Master in Innovation and Research in Informatics - MIRI
Recent Submissions
-
Sleep scoring a neural network approach
(Universitat Politècnica de Catalunya, 2019-05)
Master thesis
Open Access
Covenantee: Georgia Institute of TechnologyIn this work, we explore two solutions using neural networks to solve this problem. The first one using only one EEG signal to automatically score the sleep session. In the second, we explore the possibility of doing all ... -
Material aging for game environments
(Universitat Politècnica de Catalunya, 2019-06)
Master thesis
Open AccessWe propose a software and work-flow to create 3D model imperfections and aging effects while adding very little overhead to the rendering time of the models. Specifically, we implement a tool that adds all these effects ... -
Monitoring the Quality of Information (QoI) for low-cost sensor networks
(Universitat Politècnica de Catalunya, 2019-04-25)
Master thesis
Open AccessMantaining a low cost sensor network calibrated for a certain time having trustable information is a highly complicated task. Thanks to redundant sensors in sensing devices and a sensor network, statistical tests can be ... -
Workflows management for HPC environments
(Universitat Politècnica de Catalunya, 2019-04-25)
Master thesis
Open AccessI present EVA, an extensible platform that allows the developer to build a numerical simulation kernel that can run on a heterogeneous HPC environment, hiding various problems associated with parallelism such as the passing ... -
A machine learning approach to stock screening with fundamental analysis
(Universitat Politècnica de Catalunya, 2019-04-15)
Master thesis
Open AccessWe present HPC.FASSR, a High-Performance Computation Fundamental Analysis Stock Screener and Ranker to compare many ML models and the criteria of famous Benjamin Graham for stock investing using fundamental data. FASSR is ... -
Scheduling policies for Big Data workflows
(Universitat Politècnica de Catalunya, 2019-04)
Master thesis
Open AccessThe aim of this master thesis is to both give the programmer some guidelines to achieve good scalabilities with tasked based programming models and to improve the COMPSs runtime scheduler the capabilities to reach this ... -
Wav2Pix Enhancement and evaluation of a speech-conditioned image generator
(Universitat Politècnica de Catalunya, 2019-01)
Master thesis
Restricted access - confidentiality agreementWe propose the enhancement and evaluation of a deep neural network that is trained from scratch in an end-to-end fashion, generating a face directly from the raw speech waveform without any additional identity information ... -
Multivariate statistical analysis of urban environments applied to New York City
(Universitat Politècnica de Catalunya, 2019-01-30)
Master thesis
Open AccessDense smart cities are complex environments, characterized by a heterogeneous set of co-varying signals, which constitute urban semantics. We conduct a multivariate analysis to visualize hidden data structure, and propose ... -
An architectural journey into RISC architectures for HPC workloads
(Universitat Politècnica de Catalunya, 2019-01-28)
Master thesis
Open Access
Covenantee: Barcelona Supercomputing CentreThe thesis evaluates the current state-of-the-art of RISC architectures in HPC. Studying the performance, power, and energy to solution in heterogeneous SoCs. For the evaluation 2 arm platforms (CPU+GPU, CPU+FPGA), 1 RISC-V ... -
Traffic forecasting in smart cities
(Universitat Politècnica de Catalunya, 2019-01-15)
Master thesis
Open AccessTraffic forecasting plays a key role in mitigating traffic congestion in urban areas. The main objective of this master thesis is to perform traffic forecasting in urban contexts by developing machine learning models trained ... -
Automatically extracting relations between clinical findingand treatment from clinical texts
(Universitat Politècnica de Catalunya, 2019-01)
Master thesis
Open AccessAutomatically extracted the relations between the clinical findings and treatments in EMRs. It is a process of detecting and classifying the semantic relation between entities in medical texts. We built a supervised SVM ... -
Predicting financial distress through machine learning
(Universitat Politècnica de Catalunya, 2019-01-19)
Master thesis
Open AccessThe main question which will be raised in this thesis is - whether we can predict future financial distress of a company based on the changes in historical financial results using different machine learning techniques.
RSS 1.0