In today’s society a large portion of the worlds population get their news on electronicdevices. This opens up the possibility to enhance their reading experience bypersonalizing news for the readers based on their previous preferences. We have conductedan experiment to find out how accurately a Naïve Bayes classifier can selectarticles that a user might find interesting. Our experiments was done on two userswho read and classified 200 articles as interesting or not interesting. Those articleswere divided into four datasets with the sizes 50, 100, 150 and 200. We used a NaïveBayes classifier with 16 different settings configurations to classify the articles intotwo categories. From these experiments we could find several settings configurationsthat showed good results. One settings configuration was chosen as a good generalsetting for this kind of problem. We found that for datasets with a size larger than 50there were no significant increase in classification confidence.

In this work we made an attempt to predict the upwards/downwards movement of the S&P 500 index from the news articles published by Bloomberg and Reuters. We employed the SVM classifier and conducted multiple experiments aiming at understanding the shape of the data and the specifics of the task better. As a result, we established the common evaluation settings for all our subsequent experiments. After that we tried incorporating various features into the model and also replicated several approaches previously suggested in the literature. We were able to identify some non-trivial dependencies in the data which helped us achieve a high accuracy on the development set. However, none of the models that we built showed comparable performance on the test set. We have come to the conclusion that whereas some trends or patterns can be identified in a particular dataset, such findings are usually barely transferable to other data. The experiments that we conducted support the idea that the stock market is changing at random and a high quality of prediction may only be achieved on particular sets of data and under very special settings, but not for the task of stock market prediction in general. 1

Denna studie syftade till att undersöka hur svenska kvällstidningar porträtterar manliga och kvinnliga våldsbrottsförövare. Vidare syftade studien till att undersöka mönster kring de svenska kvällstidningarnas språkbruk som appliceras när kvinnliga respektive manliga våldsbrottsförövare beskrivs. Data samlades in genom en strukturerad granskning av 114 nyhetsartiklar i Sveriges två största kvällstidningar. Därefter analyserades data genom en kvalitativ tematisk analys. Resultatet av studien visade att de svenska kvällstidningarna porträtterar kvinnliga och manliga våldsbrottsförövare på olika vis. Kvällstidningarna var mer benägna att använda ett mildare språkbruk och positivare begrepp när man beskrev kvinnliga förövare än när manliga förövare porträtterades. Slutsatserna av denna studie var att kvällstidningarna bidrar till att upprätthålla de könsnormer som råder i samhället, där män beskrivs som starka och försörjande medan kvinnor beskrivs som utsatta och omhändertagande, även när man beskriver personer som begått grova brott. Diskussionen förs kring vad dessa skillnader kan innebära ur ett kriminologiskt sammanhang. Bland annat angående hur kvällstidningar kan påverka samhällets uppfattningar om kvinnliga respektive manliga våldsbrottsförövare och vad denna porträttering kan medföra för konsekvenser i en kriminalpolitisk kontext / The purpose of this study is to highlight how the Swedish newspapers portray male and female violence offenders and to investigate the language used when describing and writing about male and female violence offenders. Data was gathered through a structured investigation of 114 articles from two of the biggest newspapers in Sweden. The method used was a qualitative theme analysis. The result of this study shows that Swedish newspapers portray male and female violence offenders differently. The Swedish newspapers were more likely to use a softer and more positive tone when describing female offenders compared to when describing male offenders. The conclusion of this study is that Swedish newspapers are entertaining the gender norms set by society, where men are commonly described as strong breadwinners and women described in terms of being vulnerable and caring, even when describing violence offenders. The discussion in this study revolves around the impact these differences can have from a criminology perspective. The study also discusses the potential impact and effect the newspapers can have on society’s understanding of female and male violence offenders

