Understanding and characterizing the processes driving social interactions is one of the fundamental problems in social network research. A particular instance of this problem, known as link prediction, has recently attracted considerable attention in various research communities. Link prediction has many important commercial applications, e.g., recommending friends in an online social network such as Facebook and suggesting interesting pins in a collection sharing network such as Pinterest. This work is focused on the temporal link prediction problem: Given a sequence of graph snapshots G1, · ··, Gt from time 1 to t, how do we predict links in future time t + 1? To perform link prediction in a network, one needs to construct models for link probabilities between pairs of nodes. A temporal latent space model is proposed that is built upon latent homophily assumption and temporal smoothness assumption. First, the proposed modeling allows to naturally incorporate the well-known homophily effect (birds of a feather flock together). Namely, each dimension of the latent space characterizes an unobservable homogeneous attribute, and shared attributes tend to create a link in a network.

Submitted by Maicon Juliano Schmidt (maicons) on 2015-05-07T16:24:42Z No. of bitstreams: 1 João Henrique da Rosa.pdf: 3385861 bytes, checksum: f01fdf2a57917563fe8036952b98c2b8 (MD5) / Made available in DSpace on 2015-05-07T16:24:42Z (GMT). No. of bitstreams: 1 João Henrique da Rosa.pdf: 3385861 bytes, checksum: f01fdf2a57917563fe8036952b98c2b8 (MD5) Previous issue date: 2013-01-01 / CNPQ – Conselho Nacional de Desenvolvimento Científico e Tecnológico / Contexts prediction has been receiving considerable attention in the last years. Furthermore, this area seems to be the next logical step in context-aware computing, which, until a few years ago, had been concerned more with the present and the past temporal dimensions. There are many works regarding models for contexts prediction. Nevertheless, most of them employ the same algorithm for all cases. In other words, we did not find any approach that automatically decides the best prediction method according to the situation. Therefore, we propose the ORACON model. ORACON adapts itself in order to apply the best algorithm to the case. Moreover, the model supports other important aspects of ubicomp, such as, context formal representation and privacy. In this thesis, we describe the ORACON design and evaluate the model through two experiments, one using real data and the other employing simulated information.

Modern applications of mobile robots require them to have the ability to safely and effectively navigate in human environments. New challenges arise when these robots must plan their motion in a human-aware fashion. Current methods addressing this problem have focused mainly on the activity forecasting aspect, aiming at improving predictions without considering the active nature of the interaction, i.e. the robot’s effect on the environment and consequent issues such as reciprocity. Furthermore, many methods rely on computationally expensive offline training of predictive models that may not be well suited to rapidly evolving dynamic environments. This thesis presents a novel approach for enabling autonomous robots to navigate socially in environments with humans. Following formulations of the inverse planning problem, agents reason about the intentions of other agents and make predictions about their future interactive motion. A technique is proposed to implement counterfactual reasoning over a parametrised set of light-weight reciprocal motion models, thus making it more tractable to maintain beliefs over the future trajectories of other agents towards plausible goals. The speed of inference and the effectiveness of the algorithms is demonstrated via physical robot experiments, where computationally constrained robots navigate amongst humans in a distributed multi-sensor setup, able to infer other agents’ intentions as fast as 100ms after the first observation. While intention inference is a key aspect of successful human-robot interaction, executing any task requires planning that takes into account the predicted goals and trajectories of other agents, e.g., pedestrians. It is well known that robots demonstrate unwanted behaviours, such as freezing or becoming sluggishly responsive, when placed in dynamic and cluttered environments, due to the way in which safety margins according to simple heuristics end up covering the entire feasible space of motion. The presented approach makes more refined predictions about future movement, which enables robots to find collision-free paths quickly and efficiently. This thesis describes a novel technique for generating "interactive costmaps", a representation of the planner’s costs and rewards across time and space, providing an autonomous robot with the information required to navigate socially given the estimate of other agents’ intentions. This multi-layered costmap deters the robot from obstructing while encouraging social navigation respectful of other agents’ activity. Results show that this approach minimises collisions and near-collisions, minimises travel times for agents, and importantly offers the same computational cost as the most common costmap alternatives for navigation. A key part of the practical deployment of such technologies is their ease of implementation and configuration. Since every use case and environment is different and distinct, the presented methods use online adaptation to learn parameters of the navigating agents during runtime. Furthermore, this thesis includes a novel technique for allocating tasks in distributed robotics systems, where a tool is provided to maximise the performance on any distributed setup by automatic parameter tuning. All of these methods are implemented in ROS and distributed as open-source. The ultimate aim is to provide an accessible and efficient framework that may be seamlessly deployed on modern robots, enabling widespread use of intention prediction for interactive navigation in distributed robotic systems.

