What can text corpora tell us about society? How can automatic text analysis algorithms efficiently and reliably analyze the social processes revealed in language production? This work develops statistical text analyses of dynamic social and news media datasets to extract indicators of underlying social phenomena, and to reveal how social factors guide linguistic production. This is illustrated through three case studies: first, examining whether sentiment expressed in social media can track opinion polls on economic and political topics (Chapter 3); second, analyzing how novel online slang terms can be very specific to geographic and demographic communities, and how these social factors affect their transmission over time (Chapters 4 and 5); and third, automatically extracting political events from news articles, to assist analyses of the interactions of international actors over time (Chapter 6). We demonstrate a variety of computational, linguistic, and statistical tools that are employed for these analyses, and also contribute MiTextExplorer, an interactive system for exploratory analysis of text data against document covariates, whose design was informed by the experience of researching these and other similar works (Chapter 2). These case studies illustrate recurring themes toward developing text analysis as a social science methodology: computational and statistical complexity, and domain knowledge and linguistic assumptions.

O objetivo deste trabalho é analisar a atuação regional do Brasil através de uma análise de mídia. Em primeiro lugar, realizamos um debate teórico focado no conceito de liderança e ressaltamos que geralmente este termo é utilizado de forma errônea, sem levar em conta as distorções pertinentes entre diferentes subtipos. Posteriormente, para observar como a atuação brasileira é percebida na região, analisamos as notícias publicadas por cinco grandes jornais sul-americanos em relação ao Brasil. Através do estudo feito durante um ano, concluímos finalmente que este país é percebido como um líder situacional e instrumental. Para além disso, no caso da Argentina, a imagem brasileira é utilizada como ferramenta para tirar proveito nos debates domésticos, tanto por opositores assim como por membros do governo. / The objective of this work is to analyze the Brazilian regional operations through a media analysis. Firstly, we conducted a theoretical debate focused on the concept of leadership and stressed that this term is often used erroneously, without taking into account the relevant distortions between different subtypes. Subsequently, to observe how Brazil\'s performance is perceived in the region, we analyzed the articles published by five major South American newspapers in relation to Brazil. Through the study during a year, we finally concluded that this country is perceived as a situational and instrumental leader. Furthermore, in the case of Argentina, Brazilian image is used as a tool to take advantage in domestic debates, both by opponents as well as by government officials.

O objetivo deste trabalho é analisar a atuação regional do Brasil através de uma análise de mídia. Em primeiro lugar, realizamos um debate teórico focado no conceito de liderança e ressaltamos que geralmente este termo é utilizado de forma errônea, sem levar em conta as distorções pertinentes entre diferentes subtipos. Posteriormente, para observar como a atuação brasileira é percebida na região, analisamos as notícias publicadas por cinco grandes jornais sul-americanos em relação ao Brasil. Através do estudo feito durante um ano, concluímos finalmente que este país é percebido como um líder situacional e instrumental. Para além disso, no caso da Argentina, a imagem brasileira é utilizada como ferramenta para tirar proveito nos debates domésticos, tanto por opositores assim como por membros do governo. / The objective of this work is to analyze the Brazilian regional operations through a media analysis. Firstly, we conducted a theoretical debate focused on the concept of leadership and stressed that this term is often used erroneously, without taking into account the relevant distortions between different subtypes. Subsequently, to observe how Brazil\'s performance is perceived in the region, we analyzed the articles published by five major South American newspapers in relation to Brazil. Through the study during a year, we finally concluded that this country is perceived as a situational and instrumental leader. Furthermore, in the case of Argentina, Brazilian image is used as a tool to take advantage in domestic debates, both by opponents as well as by government officials.

