The sensitivity of mapping methods to reference data quality:training supervised image classifications with imperfect reference data

The accuracy of a map is dependent on the reference dataset used in its construction. Classification analyses used in thematic mapping can, for example, be sensitive to a range of sampling and data quality concerns. With particular focus on the latter, the effects of reference data quality on land cover classifications from airborne thematic mapper data are explored. Variations in sampling intensity and effort are highlighted in a dataset that is widely used in mapping and modelling studies; these may need accounting for in analyses. The quality of the labelling in the reference dataset was also a key variable influencing mapping accuracy. Accuracy varied with the amount and nature of mislabelled training cases with the nature of the effects varying between classifiers. The largest impacts on accuracy occurred when mislabelling involved confusion between similar classes. Accuracy was also typically negatively related to the magnitude of mislabelled cases and the support vector machine (SVM), which has been claimed to be relatively insensitive to training data error, was the most sensitive of the set of classifiers investigated, with overall classification accuracy declining by 8% (significant at 95% level of confidence) with the use of a training set containing 20% mislabelled cases.

Detecting expressions of blame or praise in text

The growth of social networking platforms has drawn a lot of attentions to the need for social computing. Social computing utilises human insights for computational tasks as well as design of systems that support social behaviours and interactions. One of the key aspects of social computing is the ability to attribute responsibility such as blame or praise to social events. This ability helps an intelligent entity account and understand other intelligent entities’ social behaviours, and enriches both the social functionalities and cognitive aspects of intelligent agents. In this paper, we present an approach with a model for blame and praise detection in text. We build our model based on various theories of blame and include in our model features used by humans determining judgment such as moral agent causality, foreknowledge, intentionality and coercion. An annotated corpus has been created for the task of blame and praise detection from text. The experimental results show that while our model gives similar results compared to supervised classifiers on classifying text as blame, praise or others, it outperforms supervised classifiers on more finer-grained classification of determining the direction of blame and praise, i.e., self-blame, blame-others, self-praise or praise-others, despite not using labelled training data.

Dynamic Bayesian network for semantic place classification in mobile robotics

In this paper, the problem of semantic place categorization in mobile robotics is addressed by considering a time-based probabilistic approach called dynamic Bayesian mixture model (DBMM), which is an improved variation of the dynamic Bayesian network. More specifically, multi-class semantic classification is performed by a DBMM composed of a mixture of heterogeneous base classifiers, using geometrical features computed from 2D laserscanner data, where the sensor is mounted on-board a moving robot operating indoors. Besides its capability to combine different probabilistic classifiers, the DBMM approach also incorporates time-based (dynamic) inferences in the form of previous class-conditional probabilities and priors. Extensive experiments were carried out on publicly available benchmark datasets, highlighting the influence of the number of time-slices and the effect of additive smoothing on the classification performance of the proposed approach. Reported results, under different scenarios and conditions, show the effectiveness and competitive performance of the DBMM.

Adverse drug reaction classification with deep neural networks

We study the problem of detecting sentences describing adverse drug reactions (ADRs) and frame the problem as binary classification. We investigate different neural network (NN) architectures for ADR classification. In particular, we propose two new neural network models, Convolutional Recurrent Neural Network (CRNN) by concatenating convolutional neural networks with recurrent neural networks, and Convolutional Neural Network with Attention (CNNA) by adding attention weights into convolutional neural networks. We evaluate various NN architectures on a Twitter dataset containing informal language and an Adverse Drug Effects (ADE) dataset constructed by sampling from MEDLINE case reports. Experimental results show that all the NN architectures outperform the traditional maximum entropy classifiers trained from n-grams with different weighting strategies considerably on both datasets. On the Twitter dataset, all the NN architectures perform similarly. But on the ADE dataset, CNN performs better than other more complex CNN variants. Nevertheless, CNNA allows the visualisation of attention weights of words when making classification decisions and hence is more appropriate for the extraction of word subsequences describing ADRs.