Large data sets are difficult to visualize. For a human to find structures and understand the data, good visualization tools are required. In this project a technique will be developed that makes it possible for a user to look at complex data at different scales. This technique is obvious when viewing geographical data where zooming in and out gives a good feeling for the spatial relationships in map data or satellite images. However, for other types of data it is not obvious how much scaling should be done. In this project, an experimental application is developed that visualizes data in multiple dimensions from a large news article database. Using this experimental application, the user can select multiple keywords on different axis and then can create a visualization containing news articles with those keywords. The user is able to move around the visualization. If the camera is far away from the document icons then they are clustered using red coloured spheres. If the user moves the camera closer to the clusters they will pop up into single document icons. If the camera is very close to the document icons it is possible to read the news articles

With the rise of internet literacy across the world, men and women on the street are increasingly participating in the news media more than ever before. Early speculations about the influence of citizen journalism imbued the practice with an almost messianic ability to save both journalism and democracy. Whilst these suggestions were influenced by a small amount of data analysis, mainly from Western countries, they were encouraging and demonstrated the potential of citizen journalism in representing the voice of ordinary people. This thesis suggests that citizen journalism is not only promoting the perspective of ordinary citizens, but is also supplementing the coverage of the mainstream media, building relationships, shaping the public sphere, and fulfilling the critical role of a watchdog. Analysing data from a sample of twenty-four different English language citizen journalism sites, this thesis examines the phenomenon of citizen journalism, focusing on the member countries of the South Asian Association for Regional Co-operation. Employing a mixed methods approach, quantitative and qualitative analyses were undertaken of the data set. The results show that citizen journalism sites in the larger and more developed SAARC countries provide more coverage of news than those in the smaller and underdeveloped countries. Political news is given the highest priority by the majority of the sites whilst news about war and terrorism is given the least. The analysis has also discovered that the sites function as a bridge, bringing people living in different parts of the world together and enabling them to engage in political discourse and the sharing of knowledge and experience. Moreover, citizen journalism is helping people to educate themselves about the culture and political systems of their new countries while also forming their own community online. This was particularly the case with the sites that were owned and operated by the diaspora people living in the West. In addition, with a few exceptions, the majority of the sites make substantial use of supplementary materials to enhance news articles, encouraging readers to participate in interactive news activities, such as posting comments. The study has also found that citizen journalists come from a wide range of backgrounds, from politicians acting as citizen journalists to students aspiring to generate revenues through commercial advertising on the Internet. However, they differ from each other in terms of their news values and news presentation — some of the sites offer more political news than others whilst others behave more like the mainstream media, providing a wide range of news articles. On the other hand, a few of the sites are less active and provide fewer news articles than others. The study has also found that citizen journalists from the SAARC countries include works of fiction as part of their news output, thus offering the slightly different definition of citizen journalism from that in the West.

