Sökning: "Hybrid Network"

Visar resultat 21 - 25 av 204 uppsatser innehållade orden Hybrid Network.

1. 21. Transformer-Based Multi-scale Technical Reports Analyser for Science Projects Cost Prediction

   Master-uppsats, KTH/Skolan för elektroteknik och datavetenskap (EECS)

   Författare :Thomas Bouquet; [2023]
   Nyckelord :Natural Language Processing; Transformers; Deep Learning; Cost Prediction; Traitement Automatique du Langage; Transformers; Apprentissage Profond; Prédiction de coûts; Behandling av naturligt språk; Transformers; Djupinlärning; Kostnadsförutsägelser;

   Sammanfattning : Intrinsic value prediction is a Natural Language Processing (NLP) problem consisting in determining a numerical value contained implicitly and non-trivially in a text. In this project, we introduce the SWORDSMAN model (Sentence and Word-level Oracle for Research Documents by Semantic Multi-scale ANalysis), a deep neural network architecture based on transformers whose goal is to predict the cost of research projects from the analysis of their abstract. LÄS MER

3. 22. Challenges in establishing flexible work arrangements: an exploratory study on managers' and employees' perspectives

   Magister-uppsats, Lunds universitet/Företagsekonomiska institutionen

   Författare :Liana Lammerant; Viet Tran; [2023]
   Nyckelord :flexible work arrangements; alternative work arrangements; new ways of working NWW ; hybrid work; COVID-19 pandemic; work policies; challenges; managers; employees; Sweden; Business and Economics;

   Sammanfattning : The increased adoption of flexible work arrangements and changes in the habits and behaviors of employees since the COVID-19 pandemic have prompted companies to adapt their policies to accommodate these developments in the conditions of work. Despite being pioneers in flexible work arrangements, companies in Sweden still need to continuously refine their policies to meet the changing needs of their managers and employees. LÄS MER

4. 23. Data-driven Fault Detection and Diagnosis inHardware-in-the-loop Environment

   Master-uppsats, Uppsala universitet/Institutionen för informationsteknologi

   Författare :Arya Farhang; [2023]
   Nyckelord :;

   Sammanfattning : The automotive industry has grown rapidly in the previous decades, and with the move toward electrification and driving assistance, the need for faster testing and development is rising.Model-based design(MBD) development cycles partly address the issues by allowing testing to happen before the physical product is created. LÄS MER

5. 24. Team Leadership in remote settings An Actor-Network Theory Analysis

   Master-uppsats, Göteborgs universitet/Graduate School

   Författare :Maria Gordou; Elisa Moreira Carqueijó; [2022-06-23]
   Nyckelord :team leadership; remote setting; practice; Actor-Network Theory; networks; actants; hybrid collectifs; translation; blackbox; presence;

   Sammanfattning : Team Leadership (TL) is a topic that has been widely studied throughout the years. Previous studies on TL were focused on the individual and collective perspective. In recent years, technologies were included in the collective perspectives of TL under a remote setting by with the formation of Virtual Teams (VTs). LÄS MER

7. 25. Establishment of a deep learning algorithm for dosimetry of radiopharmaceuticals

   Master-uppsats,

   Författare :Dennis Lipnicevic; [2022-01-13]
   Nyckelord :Medical physics; automated segmentation; dosimetry; kidney; nuclear medicine dosimetry; deep learning; convolutional neural network;

   Sammanfattning : Purpose: The aim of this study was to introduce a volumetric convolutional neural network for segmentation of the kidneys in SPECT images and to apply it in the dosimetry of radiopharmaceuticals of this organ, in order to decrease segmentation time and to standardize the segmentation of the kidneys. Method: Three networks were trained using two network architectures and a total of 216 retrospectively collected images from patients that underwent imaging procedures at Sahlgrenska University Hospital between 2009 and 2018. LÄS MER