ATTOS: Análisis de Tendencias y Temáticas a través de Opiniones y Sentimientos

El proyecto ATTOS centra su actividad en el estudio y desarrollo de técnicas de análisis de opiniones, enfocado a proporcionar toda la información necesaria para que una empresa o una institución pueda tomar decisiones estratégicas en función a la imagen que la sociedad tiene sobre esa empresa, producto o servicio. El objetivo último del proyecto es la interpretación automática de estas opiniones, posibilitando así su posterior explotación. Para ello se estudian parámetros tales como la intensidad de la opinión, ubicación geográfica y perfil de usuario, entre otros factores, para facilitar la toma de decisiones. El objetivo general del proyecto se centra en el estudio, desarrollo y experimentación de técnicas, recursos y sistemas basados en Tecnologías del Lenguaje Humano (TLH), para conformar una plataforma de monitorización de la Web 2.0 que genere información sobre tendencias de opinión relacionadas con un tema.

Application of Text Summarization techniques to the Geographical Information Retrieval task

Automatic Text Summarization has been shown to be useful for Natural Language Processing tasks such as Question Answering or Text Classification and other related fields of computer science such as Information Retrieval. Since Geographical Information Retrieval can be considered as an extension of the Information Retrieval field, the generation of summaries could be integrated into these systems by acting as an intermediate stage, with the purpose of reducing the document length. In this manner, the access time for information searching will be improved, while at the same time relevant documents will be also retrieved. Therefore, in this paper we propose the generation of two types of summaries (generic and geographical) applying several compression rates in order to evaluate their effectiveness in the Geographical Information Retrieval task. The evaluation has been carried out using GeoCLEF as evaluation framework and following an Information Retrieval perspective without considering the geo-reranking phase commonly used in these systems. Although single-document summarization has not performed well in general, the slight improvements obtained for some types of the proposed summaries, particularly for those based on geographical information, made us believe that the integration of Text Summarization with Geographical Information Retrieval may be beneficial, and consequently, the experimental set-up developed in this research work serves as a basis for further investigations in this field.

Noise-tolerance feasibility for restricted-domain Information Retrieval systems

Information Retrieval systems normally have to work with rather heterogeneous sources, such as Web sites or documents from Optical Character Recognition tools. The correct conversion of these sources into flat text files is not a trivial task since noise may easily be introduced as a result of spelling or typeset errors. Interestingly, this is not a great drawback when the size of the corpus is sufficiently large, since redundancy helps to overcome noise problems. However, noise becomes a serious problem in restricted-domain Information Retrieval specially when the corpus is small and has little or no redundancy. This paper devises an approach which adds noise-tolerance to Information Retrieval systems. A set of experiments carried out in the agricultural domain proves the effectiveness of the approach presented.

International conference paper presentations: A multimodal analysis to determine effectiveness

International conference presentations represent one of the biggest challenges for academics using English as a Lingua Franca (ELF). This paper aims to initiate exploration into the multimodal academic discourse of oral presentations, including the verbal, written, non-verbal material (NVM) and body language modes. It offers a Systemic Functional Linguistic (SFL) and multimodal framework of presentations to enhance mixed-disciplinary ELF academics' awareness of what needs to be taken into account to communicate effectively at conferences. The model is also used to establish evaluation criteria for the presenters' talks and to carry out a multimodal discourse analysis of four well-rated 20-min talks, two from the technical sciences and two from the social sciences in a workshop scenario. The findings from the analysis and interviews indicate that: (a) a greater awareness of the mode affordances and their combinations can lead to improved performances; (b) higher reliance on the visual modes can compensate for verbal deficiencies; and (c) effective speakers tend to use a variety of modes that often overlap but work together to convey specific meanings. However, firm conclusions cannot be drawn on the basis of workshop presentations, and further studies on the multimodal analysis of ‘real conferences’ within specific disciplines are encouraged.

Retrieval of phenological stages of onion fields during the first year of growth by means of C-band polarimetric SAR measurements

The phenological stages of onion fields in the first year of growth are estimated using polarimetric observables and single-polarization intensity channels. Experiments are undertaken on a time series of RADARSAT-2 C-band full-polarimetric synthetic aperture radar (SAR) images collected in 2009 over the Barrax region, Spain, where ground truth information about onion growth stages is provided by the European Space Agency (ESA)-funded agricultural bio/geophysical retrieval from frequent repeat pass SAR and optical imaging (AgriSAR) field campaign conducted in that area. The experimental results demonstrate that polarimetric entropy or copolar coherence when used jointly with the cross-polarized intensity allows unambiguously distinguishing three phenological intervals.

