Online Learning-based Robust Visual Tracking for Autonomous Landing of Unmanned Aerial Vehicles

Autonomous landing is a challenging and important technology for both military and civilian applications of Unmanned Aerial Vehicles (UAVs). In this paper, we present a novel online adaptive visual tracking algorithm for UAVs to land on an arbitrary field (that can be used as the helipad) autonomously at real-time frame rates of more than twenty frames per second. The integration of low-dimensional subspace representation method, online incremental learning approach and hierarchical tracking strategy allows the autolanding task to overcome the problems generated by the challenging situations such as significant appearance change, variant surrounding illumination, partial helipad occlusion, rapid pose variation, onboard mechanical vibration (no video stabilization), low computational capacity and delayed information communication between UAV and Ground Control Station (GCS). The tracking performance of this presented algorithm is evaluated with aerial images from real autolanding flights using manually- labelled ground truth database. The evaluation results show that this new algorithm is highly robust to track the helipad and accurate enough for closing the vision-based control loop.

Online Learning-based Robust Visual Tracking for Autonomous Landing of Unmanned Aerial Vehicles

Autonomous landing is a challenging and important technology for both military and civilian applications of Unmanned Aerial Vehicles (UAVs). In this paper, we present a novel online adaptive visual tracking algorithm for UAVs to land on an arbitrary field (that can be used as the helipad) autonomously at real-time frame rates of more than twenty frames per second. The integration of low-dimensional subspace representation method, online incremental learning approach and hierarchical tracking strategy allows the autolanding task to overcome the problems generated by the challenging situations such as significant appearance change, variant surrounding illumination, partial helipad occlusion, rapid pose variation, onboard mechanical vibration (no video stabilization), low computational capacity and delayed information communication between UAV and Ground Control Station (GCS). The tracking performance of this presented algorithm is evaluated with aerial images from real autolanding flights using manually- labelled ground truth database. The evaluation results show that this new algorithm is highly robust to track the helipad and accurate enough for closing the vision-based control loop.

Engineering Computational Emotion - A Reference Model for Emotion in Artificial Systems

Emotion is generally argued to be an influence on the behavior of life systems, largely concerning flexibility and adaptivity. The way in which life systems acts in response to a particular situations of the environment, has revealed the decisive and crucial importance of this feature in the success of behaviors. And this source of inspiration has influenced the way of thinking artificial systems. During the last decades, artificial systems have undergone such an evolution that each day more are integrated in our daily life. They have become greater in complexity, and the subsequent effects are related to an increased demand of systems that ensure resilience, robustness, availability, security or safety among others. All of them questions that raise quite a fundamental challenges in control design. This thesis has been developed under the framework of the Autonomous System project, a.k.a the ASys-Project. Short-term objectives of immediate application are focused on to design improved systems, and the approaching of intelligence in control strategies. Besides this, long-term objectives underlying ASys-Project concentrate on high order capabilities such as cognition, awareness and autonomy. This thesis is placed within the general fields of Engineery and Emotion science, and provides a theoretical foundation for engineering and designing computational emotion for artificial systems. The starting question that has grounded this thesis aims the problem of emotion--based autonomy. And how to feedback systems with valuable meaning has conformed the general objective. Both the starting question and the general objective, have underlaid the study of emotion, the influence on systems behavior, the key foundations that justify this feature in life systems, how emotion is integrated within the normal operation, and how this entire problem of emotion can be explained in artificial systems. By assuming essential differences concerning structure, purpose and operation between life and artificial systems, the essential motivation has been the exploration of what emotion solves in nature to afterwards analyze analogies for man--made systems. This work provides a reference model in which a collection of entities, relationships, models, functions and informational artifacts, are all interacting to provide the system with non-explicit knowledge under the form of emotion-like relevances. This solution aims to provide a reference model under which to design solutions for emotional operation, but related to the real needs of artificial systems. The proposal consists of a multi-purpose architecture that implement two broad modules in order to attend: (a) the range of processes related to the environment affectation, and (b) the range or processes related to the emotion perception-like and the higher levels of reasoning. This has required an intense and critical analysis beyond the state of the art around the most relevant theories of emotion and technical systems, in order to obtain the required support for those foundations that sustain each model. The problem has been interpreted and is described on the basis of AGSys, an agent assumed with the minimum rationality as to provide the capability to perform emotional assessment. AGSys is a conceptualization of a Model-based Cognitive agent that embodies an inner agent ESys, the responsible of performing the emotional operation inside of AGSys. The solution consists of multiple computational modules working federated, and aimed at conforming a mutual feedback loop between AGSys and ESys. Throughout this solution, the environment and the effects that might influence over the system are described as different problems. While AGSys operates as a common system within the external environment, ESys is designed to operate within a conceptualized inner environment. And this inner environment is built on the basis of those relevances that might occur inside of AGSys in the interaction with the external environment. This allows for a high-quality separate reasoning concerning mission goals defined in AGSys, and emotional goals defined in ESys. This way, it is provided a possible path for high-level reasoning under the influence of goals congruence. High-level reasoning model uses knowledge about emotional goals stability, letting this way new directions in which mission goals might be assessed under the situational state of this stability. This high-level reasoning is grounded by the work of MEP, a model of emotion perception that is thought as an analogy of a well-known theory in emotion science. The work of this model is described under the operation of a recursive-like process labeled as R-Loop, together with a system of emotional goals that are assumed as individual agents. This way, AGSys integrates knowledge that concerns the relation between a perceived object, and the effect which this perception induces on the situational state of the emotional goals. This knowledge enables a high-order system of information that provides the sustain for a high-level reasoning. The extent to which this reasoning might be approached is just delineated and assumed as future work. This thesis has been studied beyond a long range of fields of knowledge. This knowledge can be structured into two main objectives: (a) the fields of psychology, cognitive science, neurology and biological sciences in order to obtain understanding concerning the problem of the emotional phenomena, and (b) a large amount of computer science branches such as Autonomic Computing (AC), Self-adaptive software, Self-X systems, Model Integrated Computing (MIC) or the paradigm of models@runtime among others, in order to obtain knowledge about tools for designing each part of the solution. The final approach has been mainly performed on the basis of the entire acquired knowledge, and described under the fields of Artificial Intelligence, Model-Based Systems (MBS), and additional mathematical formalizations to provide punctual understanding in those cases that it has been required. This approach describes a reference model to feedback systems with valuable meaning, allowing for reasoning with regard to (a) the relationship between the environment and the relevance of the effects on the system, and (b) dynamical evaluations concerning the inner situational state of the system as a result of those effects. And this reasoning provides a framework of distinguishable states of AGSys derived from its own circumstances, that can be assumed as artificial emotion.

