EmotiBlog: a model to learn subjective information detection in the new textual genres of the Web 2.0 -a multilingual and multi-genre approach-

Tesis doctoral con mención europea en procesamiento del lenguaje natural realizada en la Universidad de Alicante por Ester Boldrini bajo la dirección del Dr. Patricio Martínez-Barco. El acto de defensa de la tesis tuvo lugar en la Universidad de Alicante el 23 de enero de 2012 ante el tribunal formado por los doctores Manuel Palomar (Universidad de Alicante), Dr. Paloma Moreda (UA), Dr. Mariona Taulé (Universidad de Barcelona), Dr. Horacio Saggion (Universitat Pompeu Fabra) y Dr. Mike Thelwall (University of Wolverhampton). Calificación: Sobresaliente Cum Laude por unanimidad.

An approach based on Logic Forms and WordNet relationships to Textual Entailment performance

This paper outlines the approach adopted by the PLSI research group at University of Alicante in the PASCAL-2006 second Recognising Textual Entailment challenge. Our system is composed of several components. On the one hand, the first component performs the derivation of the logic forms of the text/hypothesis pairs and, on the other hand, the second component provides us with a similarity score given by the semantic relations between the derived logic forms. In order to obtain this score we apply several measures of similitude and relatedness based on the structure and content of WordNet.

EmotiBlog: a fine-grained model for emotion detection in non-traditional textual genres

This paper presents the first version of EmotiBlog, an annotation scheme for emotions in non-traditional textual genres such as blogs or forums. We collected a corpus composed by blog posts in three languages: English, Spanish and Italian and about three topics of interest. Subsequently, we annotated our collection and carried out the inter-annotator agreement and a ten-fold cross-validation evaluation, obtaining promising results. The main aim of this research is to provide a finer-grained annotation scheme and annotated data that are essential to perform evaluation focused on checking the quality of the created resources.

Machine learning techniques for automatic opinion detection in non-traditional textual genres

This paper presents a preliminary study in which Machine Learning experiments applied to Opinion Mining in blogs have been carried out. We created and annotated a blog corpus in Spanish using EmotiBlog. We evaluated the utility of the features labelled firstly carrying out experiments with combinations of them and secondly using the feature selection techniques, we also deal with several problems, such as the noisy character of the input texts, the small size of the training set, the granularity of the annotation scheme and the language object of our study, Spanish, with less resource than English. We obtained promising results considering that it is a preliminary study.

A knowledge-based textual entailment approach applied to the QA answer validation at CLEF 2006

The Answer Validation Exercise (AVE) is a pilot track within the Cross-Language Evaluation Forum (CLEF) 2006. The AVE competition provides an evaluation frame- work for answer validations in Question Answering (QA). In our participation in AVE, we propose a system that has been initially used for other task as Recognising Textual Entailment (RTE). The aim of our participation is to evaluate the improvement our system brings to QA. Moreover, due to the fact that these two task (AVE and RTE) have the same main idea, which is to find semantic implications between two fragments of text, our system has been able to be directly applied to the AVE competition. Our system is based on the representation of the texts by means of logic forms and the computation of semantic comparison between them. This comparison is carried out using two different approaches. The first one managed by a deeper study of the Word- Net relations, and the second uses the measure defined by Lin in order to compute the semantic similarity between the logic form predicates. Moreover, we have also designed a voting strategy between our system and the MLEnt system, also presented by the University of Alicante, with the aim of obtaining a joint execution of the two systems developed at the University of Alicante. Although the results obtained have not been very high, we consider that they are quite promising and this supports the fact that there is still a lot of work on researching in any kind of textual entailment.

Textual Information Extraction

Material completo EIT

Topological structures of complex belief systems (II): Textual materialization

Mythical and religious belief systems in a social context can be regarded as a conglomeration of sacrosanct rites, which revolve around substantive values that involve an element of faith. Moreover, we can conclude that ideologies, myths and beliefs can all be analyzed in terms of systems within a cultural context. The significance of being able to define ideologies, myths and beliefs as systems is that they can figure in cultural explanations. This, in turn, means that such systems can figure in logic-mathematical analyses.

