56 research outputs found
SET UP OF A DCD PIG MODEL OF NORMOTHERMIC MACHINE PERFUSION ASSOCIATED TO A MESENCHYMAL STROMAL CELL COATED BIOREACTOR TO IMPROVE ORGAN PRESERVATION
No full textSET UP OF A DCD PIG MODEL OF NORMOTHERMIC MACHINE PERFUSION ASSOCIATED TO A MESENCHYMAL STROMAL CELL COATED BIOREACTOR TO IMPROVE ORGAN PRESERVATION
No full textSymbolic knowledge extraction for explainable nutritional recommenders
No full textBackground and objective:This paper focuses on nutritional recommendation systems (RS), i.e. AI-powered automatic systems providing users with suggestions about what to eat to pursue their weight/body shape goals. A trade-off among (potentially) conflictual requirements must be taken into account when designing these kinds of systems, there including: (i) adherence to experts’ prescriptions, (ii) adherence to users’ tastes and preferences, (iii) explainability of the whole recommendation process. Accordingly, in this paper we propose a novel approach to the engineering of nutritional RS, combining machine learning and symbolic knowledge extraction to profile users—hence harmonising the aforementioned requirements. MethodsOur contribution focuses on the data processing workflow. Stemming from neural networks (NN) trained to predict user preferences, we use CART Breiman et al.(1984) to extract symbolic rules in Prolog Körner et al.(2022) form, and we combine them with expert prescriptions brought in similar form. We can then query the resulting symbolic knowledge base via logic solvers, to draw explainable recommendations. ResultsExperiments are performed involving a publicly available dataset of 45,723 recipes, plus 12 synthetic datasets about as many imaginary users, and 6 experts’ prescriptions. Fully-connected 4-layered NN are trained on those datasets, reaching ∼86% test-set accuracy, on average. Extracted rules, in turn, have ∼80% fidelity w.r.t. those NN. The resulting recommendation system has a test-set precision of ∼74%. The symbolic approach makes it possible to devise how the system draws recommendations. ConclusionsThanks to our approach, intelligent agents may learn users’ preferences from data, convert them into symbolic form, and extend them with experts’ goal-directed prescriptions. The resulting recommendations are then simultaneously acceptable for the end user and adequate under a nutritional perspective, while the whole process of recommendation generation is made explainable.Interactive Intelligenc
A Computational Approach to Poetic Structure, Rhythm and Rhyme
No full textIn this paper we present SPARSAR, a system for the automatic analysis of English and Italian poetry. The system can work on any type of poem and produces a set of parameters that are then used to compare poems with one another, of the same author or of different authors. In this paper, we will concentrate on the second module, which is a rule-based system to represent and analyze poetic devices. Evaluation of the system on the basis of a manually created dataset - including poets from Shakespeare's time down to T.S.Eliot and Sylvia Plath - has shown its high precision and accuracy approximating 90%
A proof-of-concept study in small and large animal models for coupling liver normothermic machine perfusion with mesenchymal stromal cell bioreactors
No full textTo fully harness mesenchymal-stromal-cells (MSCs)’ benefits during Normothermic Machine Perfusion (NMP), we developed an advanced NMP platform coupled with a MSC-bioreactor and investigated its bio-molecular effects and clinical feasibility using rat and porcine models. The study involved three work packages: 1) Development (n = 5): MSC-bioreactors were subjected to 4 h-liverless perfusion; 2) Rat model (n = 10): livers were perfused for 4 h on the MSC-bioreactor-circuit or with the standard platform; 3) Porcine model (n = 6): livers were perfused using a clinical device integrated with a MSC-bioreactor or in its standard setup. MSCs showed intact stem-core properties after liverless-NMP. Liver NMP induced specific, liver-tailored, changes in MSCs’ secretome. Rat livers exposed to bioreactor-based perfusion produced more bile, released less damage and pro-inflammatory biomarkers, and showed improved mithocondrial function than those subjected to standard NMP. MSC-bioreactor integration into a clinical device resulted in no machine failure and perfusion-related injury. This proof-of-concept study presents a novel MSC-based liver NMP platform that could reduce the deleterious effects of ischemia/reperfusion before transplantation
Desambiguación de topónimos en la recuperación de información