The orchestration of modes and EFL audio-visual comprehension: A multimodal discourse analysis of vodcasts

This thesis explores the role of multimodality in language learners’ comprehension, and more specifically, the effects on students’ audio-visual comprehension when different orchestrations of modes appear in the visualization of vodcasts. Firstly, I describe the state of the art of its three main areas of concern, namely the evolution of meaning-making, Information and Communication Technology (ICT), and audio-visual comprehension. One of the most important contributions in the theoretical overview is the suggested integrative model of audio-visual comprehension, which attempts to explain how students process information received from different inputs. Secondly, I present a study based on the following research questions: ‘Which modes are orchestrated throughout the vodcasts?’, ‘Are there any multimodal ensembles that are more beneficial for students’ audio-visual comprehension?’, and ‘What are the students’ attitudes towards audio-visual (e.g., vodcasts) compared to traditional audio (e.g., audio tracks) comprehension activities?’. Along with these research questions, I have formulated two hypotheses: Audio-visual comprehension improves when there is a greater number of orchestrated modes, and students have a more positive attitude towards vodcasts than traditional audios when carrying out comprehension activities. The study includes a multimodal discourse analysis, audio-visual comprehension tests, and students’ questionnaires. The multimodal discourse analysis of two British Council’s language learning vodcasts, entitled English is GREAT and Camden Fashion, using ELAN as the multimodal annotation tool, shows that there are a variety of multimodal ensembles of two, three and four modes. The audio-visual comprehension tests were given to 40 Spanish students, learning English as a foreign language, after the visualization of vodcasts. These comprehension tests contain questions related to specific orchestrations of modes appearing in the vodcasts. The statistical analysis of the test results, using repeated-measures ANOVA, reveal that students obtain better audio-visual comprehension results when the multimodal ensembles are constituted by a greater number of orchestrated modes. Finally, the data compiled from the questionnaires, conclude that students have a more positive attitude towards vodcasts in comparison to traditional audio listenings. Results from the audio-visual comprehension tests and questionnaires prove the two hypotheses of this study.

The importance of promoting multimodal teaching in the foreign language classroom for the acquisition of social competences: practical examples

This article analyses the way in which the subject English Language V of the degree English Studies (English Language and Literature) combines the development of the five skills (listening, speaking, reading, writing and interacting) with the use of multimodal activities and resources in the teaching-learning process so that students increase their motivation and acquire different social competences that will be useful for the labour market such as communication, cooperation, leadership or conflict management. This study highlights the use of multimodal materials (texts, videos, etc.) on social topics to introduce cultural aspects in a language subject and to deepen into the different social competences university students can acquire when they work with them. The study was guided by the following research questions: how can multimodal texts and resources contribute to the development of the five skills in a foreign language classroom? What are the main social competences that students acquire when the teaching-learning process is multimodal? The results of a survey prepared at the end of the academic year 2015-2016 point out the main competences that university students develop thanks to multimodal teaching. For its framework of analysis, the study draws on the main principles of visual grammar (Kress & van Leeuwen, 2006) where students learn how to analyse the main aspects in multimodal texts. The analysis of the different multimodal activities described in the article and the survey reveal that multimodality is useful for developing critical thinking, for bringing cultural aspects into the classroom and for working on social competences. This article will explain the successes and challenges of using multimodal texts with social content so that students can acquire social competences while learning content. Moreover, the implications of using multimodal resources in a language classroom to develop multiliteracies will be observed.

Multimodal analysis of a sample of political posters in Ireland during and after the Celtic Tiger

The aim of this research paper is to analyse the key political posters made for the campaigns of Irish political party Fianna Fáil framed in the Celtic Tiger (1997-2008) and post-Celtic Tiger years (2009-2012). I will then focus on the four posters of the candidate in the elections that took place in 1997, 2002, 2007 and 2011 with the intention of observing first how the leader is represented, and later on pinpointing the similarities and possible differences between each. This is important in order to observe the main linguistic and visual strategies used to persuade the audience to vote that party and to highlight the power of the politician. Critical discourse analysis tools will be helpful to identify the main discursive strategies employed to persuade the Irish population to vote in a certain direction. Van Leeuwen’s (2008) social actor theory will facilitate the understanding of how participants are represented in the corpus under analysis. Finally, the main tools of Kress and van Leeuwen’s visual grammar (2006) will be applied for the analysis of the images. The study reveals that politicians are represented in a consistently positive way, with status and formal appearance so that people are persuaded to vote for the party they represent because they trust them as political leaders. The study, thus, points out that the poster is a powerful tool used in election campaigns to highlight the power of political parties.

