The role of context in image annotation and recommendation

With the rise of smart phones, lifelogging devices (e.g. Google Glass) and popularity of image sharing websites (e.g. Flickr), users are capturing and sharing every aspect of their life online producing a wealth of visual content. Of these uploaded images, the majority are poorly annotated or exist in complete semantic isolation making the process of building retrieval systems difficult as one must firstly understand the meaning of an image in order to retrieve it. To alleviate this problem, many image sharing websites offer manual annotation tools which allow the user to “tag” their photos, however, these techniques are laborious and as a result have been poorly adopted; Sigurbjörnsson and van Zwol (2008) showed that 64% of images uploaded to Flickr are annotated with < 4 tags. Due to this, an entire body of research has focused on the automatic annotation of images (Hanbury, 2008; Smeulders et al., 2000; Zhang et al., 2012a) where one attempts to bridge the semantic gap between an image’s appearance and meaning e.g. the objects present. Despite two decades of research the semantic gap still largely exists and as a result automatic annotation models often offer unsatisfactory performance for industrial implementation. Further, these techniques can only annotate what they see, thus ignoring the “bigger picture” surrounding an image (e.g. its location, the event, the people present etc). Much work has therefore focused on building photo tag recommendation (PTR) methods which aid the user in the annotation process by suggesting tags related to those already present. These works have mainly focused on computing relationships between tags based on historical images e.g. that NY and timessquare co-exist in many images and are therefore highly correlated. However, tags are inherently noisy, sparse and ill-defined often resulting in poor PTR accuracy e.g. does NY refer to New York or New Year? This thesis proposes the exploitation of an image’s context which, unlike textual evidences, is always present, in order to alleviate this ambiguity in the tag recommendation process. Specifically we exploit the “what, who, where, when and how” of the image capture process in order to complement textual evidences in various photo tag recommendation and retrieval scenarios. In part II, we combine text, content-based (e.g. # of faces present) and contextual (e.g. day-of-the-week taken) signals for tag recommendation purposes, achieving up to a 75% improvement to precision@5 in comparison to a text-only TF-IDF baseline. We then consider external knowledge sources (i.e. Wikipedia & Twitter) as an alternative to (slower moving) Flickr in order to build recommendation models on, showing that similar accuracy could be achieved on these faster moving, yet entirely textual, datasets. In part II, we also highlight the merits of diversifying tag recommendation lists before discussing at length various problems with existing automatic image annotation and photo tag recommendation evaluation collections. In part III, we propose three new image retrieval scenarios, namely “visual event summarisation”, “image popularity prediction” and “lifelog summarisation”. In the first scenario, we attempt to produce a rank of relevant and diverse images for various news events by (i) removing irrelevant images such memes and visual duplicates (ii) before semantically clustering images based on the tweets in which they were originally posted. Using this approach, we were able to achieve over 50% precision for images in the top 5 ranks. In the second retrieval scenario, we show that by combining contextual and content-based features from images, we are able to predict if it will become “popular” (or not) with 74% accuracy, using an SVM classifier. Finally, in chapter 9 we employ blur detection and perceptual-hash clustering in order to remove noisy images from lifelogs, before combining visual and geo-temporal signals in order to capture a user’s “key moments” within their day. We believe that the results of this thesis show an important step towards building effective image retrieval models when there lacks sufficient textual content (i.e. a cold start).

Mesa redonda: Futuro del profesional en bibliotecología

Esta actividad tiene como propósito plantear soluciones a la situación profesional a un futuro de 10 años, al año 2005.Predecir el futuro siempre encierra muchos riesgos, pues todo alrededor está cambiando en forma acelerada. Ya no podemos saber con certeza que el conocimiento adquirido el año pasado pueda resolver un problema este año. No cabe duda que las bibliotecas se han ido transformando en nuestro medio y desde los últimos diez años. Estamos en un momento de transición al cual se debe responder efectivamente con los usuarios tradicionales y con los usuarios familiarizados con medios automatizados.

Improving drug discovery using a neural networks based parallel scoring function

Virtual Screening (VS) methods can considerably aid clinical research, predicting how ligands interact with drug targets. Most VS methods suppose a unique binding site for the target, but it has been demonstrated that diverse ligands interact with unrelated parts of the target and many VS methods do not take into account this relevant fact. This problem is circumvented by a novel VS methodology named BINDSURF that scans the whole protein surface to find new hotspots, where ligands might potentially interact with, and which is implemented in massively parallel Graphics Processing Units, allowing fast processing of large ligand databases. BINDSURF can thus be used in drug discovery, drug design, drug repurposing and therefore helps considerably in clinical research. However, the accuracy of most VS methods is constrained by limitations in the scoring function that describes biomolecular interactions, and even nowadays these uncertainties are not completely understood. In order to solve this problem, we propose a novel approach where neural networks are trained with databases of known active (drugs) and inactive compounds, and later used to improve VS predictions.

