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.
A computational approach to the simulation of cognitive models of conceptual change in children learning elementary physics is presented. The student's mental model is inferred from a sequence of interviews collected along a period of eleven teaching sessions. Goal of the simulation is to support the cognitive scientist's investigation of learning in humans. The hypothesized cognitive models are based...
This paper presents an experimental analysis of a method recently proposed for refining knowledge bases expressed in a first order logic language. The method consists in transforming a classification theory into a neural network, called First Order logic Neural Network (FONN), by replacing predicate semantic functions and logical connectives with continuous-valued derivable functions. In this way...
Handling numerical information is one of the most important research issues for practical applications of first-order learning systems. This paper is concerned with the problem of inducing first-order classification rules from both numeric and symbolic data. We propose a specialization operator that discretizes continuous data during the learning process. The heuristic function used to choose among...
Recurrent neural networks are powerful learning machines capable of processing sequences. A recent extension of these machines can conveniently be used to process also general data structures like trees and graphs, which opens the doors to a number of new very interesting applications previously unexplored. In this paper, we show that when the problem of learning is restricted to purely symbolic...
In this paper, we describe a plan-based model for the treatment of misunderstandings in NL cooperative dialogue: an utterance is considered coherent if it is related to some of the interactants' intentions by a relation of goal adherence, goal adoption or plan continuation. If none of them is fully satisfied, a misunderstanding is hypothesized and the goal of realigning the interactants' interpretation...
This paper proposes a hierarchical organization of the linguistic knowledge, that views grammar as an abstraction of item-dependent information (in particular, an abstraction of subcategorization frames into a hierarchy of classes). The formalism has been successfully applied to a classification of 105 Italian verbal frames, developed by analysing a corpus of about 500,000 words. The proposed...
In this paper we present an architecture for choosing a flexible response in a natural language system involved in information-seeking tasks. Our work considers the crucial issue of choosing what information to provide and how to structure it, considering from the generation perspective a model of dialogue that was previously developed to study the recognition activity of an agent. In such a model...
The Earley algorithm is a widely used parsing method in natural language processing applications. We introduce a variant of Earley parsing that is based on a “delayed” recognition of constituents. This allows us to start the recognition of a constituent only in cases in which all of its subconstituents have been found within the input string. This is particularly advantageous in several cases in which...
The paper discusses the architecture of a Reactive Agent, capable of carrying out autonomous navigation. The agent extends the artificial potential field approach, used for trajectory formation, to environment exploration and symbolic feature detection. The agent's capabilities range from obstacle avoidance to maze navigation, carried out autonomously or under the supervision of higher cognitive levels.
A new hybrid approach for autonomous agents is described. The approach integrates in a principled way the functional and the behavioral approaches of agent design. The integration is based on the introduction of a conceptual space representation that links the subsymbolic level, which is a repository of reactive modules, with the symbolic level, in which rich symbolic descriptions of the agent environment...
in this paper, we describe a 3-D facets construction system which relies on the simultanueous analysis of regions and contours of stereo-images in order to reconstruct 3-D indoor scenes. Firstly, a Split-And-Merge algorithm cooperating with an edge extractor is used to compute an initial region segmentation which is improved by using a rule-based system. Then, the system matches relevant regions in...
In this paper we face the problem of finding characteristic information about. images of different objects, showing that the fractal encoding based on Iterated Function Systems, besides allowing very high compression rates, can be successfully applied also for capturing discriminatory features that can be exploited for non-fractalimage classification. An original feature extraction algorithm was developed...
In this paper, an alternative approach to the induction of relational concepts is presented. The underlying framework relies on the concept of exception, an exception being a counterexample left within the scope of a description devoted to classifying examples of the given target concept. While trying to characterize the target concept, first an initial description is searched for. Such a solution...
An implication rule Q→R is roughly a statement of the form “form all objects in the database, if an object has Q then it has also R”.We introduce a definition of minimal cover for the set of implication rules that hold in a relation, by analogy with earlier work on functional dependencies, and present an approach to computing it. The core of the proposed approach is an algorithm for inferring a reduced...
The behavior of verbs in sublanguages is highly specific and does not follow general principles of lexical decomposition. NLP applications require specific lexicons for tasks like surface parsing and shallow semantic interpretation. The reduced set of verbal senses specific to a given domain is more appropriate for efficient processing in real world tasks (e.g. information extraction and retrieval)...
In this paper a learning system is presented that is able to learn both the syntax (from an over-generalized grammar) and semantic rules (containing threshold values and relations) of an ECG grammar. These rules are used to direct the classification of QRS complexes and to distinquish between QRS and non-QRS patterns. The system demonstrates how a theory revision method can be used to refine large...
We propose an approach for the integration of abduction and induction in Logic Programming. In particular, we show how it is possible to learn an abductive logic program starting from an abductive background knowledge and a set of examples. By integrating Inductive Logic Programming with Abductive Logic Programming we can learn in presence of incomplete knowledge. Incomplete knowledge is handled by...
In this paper, we consider the following form of temporal abduction: given a domain theory where each explanatory formula is augmented with a set of temporal constraints on the atoms occurring in the formula, and given a set of observed atoms, with associated temporal constraints, the goal is the generation of a temporally consistent abductive explanation of the observations. Temporal abduction is...
Although Knowledge Representation is full of reasoning problems that have been formally proved to be both NP-hard and coNP-hard, the experimental analysis has largely focused on problems belonging to either NP or coNP. We still do not know, for example, whether well studied phenomena such as “phase transition”, which show up for many NP-complete problems (e.g., sat) happen for Σp ...
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.