Le Center for Data Science de l’Université Paris-Saclay a déployé une plateforme compatible avec le Linked Data en 2016. Or, les chercheurs rencontrent face à ces technologies de nombreuses difficultés. Pour surmonter celles-ci, une approche et une plateforme appelée LinkedWiki, ont été conçues et expérimentées au-dessus du cloud de l’université (IAAS) pour permettre la création d’environnements virtuels de recherche (VRE) modulaires et compatibles avec le Linked Data. Nous avons ainsi pu proposer aux chercheurs une solution pour découvrir, produire et réutiliser les données de la recherche disponibles au sein du Linked Open Data, c’est-à-dire du système global d’information en train d’émerger à l’échelle du Web. Cette expérience nous a permis de montrer que l’utilisation opérationnelle du Linked Data au sein d’une université est parfaitement envisageable avec cette approche. Cependant, certains problèmes persistent, comme (i) le respect des protocoles du Linked Data et (ii) le manque d’outils adaptés pour interroger le Linked Open Data avec SPARQL. Nous proposons des solutions à ces deux problèmes. Afin de pouvoir vérifier le respect d’un protocole SPARQL au sein du Linked Data d’une université, nous avons créé l’indicateur SPARQL Score qui évalue la conformité des services SPARQL avant leur déploiement dans le système d’information de l’université. De plus, pour aider les chercheurs à interroger le LOD, nous avons implémenté le démonstrateur SPARQLets-Finder qui démontre qu’il est possible de faciliter la conception de requêtes SPARQL à l’aide d’outils d’autocomplétion sans connaissance préalable des schémas RDF au sein du LOD. / The Center for Data Science of the University of Paris-Saclay deployed a platform compatible with Linked Data in 2016. Because researchers face many difficulties utilizing these technologies, an approach and then a platform we call LinkedWiki were designed and tested over the university’s cloud (IAAS) to enable the creation of modular virtual search environments (VREs) compatible with Linked Data. We are thus able to offer researchers a means to discover, produce and reuse the research data available within the Linked Open Data, i.e., the global information system emerging at the scale of the internet. This experience enabled us to demonstrate that the operational use of Linked Data within a university is perfectly possible with this approach. However, some problems persist, such as (i) the respect of protocols and (ii) the lack of adapted tools to interrogate the Linked Open Data with SPARQL. We propose solutions to both these problems. In order to be able to verify the respect of a SPARQL protocol within the Linked Data of a university, we have created the SPARQL Score indicator which evaluates the compliance of the SPARQL services before their deployments in a university’s information system. In addition, to help researchers interrogate the LOD, we implemented a SPARQLets-Finder, a demonstrator which shows that it is possible to facilitate the design of SPARQL queries using autocompletion tools without prior knowledge of the RDF schemas within the LOD.

Automotive technologies and fully autonomous driving have seen a tremendous growth in recent times and have benefitted from extensive deep learning research. State-of-the-art deep learning methods are largely supervised and require labelled data for training. However, the annotation process for image data is time-consuming and costly in terms of human efforts. It is of interest to find informative samples for labelling by Content Based Image Retrieval (CBIR). Generally, a CBIR method takes a query image as input and returns a set of images that are semantically similar to the query image. The image retrieval is achieved by transforming images to feature representations in a latent space, where it is possible to reason about image similarity in terms of image content. In this thesis, a self-supervised method is developed to learn feature representations of road scenes images. The self-supervised method learns feature representations for images by adapting intermediate convolutional features from an existing deep Convolutional Neural Network (CNN). A contrastive approach based on Noise Contrastive Estimation (NCE) is used to train the feature learning model. For complex images like road scenes where mutiple image aspects can occur simultaneously, it is important to embed all the salient image aspects in the feature representation. To achieve this, the output feature representation is obtained as an ensemble of feature embeddings which are learned by focusing on different image aspects. An attention mechanism is incorporated to encourage each ensemble member to focus on different image aspects. For comparison, a self-supervised model without attention is considered and a simple dimensionality reduction approach using SVD is treated as the baseline. The methods are evaluated on nine different evaluation datasets using CBIR performance metrics. The datasets correspond to different image aspects and concern the images at different spatial levels - global, semi-global and local. The feature representations learned by self-supervised methods are shown to perform better than the SVD approach. Taking into account that no labelled data is required for training, learning representations for road scenes images using self-supervised methods appear to be a promising direction. Usage of multiple query images to emphasize a query intention is investigated and a clear improvement in CBIR performance is observed. It is inconclusive whether the addition of an attentive mechanism impacts CBIR performance. The attention method shows some positive signs based on qualitative analysis and also performs better than other methods for one of the evaluation datasets containing a local aspect. This method for learning feature representations is promising but requires further research involving more diverse and complex image aspects.

