Enric Plaza | CSIC (Consejo Superior de Investigaciones Científicas-Spanish National Research Council) (original) (raw)

Papers by Enric Plaza

A CBR system needs a good case retention strategy to decide which cases to incorporate into the c... more A CBR system needs a good case retention strategy to decide which cases to incorporate into the case base in order to maximize the performance of the system. In this work we present a collaborative case retention strategy, designed for multiagent CBR systems, called the Collaborative Case Bargaining strategy. The CCB strategy is a bargaining mechanism in which each CBR agent tries to maximize the utility of the cases it retains. We will present a case utility measure called the Justification-based Case Utility (JCU) based upon the ability of the individual CBR agents to provide justifications of their own results. An empirical evaluation of the CCB strategy shows the benefits for CBR agents to use this strategy: individual and collective accuracy are increased while the size of the case bases is decreased.

ECCBR 2004 Workshop Proceedings, Jan 1, 2004

Maintaining compact and competent case bases has become a main topic of Case Based Reasoning (CBR... more Maintaining compact and competent case bases has become a main topic of Case Based Reasoning (CBR) research. The main goal is to obtain a compact case base (with a reduced number of cases) without losing accuracy. In this work we present JUST, a technique to reduce the size of a case base while maintaining the classification accuracy of the CBR system. JUST uses justifications in order to select a subset of cases from the original case base that will form the new reduced case base. A justification is an explanation that the CBR system generates to justify the solution found for a given problem. Moreover, we present empirical evaluation in various data sets showing that JUST is an effective case base reduction technique that maintains the classification accuracy of the case base.

Proc. Workshop on Learning Agents …, Jan 1, 2001

Ecological Entomology, 1999

Abstract. Problem-solving methods provide reusable architectures and components for implementing ... more Abstract. Problem-solving methods provide reusable architectures and components for implementing the reasoning part of knowledge-based systems. The Unified Problem-solving Method description Language UPML has been developed to describe and implement such architectures and components to facilitate their semiautomatic reuse and adaptation. Spoken in a nutshell, UPML is a framework for developing knowledge-intensive reasoning systems based on libraries of generic problem-solving

WWW Proceedings of the 1st Working IFIP Conference on Software Architectures (WICSA1), Feb 22, 1999

Abstract. Problem-solving methods provide reusable architectures and components for implementing ... more Abstract. Problem-solving methods provide reusable architectures and components for implementing the reasoning part of knowledge-based systems. The Unified Problem-solving Method description Language UPML has been developed to describe such architectures and components to facilitate their semiautomatic reuse and adaptation. This paper sketches the components and connectors provided by UPML.

Case-Based Reasoning (CBR) can give agents the capability of learning from their own experience a... more Case-Based Reasoning (CBR) can give agents the capability of learning from their own experience and solve new problems, however, in a multi-agent system, the ability of agents to collaborate is also crucial. In this paper we present an argumentation framework (AMAL) designed to provide learning agents with collaborative problem solving (joint deliberation) and information sharing capabilities (learning from communication). We will introduce the idea of CBR multi-agent systems ( mathcalMnormalfonttextsfAC\mathcal{M}{\normalfont \textsf{AC}}mathcalMnormalfonttextsfAC systems), outline our argumentation framework and provide several examples of new tasks that agents in a mathcalMnormalfonttextsfAC\mathcal{M}\normalfont \textsf{AC}mathcalMnormalfonttextsfAC system can undertake thanks to the argumentation processes.

ABSTRACT Mulliageril systems offer a new paradigm I, о organize AT applications. Our goal is to d... more ABSTRACT Mulliageril systems offer a new paradigm I, о organize AT applications. Our goal is to develop techniques to integrate CRR into applications that are developed as multiagent systems. CBR offers the multiagent systems paradigm the capability of autonomously learning from experience. In filis paper&amp;amp;amp;quot; we present a framework for&amp;amp;amp;quot; collaboration among agents that use CBR and some experiments illustrating the framework. We focus on three collaboration policies for CRR agents: Peer Counsel, Bounded Counsel and Committee ...

Most CBR systems try to solve problems in one shot neglecting the sequential behavior of most rea... more Most CBR systems try to solve problems in one shot neglecting the sequential behavior of most real world domains and the simultaneous occurrence of interleaved problems proper to multi-agent settings. This article provides a first answer to the following question: how can the CBR paradigm be enriched to support the analysis of unsegmented sequences of observational data stemming from multiple coincidental sources? We propose Ceaseless CBR, a new model that considers the CBR task as on-going rather than one-shot and aims at finding the best explanation of an unsegmented sequence of alerts with the purpose of pinpointing whether undesired situations have occurred or not and, if so, indicating the multiple responsible sources or at least which ones are the most plausible.

Predictive toxicology is concerned with the task of building models capable of determining, with ... more Predictive toxicology is concerned with the task of building models capable of determining, with a certain degree of accuracy, the toxicity of chemical compounds. We discuss several machine learning methods that have been applied to build predictive toxicology models. In particular, we present two lazy learning lazy learning techniques applied to the task of predictive toxicology. While most ML techniques use structure relationship models to represent chemical compounds, we introduce a new approach based on the chemical nomenclature to represent chemical compounds. In our experiments we show that both models, SAR and ontology-based, have comparable results for the predictive toxicology task.

Abstract Social processes and agent interaction always take place in a speci c context, and there... more Abstract Social processes and agent interaction always take place in a speci c context, and there is a school of thought in social studies that analyzes them in the framework of institutions 2]. We will present the notions ofagent-mediated institution andstruc-tured ...