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.
Multiagent settings are usually characterized by numerous goals, diverse opinions and conflicts of interest. In order to reach understanding and achieve cooperation, agents need a means of expressing their individual arguments which may contain explanations, justifications or any other kind of information. Furthermore, existing information may usually be incomplete, inconsistent and expressed in qualitative...
The field of nonmonotonic logic, sixteen years old now, is devoted to solve the problem of reasoning under incomplete knowledge, whose good understanding is essential to the construction of AI as a science and whose relevance reaches far beyond AI applications. During these years, many insights have been accumulated in the form of desirable properties the proposed formalisms should exhibit and of...
This paper develops two conditional logics that allows one to reason about the belief of another agent (which reasons in conditional logic himself). The desirable properties of such logics that reason about the beliefs of another non-monotonic agent are discussed and the two logics proposed here are shown to have those properties.
There has been great interest in the formalization and development of systems that treat plan recognition process in Artificial Intelligence. In this paper, we develop a theoretical framework, and implement a plan recognition system. Therefore, our research work can be divided in two parts: a hybrid theory, and a feasible implementation. In the theoretical part, we develop a Hybrid Formal Theory —...
We formally extend various works dealing with Heuristic Search in state graphs, focusing on 1) the notion of path length, 2) the characteristics of the state graphs, 3) the procedures that control the choices of the states to be expanded, 4) the rules that govern the update operations, 5) the properties of the evaluation functions. We present new general theorems concerning the termination at a goal...
We present a methodology that enables the use of existent classification inductive learning systems on problems of regression. We achieve this goal by transforming regression problems into classification problems. This is done by transforming the range of continuous goal variable values into a set of intervals that will be used as discrete classes. We provide several methods for discretizing the goal...
The paper presents an approach to building multi-agent systems. We are interested in complex agents able to reason about their tasks, and to display a proactive behavior, when installed on a network of heterogeneous computers. We developed the concept of a generic agent (GAg), equipped with the basic communication and ”mental” structure, but ignorant, i.e., not containing any application expertise,...
Approaches which tackle coordination in multi-agent systems have mostly taken communication for granted. In societies of individually-motivated agents where the communication costs are prohibitive, there should be other mechanisms to allow them to coordinate when interacting. In this paper game theory is used as a mathematical tool for modelling the interactions among agents and as a mechanism for...
An adaptive system called GBHCPR (Genetic Based Hierarchical Censored Production Rule) system based on Hierarchical Censored Production Rule (HCPR) system is presented that relies on development of some ties between Genetic Based Machine Learning (GBML) and symbolic machine learning. Several genetic operators are suggested that include advanced genetic operators, namely, Fusion and Fission. An appropriate...
In order to apply genetic algorithms (GAs) successfully to a given problem one has to find a good representation for potential solutions to that problem. Roughly speaking a good representation is one where building blocks for the problem's solution are relatively insensitive to crossover disruption, i.e., the building blocks have short defining lengths. However, such a representation is difficult...
This paper is concerned with the problem of resource-constrained project scheduling under multiple performing modes. A heuristic solution based on Genetic Algorithms is proposed. Furthermore, the resource leveling concept is incorporated into the procedure as a form of guidance in the search of a solution that yields not only the shortest duration, but a leveled solution as well. A comparison with...
This paper presents the first quantitative, corpus-based evaluation of the same-domain robustness and scalability of a new revision-based language generation model, in comparison to the traditional single pass pipeline model. Robustness is defined as the proportion of sentences, in a given corpus test sample that can be generated using only knowledge structures abstracted from another sample. Scalability...
This paper describes the pronominal anaphora resolution module of SQUIRREL system, a portable natural language front end to databases originally conceived to treat single sentences. The incorporation of the Discourse Module aimed to provide continuous consultations, however preserving the system's overall portability. Candidate antecedents, selected from the user's queries and database answers...
This paper describes an approach to gist preservation during automatic summarization whereby the source is a complex information structure which must be “pruned” and organized in such a way as to make it appropriate for textual expression. Based on a discourse model, we propose a process whereby gist is guaranteed at the deep level according to communicative and rhetorical settings. The main function...
This paper describes how the early visual process of contour organisation can be realised using the EM algorithm of Dempster, Laird and Rubin [2]. The underlying computational representation is based on Zucker's idea of fine spline coverings [17]. According to our EM approach the adjustment of spline parameters draws on an iterative weighted least-squares fitting process. The expectation step of our...
We consider the problem of finding a known two- dimensional object in an image, or verifying that it does not appear in the image. We adopt the strategy of doing a fast scan for potential places in the image where the object could be; we call this scan pose proposal. Each pose hypothesis is a set of edges that correspond to a subset of the transformed object bouńdary. Our algorithm works by finding...
TARCA is an Integrated Computer System which includes an Expert System for diagnosis, prevention, therapy planning, as well as monitoring, of cardiac arrhythmias. TARCA is under development at Oporto University in cooperation with S. João Hospital. TARCA aims towards an autonomous medical aid system for the treatment of cardiac patients from the automatic extraction of the electrocardiogram...
This article discusses the development of an Intelligent Distributed Environmental Decision Support System, built upon the association of a Multi-agent Belief Revision System with a Geographical Information System (GIS). The inherent multidisciplinary features of the involved expertises in the field of environmental management, the need to define clear policies that allow the synthesis of divergent...
The problem of combining dichotomic belief functions over a hierarchical structure of propositions can be viewed as a problem of updating local data in objects and exchanging messages among objects. Such approach is proposed in this paper. A set of propositions is given in a hierarchical structure so that each node represents a proposition. Nodes in the same level represent disjunctive propositions...
The comparatively small size and clean declarative semantics of logic programs hide performance issues that must be accounted for if the programs are to be executed using the currently available technology. Proposals have been made to tackle these computational issues by analysing a program and trying to optimise its code. We introduce a bottom-up method for analysing logic programs. The method incrementally...
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.