No full textTesis doctoral (con mención de doctorado europeo) en Informática realizada por Davide Buscaldi y dirigida por el doctor Paolo Rosso (Univ. Politécnica de Valencia). El acto de defensa de la tesis tuvo lugar en Valencia en Octubre de 2010 ante el tribunal formado por los doctores: Paul David Clough (University of Sheffield), Ross Purves (Universität Zürich), Emilio Sanchis Arnal (Univ. Politécnica de Valencia), Mark Sanderson (Royal Melbourne Institute of Technology), Diana Santos (Sintef-ICT, Oslo). La mención europea se obtuvo a través de una estancia en el FBK-IRST (Italia) bajo la dirección de Bernardo Magnini. La calificación obtenida fue de Sobresaliente Cum Laude.Ph.D. thesis (European doctorate mention) in Computer Science written by Davide Buscaldi under the supervision of Dr. Paolo Rosso (Univ. Politécnica de Valencia). The author was examined in Valencia in October 2010 by a panel composed by the following doctors: Paul David Clough (University of Sheffield), Ross Purves (Universität Zürich), Emilio Sanchis Arnal (Univ. Politécnica de Valencia), Mark Sanderson (Royal Melbourne Institute of Technology), Diana Santos (Sintef-ICT, Oslo). The European mentions was received after a 3 months stage at the FBK-IRST (Italy) under the guidance of Bernardo Magnini. The obtained grade was Cum Laude.Thesis supported by a FPI Grant of the Valencian government (ref. BFPI06/97)
Resolución de la ambigüedad semántica de las palabras mediante modelos de probabilidad de máxima entropía
No full textTesis doctoral en Informática realizada por Armando Suárez Cueto bajo
la dirección de los doctores Manuel Palomar Sanz, de la Universidad de Alicante, y
German Rigau Claramunt, de la Universidad del País Vasco. El acto de defensa de la
tesis tuvo lugar el 28 de junio de 2004 ante el tribunal formado por los doctores Lluís
Padró Cirera (Univ. Politécnica de Cataluña) , Andrés Montoyo Guijarro (Univ. de
Alicante), Eneko Agirre Bengoa (Univ. del País Vasco), Alfonso Ureña López (Univ.
de Jaén) y Bernardo Magnini (Istituto Trentino di Cultura) . La calificación obtenida
fue Sobresaliente Cum Laude por unanimidad.PhD Thesis in Computer Science written by Armando Suárez Cueto under
the supervision of Dr. Manuel Palomar Sanz, (Univ. of Alicante), and German
Rigau Claramunt (Univ. of Basque Country). The author was examined in June
28th, 2004 by the commitee formed by Dr. Lluís Padró Cirera (Politechnic University
of Cataluña), Andrés Montoyo Guijarro (Univ. of Alicante), Eneko Agirre
Bengoa (Univ. of Basque Country), Alfonso Ureña López (Univ. of Jaén) y Bernardo
Magnini (Istituto Trentino di Cultura). The grade obtained was Sobresaliente Cum
Laude
Archaeology and Object-Based Image Analysis: potential and issues for the application of semantic models
No full textInternational audienceIn recent years, the field of remote sensing experienced an incredible growth thanks to the increasing quality and variety of sensors and the reduction of instrumental costs. The benefits for archaeology were soon apparent. So far, data interpretation remains essentially a prerogative of the human operator and is mediated by his skills and experiences. However, the continuous increase of datasets volume and the increasing necessity to work on large scale projects require an overall revision of the methods traditionally used in the archaeological field. With the progressive diffusion of OBIA in archeology during the last few years, it is now essential to find a shared language and a common protocol of investigation (ideally passing from operational practice to operational routine), necessary to allow the comparability of data. This presentation will offer a general review on the topic considering both the results published in the literature and new case studies developed by the author and his research group, which will be used to illustrate the possibilities of the method in archeology. The case studies will comprise object-based procedures for the treatment of aerial/ satellite imaging, such as VHR multispectral data, and digital elevation models acquired by airborne and terrestrial laser scanning. These data will serve as a starting point to discuss the limits and the potential of OBIA in the archaeological field, with specific reference to the prospects for the future in light of the recent developments in other disciplines such as environmental and biomedical sciences. The objective is therefore to examine a series of crucial methodological issues linked to the residuality of the archaeological remains, also through the exemplification of practical results. In fact, the archaeological remains are always affected by post-depositional processes which have progressively altered, with different degrees of impact, their original characteristics, leading to a high variability within the same category of evidences. Diachronic semantic models were thus developed to fine-tune the classification tree and obtain a better accuracy of the results. For the same purpose, a systematic integration of OBIA and targeted ground surveys seems to be the best option of a cross-validation of the results, in order to achieve an appropriate balance between processing speed and reliability of data