Especificaciones eMathTeacher, creación del Modelo Granular Lingüístico de la evaluación del aprendizaje humano, y diseño de sistemas que implementan estos conceptos

El concepto de algoritmo es básico en informática, por lo que es crucial que los alumnos profundicen en él desde el inicio de su formación. Por tanto, contar con una herramienta que guíe a los estudiantes en su aprendizaje puede suponer una gran ayuda en su formación. La mayoría de los autores coinciden en que, para determinar la eficacia de una herramienta de visualización de algoritmos, es esencial cómo se utiliza. Así, los estudiantes que participan activamente en la visualización superan claramente a los que la contemplan de forma pasiva. Por ello, pensamos que uno de los mejores ejercicios para un alumno consiste en simular la ejecución del algoritmo que desea aprender mediante el uso de una herramienta de visualización, i. e. consiste en realizar una simulación visual de dicho algoritmo. La primera parte de esta tesis presenta los resultados de una profunda investigación sobre las características que debe reunir una herramienta de ayuda al aprendizaje de algoritmos y conceptos matemáticos para optimizar su efectividad: el conjunto de especificaciones eMathTeacher, además de un entorno de aprendizaje que integra herramientas que las cumplen: GRAPHs. Hemos estudiado cuáles son las cualidades esenciales para potenciar la eficacia de un sistema e-learning de este tipo. Esto nos ha llevado a la definición del concepto eMathTeacher, que se ha materializado en el conjunto de especificaciones eMathTeacher. Una herramienta e-learning cumple las especificaciones eMathTeacher si actúa como un profesor virtual de matemáticas, i. e. si es una herramienta de autoevaluación que ayuda a los alumnos a aprender de forma activa y autónoma conceptos o algoritmos matemáticos, corrigiendo sus errores y proporcionando pistas para encontrar la respuesta correcta, pero sin dársela explícitamente. En estas herramientas, la simulación del algoritmo no continúa hasta que el usuario introduce la respuesta correcta. Para poder reunir en un único entorno una colección de herramientas que cumplan las especificaciones eMathTeacher hemos creado GRAPHs, un entorno ampliable, basado en simulación visual, diseñado para el aprendizaje activo e independiente de los algoritmos de grafos y creado para que en él se integren simuladores de diferentes algoritmos. Además de las opciones de creación y edición del grafo y la visualización de los cambios producidos en él durante la simulación, el entorno incluye corrección paso a paso, animación del pseudocódigo del algoritmo, preguntas emergentes, manejo de las estructuras de datos del algoritmo y creación de un log de interacción en XML. Otro problema que nos planteamos en este trabajo, por su importancia en el proceso de aprendizaje, es el de la evaluación formativa. El uso de ciertos entornos e-learning genera gran cantidad de datos que deben ser interpretados para llegar a una evaluación que no se limite a un recuento de errores. Esto incluye el establecimiento de relaciones entre los datos disponibles y la generación de descripciones lingüísticas que informen al alumno sobre la evolución de su aprendizaje. Hasta ahora sólo un experto humano era capaz de hacer este tipo de evaluación. Nuestro objetivo ha sido crear un modelo computacional que simule el razonamiento del profesor y genere un informe sobre la evolución del aprendizaje que especifique el nivel de logro de cada uno de los objetivos definidos por el profesor. Como resultado del trabajo realizado, la segunda parte de esta tesis presenta el modelo granular lingüístico de la evaluación del aprendizaje, capaz de modelizar la evaluación y generar automáticamente informes de evaluación formativa. Este modelo es una particularización del modelo granular lingüístico de un fenómeno (GLMP), en cuyo desarrollo y formalización colaboramos, basado en la lógica borrosa y en la teoría computacional de las percepciones. Esta técnica, que utiliza sistemas de inferencia basados en reglas lingüísticas y es capaz de implementar criterios de evaluación complejos, se ha aplicado a dos casos: la evaluación, basada en criterios, de logs de interacción generados por GRAPHs y de cuestionarios de Moodle. Como consecuencia, se han implementado, probado y utilizado en el aula sistemas expertos que evalúan ambos tipos de ejercicios. Además de la calificación numérica, los sistemas generan informes de evaluación, en lenguaje natural, sobre los niveles de competencia alcanzados, usando sólo datos objetivos de respuestas correctas e incorrectas. Además, se han desarrollado dos aplicaciones