Algorithmique et complexité des systèmes à compteurs

Réalisé en cotutelle avec l'École normale supérieure de Cachan – Université Paris-Saclay

ADVERTISEMENT ALLOCATION AND TRUST MECHANISMS DESIGN IN SOCIAL NETWORKS

Social network sites (SNS), such as Facebook, Google+ and Twitter, have attracted hundreds of millions of users daily since their appearance. Within SNS, users connect to each other, express their identity, disseminate information and form cooperation by interacting with their connected peers. The increasing popularity and ubiquity of SNS usage and the invaluable user behaviors and connections give birth to many applications and business models. We look into several important problems within the social network ecosystem. The first one is the SNS advertisement allocation problem. The other two are related to trust mechanisms design in social network setting, including local trust inference and global trust evaluation. In SNS advertising, we study the problem of advertisement allocation from the ad platform's angle, and discuss its differences with the advertising model in the search engine setting. By leveraging the connection between social networks and hyperbolic geometry, we propose to solve the problem via approximation using hyperbolic embedding and convex optimization. A hyperbolic embedding method, \hcm, is designed for the SNS ad allocation problem, and several components are introduced to realize the optimization formulation. We show the advantages of our new approach in solving the problem compared to the baseline integer programming (IP) formulation. In studying the problem of trust mechanisms in social networks, we consider the existence of distrust (i.e. negative trust) relationships, and differentiate between the concept of local trust and global trust in social network setting. In the problem of local trust inference, we propose a 2-D trust model. Based on the model, we develop a semiring-based trust inference framework. In global trust evaluation, we consider a general setting with conflicting opinions, and propose a consensus-based approach to solve the complex problem in signed trust networks.

Consumption of sweetened beverages among schoolgoing children in a densely populated township in Lilongwe, Malawi

Background The growing global childhood obesity pandemic has not spared lowincome countries like Malawi, where 8% of children below the age of five years are overweight. Globally, regular consumption of sweetened beverages is implicated among the factors that fuel childhood obesity. Despite the growing problem, there are no local studies on any aspect of sweetened beverage consumption among children in Malawi that could help in guiding interventions and public health nutrition policies. Aim We aimed to assess sweetened beverage consumption among school-going children in Chilinde, a densely populated township in Lilongwe, the capital city of Malawi. Methods A total of 60 school-going children whose caregivers gave verbal consent were included, and a structured questionnaire was administered to the caregiver (or other knowledgeable and responsible member of the household) of each eligible child. Results Our results showed that 50 of the 60 children sampled were consuming a wide-range of sweetened beverages on a regular basis on any day of the week, mostly during meal times (n = 23), before going to school (n = 22), and after school (n = 19). One-third of the children were reportedly consuming up to 300 mL of several sweetened beverages per day. Conclusion Like in many countries around the world, consumption of sweetened beverages appears to be common among young school-going children in this urban setting in Malawi. As the country builds public health responses to the growing problem of non-communicable diseases, early preventive interventions among children should be given priority.

Integration of Social Concerns in Collaborative Logistics and Transportation Networks

Part 21: Mobility and Logistics

THE STOCHASTIC NAVIER STOKES EQUATIONS FOR HEAT CONDUCTING, COMPRESSIBLE FLUIDS

This dissertation is devoted to the equations of motion governing the evolution of a fluid or gas at the macroscopic scale. The classical model is a PDE description known as the Navier-Stokes equations. The behavior of solutions is notoriously complex, leading many in the scientific community to describe fluid mechanics using a statistical language. In the physics literature, this is often done in an ad-hoc manner with limited precision about the sense in which the randomness enters the evolution equation. The stochastic PDE community has begun proposing precise models, where a random perturbation appears explicitly in the evolution equation. Although this has been an active area of study in recent years, the existing literature is almost entirely devoted to incompressible fluids. The purpose of this thesis is to take a step forward in addressing this statistical perspective in the setting of compressible fluids. In particular, we study the well posedness for the corresponding system of Stochastic Navier Stokes equations, satisfied by the density, velocity, and temperature. The evolution of the momentum involves a random forcing which is Brownian in time and colored in space. We allow for multiplicative noise, meaning that spatial correlations may depend locally on the fluid variables. Our main result is a proof of global existence of weak martingale solutions to the Cauchy problem set within a bounded domain, emanating from large initial datum. The proof involves a mix of deterministic and stochastic analysis tools. Fundamentally, the approach is based on weak compactness techniques from the deterministic theory combined with martingale methods. Four layers of approximate stochastic PDE's are built and analyzed. A careful study of the probability laws of our approximating sequences is required. We prove appropriate tightness results and appeal to a recent generalization of the Skorohod theorem. This ultimately allows us to deduce analogues of the weak compactness tools of Lions and Feireisl, appropriately interpreted in the stochastic setting.

