This thesis explores the topic of how one can streamline a data journalists analytical workflow in a graph visualization environment. Interactive graph visualizations have been used recently by data journalists to investigate the biggest leaks of data in history. Graph visualizations empower users to find patterns in their connected data, and as the world continuously produces more data, the more important it becomes to make sense of it. The exploration was done by conducting semi-structured interviews with users, which illuminated three categories of insights called Graph Readability, Charts in Graphs and Temporality. Graph Readability was the category that were conceptualized and designed by integrating user research and data visualization best practises. The design process was concluded with a usability test with graph visualization developers, followed by a final iteration of the concept. The outcome resulted in a module that lets users simplify their graph and preserve information by aggregating nodes with similar attributes.

Now when information is abundant, practice of Data Journalism is quickly becoming a core technique of the 21st century newsrooms. The diploma thesis "Data Journalism aimed by Datablog IHNED.cz" is focused on introducing Data Journalism as a genre responding to the changes in the society and the technology development. The theoretical part concerns with the history of the subject, the workflow of Data Journalism in the newsrooms and the limitations; such as inappropriate business models, difficult process of collecting data or the lack of training. The main goal of the research is to map the work of the Czech data team in the IHNED.cz and compare the results with the work of teams in Australia. The research methods are combination of quantitative and qualitative analyses. The main source of information are interviews conducted with data journalists from the Czech Republic and Australia. On top of that the research gives deeper understanding of how the integration of data journalists in the newsrooms affect the way journalism can support the existence of media organizations and contribute to the public good. The last part offers predictions of the future of Data Journalism.

The WikiLeaks Afghanistan war logs contain more than 76 000 reports about fatalities and their circumstances in the US led Afghanistan war, covering the period from January 2004 to December 2009. In this paper we use those reports to build statistical models to help us understand the mortality rates associated with specific circumstances. We choose an approach that combines Latent Dirichlet Allocation (LDA) with negative binomial based recursive partitioning. LDA is used to process the natural language information contained in each report summary. We estimate latent topics and assign each report to one of them. These topics - in addition to other variables in the data set - subsequently serve as explanatory variables for modeling the number of fatalities of the civilian population, ISAF Forces, Anti-Coalition Forces and the Afghan National Police or military as well as the combined number of fatalities. Modeling is carried out with manifest mixtures of negative binomial distributions estimated with model-based recursive partitioning. For each group of fatalities, we identify segments with different mortality rates that correspond to a small number of topics and other explanatory variables as well as their interactions. Furthermore, we carve out the similarities between segments and connect them to stories that have been covered in the media. This provides an unprecedented description of the war in Afghanistan covered by the war logs. Additionally, our approach can serve as an example as to how modern statistical methods may lead to extra insight if applied to problems of data journalism. (author's abstract) / Series: Research Report Series / Department of Statistics and Mathematics

Preserving dynamic, born-digital data journalism requires more than web scraping, as news stories today are built from much more than just text and images. Data journalism projects like news applications, or “news apps,” are composed of a database, the data in the database, the graphical interface that appears in the browser, accompanying text, and often images, videos, audio, and other multimedia components. Existing Internet archiving methods are not sufficient to save these data journalism projects for the future. This paper summarizes the context and history of news apps archiving, and describes the development of a survey of news applications. This survey will be used to create a working list of news organizations that are producing data journalism in the United States and a better sense of how and where these projects are currently being stored. The results of the survey will inform recommendations and processes for archiving dynamic data journalism.

The thesis deals with the presence of the technology team in various types of newsrooms of investigative journalism. The aim of the work is to find out the degree of representation of the technological component in the Slovak and Czech newsrooms. In case of their absence to find out the work process of journalists in more technologically demanding tasks. The theoretical part introduces the basic terminology of this field and this part is based on the data from quantitative study from 2019 conducted under the auspices of the ICFJ. In the practical part, qualitative research will be carried out in the form of semi-structured interviews with representatives of selected newsrooms.