Spectral-domain optical coherence tomography study of macular structure as prognostic and determining factor for macular hole surgery outcome

Purpose: To evaluate postoperative spectral-domain optical coherence tomography findings after macular hole surgery. Methods: Retrospective, interventional, nonrandomized study. Overall, 164 eyes of 157 patients diagnosed with macular hole were operated on by vitrectomy and internal limiting membrane peeling. Preoperative and postoperative best-corrected visual acuity and spectral-domain optical coherence tomography images were obtained. Two groups were considered on the basis of the postoperative integrity of the back reflection line from the ellipsoid portion of the photoreceptor inner segment: group A (disruption of ellipsoid portion of the inner segment line, 60 eyes) and group B (restoration of ellipsoid portion of the inner segment line, 104 eyes). Results: Logarithm of the minimum angle of resolution best-corrected visual acuity improved significantly after the surgery of macular hole from a mean preoperative value of 0.79 ± 0.37 (range, 0.15–2.00) to a mean postoperative value of 0.35 ± 0.31 (range, 0.00–1.30) at the last follow-up visit (P < 0.01). Best-corrected visual acuity improved significantly in the 2 groups analyzed (all P < 0.01). A larger improvement was found in group B than in group A (P < 0.01). Conclusion: Ellipsoid portion of the inner segment line reconstruction seems to be a good prognostic factor for visual rehabilitation after macular hole surgery.

Saint Mathew Law and Bonini Paradox in Textual Theory of Complex Models

The mathematical models of the complex reality are texts belonging to a certain literature that is written in a semi-formal language, denominated L(MT) by the authors whose laws linguistic mathematics have been previously defined. This text possesses linguistic entropy that is the reflection of the physical entropy of the processes of real world that said text describes. Through the temperature of information defined by Mandelbrot, the authors begin a text-reality thermodynamic theory that drives to the existence of information attractors, or highly structured point, settling down a heterogeneity of the space text, the same one that of ontologic space, completing the well-known law of Saint Mathew, of the General Theory of Systems and formulated by Margalef saying: “To the one that has more he will be given, and to the one that doesn't have he will even be removed it little that it possesses.

Long time remodeling during retinal degeneration evaluated by optical coherence tomography, immunocytochemistry and fundus autofluorescence

Purpose: To characterize the relationship between fundus autofluorescence (FAF), Optical Coherence Tomography (OCT) and immunohistochemistry (IHC) over the course of chronic retinal degeneration in the P23H rat. Methods: Homozygous albino P23H rats, Sprague–Dawley (SD) rats as controls and pigmented Long Evans (LE) rats were used. A Spectralis HRA OCT system was used for scanning laser ophthalmoscopy (SLO) imaging OCT and angiography. To determine FAF, fluorescence was excited using diode laser at 488 nm. A fast retina map OCT was performed using the optic nerve as a landmark. IHC was performed to correlate with the findings of OCT and FAF changes. Results: During the course of retinal degeneration, the FAF pattern evolved from some spotting at 2 months old to a mosaic of hyperfluorescent dots in rats 6 months and older. Retinal thicknesses progressively diminished over the course of the disease. At later stages of degeneration, OCT documented changes in the retinal layers, however, IHC better identified the cell loss and remodeling changes. Angiography revealed attenuation of the retinal vascular plexus with time. Conclusion: We provide for the first time a detailed long-term analysis of the course of retinal degeneration in P23H rats using a combination of SLO and OCT imaging, angiography, FAF and IHC. Although, the application of noninvasive methods enables longitudinal studies and will decrease the number of animals needed for a study, IHC is still an essential tool to identify retinal changes at the cellular level.

A semantic framework for textual data enrichment

In this work we present a semantic framework suitable of being used as support tool for recommender systems. Our purpose is to use the semantic information provided by a set of integrated resources to enrich texts by conducting different NLP tasks: WSD, domain classification, semantic similarities and sentiment analysis. After obtaining the textual semantic enrichment we would be able to recommend similar content or even to rate texts according to different dimensions. First of all, we describe the main characteristics of the semantic integrated resources with an exhaustive evaluation. Next, we demonstrate the usefulness of our resource in different NLP tasks and campaigns. Moreover, we present a combination of different NLP approaches that provide enough knowledge for being used as support tool for recommender systems. Finally, we illustrate a case of study with information related to movies and TV series to demonstrate that our framework works properly.

Complexidade textual e progressão escolar em dois registos : um estudo de correlação baseado em um corpus quasi-longitudinal

Tese de doutoramento, Linguística (Linguística Educacional), Universidade de Lisboa, Faculdade de Letras, 2016