Ribonucleic acid (RNA) is an essential molecule, which carries out a wide variety of functions within the cell, from its crucial involvement in protein synthesis to catalysing biochemical reactions and regulating gene expression. Such diverse functional repertoire is indebted to complex structures that RNA can adopt and its flexibility as an interacting molecule. It has become possible to experimentally measure these two crucial aspects of RNA regulatory role with such technological advancements as next-generation sequencing (NGS). NGS methods can rapidly obtain the nucleotide sequence of many molecules in parallel. Designing experiments, where only the desired parts of the molecule (or specific parts of the transcriptome) are sequenced, allows to study various aspects of RNA biology. Analysis of NGS data is insurmountable without computational methods. One such experimental method is RNA structure probing, which aims to infer RNA structure from sequencing chemically altered transcripts. RNA structure probing data is inherently noisy, affected both by technological biases and the stochasticity of the underlying process. Most existing methods do not adequately address the issue of noise, resorting to heuristics and limiting the informativeness of their output. In this thesis, a statistical pipeline was developed for modelling RNA structure probing data, which explicitly captures biological variability, provides automated bias-correcting strategies, and generates a probabilistic output based on experimental measurements. The output of our method agrees with known RNA structures, can be used to constrain structure prediction algorithms, and remains robust to reduced sequence coverage, thereby increasing sensitivity of the technology. Another recent experimental innovation maps RNA-protein interactions at very high temporal resolution, making it possible to study rapid binding events happening on a minute time scale. In this thesis, a non-parametric algorithm was developed for identifying significant changes in RNA-protein binding time-series between different conditions. The method was applied to novel yeast RNA-protein binding time-course data to study the role of RNA degradation in stress response. It revealed pervasive changes in the binding to the transcriptome of the yeast transcription termination factor Nab3 and the cytoplasmic exoribonuclease Xrn1 under nutrient stress. This challenged the common assumption of viewing transcriptional changes as the major driver of changes in RNA expression during stress and highlighted the importance of degradation. These findings inspired a dynamical model for RNA expression, where transcription and degradation rates are modelled using RNA-protein binding time-series data.

<div> <div> <div> <p>Proteins are drivers of almost all biological processes in the cell. The functions of a protein are dependent on their three-dimensional structure and elucidating the structure and function of proteins is key to understanding how a biological system operates. In this research, we developed computational methods using machine learning techniques to predicts the structure and function of proteins. Protein 3D structure prediction has advanced significantly in recent years, largely due to deep learning approaches that predict inter-residue contacts and, more recently, distances using multiple sequence alignments (MSAs). The performance of these models depends on the number of similar protein sequences to the query protein, wherein some cases similar sequences are few but dissimilar sequences with local similarities are more and can be helpful. We have developed a novel deep learning-based approach AttentiveDist which further improves over the previous state of art. We added an attention mechanism where dis-similar sequences are also used (increasing number of sequences) and the model itself determines which information from such sequences it should attend to. We showed that the improvement of distance predictions was successfully transferred to achieve better protein tertiary structure modeling. We also show that structure prediction from a predicted distance map can be further enhanced by using predicted inter-residue sidechain center distances and main-chain hydrogen-bonds. Protein function prediction is another avenue we explored where we want to predict the function that a protein will perform. The crux of the approach is to predict the function of protein based on the function of similar sequences. Here, we developed a method where we use dissimilar sequences to extract additional information and improve performance over the previous approaches. We used phylogenetic analysis to determine if a dissimilar sequence can be close to the query sequence and thus can provide functional information. Our method was ranked highly in worldwide protein function prediction competition CAFA3 (2016-2019). Further, we expanded the method with a neural network to predict protein toxicity that can be used as a safety check for human-designed protein sequences.</p></div></div></div>