Sensitivity analysis in optimized parametric curve fitting

Purpose – Curve fitting from unordered noisy point samples is needed for surface reconstruction in many applications -- In the literature, several approaches have been proposed to solve this problem -- However, previous works lack formal characterization of the curve fitting problem and assessment on the effect of several parameters (i.e. scalars that remain constant in the optimization problem), such as control points number (m), curve degree (b), knot vector composition (U), norm degree (k), and point sample size (r) on the optimized curve reconstruction measured by a penalty function (f) -- The paper aims to discuss these issues -- Design/methodology/approach - A numerical sensitivity analysis of the effect of m, b, k and r on f and a characterization of the fitting procedure from the mathematical viewpoint are performed -- Also, the spectral (frequency) analysis of the derivative of the angle of the fitted curve with respect to u as a means to detect spurious curls and peaks is explored -- Findings - It is more effective to find optimum values for m than k or b in order to obtain good results because the topological faithfulness of the resulting curve strongly depends on m -- Furthermore, when an exaggerate number of control points is used the resulting curve presents spurious curls and peaks -- The authors were able to detect the presence of such spurious features with spectral analysis -- Also, the authors found that the method for curve fitting is robust to significant decimation of the point sample -- Research limitations/implications - The authors have addressed important voids of previous works in this field -- The authors determined, among the curve fitting parameters m, b and k, which of them influenced the most the results and how -- Also, the authors performed a characterization of the curve fitting problem from the optimization perspective -- And finally, the authors devised a method to detect spurious features in the fitting curve -- Practical implications – This paper provides a methodology to select the important tuning parameters in a formal manner -- Originality/value - Up to the best of the knowledge, no previous work has been conducted in the formal mathematical evaluation of the sensitivity of the goodness of the curve fit with respect to different possible tuning parameters (curve degree, number of control points, norm degree, etc.)

Ansiedade na infância

Dissertação para obtenção do grau de Mestre em Design da Comunicação, apresentada na Universidade de Lisboa - Faculdade de Arquitetura.

Exact solutions for the agricultural and the two-dimensional packing problems

Objectives and study method: The objective of this study is to develop exact algorithms that can be used as management tools for the agricultural production planning and to obtain exact solutions for two of the most well known twodimensional packing problems: the strip packing problem and the bin packing problem. For the agricultural production planning problem we propose a new hierarchical scheme of three stages to improve the current agricultural practices. The objective of the first stage is to delineate rectangular and homogeneous management zones into the farmer’s plots considering the physical and chemical soil properties. This is an important task because the soil properties directly affect the agricultural production planning. The methodology for this stage is based on a new method called “Positions and Covering” that first generates all the possible positions in which the plot can be delineated. Then, we use a mathematical model of linear programming to obtain the optimal physical and chemical management zone delineation of the plot. In the second stage the objective is to determine the optimal crop pattern that maximizes the farmer’s profit taken into account the previous management zones delineation. In this case, the crop pattern is affected by both management zones delineation, physical and chemical. A mixed integer linear programming is used to solve this stage. The objective of the last stage is to determine in real-time the amount of water to irrigate in each crop. This stage takes as input the solution of the crop planning stage, the atmospheric conditions (temperature, radiation, etc.), the humidity level in plots, and the physical management zones of plots, just to name a few. This procedure is made in real-time during each irrigation period. A linear programming is used to solve this problem. A breakthrough happen when we realize that we could propose some adaptations of the P&C methodology to obtain optimal solutions for the two-dimensional packing problem and the strip packing. We empirically show that our methodologies are efficient on instances based on real data for both problems: agricultural and two-dimensional packing problems. Contributions and conclusions: The exact algorithms showed in this study can be used in the making-decision support for agricultural planning and twodimensional packing problems. For the agricultural planning problem, we show that the implementation of the new hierarchical approach can improve the farmer profit between 5.27% until 8.21% through the optimization of the natural resources. An important characteristic of this problem is that the soil properties (physical and chemical) and the real-time factors (climate, humidity level, evapotranspiration, etc.) are incorporated. With respect to the two-dimensional packing problems, one of the main contributions of this study is the fact that we have demonstrate that many of the best solutions founded in literature by others approaches (heuristics approaches) are the optimal solutions. This is very important because some of these solutions were up to now not guarantee to be the optimal solutions.

Sistema supervisório para poços de petróleo baseados no método de elevação artificial Plunger Lift

The several existing methods for oil artificial lifting and the variety of automation equipment for these methods many times lead the supervisory systems to be dedicated to a unique method and/or to a unique manufacturer. To avoid this problem, it has been developed the supervisory system named SISAL, conceived to supervise wells with different lifting methods and different automation equipments. The SISAL system is working in several Brazilian states but, nowadays, it is only supervising rod pump-based wells. The objective of this work is the development of a supervision module to the plunger lift artificial lift method. The module will have the same characteristics of working with automation hardware of many manufacturers. The module will be integrated to the SISAL system, incorporating the capacity to supervise the plunger lift artificial lift method.