[pt] A transformação nos modos de produção impulsionada pela internet favorece ovos arranjos produtivos que impactam a centralidade das grandes organizações jornalísticas. A autoridade profissional, por sua vez, é posta em xeque em um ambiente de crescente desinformação instigado pelos algoritmos, que geram descrédito frente à mídia. Inspiradas na filosofia do movimento open source e amparadas no uso do big data, agências de jornalismo sem fins lucrativos expandem seu território nas plataformas virtuais com a adoção do jornalismo de dados abertos – que propomos como nomenclatura –, o que sugere o fortalecimento da credibilidade a partir da transparência. Pautadas pelo paradigma do interesse público, as agências The Bureau of Investigative Journalism, Gênero e Número e Agência Pública adotam esta prática. A partir de metodologia experimental, alicerçada em indicadores originalmente desenvolvidos, inferiu-se em que medida o jornalismo praticado pelas agências oferece transparência e possibilita averiguar o método empregado no que tange o rigor. Com fulcro na teoria do newsmaking, o modelo criado serviu como parâmetro para avaliação de atributos e observação de esforços na construção de narrativas confiáveis. Para realização da pesquisa, foram analisadas 135 reportagens e conduzidas dezesseis entrevistas em profundidade. A análise de conteúdo indicou quatro categorias: verificação como fator de precisão, rigor na coleta, valor documental e checagem ao alcance do leitor. A última traz, em seu bojo, a hipótese desta tese, ou seja, a verificação da qualidade com base nas fontes abertas, que foi mensurada a partir da denominada metodologia da transparência. O resultado revelou alta transparência para a Agência Pública, média transparência para a Gênero e Número e baixa transparência para o TBIJ. / [en] The internet-driven transformation in journalistic modes of production has favored new arrangements that impact the centrality of larger organizations. The authority of professionalism has been undermined in an environment of growing and proactive disinformation, driven by algorithms that disrupt the trustworthiness of established media. Meanwhile, inspired by the philosophy of the open-source movement and by the possibilities offered by big data, non-profit journalism agencies have also expanded their territory to digital platforms adopting open data journalism – which we propose as a nomenclature – as a means of reinforcing credibility through transparency. Guided by the paradigm of public interest, the agencies The Bureau of Investigative Journalism, Gênero e Número and Agência Pública are examples of organizations that have taken this road. Based on experimental methodology, anchored in indicators developed for the purposes of the research, we drew inferences on the extent to which these agencies offer transparency, leading to further findings about newsgathering methods based on the criteria regarding rigor and evidence. Supported by the newsmaking theory, the model served as a parameter for assessment of attributes and the observation of efforts to build reliable narratives. To carry out the study, 135 reporting fields were analyzed and six in-depth interviews were conducted. The content analysis indicated four categories: documental value; fact-checking within the readership. The latter takes us to the heart of the hypothesis, namely, the verification based on open sources, measured from what can be called methodology of transparency. The result revealed high transparency for Agência Pública, medium transparency for Gênero e Número and low transparency for TBIJ.

The diploma thesis titled Transformation of Journalistic Practices Relating to Advent of Data Journalism concerns the current situation in Czech data journalism and its devel- opment. It examines how the newsroom, the journalists and the readers cope with new technology, big data and the related advent of data journalism. The theoretical part of the text explains the relationship between classical journalism and data-driven journal- ism and compares the so-called "narrative" and "interactive" approaches as forms of data-driven journalism. Further it gives a deeper understanding of the relation between data journalism and the new concept of objectivity and it deals with the cultural, social and technical preconditions of data journalism. The thesis also describes the historical aspects of the topic comparing computer-assisted reporting to current practice. Sub- sequent chapters deal with the visualizations, infographics, amateur data journalism and open data which all play a key role in data-driven journalism. The practical part intro- duces the Czech data team. It was originally established within the publishing house Economia and then moved to the office of Czech Radio (Český rozhlas). The final parts discuss the cooperation model of the Czech data team, the use of visualization in their projects,...

Intresset för datajournalistik har ökat både på redaktionerna och inom forskningen det senaste decenniet i takt med att tillgången på data har ökat och därmed behovet av att kunna hitta nyheter och berätta med hjälp av data. De flesta tidigare studier om datajournalistik har utgjorts av intervjuer med verksamma datajournalister och bland annat undersökt olika ideal. Men det finns få kvantitativa studier av vad datajournalisterna faktiskt gör, och av svensk datajournalistik. Den här uppsatsen syftar till att med innehållsanalytiska metoder beskriva och analysera datajournalistik som publicerats i svenska medier de senaste fem åren, och som journalister själva lyfter fram som exempel på datajournalistik. Urvalet är 91 datajournalistiska projekt som lämnats in som bidrag till NODA (Nordic Data Journalism Awards) eller som publicerats i Facebookgruppen Datajournalistik mellan 2013 och 2017. Ur en genomgång av tidigare forskning om datajournalistik har fyra olika ideal identifierats: ett undersökande, ett vetenskapligt, ett entreprenöriellt och ett från hackerkulturen, vilka utgör grunden för en analysmodell. Dessutom undersöks vilka producenterna är, vilka ämnen som är vanligast, vilka metoder och datakällor som används och hur publiken inbjuds att ta del av data och att vara delaktiga utifrån teorier om interaktivitet. Resultatet visar att datajournalistiken i denna undersökning ofta bygger på unika data, det vill säga självständig research, och att det undersökande idealet är relativt starkt. Trots att datajournalistik sägs präglas av hackerkulturens ideal om samarbete och deltagande är det ofta få producenter och yrkesroller bakom varje publicering och publiken bjuds sällan in att delta i processen.