Reoccurrence of deformity can affect upwards of 64% of children with clubfoot. The ability to use foot function as a measure of reoccurrence has not been previously assessed. The purpose of this investigation was to utilize foot pressure analysis to predict the probability of reoccurrence in children with unilateral clubfoot. Retrospective foot pressure data revealed predictive algorithms detecting the probability of experiencing any type of reoccurrence (overall reoccurrence) and for experiencing a tibialis anterior tendon transfer (TATT). The equation for overall reoccurrence reported sensitivity and specificity of 0.82 and 0.81 and the equation for TATT reported values of 0.81 and 0.84. These algorithms were then applied prospectively to a cohort of children with unilateral clubfoot. Interim sensitivity and specificity results at a 1.5-year follow-up demonstrate that the equations for overall reoccurrence and TATT were highly specific but not sensitive (0.84, 0.73 specificity; 0.11, 0 sensitivity). This is an indication that these algorithms were more accurate when identifying the absence of reoccurrence. However, these results may change as the prospective subjects continue to age. Overall, the results of this investigation show that foot pressure analysis can predict the presence/absence of reoccurrence. The algorithms developed herein have the potential to improve long and short-term outcomes for children with clubfoot. Providing clinicians with the probability of reoccurrence will improve their ability to be proactive during the treatment decision making process.

Submitted by Geandra Rodrigues (geandrar@gmail.com) on 2018-07-20T12:18:30Z No. of bitstreams: 1 camilaacaciodepaiva.pdf: 9167548 bytes, checksum: 23bc00520b1453f40c0c4c739227c90b (MD5) / Approved for entry into archive by Adriana Oliveira (adriana.oliveira@ufjf.edu.br) on 2018-07-23T15:21:00Z (GMT) No. of bitstreams: 1 camilaacaciodepaiva.pdf: 9167548 bytes, checksum: 23bc00520b1453f40c0c4c739227c90b (MD5) / Made available in DSpace on 2018-07-23T15:21:00Z (GMT). No. of bitstreams: 1 camilaacaciodepaiva.pdf: 9167548 bytes, checksum: 23bc00520b1453f40c0c4c739227c90b (MD5) Previous issue date: 2018-06-27 / CAPES - Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / O software está cada vez mais presente no cotidiano das pessoas. Vários setores ou aspectos do ambiente são influenciados por ele. Desta forma, o desenvolvimento de software torna-se uma atividade crítica. Assim, o processo de teste se torna crucialmente importante, pois qualquer negligência pode refletir na qualidade do produto. Contudo, o cenário de desenvolvimento de software vem sofrendo mudanças a partir da necessidade de suprir demandas com maior agilidade e as exigências do mercado. É fundamental haver uma visão holística dos processos de desenvolvimento do software com o objetivo de gerar um ciclo de melhoria contínua. Tal visão é denominada como Engenharia de Software Contínua. A Engenharia de Software Contínua é caracterizada pelo uso do feedback de execuções para alcançar uma melhoria contínua e pela realização das atividades de maneira contínua. Feedback esse que pode ser fornecido através da proveniência de dados: descrição das origens de um dado e os processos pelos quais passou. Diante disso, este trabalho apresenta uma infraestrutura que tem como foco a captura e o armazenamento do histórico dos dados de execução do projeto, a previsão dos resultados dos testes de unidade através de algoritmos de previsão Logistic Regression, Naive Bayes e C4.5 Algorithm, e a disponibilização dos dados para aplicações externas. Além disso, oferece elementos de visualização que auxiliam na compreensão dos dados. Um experimento com dados de um projeto real foi realizado com o intuito de identificar a acurácia das previsões. / Software is increasingly present in people's daily lives. Various sectors and the environment are influenced by them. In this way, the development of software becomes a critical activity. Thus, the testing process becomes crucially important because any negligence can affect the quality of the product and the insecurity related to the use of the software. However, the software development scenario has undergone changes from the need to meet demands with greater agility and the demands of the market. It is fundamental to have a holistic view of the software development processes in order to generate a cycle of continuous improvement. Such a view is referred to as Continuous Software Engineering. Continuous Software Engineering is characterized by the use of feedback from executions to achieve continuous improvement and by performing activities on an ongoing basis. Feedback this can be provided through the provenance of data: description of the origins of a given and the processes by which it passed. This work presents an infrastructure that focuses on the capture and storage of the project execution data history, the prediction of the results of the unit tests through prediction algorithms: Logistic Regression, Naive Bayes and C4.5 Algorithm, and the provision of data for external applications. It also provides preview elements which help in understanding the data. An experiment with data from a real project was carried out in order to identify the accuracy of the prediction.