MobiMed: Framework for Rapid Application Development of Medical Mobile Apps

In the medical field images obtained from high definition cameras and other medical imaging systems are an integral part of medical diagnosis. The analysis of these images are usually performed by the physicians who sometimes need to spend long hours reviewing the images before they are able to come up with a diagnosis and then decide on the course of action. In this dissertation we present a framework for a computer-aided analysis of medical imagery via the use of an expert system. While this problem has been discussed before, we will consider a system based on mobile devices. Since the release of the iPhone on April 2003, the popularity of mobile devices has increased rapidly and our lives have become more reliant on them. This popularity and the ease of development of mobile applications has now made it possible to perform on these devices many of the image analyses that previously required a personal computer. All of this has opened the door to a whole new set of possibilities and freed the physicians from their reliance on their desktop machines. The approach proposed in this dissertation aims to capitalize on these new found opportunities by providing a framework for analysis of medical images that physicians can utilize from their mobile devices thus remove their reliance on desktop computers. We also provide an expert system to aid in the analysis and advice on the selection of medical procedure. Finally, we also allow for other mobile applications to be developed by providing a generic mobile application development framework that allows for access of other applications into the mobile domain. In this dissertation we outline our work leading towards development of the proposed methodology and the remaining work needed to find a solution to the problem. In order to make this difficult problem tractable, we divide the problem into three parts: the development user interface modeling language and tooling, the creation of a game development modeling language and tooling, and the development of a generic mobile application framework. In order to make this problem more manageable, we will narrow down the initial scope to the hair transplant, and glaucoma domains.

Distributed optimization in transportation and logistics networks

Many important problems in communication networks, transportation networks, and logistics networks are solved by the minimization of cost functions. In general, these can be complex optimization problems involving many variables. However, physicists noted that in a network, a node variable (such as the amount of resources of the nodes) is connected to a set of link variables (such as the flow connecting the node), and similarly each link variable is connected to a number of (usually two) node variables. This enables one to break the problem into local components, often arriving at distributive algorithms to solve the problems. Compared with centralized algorithms, distributed algorithms have the advantages of lower computational complexity, and lower communication overhead. Since they have a faster response to local changes of the environment, they are especially useful for networks with evolving conditions. This review will cover message-passing algorithms in applications such as resource allocation, transportation networks, facility location, traffic routing, and stability of power grids.

Etiologia da recessão gengival

Introdução: A Periodontologia é um ramo da Medicina Dentária que tem como objetivo manter o periodonto saudável. A recessão gengival tem vindo a ser estudada, tanto em populações com pobre controlo de placa bacteriana, quanto naquelas com boa Higiene Oral. Os médicos dentistas desconhecem ainda muitos dos aspetos da etiologia da recessão gengival e como tal, este assunto foi objeto de muitas conjeturas, nomeadamente a causa da mesma, sendo ainda mais importante, o controle deste problema. Sendo assim, persiste a confusão levantada por várias opiniões e pontos de vista contraditórios, sendo alvo desta dissertação. Objetivos: Esta dissertação tem como objetivo analisar, e verificar a Etiologia da Recessão Gengival, mais precisamente, a sua origem e os fatores que predispõem a mesma. Tendo sido assim, realizada uma revisão bibliográfica, de modo a verificar: quais as causas desta patologia e as suas limitações. Materiais e Métodos: Para a obtenção da informação necessária na realização da presente dissertação, foi efetuada uma pesquisa bibliográfica nas bases de dados da Pubmed, Scielo, o livro Tratado de Peridontia Clinica e Implantologia Oral, o Jornal da Associação Dentária Americana, o livro Peridontia Clinica, o livro de Histologia Básica e o livro Anatomia, Embriologia e Histologia Oral. Para tal, foi realizada a investigação através das seguintes palavras-chaves: “Etiology”, Gingival Recession”, “Periodontitis”, “Dental plaque”, e “Prevalence”. Conclusão: No trabalho realizado, é possível concluir que a Etiologia da recessão é multifatorial e raramente leva à perda do elemento dentário, embora, cause muitos danos por provocar a sensibilidade dentária, devido a perda e retração da gengiva. Esta etiologia leva a uma maior incidência de cáries radiculares, sacrificando o aspecto estético do paciente, e consequentemente, leva a um desconforto psicológico. Subsequentemente a recessão gengival e as suas múltiplas causas, existem atualmente métodos e técnicas, que nos permitem a resolução de alguns dos danos provocados por esta, com o objetivo de criar uma maior probabilidade de eliminação dos fatores causais da mesma.