capaces de ser configuradas para implementar los sistemas expertos mencionados. Una procesa los archivos producidos por GRAPHs y la otra, integrable en Moodle, evalúa basándose en los resultados de los cuestionarios. ABSTRACT The concept of algorithm is one of the core subjects in computer science. It is extremely important, then, for students to get a good grasp of this concept from the very start of their training. In this respect, having a tool that helps and shepherds students through the process of learning this concept can make a huge difference to their instruction. Much has been written about how helpful algorithm visualization tools can be. Most authors agree that the most important part of the learning process is how students use the visualization tool. Learners who are actively involved in visualization consistently outperform other learners who view the algorithms passively. Therefore we think that one of the best exercises to learn an algorithm is for the user to simulate the algorithm execution while using a visualization tool, thus performing a visual algorithm simulation. The first part of this thesis presents the eMathTeacher set of requirements together with an eMathTeacher-compliant tool called GRAPHs. For some years, we have been developing a theory about what the key features of an effective e-learning system for teaching mathematical concepts and algorithms are. This led to the definition of eMathTeacher concept, which has materialized in the eMathTeacher set of requirements. An e-learning tool is eMathTeacher compliant if it works as a virtual math trainer. In other words, it has to be an on-line self-assessment tool that helps students to actively and autonomously learn math concepts or algorithms, correcting their mistakes and providing them with clues to find the right answer. In an eMathTeacher-compliant tool, algorithm simulation does not continue until the user enters the correct answer. GRAPHs is an extendible environment designed for active and independent visual simulation-based learning of graph algorithms, set up to integrate tools to help the user simulate the execution of different algorithms. Apart from the options of creating and editing the graph, and visualizing the changes made to the graph during simulation, the environment also includes step-by-step correction, algorithm pseudo-code animation, pop-up questions, data structure handling and XML-based interaction log creation features. On the other hand, assessment is a key part of any learning process. Through the use of e-learning environments huge amounts of data can be output about this process. Nevertheless, this information has to be interpreted and represented in a practical way to arrive at a sound assessment that is not confined to merely counting mistakes. This includes establishing relationships between the available data and also providing instructive linguistic descriptions about learning evolution. Additionally, formative assessment should specify the level of attainment of the learning goals defined by the instructor. Till now, only human experts were capable of making such assessments. While facing this problem, our goal has been to create a computational model that simulates the instructor’s reasoning and generates an enlightening learning evolution report in natural language. The second part of this thesis presents the granular linguistic model of learning assessment to model the assessment of the learning process and implement the automated generation of a formative assessment report. The model is a particularization of the granular linguistic model of a phenomenon (GLMP) paradigm, based on fuzzy logic and the computational theory of perceptions, to the assessment phenomenon. This technique, useful for implementing complex assessment criteria using inference systems based on linguistic rules, has been applied to two particular cases: the assessment of the interaction logs generated by GRAPHs and the criterion-based assessment of Moodle quizzes. As a consequence, several expert systems to assess different algorithm simulations and Moodle quizzes have been implemented, tested and used in the classroom. Apart from the grade, the designed expert systems also generate natural language progress reports on the achieved proficiency level, based exclusively on the objective data gathered from correct and incorrect responses. In addition, two applications, capable of being configured to implement the expert systems, have been developed. One is geared up to process the files output by GRAPHs and the other one is a Moodle plug-in set up to perform the assessment based on the quizzes results.