Feature selection or variable selection is an important step in different machine learning tasks. In a traditional approach, users specify the amount of features, which shall be selected. Afterwards, algorithm select features by using scores like the Joint Mutual Information (JMI). If users do not know the exact amount of features to select, they need to evaluate the full learning chain for different feature counts in order to determine, which amount leads to the lowest training error. To overcome this drawback, we extend the JMI score and mitigate the flaw by introducing a stopping criterion to the selection algorithm that can be specified depending on the learning task. With this, we enable developers to carry out the feature selection task before the actual learning is done. We call our new score Historical Joint Mutual Information (HJMI). Additionally, we compare our new algorithm, using the novel HJMI score, against traditional algorithms, which use the JMI score. With this, we demonstrate that the HJMI-based algorithm is able to automatically select a reasonable amount of features: Our approach delivers results as good as traditional approaches and sometimes even outperforms them, as it is not limited to a certain step size for feature evaluation.

Das drahtlose Sensornetzwerk (WSN) ist eine Technologie, die Umgebungsbedingungen oder physikalische Parameter misst, weiterleitet und per Fernüberwachung zur Verfügung stellt. Normalerweise werden die Sensorknoten, die diese Netzwerke bilden, von Batterien gespeist. Diese sollen aus verschiedenen Gründen nicht mehr verwendet werden, sondern es wird auf eine eigenständige Stromversorgung gesetzt. Dies soll den aufwendigen Austausch und die Wartung minimieren. Energy Harvesting kann mit den Knoten verwendet werden, um die Batterien zu unterstützen und die Lebensdauer der Netzwerke zu verlängern. Aufgrund der hohen Leistungsdichte der Solarenergie im Vergleich zu verschiedenen anderen Umweltenergien sind Solarzellen die am häufigsten eingesetzten Wandler, allerdings stellt die schwankende und intermittierende Natur der Solarenergie eine Herausforderung dar, einen funktionalen und zuverlässigen Sensorknoten zu versorgen. Um den Sensorknoten effektiv zu betreiben, sollte sein Energieverbrauch sinnvoll gesteuert werden. Ein interessanter Ansatz zu diesem Zweck ist die Steuerung der Aktivitäten des Knotens in Abhängigkeit von der zukünftig verfügbaren Energie. Dies erfordert eine Vorhersage der wandelbaren Sonnenenergie für die kommenden Betriebszeiten einschließlich der freien Zeiten der Sonne. Einige Vorhersagealgorithmen wurden mit stochastischen und statistischen Prinzipien sowie mit Methoden der künstlichen Intelligenz (KI) erstellt. Durch diese Algorithmen bleibt ein erheblicher Vorhersagefehler von 5-70%, der den zuverlässigen Betrieb der Knoten beeinträchtigt. Beispielsweise verwenden die stochastischen Methoden einen diskreten Energiezustand, der meist nicht zu den tatsächlichen Messwerten passt. Die statistischen Methoden verwenden einen Gewichtungsfaktor für die zuvor registrierten Messwerte. Daher sind sie nur geeignet, um Energieprofile bei konstanten Wetterbedingungen vorherzusagen. KI-Methoden erfordern große Beobachtungen im Trainingsprozess, die den benötigten Speicherplatz erhöhen. Dementsprechend ist die Leistung hinsichtlich der Vorhersagegenauigkeit dieser Algorithmen nicht ausreichend. In dieser Arbeit wird ein Vorhersagealgorithmus mit einem neuronalen Netzwerk entwickelt und eingebunden in einen Mikrocontroller, um die Verwaltung des Energieverbrauchs von solarzellengesteuerten Sensorknoten zu optimieren. Das verwendete neuronale Netzwerk wurde mit einer Kombination aus meteorologischen und statistischen Eingangsparametern realisiert. Dies hat zum Ziel, die erforderlichen Designkriterien für Sensorknoten zu erfüllen und eine Leistung zu erreichen, die in ihrer Genauigkeit die Leistung der oben genannten traditionellen Algorithmen übersteigt. Die Vorhersagegenauigkeit die durch den Korrelationskoeffizienten repräsentiert wird, wurde für das entwickelte neuronale Netzwerk auf 0,992 bestimmt. Das genaueste traditionelle Netzwerk erreicht nur einen Wert von 0,963. Das entwickelte neuronale Netzwerk wurde in einen Prototyp eines Sensorknotens integriert, um die Betriebszustände oder -modi über einen Simulationszeitraum von einer Woche anzupassen. Während dieser Zeit hat der Sensorknoten 6 Stunden zusätzlich im Normalbetrieb gearbeitet. Dies trug dazu bei, eine effektive Nutzung der verfügbaren Energie um ca. 3,6% besser zu erfüllen als das genaueste traditionelle Netz. Dadurch wird eine längere Lebensdauer und Zuverlässigkeit des Sensorknotens erreicht. / Wireless Sensor Network (WSN) is a technology that measures an environmental or physical parameters in order to use them by decision makers with a possibility of remote monitoring. Normally, sensor nodes that compose these networks are powered by batteries which are no longer feasible, especially when they used as fixed and standalone power source. This is due to the costly replacement and maintenance. Ambient energy harvesting systems can be used with these nodes to support the batteries and to prolong the lifetime of these networks. Due to the high power density of solar energy in comparison with different environmental energies, solar cells are the most utilized harvesting systems. Although that, the fluctuating and intermittent nature of solar energy causes a real challenge against fulfilling a functional and reliable sensor node. In order to operate the sensor node effectively, its energy consumption should be well managed. One interesting approach for this purpose is to control the future node’s activities according to the prospective energy available. This requires performing a prior prediction of the harvestable solar energy for the upcoming operation periods including the sun’s free times. A few prediction algorithms have been created using stochastic and statistical principles as well as artificial intelligence (AI) methods. A considerable prediction error of 5-70% is realized by these algorithms affecting the reliable operation of the nodes. For example, the stochastic ones use a discrete energy states which are mostly do not fit the actual readings. The statistical methods use a weighting factors for the previous registered readings. Thus, they are convenient only to predict energy profiles under consistent weather conditions. AI methods require large observations to be used in the training process which increase the memory space needed. Accordingly, the performance concerning the prediction accuracy of these algorithms is not sufficient. In this thesis, a prediction algorithm using a neural network has been proposed and implemented in a microcontroller for managing energy consumption of solar cell driven sensor nodes. The utilized neural network has been developed using a combination of meteorological and statistical input parameters. This is to meet a required design criteria for the sensor nodes and to fulfill a performance exceeds in its accuracy the performance of aforementioned traditional algorithms. The prediction accuracy represented by the correlation coefficient has been registered for the developed neural network to be 0.992, which increases the most accurate traditional network which has a value 0.963. The developed neural network has been embedded into a sensor node prototype to adjust the operating states or modes over a simulation period of one week. During this period, the sensor node has worked 6 hours more towards normal operation mode. This in its role helped to fulfill an effective use of available energy approximately 3.6% better than the most accurate traditional network. Thus, longer lifetime and more reliable sensor node.