Towards person-centered neuroimaging markers for resilience and vulnerability in Bipolar Disorder

Improved clinical care for Bipolar Disorder (BD) relies on the identification of diagnostic markers that can reliably detect disease-related signals in clinically heterogeneous populations. At the very least, diagnostic markers should be able to differentiate patients with BD from healthy individuals and from individuals at familial risk for BD who either remain well or develop other psychopathology, most commonly Major Depressive Disorder (MDD). These issues are particularly pertinent to the development of translational applications of neuroimaging as they represent challenges for which clinical observation alone is insufficient. We therefore applied pattern classification to task-based functional magnetic resonance imaging (fMRI) data of the n-back working memory task, to test their predictive value in differentiating patients with BD (n=30) from healthy individuals (n=30) and from patients' relatives who were either diagnosed with MDD (n=30) or were free of any personal lifetime history of psychopathology (n=30). Diagnostic stability in these groups was confirmed with 4-year prospective follow-up. Task-based activation patterns from the fMRI data were analyzed with Gaussian Process Classifiers (GPC), a machine learning approach to detecting multivariate patterns in neuroimaging datasets. Consistent significant classification results were only obtained using data from the 3-back versus 0-back contrast. Using contrast, patients with BD were correctly classified compared to unrelated healthy individuals with an accuracy of 83.5%, sensitivity of 84.6% and specificity of 92.3%. Classification accuracy, sensitivity and specificity when comparing patients with BD to their relatives with MDD, were respectively 73.1%, 53.9% and 94.5%. Classification accuracy, sensitivity and specificity when comparing patients with BD to their healthy relatives were respectively 81.8%, 72.7% and 90.9%. We show that significant individual classification can be achieved using whole brain pattern analysis of task-based working memory fMRI data. The high accuracy and specificity achieved by all three classifiers suggest that multivariate pattern recognition analyses can aid clinicians in the clinical care of BD in situations of true clinical uncertainty regarding the diagnosis and prognosis.

L’influence des capacités cognitives mâles et femelles sur le choix de partenaire chez le diamant mandarin (Taeniopygia guttata)

S’approvisionner en nourriture est essentiel à la survie et au succès reproducteur. Lorsque les animaux font face à des changements environnementaux brutaux, ils doivent s’ajuster rapidement à leur nouvel environnement et parfois même innover dans leur façon de s’approvisionner. Des processus comportementaux et cognitifs, tels que l’innovation et l’apprentissage, permettent aux animaux d'intégrer de nouveaux comportements à leur répertoire comportemental afin de s'adapter de façon optimale. Les performances cognitives varient entre les individus d’une même population et bien que des études récentes se soient intéressées aux causes de ce phénomène, de convaincantes évidences sont manquantes afin d’expliquer pourquoi ces variations sont maintenues. Au cours de ce mémoire, les questions des pressions de sélection s'exerçant sur les performances d’alimentation par une tâche motrice nouvelle sont abordées afin de mieux comprendre l'évolution des capacités cognitives au sein d'une population captive de diamants mandarins (Taeniopygia guttata). Nous avons tout d'abord testé si les femelles diamants mandarins modifient leurs préférences d'accouplement après avoir observé la performance d'alimentation par une tâche motrice nouvelle des mâles. Afin de déterminer si les femelles sont capables de discriminer entre les mâles sur la base de leur capacité cognitive, nous avons également évalué les performances d’apprentissage de chacune d’elles. En effet, des études ont suggéré qu’il peut être coûteux, spécialement en terme de temps, de discriminer entre des partenaires potentiels sur cette base. La généralisation d’une préférence pour un mâle performant à d’autres mâles possédant le même phénotype permettrait la réduction de ces coûts. Nous avons donc finalement testé si les femelles diamants mandarins peuvent généraliser leur préférence après avoir observé les performances d’alimentation pour une tâche motrice nouvelle d’un mâle. Nos résultats suggèrent que les femelles diamants mandarins ne peuvent évaluer les capacités cognitives d’un mâle par l’intermédiaire de traits indicateurs. Toutefois, nous avons démontré qu’une observation directe des performances d’alimentation d’un mâle guide le choix d’appariement des femelles. Également, nous avons montré que les femelles peuvent généraliser l’apparence du mâle le plus performant et utiliser cette information lors de l’évaluation de nouveaux mâles. La relation entre les performances cognitives et le choix de partenaire pourraient s’expliquer par exemple par une meilleure exploitation de l’habitat, mais nécessite des études plus approfondies.

La notion de l’explosion de la coopération et ses implications chez Kim Sterelny

Le concept de coopération est souvent utilisé dans le domaine de l’éthique et de la politique pour illustrer et comprendre l’alignement des comportements associatifs entre les êtres humains. En lien avec ce concept, notre recherche portera sur la première question de savoir si Kim Sterelny (2003) réussit à produire un modèle théorique permettant d’expliquer les origines et les mécanismes de la coopération humaine. Notre recherche portera aussi sur la deuxième question de savoir s’il arrive à se servir de ce modèle pour infirmer la thèse de la modularité massive. Ainsi, ce mémoire traitera successivement du problème de la coopération, de la théorie de la sélection de groupe, du déclencheur écologique de la coopération des hominidés, des notions de coalition, d’exécution et d’engagement et finalement de la thèse de la modularité massive. Par l’examen de ces sujets, nous souhaitons démontrer que Sterelny n’arrive qu’à fournir une esquisse probable des origines et du développement de la coopération humaine et que sa critique de la thèse de la modularité massive n’arrive pas à infirmer cette dernière.

