Entity

Time filter

Source Type


Destercke S.,Montpellier SupAgro | Dubois D.,CNRS Toulouse Institute in Information Technology
Information Sciences | Year: 2011

When conjunctively merging two belief functions concerning a single variable but coming from different sources, Dempster rule of combination is justified only when information sources can be considered as independent. When dependencies between sources are ill-known, it is usual to require the property of idempotence for the merging of belief functions, as this property captures the possible redundancy of dependent sources. To study idempotent merging, different strategies can be followed. One strategy is to rely on idempotent rules used in either more general or more specific frameworks and to study, respectively, their particularization or extension to belief functions. In this paper, we study the feasibility of extending the idempotent fusion rule of possibility theory (the minimum) to belief functions. We first investigate how comparisons of information content, in the form of inclusion and least-commitment, can be exploited to relate idempotent merging in possibility theory to evidence theory. We reach the conclusion that unless we accept the idea that the result of the fusion process can be a family of belief functions, such an extension is not always possible. As handling such families seems impractical, we then turn our attention to a more quantitative criterion and consider those combinations that maximize the expected cardinality of the joint belief functions, among the least committed ones, taking advantage of the fact that the expected cardinality of a belief function only depends on its contour function. © 2011 Elsevier Inc. All rights reserved.


Cabanac G.,CNRS Toulouse Institute in Information Technology
Scientometrics | Year: 2011

Scientific literature recommender systems (SLRSs) provide papers to researchers according to their scientific interests. Systems rely on inter-researcher similarity measures that are usually computed according to publication contents (i.e., by extracting paper topics and citations). We highlight two major issues related to this design. The required full-text access and processing are expensive and hardly feasible. Moreover, clues about meetings, encounters, and informal exchanges between researchers (which are related to a social dimension) were not exploited to date. In order to tackle these issues, we propose an original SLRS based on a threefold contribution. First, we argue the case for defining inter-researcher similarity measures building on publicly available metadata. Second, we define topical and social measures that we combine together to issue socio-topical recommendations. Third, we conduct an evaluation with 71 volunteer researchers to check researchers' perception against socio-topical similarities. Experimental results show a significant 11.21% accuracy improvement of socio-topical recommendations compared to baseline topical recommendations. © 2011 Akadémiai Kiadó, Budapest, Hungary.


Demolombe R.,CNRS Toulouse Institute in Information Technology
Artificial Intelligence and Law | Year: 2011

The paper presents a logical framework for the representation of interactions between institutional agents, human agents and software agents. A case study is used to analyze how obligations on institutional agents are ''propagated'' to human and software agents, and how actions performed by these agents count as actions that satisfy the obligations imposed to institutional agents. It is shown that the relationship between the different kinds of obligations and actions can be represented in terms of the concept of ''count as'' proposed by Searle, of role and of causality. The logical framework focus on those three concepts. © 2011 Springer Science+Business Media B.V.


Cabanac G.,CNRS Toulouse Institute in Information Technology
Scientometrics | Year: 2013

Schubert introduced the partnership ability φ-index relying on a researcher's number of co-authors and collaboration rate. As a Hirsch-type index, φ was expected to be consistent with Schubert-Glänzel's model of h-index. Schubert demonstrated this relationship with the 34 awardees of the Hevesy medal in the field of nuclear and radiochemistry (r2=0.8484). In this paper, we upscale this study by testing the φ-index on a million researchers in computer science. We found that the Schubert-Glänzel's model correlates with the million empirical φ values (r2=0.8695). In addition, machine learning through symbolic regression produces models whose accuracy does not exceed a 6.1 % gain (r2=0.9227). These results suggest that the Schubert-Glänzel's model of φ-index is accurate and robust on the domain-wide bibliographic dataset of computer science. © 2012 Akadémiai Kiadó, Budapest, Hungary.


Lorini E.,CNRS Toulouse Institute in Information Technology
Journal of Logic, Language and Information | Year: 2010

We continue the work initiated in Herzig and Lorini (J Logic Lang Inform, in press) whose aim is to provide a minimalistic logical framework combining the expressiveness of dynamic logic in which actions are first-class citizens in the object language, with the expressiveness of logics of agency such as STIT and logics of group capabilities such as CL and ATL. We present a logic called DDLA (Deterministic Dynamic logic of Agency) which supports reasoning about actions and joint actions of agents and coalitions, and agentive and coalitional capabilities. In DDLA it is supposed that, once all agents have selected a joint action, the effect of this joint action is deterministic. In order to assess DDLA we prove that it embeds Coalition Logic. We then extend DDLA with modal operators for agents' preferences, and show that the resulting logic is sufficiently expressive to capture the game-theoretic concepts of best response and Nash equilibrium. © Springer Science+Business Media B.V. 2009.

Discover hidden collaborations