Among the technologies that have modified the digital journalism since its inception, there are two that can be highlighted: 1) the World Wide Web (Web), a network of digital documents that has being used as a platform to the practice of journalism on the Internet and that determined the three generations of digital journalism; and 2) the databases aggregate to the Web, that have become the main technology behind the structuring of journalistic products in the transition between the third and fourth generation of digital journalism. In 2001, the scientist Tim Berners-Lee, inventor of the web, published a paper with a proposal of an extension to this network, which was called the Semantic Web. The paper proposed a change in the concept of the current web: from the traditional network made of documents to a network made of data, plus the technical ability to represent real concepts, such as people, places and objects. A great advantage of this proposal is that computers would be able to understand the data and identify their meanings. With a semantic network, the information could be organized and managed more efficiently and in an automated way, and the connections between the data would be richer than the current hyperlinks between documents. The concept of the Semantic Web is still maturing, but it is currently possible to find digital products that implement this concept. This research aims to analyze two real cases that apply the concept of the Semantic Web in digital journalism, specifically in the organization and management of the newspaper reports. For the theoretical background of research, we conducted a literature review on digital journalism, paradigm of the Digital Journalism on Databases (JDBD) and how the standard technologies of the Semantic Web work, such as RDF and ontologies. This is an exploratory research and it uses the case study as a method. The cases are the site 'World Cup 2010 BBC' and the site 'BBC Wildlife'. The analysis was performed using eight categories applicable to the study of JDBD. Among the results, it is found that the Semantic Web improve some of the characteristics of JDBD, mainly due to the automation on management tasks. Moreover, it identified that automated interoperability was the more advantageous benefit of Semantic Web to both digital journalism cases, and that it can become a potential rupture if the Semantic Web project come to succeed. / Entre as tecnologias que transformaram o jornalismo digital desde o seu surgimento, destacam-se duas: a World Wide Web (web), rede de documentos digitais que serviu como plataforma à prática jornalística na internet e determinou as três fases evolutivas do jornalismo digital; e as bases de dados, que, agregadas à web, se tornaram a principal tecnologia estruturante dos produtos jornalísticos na fase de transição entre a terceira e a quarta geração do jornalismo digital. No ano de 2001, o cientista Tim Berners-Lee, inventor da web, publicou um artigo com a proposta de uma expansão para esta rede, a qual foi denominada Web Semântica. O artigo propunha uma mudança no conceito da web: da tradicional rede de documentos para uma rede de dados, com capacidade para representar conceitos reais, como pessoas, lugares e objetos. Um grande diferencial desta proposta é que os computadores teriam capacidade para interpretar tais dados e identificar seus significados. Em uma rede semântica, as informações poderiam ser organizadas e gerenciadas de forma mais eficiente e automatizada, e as conexões entre dados seriam mais ricas do que através dos atuais links entre documentos. O conceito de Web Semântica ainda está em fase de amadurecimento, mas já é possível encontrar em funcionamento produtos digitais que aplicam tal conceito. A proposta desta pesquisa é analisar dois casos que aplicam o conceito da Web Semântica no jornalismo digital, mais especificamente na organização e no gerenciamento das informações jornalísticas. Para o embasamento teórico da investigação, foi realizada uma revisão bibliográfica sobre o jornalismo digital, sobre o paradigma do Jornalismo Digital em Base de Dados (JDBD) e sobre o funcionamento das tecnologias empregadas na Web Semântica, tais como o RDF e as ontologias. A pesquisa apresenta caráter exploratório e emprega como estratégia de investigação o estudo de caso, especificamente dos sites BBC World Cup 2010 e BBC Wildlife. A análise foi realizada a partir de oito categorias aplicáveis ao estudo do JDBD. Entre os resultados, é constatado que a Web Semântica potencializa algumas das características do JDBD, principalmente devido à automatização. Além disso, foi identificado nos casos estudados que a interoperabilidade automatizada foi o benefício mais vantajoso da Web Semântica em relação às tecnologias até então utilizadas no jornalismo digital, e que pode se tornar uma ruptura caso o projeto de Web Semântica obtenha êxito.

[pt] Devido ao crescimento da produção e disponibilidade de dados, houve uma consequente necessidade de explorá-los em busca de entendê-los, comunicar as informações obtidas a partir deles e gerar novos conhecimentos. Diante deste cenário de produção de informação a partir do grande volume de dados surgiu o Jornalismo de Dados, campo que busca não apenas construir visualizações claras e eficientes, mas principalmente contar histórias baseadas em dados e fatos, de forma a facilitar a compreensão do público. Entretanto, há uma necessidade de desenvolver e melhorar as ferramentas de análise visual com suporte a narrativas. Neste documento, descrevemos a nossa proposta de pesquisa, que consiste na expansão da ferramenta de visualização de dados VisMaker, integrando a possibilidade de criação de data stories para as visualizações, em um novo módulo denominado StoryMaker. / [en] Due to the growth of data production and availability, there was a consequent need to explore them to understand them, communicate the information obtained from them, and generate novel knowledge. Faced with this scenario of production of information from the large volume of data, Data Journalism emerged as a field that seeks not only to include clear and efficient visualizations, but mainly to tell stories based on data and facts to facilitate the public’s understanding. However, there is a need to develop and improve visual analysis tools that support narratives. In this document, we describe our research proposal, which consists of expanding the data visualization tool VisMaker, integrating the possibility of creating data stories for visualizations, in a new module named StoryMaker.