Archaeology and Object-Based Image Analysis: potential and issues for the application of semantic models
No full textInternational audienceIn recent years, the field of remote sensing experienced an incredible growth thanks to the increasing quality and variety of sensors and the reduction of instrumental costs. The benefits for archaeology were soon apparent. So far, data interpretation remains essentially a prerogative of the human operator and is mediated by his skills and experiences. However, the continuous increase of datasets volume and the increasing necessity to work on large scale projects require an overall revision of the methods traditionally used in the archaeological field. With the progressive diffusion of OBIA in archeology during the last few years, it is now essential to find a shared language and a common protocol of investigation (ideally passing from operational practice to operational routine), necessary to allow the comparability of data. This presentation will offer a general review on the topic considering both the results published in the literature and new case studies developed by the author and his research group, which will be used to illustrate the possibilities of the method in archeology. The case studies will comprise object-based procedures for the treatment of aerial/ satellite imaging, such as VHR multispectral data, and digital elevation models acquired by airborne and terrestrial laser scanning. These data will serve as a starting point to discuss the limits and the potential of OBIA in the archaeological field, with specific reference to the prospects for the future in light of the recent developments in other disciplines such as environmental and biomedical sciences. The objective is therefore to examine a series of crucial methodological issues linked to the residuality of the archaeological remains, also through the exemplification of practical results. In fact, the archaeological remains are always affected by post-depositional processes which have progressively altered, with different degrees of impact, their original characteristics, leading to a high variability within the same category of evidences. Diachronic semantic models were thus developed to fine-tune the classification tree and obtain a better accuracy of the results. For the same purpose, a systematic integration of OBIA and targeted ground surveys seems to be the best option of a cross-validation of the results, in order to achieve an appropriate balance between processing speed and reliability of data
Automatically producing semantically tagged bilingual terminologies
No full textEven though many NLP resources and tools claim to be domain independent, their application to specifc tasks is restricted to
some specifc domain, otherwise their performance degrade notably. As the accuracy of NLP resources drops heavily when
applied in environments diferent from which they were built a tuning to the new environment is needed. This paper proposes
a method for automatically compile terminologies from potentially any domain. The proposed method takes as reference the
set of domains defned by Magnini, the Multilingual Central Repository (a resource based on WordNet 3.0) together with
DBpedia, an open knowledge source that had proven to be reliable for restricted domains. Using the method described in
this article, we have produced a big set of reliable terminologies for 164 domains and 2 languages totalling 635,527 terms.
The proposed method has been applied to English and Spanish languages but it is potentially applicable to any language that
has its own a DBpedia evolved enough. The obtained results have been intensively evaluated in several ways.The author Jorge Vivaldi was partially funded by
the public supported project TERMMED (FFI2017-88100-P, MINECO). The author Horacio Rodríguez was partially supported by the
public funded project GRAPHMED (TIN2016-77820-C3-3R)
- …