GESTÃO DO CONHECIMENTO: UM ESTUDO SOBRE A RELAÇÃO ENTRE A LIDERANÇA TRANSFORMACIONAL, A CONVERSÃO DO CONHECIMENTO E A EFICÁCIA ORGANIZACIONAL

Este estudo teve como objetivo principal analisar a relação entre a Liderança Transformacional, a Conversão do Conhecimento e a Eficácia Organizacional. Foram considerados como pressupostos teóricos conceitos consolidados sobre os temas desta relação, além de recentes pesquisas já realizadas em outros países e contextos organizacionais. Com base nisto identificou-se potencial estudo de um modelo que relacionasse estes três conceitos. Para tal considera-se que as organizações que buscam atingir Vantagem Competitiva e incorporam a Knowledge-Based View possam conquistar diferenciação frente a seus concorrentes. Nesse contexto o conhecimento ganha maior destaque e papel protagonista nestas organizações. Dessa forma criar conhecimento através de seus colaboradores, passa a ser um dos desafios dessas organizações ao passo que sugere melhoria de seus indicadores Econômicos, Sociais, Sistêmicos e Políticos, o que se define por Eficácia Organizacional. Portanto os modos de conversão do conhecimento nas organizações, demonstram relevância, uma vez que se cria e se converte conhecimentos através da interação entre o conhecimento existente de seus colaboradores. Essa conversão do conhecimento ou modelo SECI possui quatro modos que são a Socialização, Externalização, Combinação e Internalização. Nessa perspectiva a liderança nas organizações apresenta-se como um elemento capaz de influenciar seus colaboradores, propiciando maior dinâmica ao modelo SECI de conversão do conhecimento. Se identifica então na liderança do tipo Transformacional, características que possam influenciar colaboradores e entende-se que esta relação entre a Liderança Transformacional e a Conversão do Conhecimento possa ter influência positiva nos indicadores da Eficácia Organizacional. Dessa forma esta pesquisa buscou analisar um modelo que explorasse essa relação entre a liderança do tipo Transformacional, a Conversão do Conhecimento (SECI) e a Eficácia Organizacional. Esta pesquisa teve o caráter quantitativo com coleta de dados através do método survey, obtendo um total de 230 respondentes válidos de diferentes organizações. O instrumento de coleta de dados foi composto por afirmativas relativas ao modelo de relação pesquisado com um total de 44 itens. O perfil de respondentes concentrou-se entre 30 e 39 anos de idade, com a predominância de organizações privadas e de departamentos de TI/Telecom, Docência e Recursos Humanos respectivamente. O tratamento dos dados foi através da Análise Fatorial Exploratória e Modelagem de Equações Estruturais via Partial Least Square Path Modeling (PLS-PM). Como resultado da análise desta pesquisa, as hipóteses puderam ser confirmadas, concluindo que a Liderança Transformacional apresenta influência positiva nos modos de Conversão do Conhecimento e que; a Conversão do Conhecimento influencia positivamente na Eficácia Organizacional. Ainda, concluiu-se que a percepção entre os respondentes não apresenta resultado diferente sobre o modelo desta pesquisa entre quem possui ou não função de liderança.

