Sökning: "news features"

Visar resultat 26 - 30 av 94 uppsatser innehållade orden news features.

1. 26. Exploring Media Panic Discourses: News Media Attitudes toward Digital Games in China

   Master-uppsats, Uppsala universitet/Medier och kommunikation

   Författare :Shi Erchen; [2021]
   Nyckelord :digital games; game addiction; news media; moral panic; media panic; content analysis;

   Sammanfattning : Previous research demonstrated the phenomenon of moral panics on “dangerous” games mostly from Western perspectives, regarding media violence and deviant behaviour. With the development of media technology, the term “media panic” has evolved from moral panic, representing the debates and fearful emotion from the public when a new media technology has been created. LÄS MER

3. 27. Predicting the Movement Direction of OMXS30 Stock Index Using XGBoost and Sentiment Analysis

   Kandidat-uppsats, Blekinge Tekniska Högskola/Institutionen för datavetenskap

   Författare :Podasca Elena; [2021]
   Nyckelord :Machine learning; XGBoost; Sentiment analysis; Stock market prediction; OMXS30;

   Sammanfattning : Background. Stock market prediction is an active yet challenging research area. A lot of effort has been put in by both academia and practitioners to produce accurate stock market predictions models, in the attempt to maximize investment objectives. Tree-based ensemble machine learning methods such as XGBoost have proven successful in practice. LÄS MER

4. 28. Automatic Categorization of News Articles With Contextualized Language Models

   Master-uppsats, Linköpings universitet/Artificiell intelligens och integrerade datorsystem

   Författare :Lukas Borggren; [2021]
   Nyckelord :Natural Language Processing; Text Classification; Hierarchical Classification; Hierarchical Multi-label Text Classification; Domain Specialization; Metadata Features; Model Compression; Quantization; Pruning; Machine Learning; Deep Learning; Contextualized Language Models; BERT; ELECTRA; News Media;

   Sammanfattning : This thesis investigates how pre-trained contextualized language models can be adapted for multi-label text classification of Swedish news articles. Various classifiers are built on pre-trained BERT and ELECTRA models, exploring global and local classifier approaches. LÄS MER

5. 29. Extracting Salient Named Entities from Financial News Articles

   Master-uppsats, Linköpings universitet/Institutionen för datavetenskap

   Författare :David Grönberg; [2021]
   Nyckelord :named entity recognition; named entity linking; natural language processing;

   Sammanfattning : This thesis explores approaches for extracting company mentions from financial newsarticles that carry a central role in the news. The thesis introduces the task of salient named entity extraction (SNEE): extract all salient named entity mentions in a text document. LÄS MER

7. 30. COVID-19 Pandemic: To Be in the Focus of Media : A Critical Discourse Analysis of Denmark and Sweden's News Media's Reporting on the Coronavirus Pandemic in 2020

   Magister-uppsats, Högskolan i Halmstad/Akademin för hälsa och välfärd

   Författare :Sofia Ta; [2021]
   Nyckelord :COVID-19; Critical discourse analysis; Health communication; News media;

   Sammanfattning : Aim and research questions: This study investigates how the corona pandemic COVID-19 has been noticed and presented in the news reporting of a Danish newspaper and a Swedish newspaper in their digital news feed. The study is based on the following questions: ● What discursive themes can be identified in Danish and Swedish news reporting on COVID-19 in 2020? ● What are linguistic features used in the headline of the articles in Danish and Swedishnews media? ● How can the Danish and Swedish media reports on COVID-19 be explained by discourse analysis and theoretical framework? Method and material: The study has used a qualitative method with a critical discourse analysis that uses Norman Fairclough's three-dimensional analysis model to analyze a total of 48 articles from Aftonbladet and Kristeligt Dagblad. LÄS MER