This thesis has explored the development and potential effects of an intelligent decision support system (IDSS) to predict potential buyers for commercial real estate property. The overarching need for an IDSS of this type has been identified exists due to information overload, which the IDSS aims to reduce. By shortening the time needed to process data, time can be allocated to make sense of the environment with colleagues. The system architecture explored consisted of clustering commercial real estate buyers into groups based on their characteristics, and training a prediction model on historical transaction data from the Swedish market from the cadastral and land registration authority. The prediction model was trained to predict which out of the cluster groups most likely will buy a given property. For the clustering, three different clustering algorithms were used and evaluated, one density based, one centroid based and one hierarchical based. The best performing clustering model was the centroid based (K-means). For the predictions, three supervised Machine learning algorithms were used and evaluated. The different algorithms used were Naive Bayes, Random Forests and Support Vector Machines. The model based on Random Forests performed the best, with an accuracy of 99.9%. / Denna uppsats har undersökt utvecklingen av och potentiella effekter med ett intelligent beslutsstödssystem (IDSS) för att prediktera potentiella köpare av kommersiella fastigheter. Det övergripande behovet av ett sådant system har identifierats existerar på grund av informtaionsöverflöd, vilket systemet avser att reducera. Genom att förkorta bearbetningstiden av data kan tid allokeras till att skapa förståelse av omvärlden med kollegor. Systemarkitekturen som undersöktes bestod av att gruppera köpare av kommersiella fastigheter i kluster baserat på deras köparegenskaper, och sedan träna en prediktionsmodell på historiska transkationsdata från den svenska fastighetsmarknaden från Lantmäteriet. Prediktionsmodellen tränades på att prediktera vilken av grupperna som mest sannolikt kommer köpa en given fastighet. Tre olika klusteralgoritmer användes och utvärderades för grupperingen, en densitetsbaserad, en centroidbaserad och en hierarkiskt baserad. Den som presterade bäst var var den centroidbaserade (K-means). Tre övervakade maskininlärningsalgoritmer användes och utvärderades för prediktionerna. Dessa var Naive Bayes, Random Forests och Support Vector Machines. Modellen baserad p ̊a Random Forests presterade bäst, med en noggrannhet om 99,9%.