Manually reading, evaluating, and scoring motivation letters as part of the admissions process is a time-consuming and tedious task for Dalarna University's program managers. An automated scoring system would provide them with relief as well as the ability to make much faster decisions when selecting applicants for admission. The aim of this thesis was to analyse current human judgment and attempt to emulate it using machine learning techniques. We used various topic modelling methods, such as Latent Dirichlet Allocation and Non-Negative Matrix Factorization, to find the most interpretable topics, build a bridge between topics and human-defined factors, and finally evaluate model performance by predicting scoring values and finding accuracy using logistic regression, discriminant analysis, and other classification algorithms. Despite the fact that we were able to discover the meaning of almost all human factors on our own, the topic models' accuracy in predicting overall score was unexpectedly low. Setting a threshold on overall score to select applicants for admission yielded a good overall accuracy result, but did not yield a good consistent precision or recall score. During our investigation, we attempted to determine the possible causes of these unexpected results and discovered that not only is topic modelling limitation to blame, but human bias also plays a role.

The idea behind this report is based on the fact that it is not only the number of users in the traffic network that is increasing, the number of connected devices such as probe vehicles and mobile sources has increased dramatically in the last decade. These connected devices provide large-scale mobility data and new opportunities to analyze the current traffic situation as they traverse through the network and continuously send out different types of information like Global Positioning System (GPS) data and Mobile Network Data (MND). Travel demand is often described in terms of an Origin Destination (OD) matrix which represents the number of trips from an origin zone to a destination zone in a geographic area. The aim of this master thesis is to develop and evaluate a data-driven method for estimation of dynamic OD matrices using unsupervised learning, sensor fusion and large-scale mobility data. Traditionally, OD matrices are estimated based on travel surveys and link counts. The problem is that these sources of information do not provide the quality required for online control of the traffic network. A method consisting of an offline process and an online process has therefore been developed. The offline process utilizes historical large-scale mobility data to improve an inaccurate prior OD matrix. The online process utilizes the results and tuning parameters from the offline estimation in combination with real-time observations to describe the current traffic situation. A simulation study on a toy network with synthetic data was used to evaluate the data-driven estimation method. Observations based on GPS data, MND and link counts were simulated via a traffic simulation tool. The results showed that the sensor fusion algorithms Kalman filter and Kalman filter smoothing can be used when estimating dynamic OD matrices. The results also showed that the quality of the data sources used for the estimation is of high importance. Aggregating large-scale mobility data as GPS data and MND by using the unsupervised learning method Principal Component Analysis (PCA) improves the quality of the large-scale mobility data and so the estimation results. / <p>Examensarbetet är utfört vid Institutionen för teknik och naturvetenskap (ITN) vid Tekniska fakulteten, Linköpings universitet</p>

Fraud is a common crime within the insurance industry, and insurance companies want to quickly identify fraudulent claimants as they often result in higher premiums for honest customers. Due to the digital transformation where the sheer volume and complexity of available data has grown, manual fraud detection is no longer suitable. This work aims to automate the detection of fraudulent claimants and gain practical insights into fraudulent behavior using unsupervised anomaly detection, which, compared to supervised methods, allows for a more cost-efficient and practical application in the insurance industry. To obtain interpretable results and benefit from the temporal dependencies in human behavior, we propose two variations of LSTM based autoencoders to classify sequences of insurance claims. Autoencoders can provide feature importances that give insight into the models' predictions, which is essential when models are put to practice. This approach relies on the assumption that outliers in the data are fraudulent. The models were trained and evaluated on a dataset we engineered using data from a Swedish insurance company, where the few labeled frauds that existed were solely used for validation and testing. Experimental results show state-of-the-art performance, and further evaluation shows that the combination of autoencoders and LSTMs are efficient but have similar performance to the employed baselines. This thesis provides an entry point for interested practitioners to learn key aspects of anomaly detection within fraud detection by thoroughly discussing the subject at hand and the details of our work. / <p>Gjordes digitalt via Zoom. </p>

This thesis evaluates the performance of different machine learning approaches to log classification based on a dataset derived from simulating intrusive behavior towards an enterprise web application. The first experiment consists of performing attacks towards the web app in correlation with the logs to create a labeled dataset. The second experiment consists of one unsupervised model based on a variational autoencoder and four super- vised models based on both conventional feature-engineering techniques with deep neural networks and embedding-based feature techniques followed by long-short-term memory architectures and convolutional neural networks. With this dataset, the embedding-based approaches performed much better than the conventional one. The autoencoder did not perform well compared to the supervised models. To conclude, embedding-based ap- proaches show promise even on datasets with different characteristics compared to natural language.

