Time filter

Source Type

Armengol E.,IIIA Artificial Intelligence Research Institute | Torra V.,University of Skövde
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) | Year: 2016

Generalization and Suppression are two of the most used techniques to achieve k-anonymity. However, the generalization concept is also used in machine learning to obtain domain models useful for the classification task, and the suppression is the way to achieve such generalization. In this paper we want to address the anonymization of data preserving the classification task. What we propose is to use machine learning methods to obtain partial domain theories formed by partial descriptions of classes. Differently than in machine learning, we impose that such descriptions be as specific as possible, i.e., formed by the maximum number of attributes. This is achieved by suppressing some values of some records. In our method, we suppress only a particular value of an attribute in only a subset of records, that is, we use local suppression. This avoids one of the problems of global suppression that is the loss of more information than necessary. © Springer International Publishing Switzerland 2016.


Pinyol I.,Technology Center | Pinyol I.,PlastiaSite S.A. | Sabater-Mir J.,IIIA Artificial Intelligence Research Institute
Artificial Intelligence Review | Year: 2013

In open environments, agents depend on reputation and trust mechanisms to evaluate the behavior of potential partners. The scientific research in this field has considerably increased, and in fact, reputation and trust mechanisms have been already considered a key elements in the design of multi-agent systems. In this paper we provide a survey that, far from being exhaustive, intends to show the most representative models that currently exist in the literature. For this enterprise we consider several dimensions of analysis that appeared in three existing surveys, and provide new dimensions that can be complementary to the existing ones and that have not been treated directly. Moreover, besides showing the original classification that each one of the surveys provide, we also classify models that where not taken into account by the original surveys. The paper illustrates the proliferation in the past few years of models that follow a more cognitive approach, in which trust and reputation representation as mental attitudes is as important as the final values of trust and reputation. Furthermore, we provide an objective definition of trust, based on Castelfranchi's idea that trust implies a decision to rely on someone. © 2011 Springer Science+Business Media B.V.


Ontanon S.,Drexel University | Plaza E.,IIIA Artificial Intelligence Research Institute
Autonomous Agents and Multi-Agent Systems | Year: 2015

This paper focuses on coordinated inductive learning, concerning how agents with inductive learning capabilities can coordinate their learnt hypotheses with other agents. Coordination in this context means that the hypothesis learnt by one agent is consistent with the data known to the other agents. In order to address this problem, we present A-MAIL, an argumentation approach for agents to argue about hypotheses learnt by induction. A-MAIL integrates, in a single framework, the capabilities of learning from experience, communication, hypothesis revision and argumentation. Therefore, the A-MAIL approach is one step further in achieving autonomous agents with learning capabilities which can use, communicate and reason about the knowledge they learn from examples. © 2014, The Author(s).


Flaminio T.,IIIA Artificial Intelligence Research Institute | Lacasa L.G.,IIIA Artificial Intelligence Research Institute
Advances in Intelligent Systems and Computing | Year: 2013

In this paper we address the issue of providing a geometrical characterization for the decision problem of asking whether a partial assignment β:fi → αi mapping fuzzy events f i into real numbers αi (i = 1, ...,n) extends to a generalized belief function on fuzzy sets, according to a suitable definition. We will characterize this problem in a way that allows to treat it as the membership problem of a point to a specific convex set. © 2013 Springer-Verlag.


Abril D.,IIIA Artificial Intelligence Research Institute | Navarro-Arribas G.,IIIA Artificial Intelligence Research Institute | Navarro-Arribas G.,Autonomous University of Barcelona | Torra V.,IIIA Artificial Intelligence Research Institute
Information Fusion | Year: 2012

In data privacy, record linkage can be used as an estimator of the disclosure risk of protected data. To model the worst case scenario one normally attempts to link records from the original data to the protected data. In this paper we introduce a parametrization of record linkage in terms of a weighted mean and its weights, and provide a supervised learning method to determine the optimum weights for the linkage process. That is, the parameters yielding a maximal record linkage between the protected and original data. We compare our method to standard record linkage with data from several protection methods widely used in statistical disclosure control, and study the results taking into account the performance in the linkage process, and its computational effort. © 2011 Elsevier B.V. All rights reserved.


Pinyol I.,IIIA Artificial Intelligence Research Institute | Sabater-Mir J.,IIIA Artificial Intelligence Research Institute | Dellunde P.,IIIA Artificial Intelligence Research Institute | Dellunde P.,Autonomous University of Barcelona | Paolucci M.,National Research Council Italy
Autonomous Agents and Multi-Agent Systems | Year: 2012

Computational trust and reputation models have been recognized as one of the key technologies required to design and implement agent systems. These models manage and aggregate the information needed by agents to efficiently perform partner selection in uncertain situations. For simple applications, a game theoretical approach similar to that used in most models can suffice. However, if we want to undertake problems found in socially complex virtual societies, we need more sophisticated trust and reputation systems. In this context, reputation-based decisions that agents make take on special relevance and can be as important as the reputation model itself. In this paper, we propose a possible integration of a cognitive reputation model, Repage, into a cognitive BDI agent. First, we specify a belief logic capable to capture the semantics of Repage information, which encodes probabilities. This logic is defined by means of a two first-order languages hierarchy, allowing the specification of axioms as first-order theories. The belief logic integrates the information coming from Repage in terms if image and reputation, and combines them, defining a typology of agents depending of such combination. We use this logic to build a complete graded BDI model specified as a multi-context system where beliefs, desires, intentions and plans interact among each other to perform a BDI reasoning. We conclude the paper with an example and a related work section that compares our approach with current state-of-the-art models. © 2010 The Author(s).