Submitted by Morgana Silva (morgana_linhares@yahoo.com.br) on 2016-10-11T18:39:16Z No. of bitstreams: 1 arquivototal.pdf: 1029374 bytes, checksum: bb83210555827330fa8403557be7508e (MD5) / Made available in DSpace on 2016-10-11T18:39:16Z (GMT). No. of bitstreams: 1 arquivototal.pdf: 1029374 bytes, checksum: bb83210555827330fa8403557be7508e (MD5) Previous issue date: 2016-03-15 / This work is the analysis of adversative connectors which are present in the newspaper articles about soccer, in Pernambuco. To execute this, is has been used a corpus with data from the early XX century and the early XXI century. In all, sixty articles were analyzed, thirty for each time sample, in a total of a hundred and fifteen adversative items found. The study of function behavior of the items is observed by the application of functionalist theoretical support: markedness principle, iconicity principle and prototypes theory. We follow this theoretical perspective based on the studies of Givón (1990 and 1995), Hopper and Traugott (1993 and 2003), Neves (2000), Castilho and Elias (2012), Furtado da Cunha et al (2013). The analysis uses, in contrast, the consultation with traditional grammarians, as Almeida (1962), Melo (1978), Cunha (1986), Cegalla (1989), Bechara (2009), and others. Among the results, it has been verified that the item ‘mas’ has been presented as the prototypical of adversity, remaining as the most frequent in the two time samples of the research. This shows the prototypical value of the item in this function, although there has been insertion of it in contexts of innovative uses. Furthermore, it is observed that the traditional grammar hasn’t updated the list of adversative conjunctions once they haven’t adopted perspectives of observing the item in use situation. They have disregarded the language change process. / Este trabalho consiste na análise de conectores adversativos presentes em notícias jornalísticas sobre futebol no estado de Pernambuco. Para sua realização, constituímos um corpus com dados das primeiras décadas do século XX e das primeiras décadas do século XXI. Ao todo, analisamos sessenta notícias, sendo trinta de cada amostra, tendo sido identificados cento e quinze itens adversativos. O estudo do comportamento funcional dos itens parte da aplicação de suportes teóricos funcionalistas: princípio da marcação, princípio da iconicidade e teoria dos protótipos. Seguimos essa perspectiva teórica baseando-nos em estudos de Givón (1990; 1995), Hopper e Traugott (1993; 2003), Neves (2000), Castilho e Elias (2012), Furtado da Cunha et al. (2013). A análise utiliza como contraponto a consulta a gramáticos tradicionais, como Almeida (1962), Melo (1978), Cunha (1986), Cegalla (1989), Bechara (2009), dentre outros. Dentre os resultados alcançados, constatamos que o mas se apresenta como o item mais recorrente nos dois recortes da pesquisa, o que comprova seu valor de prototípico nessa função, embora registremos sua inserção em contextos de uso inovadores. Além disso, percebemos que a gramática tradicional não atualiza a lista de conjunções adversativas, uma vez que não adota perspectivas de observação dos itens em situação de uso, desconsiderando, assim, os processos de mudança que se concretizam com o passar do tempo.

Text categorisation is an important feature for organising text data and making it easier to find information on the world wide web.  The categorisation of text data can be done through the use of machine learning classifiers. These classifiers need to be trained with data in order to predict a result for future input. The authors chose to investigate how accurate two classifiers are when classifying recent news articles on a classifier model that is trained with older news articles. To reach a result the authors chose the Naive Bayes and Support Vector Machine classifiers and conducted an experiment. The experiment involved training models of both classifiers with news articles from 2005 and testing the models with news articles from 2005 and 2017 to compare the results. The results showed that both classifiers did considerably worse when classifying the news articles from 2017 compared to classifying the news articles from the same year as the training data.

The media plays a major role in today's society. What we do not experience ourselves, we can be told through the media. Seen from this, the purpose of this study is to investigate how immigrants are described in news articles in Swedish morning and evening newspapers during the period 2016-2021, with focus on the negative aspects. We will focus on answering which specific characteristics the term immigrant is associated with and in which perspectives it is written about immigrants. The study is based on a qualitative text analysis and the empirical material we have collected thus consists of 24 news articles.The articles were analyzed using post colonial theory, the dichotomy We and the Others and stereotypes.The results show that Swedish morning and evening newspapers describe immigrants in more negative than positive terms. There is also a continuous description of characteristics in the form of stereotypes. Newspapers portray, so to speak, an image of immigrants as different, deviant and foreign. At the same time, the results show that there is a marked difference between immigrants and Swedes as immigrants stand in sharp contrast to the so-called "Swedishness". The conclusions we can draw from the results is that the media differentiate between immigrants and ethnic Swedes with the help of different characteristics and perspectives to describe immigrants.