É possível acompanhar o luto? Envolvimento associativo a famílias vulneráveis em face da perda de seres íntimos

Partindo da análise das lógicas de ação solidária, propõem-se com esta comunicação refletir sobre os princípios que poderão ter contribuído para uma alteração das sensibilidades e compaixões dos cidadãos relativamente aos quadros sociais do sofrimento humano, dando lugar a diferentes quadrantes de operações criticas na prossecução de um bem comum. O sentimento de vulnerabilidade, associado às “vítimas de luto”, poderá ser um dos fatores promotores de diferentes interpretações críticas e manifestações coletivas de indignação que é denunciada publicamente pelas associações do luto, originando controvérsias, disputas e conflitos. As controvérsias públicas, que diferentes gramáticas de motivação conduzem os atores a associar-se, em consequência do cruzamento das intenções individuais e coletivas, perseguem um fim comum sujeito a um acordo (umas vezes mais precário, outras vezes menos precário). O tipo de acordo e as modalidades de cooperação da ação são aspetos fundamentais para perceber, por um lado qual a gramática política em que se baseiam na generalidade as associações, que existem atualmente na sociedade portuguesa, e por outro lado, as novas práticas sociais por elas desenvolvidas enformadas pelos princípios da solidariedade e participação. Apresentar-se-ão alguns dados preliminares de um estudo, no sentido de compreender e explicar os diferentes regimes de envolvimento associativo em torno do luto, partindo do singular para o geral: o regime familiar, o regime de plano e o regime público, diferenciados em função do julgamento feito pelo atores em situação.

Experientialist Epistemology and Classification Theory: Embodied and Dimensional Classification

What theoretical framework can help in building, maintaining and evaluating networked knowledge organization resources? Specifically, what theoretical framework makes sense of the semantic prowess of ontologies and peer-to-peer sys- tems, and by extension aids in their building, maintenance, and evaluation? I posit that a theoretical work that weds both for- mal and associative (structural and interpretive) aspects of knowledge organization systems provides that framework. Here I lay out the terms and the intellectual constructs that serve as the foundation for investigative work into experientialist classifi- cation theory, a theoretical framework of embodied, infrastructural, and reified knowledge organization. I build on the inter- pretive work of scholars in information studies, cognitive semantics, sociology, and science studies. With the terms and the framework in place, I then outline classification theory s critiques of classificatory structures. In order to address these cri- tiques with an experientialist approach an experientialist semantics is offered as a design commitment for an example: metadata in peer-to-peer network knowledge organization structures.

Machine learning techniques for implicit interaction using mobile sensors

Interactions in mobile devices normally happen in an explicit manner, which means that they are initiated by the users. Yet, users are typically unaware that they also interact implicitly with their devices. For instance, our hand pose changes naturally when we type text messages. Whilst the touchscreen captures finger touches, hand movements during this interaction however are unused. If this implicit hand movement is observed, it can be used as additional information to support or to enhance the users’ text entry experience. This thesis investigates how implicit sensing can be used to improve existing, standard interaction technique qualities. In particular, this thesis looks into enhancing front-of-device interaction through back-of-device and hand movement implicit sensing. We propose the investigation through machine learning techniques. We look into problems on how sensor data via implicit sensing can be used to predict a certain aspect of an interaction. For instance, one of the questions that this thesis attempts to answer is whether hand movement during a touch targeting task correlates with the touch position. This is a complex relationship to understand but can be best explained through machine learning. Using machine learning as a tool, such correlation can be measured, quantified, understood and used to make predictions on future touch position. Furthermore, this thesis also evaluates the predictive power of the sensor data. We show this through a number of studies. In Chapter 5 we show that probabilistic modelling of sensor inputs and recorded touch locations can be used to predict the general area of future touches on touchscreen. In Chapter 7, using SVM classifiers, we show that data from implicit sensing from general mobile interactions is user-specific. This can be used to identify users implicitly. In Chapter 6, we also show that touch interaction errors can be detected from sensor data. In our experiment, we show that there are sufficient distinguishable patterns between normal interaction signals and signals that are strongly correlated with interaction error. In all studies, we show that performance gain can be achieved by combining sensor inputs.