Learning representations for Information Retrieval

La recherche d'informations s'intéresse, entre autres, à répondre à des questions comme: est-ce qu'un document est pertinent à une requête ? Est-ce que deux requêtes ou deux documents sont similaires ? Comment la similarité entre deux requêtes ou documents peut être utilisée pour améliorer l'estimation de la pertinence ? Pour donner réponse à ces questions, il est nécessaire d'associer chaque document et requête à des représentations interprétables par ordinateur. Une fois ces représentations estimées, la similarité peut correspondre, par exemple, à une distance ou une divergence qui opère dans l'espace de représentation. On admet généralement que la qualité d'une représentation a un impact direct sur l'erreur d'estimation par rapport à la vraie pertinence, jugée par un humain. Estimer de bonnes représentations des documents et des requêtes a longtemps été un problème central de la recherche d'informations. Le but de cette thèse est de proposer des nouvelles méthodes pour estimer les représentations des documents et des requêtes, la relation de pertinence entre eux et ainsi modestement avancer l'état de l'art du domaine. Nous présentons quatre articles publiés dans des conférences internationales et un article publié dans un forum d'évaluation. Les deux premiers articles concernent des méthodes qui créent l'espace de représentation selon une connaissance à priori sur les caractéristiques qui sont importantes pour la tâche à accomplir. Ceux-ci nous amènent à présenter un nouveau modèle de recherche d'informations qui diffère des modèles existants sur le plan théorique et de l'efficacité expérimentale. Les deux derniers articles marquent un changement fondamental dans l'approche de construction des représentations. Ils bénéficient notamment de l'intérêt de recherche dont les techniques d'apprentissage profond par réseaux de neurones, ou deep learning, ont fait récemment l'objet. Ces modèles d'apprentissage élicitent automatiquement les caractéristiques importantes pour la tâche demandée à partir d'une quantité importante de données. Nous nous intéressons à la modélisation des relations sémantiques entre documents et requêtes ainsi qu'entre deux ou plusieurs requêtes. Ces derniers articles marquent les premières applications de l'apprentissage de représentations par réseaux de neurones à la recherche d'informations. Les modèles proposés ont aussi produit une performance améliorée sur des collections de test standard. Nos travaux nous mènent à la conclusion générale suivante: la performance en recherche d'informations pourrait drastiquement être améliorée en se basant sur les approches d'apprentissage de représentations.

A multimodal tablet-based interface for designing and reviewing 3D engineering models

The ability to view and interact with 3D models has been happening for a long time. However, vision-based 3D modeling has only seen limited success in applications, as it faces many technical challenges. Hand-held mobile devices have changed the way we interact with virtual reality environments. Their high mobility and technical features, such as inertial sensors, cameras and fast processors, are especially attractive for advancing the state of the art in virtual reality systems. Also, their ubiquity and fast Internet connection open a path to distributed and collaborative development. However, such path has not been fully explored in many domains. VR systems for real world engineering contexts are still difficult to use, especially when geographically dispersed engineering teams need to collaboratively visualize and review 3D CAD models. Another challenge is the ability to rendering these environments at the required interactive rates and with high fidelity. In this document it is presented a virtual reality system mobile for visualization, navigation and reviewing large scale 3D CAD models, held under the CEDAR (Collaborative Engineering Design and Review) project. It’s focused on interaction using different navigation modes. The system uses the mobile device's inertial sensors and camera to allow users to navigate through large scale models. IT professionals, architects, civil engineers and oil industry experts were involved in a qualitative assessment of the CEDAR system, in the form of direct user interaction with the prototypes and audio-recorded interviews about the prototypes. The lessons learned are valuable and are presented on this document. Subsequently it was prepared a quantitative study on the different navigation modes to analyze the best mode to use it in a given situation.

