Sökning: "domain-transfer problem"

Hittade 2 uppsatser innehållade orden domain-transfer problem.

1. 1. All Negative on the Western Front: Analyzing the Sentiment of the Russian News Coverage of Sweden with Generic and Domain-Specific Multinomial Naive Bayes and Support Vector Machines Classifiers

   Kandidat-uppsats, Uppsala universitet/Institutionen för lingvistik och filologi

   Författare :David Michel; [2021]
   Nyckelord :sentiment analysis; news sentiment; text classification; cross-domain sentiment classification; domain specificity; domain-transfer problem; transfer learning; knowledge transfer; support vector machines; SVM; multinomial naive Bayes; Sweden; Aurora 17; Russia; Russian news; RT; Sputnik; cyberwarfare; influence campaign; disinformation; fake news; propaganda;

   Sammanfattning : This thesis explores to what extent Multinomial Naive Bayes (MNB) and Support Vector Machines (SVM) classifiers can be used to determine the polarity of news, specifically the news coverage of Sweden by the Russian state-funded news outlets RT and Sputnik. Three experiments are conducted. LÄS MER

3. 2. Data Augmentation to Improve Cross-Domain Generalization in Deep Learning MRI Segmentation

   Master-uppsats, Lunds universitet/Matematik LTH

   Författare :Rasmus Helander; [2021]
   Nyckelord :deep learning; medical imaging; mri; segmentation; data augmentation; cyclegan; noisy labels; Mathematics and Statistics;

   Sammanfattning : Semantic segmentation of medical images is an important task with many applications. However, manually delineating 3D images is time-consuming and the demand for automation is high. For many image segmentation tasks, deep learning has provided state-of-the-art results. LÄS MER