Estimating depth and detection of object instances in 3D space is fundamental in autonomous navigation, localization, and mapping, robotic object manipulation, and augmented reality. RGB-D images and LiDAR point clouds are the most illustrative formats of depth information. However, depth sensors offer many shortcomings, such as low effective spatial resolutions and capturing of a scene from a single perspective. The thesis focuses on reproducing denser and comprehensive 3D scene structure for given monocular RGB images using depth and 3D object detection. The first contribution of this thesis is the pipeline for the depth estimation based on an unsupervised learning framework. This thesis proposes two architectures to analyze structure from motion and 3D geometric constraint methods. The proposed architectures trained and evaluated using only RGB images and no ground truth depth data. The architecture proposed in this thesis achieved better results than the state-of-the-art methods. The second contribution of this thesis is the application of the estimated depth map, which includes two algorithms: point cloud generation and collision avoidance. The predicted depth map and RGB image are used to generate the point cloud data using the proposed point cloud algorithm. The collision avoidance algorithm predicts the possibility of collision and provides the collision warning message based on decoding the color in the estimated depth map. This algorithm design is adaptable to different color map with slight changes and perceives collision information in the sequence of frames. Our third contribution is a two-stage pipeline to detect the 3D objects from a monocular image. The first stage pipeline used to detect the 2D objects and crop the patch of the image and the same provided as the input to the second stage. In the second stage, the 3D regression network train to estimate the 3D bounding boxes to the target objects. There are two architectures proposed for this 3D regression network model. This approach achieves better average precision than state-of-theart for truncation of 15% or fully visible objects and lowers but comparable results for truncation more than 30% or partly/fully occluded objects.

No description available.

The increase in demand for food has resulted in increased demand for tools that help streamline plant breeding process in order to create new varieties of crops. Identifying the underlying genetic mechanism of favourable characteristics is essential in order to make the best breeding decisions. In this project we have developed a modified autoencoder model which allows for lateral phenotype injection into the latent layer, in order to identify causal SNPs for phenotypes of interest in wheat. SNP and phenotype data for 500 samples of Lantmännen SW Seed provided by Lantmännen was used to train the network. Artificial phenotype created using a single SNP was used during training instead of real phenotype, since the relationship between the phenotype and SNP is already known. The modified training model with lateral phenotype injection showed significant increase in genotype concordance of the artificial phenotype when compared to the control model without phenotype injection. Causal SNP was successfully identified by using concordance terrain graph, where the difference in concordance of individual SNPs  between the modified modified model and control model was plotted against the genomic position of each SNP. The model requires further testing to elucidate its behaviour for phenotypes linked to multiple SNPs.

[pt] O Degree-Corrected Stochastic Block Model (DCSBM) é um modelo popular para geração de grafos aleatórios com estrutura de comunidade, dada uma sequência de graus esperados. O princípio básico de algoritmos que utilizam o DCSBM para detecção de comunidades é ajustar os parâmetros do modelo a dados observados, de forma a encontrar a estimativa de máxima verossimilhança, ou maximum likelihood estimate (MLE), dos parâmetros do modelo. O problema de otimização para o MLE é comumente resolvido por meio de heurísticas. Neste trabalho, propomos métodos de programação matemática, para resolver de forma exata o problema de otimização descrito, e comparamos os métodos propostos com heurísticas baseadas no algoritmo de expectation-maximization (EM). Métodos exatos são uma ferramenta fundamental para a avaliação de heurísticas, já que nos permitem identificar se uma solução heurística é sub-ótima e medir seu gap de otimalidade. / [en] The Degree-Corrected Stochastic Block Model (DCSBM) is a popular model to generate random graphs with community structure given an expected degree sequence. The standard approach of community detection algorithms based on the DCSBM is to search for the model parameters which are the most likely to have produced the observed network data, via maximum likelihood estimation (MLE). Current techniques for the MLE problem are heuristics and therefore do not guarantee convergence to the optimum. We present mathematical programming formulations and exact solution methods that can provably find the model parameters and community assignments of maximum likelihood given an observed graph. We compare the proposed exact methods with classical heuristic algorithms based on expectation-maximization (EM). The solutions given by exact methods give us a principled way of recognizing when heuristic solutions are sub-optimal and measuring how far they are from optimality.