Investigation of brain function in neuropsychiatric disorder using multimodal imaging analysis

Otto-von-Guericke-Universität Magdeburg, Fakultät für Naturwissenschaften, Dissertation, 2016

Konzeption eines selbständig lernenden Information Retrieval Systems, als ein Informationssammel- und Verwaltungsprogramm

[s.c.]

Learning representations for Information Retrieval

La recherche d'informations s'intéresse, entre autres, à répondre à des questions comme: est-ce qu'un document est pertinent à une requête ? Est-ce que deux requêtes ou deux documents sont similaires ? Comment la similarité entre deux requêtes ou documents peut être utilisée pour améliorer l'estimation de la pertinence ? Pour donner réponse à ces questions, il est nécessaire d'associer chaque document et requête à des représentations interprétables par ordinateur. Une fois ces représentations estimées, la similarité peut correspondre, par exemple, à une distance ou une divergence qui opère dans l'espace de représentation. On admet généralement que la qualité d'une représentation a un impact direct sur l'erreur d'estimation par rapport à la vraie pertinence, jugée par un humain. Estimer de bonnes représentations des documents et des requêtes a longtemps été un problème central de la recherche d'informations. Le but de cette thèse est de proposer des nouvelles méthodes pour estimer les représentations des documents et des requêtes, la relation de pertinence entre eux et ainsi modestement avancer l'état de l'art du domaine. Nous présentons quatre articles publiés dans des conférences internationales et un article publié dans un forum d'évaluation. Les deux premiers articles concernent des méthodes qui créent l'espace de représentation selon une connaissance à priori sur les caractéristiques qui sont importantes pour la tâche à accomplir. Ceux-ci nous amènent à présenter un nouveau modèle de recherche d'informations qui diffère des modèles existants sur le plan théorique et de l'efficacité expérimentale. Les deux derniers articles marquent un changement fondamental dans l'approche de construction des représentations. Ils bénéficient notamment de l'intérêt de recherche dont les techniques d'apprentissage profond par réseaux de neurones, ou deep learning, ont fait récemment l'objet. Ces modèles d'apprentissage élicitent automatiquement les caractéristiques importantes pour la tâche demandée à partir d'une quantité importante de données. Nous nous intéressons à la modélisation des relations sémantiques entre documents et requêtes ainsi qu'entre deux ou plusieurs requêtes. Ces derniers articles marquent les premières applications de l'apprentissage de représentations par réseaux de neurones à la recherche d'informations. Les modèles proposés ont aussi produit une performance améliorée sur des collections de test standard. Nos travaux nous mènent à la conclusion générale suivante: la performance en recherche d'informations pourrait drastiquement être améliorée en se basant sur les approches d'apprentissage de représentations.

Multimodal Imaging and Choroidal Volumetric Changes After Half-fluence PDT in Central Serous Chorioretinopathy.

PURPOSE The purpose of this study was to identify SD-OCT changes that correspond to leakage on fluorescein (FA) and indocyanine angiography (ICGA) and evaluate effect of half-fluence photodynamic therapy (PDT) on choroidal volume in chronic central serous choroidoretinopathy (CSC). METHODS Retrospective analysis of patients with chronic CSC who had undergone PDT. Baseline FA and ICGA images were overlaid on SD-OCT to identify OCT correlates of FA or ICGA hyperfluorescence. Choroidal volume was evaluated in a subgroup of eyes before and after PDT. RESULTS Twenty eyes were evaluated at baseline, of which seven eyes had choroidal volume evaluations at baseline and 3 months following PDT. SD-OCT changes corresponding to FA hyperfluorescence were subretinal fluid (73%), RPE microrip (50%), RPE double-layer sign (31%), RPE detachment (15%), and RPE thickening (8%). ICGA hyperfluoresence was correlated in 93% with hyperreflective spots in the superficial choroid. Choroidal volume decreased from 9.35 ± 1.99 to 8.52 ± 1.92 and 8.04 ± 1.7 mm(3) (at 1 and 3 months post PDT, respectively, p ≤ 0.001). CONCLUSIONS We identified specific OCT findings that correlate with FA and ICGA leakage sites. SD-OCT is a valuable tool to localize CSC lesions and may be useful to guide PDT treatment. Generalized choroidal volume decrease occurs following PDT and extends beyond PDT treatment site.