The vibrational energy flow transition in organic molecules: Theory meets experiment

Most large dynamical systems are thought to have ergodic dynamics, whereas small systems may not have free interchange of energy between degrees of freedom. This assumption is made in many areas of chemistry and physics, ranging from nuclei to reacting molecules and on to quantum dots. We examine the transition to facile vibrational energy flow in a large set of organic molecules as molecular size is increased. Both analytical and computational results based on local random matrix models describe the transition to unrestricted vibrational energy flow in these molecules. In particular, the models connect the number of states participating in intramolecular energy flow to simple molecular properties such as the molecular size and the distribution of vibrational frequencies. The transition itself is governed by a local anharmonic coupling strength and a local state density. The theoretical results for the transition characteristics compare well with those implied by experimental measurements using IR fluorescence spectroscopy of dilution factors reported by Stewart and McDonald [Stewart, G. M. & McDonald, J. D. (1983) J. Chem. Phys. 78, 3907–3915].

Theory for normal and impaired experience-dependent plasticity in neocortex of adult rats

We model experience-dependent plasticity in the cortical representation of whiskers (the barrel cortex) in normal adult rats, and in adult rats that were prenatally exposed to alcohol. Prenatal exposure to alcohol (PAE) caused marked deficits in experience-dependent plasticity in a cortical barrel-column. Cortical plasticity was induced by trimming all whiskers on one side of the face except two. This manipulation produces high activity from the intact whiskers that contrasts with low activity from the cut whiskers while avoiding any nerve damage. By a computational model, we show that the evolution of neuronal responses in a single barrel-column after this sensory bias is consistent with the synaptic modifications that follow the rules of the Bienenstock, Cooper, and Munro (BCM) theory. The BCM theory postulates that a neuron possesses a moving synaptic modification threshold, θM, that dictates whether the neuron's activity at any given instant will lead to strengthening or weakening of its input synapses. The current value of θM changes proportionally to the square of the neuron's activity averaged over some recent past. In the model of alcohol impaired cortex, the effective θM has been set to a level unattainable by the depressed levels of cortical activity leading to “impaired” synaptic plasticity that is consistent with experimental findings. Based on experimental and computational results, we discuss how elevated θM may be related to (i) reduced levels of neurotransmitters modulating plasticity, (ii) abnormally low expression of N-methyl-d-aspartate receptors (NMDARs), and (iii) the membrane translocation of Ca2+/calmodulin-dependent protein kinase II (CaMKII) in adult rat cortex subjected to prenatal alcohol exposure.

Computational method to reduce the search space for directed protein evolution

We introduce a computational method to optimize the in vitro evolution of proteins. Simulating evolution with a simple model that statistically describes the fitness landscape, we find that beneficial mutations tend to occur at amino acid positions that are tolerant to substitutions, in the limit of small libraries and low mutation rates. We transform this observation into a design strategy by applying mean-field theory to a structure-based computational model to calculate each residue's structural tolerance. Thermostabilizing and activity-increasing mutations accumulated during the experimental directed evolution of subtilisin E and T4 lysozyme are strongly directed to sites identified by using this computational approach. This method can be used to predict positions where mutations are likely to lead to improvement of specific protein properties.

Formation of temporal-feature maps by axonal propagation of synaptic learning

Computational maps are of central importance to a neuronal representation of the outside world. In a map, neighboring neurons respond to similar sensory features. A well studied example is the computational map of interaural time differences (ITDs), which is essential to sound localization in a variety of species and allows resolution of ITDs of the order of 10 μs. Nevertheless, it is unclear how such an orderly representation of temporal features arises. We address this problem by modeling the ontogenetic development of an ITD map in the laminar nucleus of the barn owl. We show how the owl's ITD map can emerge from a combined action of homosynaptic spike-based Hebbian learning and its propagation along the presynaptic axon. In spike-based Hebbian learning, synaptic strengths are modified according to the timing of pre- and postsynaptic action potentials. In unspecific axonal learning, a synapse's modification gives rise to a factor that propagates along the presynaptic axon and affects the properties of synapses at neighboring neurons. Our results indicate that both Hebbian learning and its presynaptic propagation are necessary for map formation in the laminar nucleus, but the latter can be orders of magnitude weaker than the former. We argue that the algorithm is important for the formation of computational maps, when, in particular, time plays a key role.

