logic cannot be applicable in a straightforward manner to the predicate-logic case. A fast hypothetical reasoning mechanism that is effective for predicate-logic knowledge (actually for function-free predicate Horn-clause knowledge) is presented. A reasoning method developed in the deductive database area is effectively applied to this mechanism. >

A Science of Reasoning.

Abstract: How can we understand reasoning in general and mathematical proofs in particular? It is argued that a high-level understanding of proofs is needed to complement the low-level understanding provided by Logic A role for computation is proposed to provide this high-level understanding, namely by the association of proof plans with proofs Criteria are given for assessing the association of a proof plan with a proof

Abductive Reasoning in a Multi-Theory Framework

Abstract: A multi-theory framework is presented where both abduction and dynamic theory composition are modelled in a logic programming setting. In this framework the evolution of knowledge can be modelled by means of theory composition, while a form of hypothetical reasoning is supported by abduction. The semantics is expressed in terms of the standard semantics of logic programming (Herbrand models), by defining a compositional model-theory. A proof-theory is given in terms of inference rules, and soundness and completeness results are stated.

Probabilistic Reasoning With Facts And Rules In Deductive Databases

Abstract: In this paper we present a new method for probabilistic reasoning with true facts and uncertain rules within a deductive database. Besides a cautious approach to inferences on uncertain rules, we show a default approach for uncertainty reasoning including factual knowledge, based on the ideas of maximal context and detachment. Integrated into a database these approaches support many important applications with probabilistic value dependencies. One sample application will be provided: Lead qualification within a marketing database.

Mechanizing skeptical abduction and its applications to artificial intelligence

Abstract: Abduction is the process of generating the best explanation as to why a fact is observed given what is already known. A real problem in this area is the selective generation of hypotheses that have some reasonable prospect of being valid. The author proposes the notion of skeptical abduction as a model to face this problem. After providing a definition of abductive reasoning, skeptical abduction and specific abduction are compared in a logical framework. The mechanism of abductive reasoning in propositional logic is investigated and its generalization to first-order logic is discussed. >

On the relationship between parsimonious covering and Boolean minimization

Abstract: The authors explain some of the relationships of the Boolean minimization problem (BMP) to a formalization of abductive inference called parsimonious covering (PC). Abductive inference often occurs in diagnostic problems such as finding the causes of circuit faults or determining the disease causing the symptoms reported by a patient. Parsimonious covering involves covering all observed facts by means of a parsimonious set of explanations that can account for the observation. It is shown that only the prime implicants of a given Boolean function in a BMP, rather than any general product terms, are considered analogous to disorders in a PC problem. >

Abductive inference models for diagnostic problem-solving by Yun Peng and James A. Reggia, Springer-Verlag, Berlin, 1990, pp 268.

Abstract: Readers who have experience either of conventional systems development or of unsuccessful real-world KBS projects are likely to recognize the relevance of KADS and to welcome this book as the first readily available exposition of the methodology for analysis. We must hope that further volumes appear covering all the other stages and issues in KBS development. The publishers are to be congratulated on a well-produced and speedily published (November 1989) book, which has a good index and very few errors.

Retrieval in Case-Based Reasoning Using Preferred Subtheories

Abstract: The goal of this work is to develop a formal logical foundation of the representation and the retrieval of cases in case-based reasoning. An adequate basis therefor provides the default logic with priorities. We present transformations which construct defaults from the memory of cases such that the retrieval of knowledge in case-based reasoning corresponds roughly to the preferred subtheory obtained by the defaults.

Using Normal Deduction Graphs in Default Reasoning

Abstract: This paper presents a formalization of commonsense reasoning by using normal deduction graphs (NDGs), which form a powerful tool for deriving Horn and non-Horn clauses, based on Kleene's three-valued logic We show how NDGs, in conjunction with default logic, can be used to answer queries of commonsense reasoning by developing a formalization with results which are consistent with Etherington's ordered network theory Index terms: Artificial intelligence, commonsense reasoning, default reasoning, first-order logic, inference, logic programming, normal deduction graph

Investigating the Relationships between Abduction and Inverse Resolution in Proposition Calculus

Abstract: In systems devoted to diagnostic tasks, the term abduction usually denotes the process of explaining a set of observed manifestations, in the light of an existing domain theory. This amounts to find a set of hypotheses such that the manifestations logically follow from them and from the theory. Then, abduction corresponds to reasoning from consequences to possible premises. On the other hand, an analogous line of reasoning has been independently followed in constructive induction, where new predicates are introduced and theories are completed by trying to invert the resolution mechanism. In this paper, links are established between these two approaches by showing that the inverse resolution operators are sound abductive inference rules, in the case of propositional calculus. Some relationships with deduction in a closed world assumption are also investigated.

Abductive Reasoning in Three-valued Logic for Knowledge Bases

Abstract: In this paper we study the case of monotonic rule-based systems which use a forward chaining algorithm to make deductions and a query algorithm to compute queries to be asked to the user in order to be closer to the expected goal (as Guru ?], IS ?] or Nexpert ?]). Unfortunately in many systems the query algorithm is not consistent with the deduction algorithm. Queries asked by these systems are often useless. We show that computing a query for those systems is equivalent to making abductive reasoning in three-valued logic and we propose a formalization of useful queries for expert systems. We give two deenitions of intelligent queries and two levels to compute them: The logical level which aims at computing a list of intelligent queries, and the heuristic level which aims at choosing which query of that list it will ask the user. These are three examples: If b then a If b then a If b and c then a If c then b If c then a If d then a If d then b If d then b If e then b If e then not c If b then e not c If f then d If b then not e If we run a mixed chaining on the rst example with the goal a then almost all the commercial expert systems will ask the query \Is e true?". Unfortunately, e does not help the system to be closer to the expected goal because e gives not c and then nothing is deduced. In this case, there is just one useful query which is \Is f true?". By the same way we can see that the only useful query on the second example is c (d is useless since \yes" to this query leads to a contradiction) and on the third example only d is useful (e is useless).

Processing Abductive Reasoning via Contextual Logic Programming

Abstract: Different extensions to logic programming have recently been introduced to deal with abductive reasoning. In this work we address the issue of how to process abductive reasoning in the field of logic programming by following a compilation-based approach. In particular, we describe how an abductive reasoning system can be translated in Contextual Logic Programming, so we can use the efficiency of already existing compilebased implementation for contexts.

Syntax and semantics of a monotonic framework for non-monotonic reasoning

Abstract: We present the proof theory and the model theory of a montonic framework for default reasoning, and we extend logic programming techniques to this framework. Standard formalizations of default reasoning do not syntactically separate between hard knowledge and conjectural knowledge. Such a separation is fundamental in our framework. To illustrate our approach, we show how it solves the Yale Shooting problem.