This study ”What’s does the local news contain? - an analysis of a week’s local journalism on the web in Västerbotten, Sweden” examines the different discourses in the local journalism of newspapers and public service media. The aim is to find what makes the local news with the use of a quantitative content analysis combined with a Critical Discourse Analysis (CDA).The study uses a theoretical framework that includes theories surrounding agenda setting, the power of the media, local journalism, news values and web based journalism. For the quantitative content analysis the online text tool Textometrica was used.The analysis found that the local news mainly consisted of news regarding sports and negative news such as crime, accidents and court cases. It also found that the local news mainly focused on a small part of the municipality of Västerbotten in its news coverage, where the biggest city in the region was of the most interest. When other parts were covered it was mostly for sport events or negative news. The qualitative analysis also found that there was a high frequency of updates on the websites and non-local news that made it into the local news due to being sensational. A difference between the newspapers and the public service media was also found where the papers focused more on local debates and public service media wasn’t as tied to earlier theories regarding news values.The study could also state that the content of the local media during the week differed from what is said to be good journalism and also what is said to be the journalistic responsibility that citizens rely on to make democratic decisions.

Reconhecendo que o discurso é um dos meios pelos quais a política se materializa, essa tese tem o objetivo de responder a seguinte questão: há indícios suficientes para afirmar que a comunicação parlamentar no âmbito da Câmara dos Deputados é governada pela relação governo-oposição, assim como constatado em sua atuação no processo decisório? Para respondê-la, foi feito o uso de técnicas computacionais e modelos estatísticos com a aplicação do expressed agenda model para análise do conteúdo de mais de 127 mil pronunciamentos realizados por mais de 2 mil oradores diferentes ao longo de mais de 15 anos de atividade parlamentar na Câmara dos Deputados. Com a apresentação de evidências robustas sobre a frequência dos discursos proferidos no Pequeno Expediente, a estratégia de concentração temática dos Deputados Federais e a ênfase por eles atribuída às agendas econômica e social, a resposta é que não. Há, na atividade parlamentar, muito mais do que a tarefa de produzir e aprovar leis. As conclusões apresentadas aprimoram o conhecimento a respeito do comportamento parlamentar no âmbito da Câmara dos Deputados e indicam a influência de outras variáveis sobre sua atuação como, por exemplo: a ideologia, o gênero, senioridade e a sua popularidade. / Recognizing that parliamentary speeches are one way in which politics is realized, this thesis aims to answer the following question: is there sufficient evidence to say that the parliamentary communication within the Chamber of Deputies is governed by the government-opposition relation, following conclusions about the decision-making process? To answer this question, I apply the expressed agenda model, computational techniques and statistical models over 127,000 speeches made by more than 2,000 different speakers in more than 15 years of parlamentary activity in the Chamber of Deputies. With robust evidence on the frequency of the speeches, thematic concentration and the expressed priorities between the emphasis attributed to economic and social agendas, the answer is no. The findings presented improve the knowledge about Brazilian parliamentary behavior and indicate the influence of other variables on its activities, for example: ideology, gender, seniority and popularity.

Reconhecendo que o discurso é um dos meios pelos quais a política se materializa, essa tese tem o objetivo de responder a seguinte questão: há indícios suficientes para afirmar que a comunicação parlamentar no âmbito da Câmara dos Deputados é governada pela relação governo-oposição, assim como constatado em sua atuação no processo decisório? Para respondê-la, foi feito o uso de técnicas computacionais e modelos estatísticos com a aplicação do expressed agenda model para análise do conteúdo de mais de 127 mil pronunciamentos realizados por mais de 2 mil oradores diferentes ao longo de mais de 15 anos de atividade parlamentar na Câmara dos Deputados. Com a apresentação de evidências robustas sobre a frequência dos discursos proferidos no Pequeno Expediente, a estratégia de concentração temática dos Deputados Federais e a ênfase por eles atribuída às agendas econômica e social, a resposta é que não. Há, na atividade parlamentar, muito mais do que a tarefa de produzir e aprovar leis. As conclusões apresentadas aprimoram o conhecimento a respeito do comportamento parlamentar no âmbito da Câmara dos Deputados e indicam a influência de outras variáveis sobre sua atuação como, por exemplo: a ideologia, o gênero, senioridade e a sua popularidade. / Recognizing that parliamentary speeches are one way in which politics is realized, this thesis aims to answer the following question: is there sufficient evidence to say that the parliamentary communication within the Chamber of Deputies is governed by the government-opposition relation, following conclusions about the decision-making process? To answer this question, I apply the expressed agenda model, computational techniques and statistical models over 127,000 speeches made by more than 2,000 different speakers in more than 15 years of parlamentary activity in the Chamber of Deputies. With robust evidence on the frequency of the speeches, thematic concentration and the expressed priorities between the emphasis attributed to economic and social agendas, the answer is no. The findings presented improve the knowledge about Brazilian parliamentary behavior and indicate the influence of other variables on its activities, for example: ideology, gender, seniority and popularity.

