Using social semantic knowledge to improve annotations in personal photo collections

Instituto Politécnico de Lisboa (IPL) e Instituto Superior de Engenharia de Lisboa (ISEL)apoio concedido pela bolsa SPRH/PROTEC/67580/2010, que apoiou parcialmente este trabalho

Implementação de um modelo de geometric semantic genetic programming para aplicação naval

Recaí sob a responsabilidade da Marinha Portuguesa a gestão da Zona Económica Exclusiva de Portugal, assegurando a sua segurança da mesma face a atividades criminosas. Para auxiliar a tarefa, é utilizado o sistema Oversee, utilizado para monitorizar a posição de todas as embarcações presentes na área afeta, permitindo a rápida intervenção da Marinha Portuguesa quando e onde necessário. No entanto, o sistema necessita de transmissões periódicas constantes originadas nas embarcações para operar corretamente – casos as transmissões sejam interrompidas, deliberada ou acidentalmente, o sistema deixa de conseguir localizar embarcações, dificultando a intervenção da Marinha. A fim de colmatar esta falha, é proposto adicionar ao sistema Oversee a capacidade de prever as posições futuras de uma embarcação com base no seu trajeto até à cessação das transmissões. Tendo em conta os grandes volumes de dados gerados pelo sistema (históricos de posições), a área de Inteligência Artificial apresenta uma possível solução para este problema. Atendendo às necessidades de resposta rápida do problema abordado, o algoritmo de Geometric Semantic Genetic Programming baseado em referências de Vanneschi et al. apresenta-se como uma possível solução, tendo já produzido bons resultados em problemas semelhantes. O presente trabalho de tese pretende integrar o algoritmo de Geometric Semantic Genetic Programming desenvolvido com o sistema Oversee, a fim de lhe conceder capacidades preditivas. Adicionalmente, será realizado um processo de análise de desempenho a fim de determinar qual a ideal parametrização do algoritmo. Pretende-se com esta tese fornecer à Marinha Portuguesa uma ferramenta capaz de auxiliar o controlo da Zona Económica Exclusiva Portuguesa, permitindo a correta intervenção da Marinha em casos onde o atual sistema não conseguiria determinar a correta posição da embarcação em questão.

BIOMedical search engine framework: lightweight and customized implementation of domain-specific biomedical search engines

The Smart Drug Search is publicly accessible at http://sing.ei.uvigo.es/sds/. The BIOMedical Search Engine Framework is freely available for non-commercial use at https://github.com/agjacome/biomsef

Robust acoustic and semantic modeling in a telephone-based spoken dialog system

Magdeburg, Univ., Fak. für Elektrotechnik und Informationstechnik, Diss., 2009

A follow-up study on the quality of alcohol dependence-related information on the web.

ABSTRACT: In order to evaluate the one-year evolution of web-based information on alcohol dependence, we re-assessed alcohol-related sites in July 2007 with the same evaluating tool that had been used to assess these sites in June 2006. Websites were assessed with a standardized form designed to rate sites on the basis of accountability, presentation, interactivity, readability, and content quality. The DISCERN scale was also used, which aimed to assist persons without content expertise in assessing the quality of written health publications. Scores were highly stable for all components of the form one year later (r = .77 to .95, p < .01). Analysis of variance for repeated measures showed no time effect, no interaction between time and scale, no interaction between time and group (affiliation categories), and no interaction between time, group, and scale. The study highlights lack of change of alcohol-dependence-related web pages across one year.

Does Semantic Context Benefit Speech Understanding through "Top-Down" Processes? Evidence from Time-resolved Sparse fMRI.

When speech is degraded, word report is higher for semantically coherent sentences (e.g., her new skirt was made of denim) than for anomalous sentences (e.g., her good slope was done in carrot). Such increased intelligibility is often described as resulting from "top-down" processes, reflecting an assumption that higher-level (semantic) neural processes support lower-level (perceptual) mechanisms. We used time-resolved sparse fMRI to test for top-down neural mechanisms, measuring activity while participants heard coherent and anomalous sentences presented in speech envelope/spectrum noise at varying signal-to-noise ratios (SNR). The timing of BOLD responses to more intelligible speech provides evidence of hierarchical organization, with earlier responses in peri-auditory regions of the posterior superior temporal gyrus than in more distant temporal and frontal regions. Despite Sentence content × SNR interactions in the superior temporal gyrus, prefrontal regions respond after auditory/perceptual regions. Although we cannot rule out top-down effects, this pattern is more compatible with a purely feedforward or bottom-up account, in which the results of lower-level perceptual processing are passed to inferior frontal regions. Behavioral and neural evidence that sentence content influences perception of degraded speech does not necessarily imply "top-down" neural processes.

