Sökning: "clustering"

Visar resultat 1 - 5 av 537 uppsatser innehållade ordet clustering.

  1. 1. Deep Neural Network Compression for Object Detection and Uncertainty Quantification

    Master-uppsats, Göteborgs universitet/Institutionen för data- och informationsteknik

    Författare :Mohammad Ahraz Asif; Georgios Tzelepis; [2019-11-21]
    Nyckelord :Deep Learning; Model Compression; Knowledge Distillation; Bayesian Deep Learning; Object Detection;

    Sammanfattning : Neural networks have been notorious for being computational expensive. Their demand for hardware resources prohibits their extensive use in embedded devices and puts restrictions on tasks like real time tracking. LÄS MER

  2. 2. Analyzing the Usage of Traceability Links Within Open Source Software Projects

    Kandidat-uppsats, Göteborgs universitet/Institutionen för data- och informationsteknik

    Författare :Sebastian Fransson; Tim Jonasson; [2019-11-18]
    Nyckelord :Traceability links; Open source development; OSS; Modeling; Hierarchical clustering; Software maintenance;

    Sammanfattning : Traceability is important when it comes to many different software engineering activities. It is used as a means of enabling smoother maintenance and maneuvering within a repository. Because of its importance, traceability within com- mercial development has long been a subject of research and development. LÄS MER

  3. 3. Identifying Related Clinical Trials by Using Graph Database and Clustering

    Master-uppsats, Göteborgs universitet/Institutionen för data- och informationsteknik

    Författare :Razan Ghzouli; [2019-10-04]
    Nyckelord :clinical trials; clustering method;

    Sammanfattning : In this work we investigate two methods to nd similar clinical trials; creating a graph database to migrate clinical trials meta-data from relational database, and clustering clinical trials. We succeeded in identifying similar groups of clinical trials using the clustering method. LÄS MER

  4. 4. Predicting Mechanisms of Toxicity for Drug Development

    Master-uppsats, Göteborgs universitet/Institutionen för data- och informationsteknik

    Författare :SÖREN RICHARD STAHLSCHMIDT; [2019-10-04]
    Nyckelord :Predictive Toxicology; Metabolomics; Semi-Supervised Learning; DoseResponse;

    Sammanfattning : The aim of this thesis is to predict dierent mechanisms of toxicity from the metabolomic response of HepG2 liver cells. In order to utilize the metabolomic data the a semisupervised machine learning approach is investigated, namely the cluster-then-label approach. The research focuses on the unsupervised part due to the centrality to this method. LÄS MER

  5. 5. Supporting regional innovation through Clusters - A multiple-case study of five clusters in Innovation Leading + European Regions

    Master-uppsats, Göteborgs universitet/Graduate School

    Författare :Prodromos Iatridis; [2019-07-09]
    Nyckelord :Business clusters; Regional innovation performance; Innovation leading regions in Europe; Contribution of clusters; Cluster activities; Introducing innovation; Facilitate SMEs collaborations; Promote sales of new to market or new to firm innovations;

    Sammanfattning : Nowadays, emerging megatrends foster a fierce competition around the world. Authors are constantly studying the impacts of clustering on regional innovation and in turn economic growth and competitiveness of regions. LÄS MER