Flaminio T.,IIIA Artificial Intelligence Research Institute | Godo L.,IIIA Artificial Intelligence Research Institute | Marchioni E.,IIIA Artificial Intelligence Research Institute
Soft Computing | Year: 2012

In this paper, we study generalized possibility and necessity measures on MV-algebras of [0, 1]-valued functions (MV-clans) in the framework of idempotent mathematics, where the usual field of reals ℝ is replaced by the max-plus semiring ℝ max We prove results about extendability of partial assessments to possibility and necessity measures, and characterize the geometrical properties of the space of homogeneous possibility measures. The aim of the present paper is also to support the idea that idempotent mathematics is the natural framework to develop the theory of possibility and necessity measures, in the same way classical mathematics serves as a natural setting for probability theory. © 2012 Springer-Verlag.


Armengol E.,IIIA Artificial Intelligence Research Institute | Torra V.,University of Skövde
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) | Year: 2015

Microaggregation is an anonymization technique consisting on partitioning the data into clusters no smaller than k elements and then replacing the whole cluster by its prototypical representant. Most of microaggregation techniques work on numerical attributes. However, many data sets are described by heterogeneous types of data, i.e., numerical and categorical attributes. In this paper we propose a new microaggregation method for achieving a compliant k-anonymous masked file for categorical microdata based on generalization. The goal is to build a generalized description satisfied by at least k domain objects and to replace these domain objects by the description. The way to construct that generalization is similar that the one used in growing decision trees. Records that cannot be generalized satisfactorily are discarded, therefore some information is lost. In the experiments we performed we prove that the new approach gives good results. © Springer International Publishing Switzerland 2015.


Armengol E.,IIIA Artificial Intelligence Research Institute | Garcia-Cerdana A.,IIIA Artificial Intelligence Research Institute
Communications in Computer and Information Science | Year: 2010

In this paper we introduce an extension of the lazy learning method called Lazy Induction of Descriptions (LID). This new version is able to deal with fuzzy cases, i.e., cases described by attributes taking continuous values represented as fuzzy sets. LID classifies new cases based on the relevance of the attributes describing them. This relevance is assessed using a distance measure that compares the correct partition (i.e., the correct classification of cases) with the partitions induced by each one of the attributes. The fuzzy version of LID introduced in this paper uses two fuzzy versions of the Rand index to compare fuzzy partitions: one proposed by Campello and another proposed by Hüllermeier and Rifqi. We experimented with both indexes on data sets from the UCI machine learning repository. © Springer-Verlag Berlin Heidelberg 2010.


Armengol E.,IIIA Artificial Intelligence Research Institute
Artificial Intelligence in Medicine | Year: 2011

Objective: Early diagnosis of melanoma is based on the ABCD rule which considers asymmetry, border irregularity, color variegation, and a diameter larger than 5. mm as the characteristic features of melanomas. When a skin lesion presents these features it is excised as prevention. Using a non-invasive technique called dermoscopy, dermatologists can give a more accurate evaluation of skin lesions, and can therefore avoid the excision of lesions that are benign. However, dermatologists need to achieve a good dermatoscopic classification of lesions prior to extraction. In this paper we propose a procedure called LazyCL to support dermatologists in assessing the classification of skin lesions. Our goal is to use LazyCL for generating a domain theory to classify melanomas in situ. Methods: To generate a domain theory, the LazyCL procedure uses a combination of two artificial intelligence techniques: case-based reasoning and clustering. First LazyCL randomly creates clusters and then uses a lazy learning method called lazy induction of descriptions (LID) with leave-one-out on them. By means of LID, LazyCL collects explanations of why the cases in the database should belong to a class. Then the analysis of relationships among explanations produces an understandable clustering of the dataset. After a process of elimination of redundancies and merging of clusters, the set of explanations is reduced to a subset of it describing classes that are " almost" discriminant. The remaining explanations form a preliminary domain theory that is the basis on which experts can perform knowledge discovery. Results: We performed two kinds of experiments. First ones consisted on using LazyCL on a database containing the description of 76 melanomas. The domain theory obtained from these experiments was compared on previous experiments performed using a different clustering method called self-organizing maps (SOM).Results of both methods, LazyCL and SOM, were similar. The second kind of experiments consisted on using LazyCL on well known domains coming from the machine learning repository of the Irvine University. Thus, since these domains have known solution classes, we can prove that the clusters build by LazyCL are correct. Conclusions: We can conclude that LazyCL that uses explained case-based reasoning for knowledge discovery is feasible for constructing a domain theory. On one hand, experiments on the melanoma database show that the domain theory build by LazyCL is easy to understand. Explanations provided by LID are easily understood by domain experts since these descriptions involve the same attributes than they used to represent domain objects. On the other hand, experiments on standard machine learning data sets show that LazyCL is a good method of clustering since all clusters produced are correct. © 2010 Elsevier B.V.

Loading IIIA Artificial Intelligence Research Institute collaborators
Loading IIIA Artificial Intelligence Research Institute collaborators