Forming classes by stimulus frequency: Behavior and theory

Visual classification is the way we relate to different images in our environment as if they were the same, while relating differently to other collections of stimuli (e.g., human vs. animal faces). It is still not clear, however, how the brain forms such classes, especially when introduced with new or changing environments. To isolate a perception-based mechanism underlying class representation, we studied unsupervised classification of an incoming stream of simple images. Classification patterns were clearly affected by stimulus frequency distribution, although subjects were unaware of this distribution. There was a common bias to locate class centers near the most frequent stimuli and their boundaries near the least frequent stimuli. Responses were also faster for more frequent stimuli. Using a minimal, biologically based neural-network model, we demonstrate that a simple, self-organizing representation mechanism based on overlapping tuning curves and slow Hebbian learning suffices to ensure classification. Combined behavioral and theoretical results predict large tuning overlap, implicating posterior infero-temporal cortex as a possible site of classification.

Stochastic game theory: For playing games, not just for doing theory

Recent theoretical advances have dramatically increased the relevance of game theory for predicting human behavior in interactive situations. By relaxing the classical assumptions of perfect rationality and perfect foresight, we obtain much improved explanations of initial decisions, dynamic patterns of learning and adjustment, and equilibrium steady-state distributions.

The neural development and organization of letter recognition: Evidence from functional neuroimaging, computational modeling, and behavioral studies

Although much of the brain’s functional organization is genetically predetermined, it appears that some noninnate functions can come to depend on dedicated and segregated neural tissue. In this paper, we describe a series of experiments that have investigated the neural development and organization of one such noninnate function: letter recognition. Functional neuroimaging demonstrates that letter and digit recognition depend on different neural substrates in some literate adults. How could the processing of two stimulus categories that are distinguished solely by cultural conventions become segregated in the brain? One possibility is that correlation-based learning in the brain leads to a spatial organization in cortex that reflects the temporal and spatial clustering of letters with letters in the environment. Simulations confirm that environmental co-occurrence does indeed lead to spatial localization in a neural network that uses correlation-based learning. Furthermore, behavioral studies confirm one critical prediction of this co-occurrence hypothesis, namely, that subjects exposed to a visual environment in which letters and digits occur together rather than separately (postal workers who process letters and digits together in Canadian postal codes) do indeed show less behavioral evidence for segregated letter and digit processing.

Image recognition: Visual grouping, recognition, and learning

Vision extracts useful information from images. Reconstructing the three-dimensional structure of our environment and recognizing the objects that populate it are among the most important functions of our visual system. Computer vision researchers study the computational principles of vision and aim at designing algorithms that reproduce these functions. Vision is difficult: the same scene may give rise to very different images depending on illumination and viewpoint. Typically, an astronomical number of hypotheses exist that in principle have to be analyzed to infer a correct scene description. Moreover, image information might be extracted at different levels of spatial and logical resolution dependent on the image processing task. Knowledge of the world allows the visual system to limit the amount of ambiguity and to greatly simplify visual computations. We discuss how simple properties of the world are captured by the Gestalt rules of grouping, how the visual system may learn and organize models of objects for recognition, and how one may control the complexity of the description that the visual system computes.

Educating a New Cadre of Experts Specializing in Digital Collections and Digital Curation: Experiential Learning in Digital Library Curriculum

Integration of experiential learning into the library and information science (LIS) courses has been a theme in LIS education, but the topic deserves renewed attention with an increasing demand for professionals in the digital library field and in light of the new initiative announced by the Library of Congress (LC) and the Institution of Museum and Library Services (IMLS) for national residency program in digital curation. The balance between theory and practice in digital library curricula, the challenges of incorporating practical projects into LIS coursework, and the current practice of teaching with hands on activities represent the primary areas of this panel discussion.

EmotiBlog: a finer-grained and more precise learning of subjectivity expression models

The exponential growth of the subjective information in the framework of the Web 2.0 has led to the need to create Natural Language Processing tools able to analyse and process such data for multiple practical applications. They require training on specifically annotated corpora, whose level of detail must be fine enough to capture the phenomena involved. This paper presents EmotiBlog – a fine-grained annotation scheme for subjectivity. We show the manner in which it is built and demonstrate the benefits it brings to the systems using it for training, through the experiments we carried out on opinion mining and emotion detection. We employ corpora of different textual genres –a set of annotated reported speech extracted from news articles, the set of news titles annotated with polarity and emotion from the SemEval 2007 (Task 14) and ISEAR, a corpus of real-life self-expressed emotion. We also show how the model built from the EmotiBlog annotations can be enhanced with external resources. The results demonstrate that EmotiBlog, through its structure and annotation paradigm, offers high quality training data for systems dealing both with opinion mining, as well as emotion detection.