Quality of web-based information on bipolar disorder

Rapport de synthèse : Introduction : Internet est une source importante d'information sur la santé mentale. Le trouble bipolaire est communément associé à un handicap, des comorbidités, un faible taux d'introspection et une mauvaise compliance au traitement. Le fardeau de la maladie, de par les épisodes dépressifs et maniaques, peut conduire les personnes (dont le diagnostic de trouble bipolaire a été déjà posé ou non), ainsi que leur famille à rechercher des informations sur Internet. De ce fait, il est important que les sites Web traitant du sujet contiennent de l'information de haute qualité, basée sur les évidences scientifiques. Objectif.: évaluer la qualité des informations consultables sur Internat au sujet du trouble bipolaire et identifier des indicateurs de qualité. Méthode: deux mots-clés : « bipolar disorder » et « manic depressive illness » ont été introduits dans les moteurs de recherche les plus souvent utilisés sur Internet. Les sites Internet ont été évalués avec un formulaire standard conçu pour noter les sites sur la base de l'auteur (privé, université, entreprise,...), la présentation, l'interactivité, la lisibilité et la qualité du contenu. Le label de qualité « Health On the Net» (HON), et l'outil DISCERN ont été utilisés pour vérifier leur efficacité comme indicateurs de la qualité. Résultats: sur les 80 sites identifiés, 34 ont été inclus. Sur la base de la mesure des résultats, la qualité du contenu des sites s'est avérée être bonne. La qualité du contenu des sites Web qui traitent du trouble bipolaire est expliquée de manière significative par la lisibilité, la responsabilité et l'interactivité aussi bien que par un score global. Conclusions: dans l'ensemble, la qualité du contenu de l'étude des sites Web traitant du trouble bipolaire est de bonne qualité.

Version brève du DISCERN, une voie d'amélioration de la recherche d'information médicale sur le net [Brief-DISCERN, a possible way to improve patient's search on the health-related web]

Internet is increasingly used as a source of information on health issues and is probably a major source of patients' empowerment. This process is however limited by the frequently poor quality of web-based health information designed for consumers. A better diffusion of information about criteria defining the quality of the content of websites, and about useful methods designed for searching such needed information, could be particularly useful to patients and their relatives. A brief, six-items DISCERN version, characterized by a high specificity for detecting websites with good or very good content quality was recently developed. This tool could facilitate the identification of high-quality information on the web by patients and may improve the empowerment process initiated by the development of the health-related web.

Semàntica del mot : propietats semàntiques de tres verbs de desplaçament

El treball té com a objectiu l'estudi de les propietats semàntiques d'un grup de verbs de desplaçament i els seus corresponents arguments. La informació sobre el tipus de complement que demana cada verb és important de cara a conèixer l'estructura sintàctica de la frase i oferir solucions pràctiques en tasques de Processament del Llenguatge Natural. L'anàlisi se centrarà en els verbs conduir, navegar i volar, a partir dels sentits bàsics que el Diccionari d'ús dels verbs catalans (DUVC) descriu per a cadascun d'aquests verbs i de les seves restriccions selectives. Comprovarem, mitjançant un centenar de frases extretes del Corpus d'Ús del Català a la Web de la Universitat Pompeu Fabra i del Corpus Textual Informatitzat de la Llengua Catalana de l'Institut d'Estudis Catalans, si en la llengua es donen només els sentits i usos descrits en el DUVC i quins són els més freqüents. Finalment, descriurem els noms que fan de nucli dels arguments en termes de trets semàntics.

Atypical association of semantic dementia, corticobasal syndrome, and 4R tauopathy.

A 57-year-old male with no family history was diagnosed with semantic dementia. He also showed some unusual cognitive features such as episodic memory and executive dysfunctions, spatial disorientation, and dyscalculia. Rapidly progressive cognitive and physical decline occurred. About 1.5 years later, he developed clinical features of a corticobasal syndrome. He died at the age of 60. Brain autopsy revealed numerous 4R-tau-positive lesions in the frontal, parietal and temporal lobes, basal ganglia, and brainstem. Neuronal loss was severe in the temporal cortex. Such association of semantic dementia with tauopathy and corticobasal syndrome is highly unusual. These findings are discussed in the light of current knowledge about frontotemporal lobar degeneration.

Engaging students in an e-learning experience : how to manage information resources through wikis and web blogs

The purpose of this paper is to describe the collaboration between librarians and scholars, from a virtual university, in order to facilitate collaborative learning on how to manage information resources. The personal information behaviour of e-learning students when managing information resources for academic, professional and daily life purposes was studied from 24 semi-structured face-to-face interviews. The results of the content analysis of the interview' transcriptions, highlighted that in the workplace and daily life contexts, competent information behaviour is always linked to a proactive attitude, that is to say, that participants seek for information without some extrinsic reward or avoiding punishment. In the academic context, it was observed a low level of information literacy and it seems to be related with a prevalent uninvolved attitude.

Categorization of bibsonomy tags to apply privacy-preserving mechanisms.

In this technical report, we approach one of the practical aspects when it comes to represent users' interests from their tagging activity, namely the categorization of tags into high-level categories of interest. The reason is that the representation of user profiles on the basis of the myriad of tags available on the Web is certainly unfeasible from various practical perspectives; mainly concerningthe unavailability of data to reliably, accurately measure interests across such fine-grained categorization, and, should the data be available, its overwhelming computational intractability. Motivated by this, our study presents the results of a categorization process whereby a collection of tags posted at BibSonomy #http://www.bibsonomy.org# are classified into 5 categories of interest. The methodology used to conduct such categorization is in line with other works in the field.

Hierarchical categorisation of tags for delicious

In the scenario of social bookmarking, a user browsing the Web bookmarks web pages and assigns free-text labels (i.e., tags) to them according to their personal preferences. In this technical report, we approach one of the practical aspects when it comes to represent users' interests from their tagging activity, namely the categorization of tags into high-level categories of interest. The reason is that the representation of user profiles on the basis of the myriad of tags available on the Web is certainly unfeasible from various practical perspectives; mainly concerning the unavailability of data to reliably, accurately measure interests across such fine-grained categorisation, and, should the data be available, its overwhelming computational intractability. Motivated by this, our study presents the results of a categorization process whereby a collection of tags posted at Delicious #http://delicious.com# are classified into 200 subcategories of interest.