The amount of published news articles is steadily increasing, and readers are shifting toward online platforms because of the convenience and affordable technology costs (Shearer, 2021). Users have become more engaged with online news articles. This engagement creates a rich corpus, which makes it a powerful means to understand public opinion, emerging events, and their evolvement. Therefore, many organizations invest in mining this large-scale user-generated content to improve their products, services, and, more importantly, their decision-making process. Studying users’ reactions to online news is essential for social scientists, policymakers, and journalists. This type of engagement is an area of study introduced previously. In the statistical and machine learning community, many survey-based studies tried to understand the users’ behavior by characterizing and categorizing comments in online news. Some studies focus on mining user opinions from social media and online news comments. Other works look into bias in the news and its influence on user-generated content. At the same time, the social network community addresses the problem of mining large-scale online news from different angles. Some work focuses on constructing knowledge graphs from the text. Others focus on building high-level graphs, where nodes are users and posts or documents, and links represent the relationship between nodes. Another line of work looked into the word level of the text. They extracted entities and topics by combining Natural Language Processing and graph techniques. From a Machine Learning perspective, there are three main challenges in all these studies 1) jointly mining massive user-generated data, 2) from multiple sources and platforms, and 3) the unpredictable quality of user-generated content. To address these issues, we tackle the problem of jointly learning and mining valuable information from online news articles and user-generated content. We start by studying and understating the relationship between users’ comments and articles in online news. Where the focus is to understand the level of relevancy between articles and their comments, we labeled a few article-comment pairs in this work. We proposed BERTAC (Alshehri et al.,2021), a BERT-based model that jointly learns article-comment embeddings and infers the relevance class of comment. However, we found that the disagreement among annotators as a part of a human (expert) labeling process produces noisy labels, which affect the performance of supervised learning algorithms. On the other hand, working only with high agreement annotations introduces another challenge: the data imbalance problem (Alshehri et al., 2022). As in many machine learning problems, labeling a sufficient number of examples is costly and time-consuming. Therefore, we propose a framework for aligning comments and news articles under a constrained budget(Alshehri et al., 2023a). The proposed model considers the data imbalanced, where we have only a few examples from one class, in addition, it considers the degrees of annotator disagreement. Within the framework, we consider two solutions, 1) semi-automatic labeling based on human-AI collaboration and 2) synthetic data augmentation. Another critical aspect of mining news articles and user-generated content is understanding emerging events and their associated entities. However, this is challenging, especially with the massive growth of online articles and user-generated content across different platforms. Therefore, we proposed MultiLayerET (Alshehri et al., 2023b), a unified representation of online news articles and comments. This work highlights the relationship between entities and topics in news articles and user-generated content. It projects entities and topics as a multi-layer graph, which gives a high-level understanding of the story behind the large pile of the corpus. We showed that such graphs enrich the textual representation and enhance the model learning performance in many downstream applications, such as media bias classification and fake news detection. / Computer and Information Science

This thesis aims to conduct a critical discourse analysis (CDA) of news articles and tweets discussing the climate protest that occurred on October 14th, 2022, where activists threw tomato soup on Van Gogh’s “Sunflower” at the London National Museum. The purpose is to investigate the media logic and the underlying social and cultural factors of iconoclastic actions that shaped the media discourse surrounding the event. The research questions this thesis aims to investigate are: RQ1: How are news values emphasized in media coverage of the tomato soup incident, as reflected in both news articles and Twitter posts? RQ2: Through the lens of media logic, how do media discourses shape perceptions of the climate action? RQ3: ​​How do affiliations with culturally significant artifacts and the climate shape the discursive representations of the protest in news articles and tweets? The sample consists of 34 tweets and 15 news articles. It is analyzed according to Van Dijk’s and Faircough’s CDA frameworks and the concepts of mediatization, media logic, iconoclasm, affiliation (as defined by Stoler (2022)) and news values. The results of this study suggest that the event had significant news value due to the iconoclastic tactics. Further, the media logic is seemingly involved in shaping the news into a sensationalistic story with little focus on the cause of the action. Lastly, the discourses surrounding the event on the platforms can be discussed in light of affiliation to provide an understanding of the discourse.