The aim of the presented dissertation thesis is to find out the answer for the question of how it is possible to process and evaluate by computer methods of data coming from psychotherapy sessions using the method of Guided Affective Imagery. The Guided Affective Imagery is a relatively young psychotherapy approach based on the European tradition of psychodynamic psychology which presumes that in assignment relatively indefinite motive in imagination during day dream comes to manifestation of a client's unconsciousness contents and conflicts, defensive mechanisms, resistance, motives and fantasy. The biofeedback device by which was the rate of heartbeat used during the whole session and computer programs for quantitative text analysis were used for verifying the usage of computer methods for the sessions' analysis. The quantitative text analysis focused on identification of primarily and secondarily measure of the item process in the preliminary part of a session, during the phase of imagination and at the final part of the session by two code systems. In a certain way the results indicate complementary progress of changes in the primary and secondary process when the highest representation of secondary process expressions was identified in the introductory part of the session whereas the highest...

Welche Kommunikationsstrategien benutzen Koalitionsparteien während ihrer Zeit im Amt? Koalitionsparteien stehen vor einem Dilemma, dass sie zwar nach aussen Einheit demonstrieren sollen, sich aber gleichzeitig von ihren Partnern differenzieren müssen. Ich argumentiere, dass politische Kommunikation eine wichtige Rolle dabei spielt, wie Parteien versuchen, ihr individuelles Profil zu erhalten. Dazu habe ich drei Hauptstrategien definiert, die Parteien benutzen können. Basierend auf einem Datensatz von über 35'000 deutschen und niederländischen Pressemitteilungen von Koalitionsparteien analysiere ich, was Parteien in ihrer Strategiewahl beeinflusst. Meine Resultate zeigen, dass Personalisierung, definiert als ein verstärkter Fokus auf Individuen, davon beeinflusst wird, wie stark sich Koalitionsparteien ideologisch unterscheiden. Ich benutze einen supervised classification algorithm, um die deutschen Pressemitteilungen in verschiedene thematische Kategorien zu klassifizieren. Ich nutze diese Klassifizierung um zu analysieren, ob Parteien während dem Wahlkampf einen besonderen Fokus auf die Themen legen, die ihnen wichtig sind. Dies ist nicht der Fall, und meine Analysen zeigen, dass Parteien kurz vor einer Wahl sogar einen geringeren Schwerpunkt auf diese Themen legen, ungeachtet ideologischer Differenzen zwischen ihnen und ihren Koalitionspartnern. Schließlich analysiere ich die Präsenz von negativem Campaigning in deutschen Pressemitteilungen und in einer Auswahl von Episoden einer politischen Talkshow. Im Falle der Pressemitteilungen stelle ich fest, dass die Parteien mit zunehmender Wahrscheinlichkeit "feindlich gesinnte" Politiker erwähnen, je näher die Wahl rückt, und dass diese Erwähnungen mit einem generell negativerem sentiment der betreffenden Pressemitteilungen korrelieren. / What communication strategies do coalition parties use during their time in office? Coalition parties face a dilemma between unity and differentiation, because while they need to keep relations with their partners stable, they also have to keep up constant efforts to differentiate themselves sufficiently from them. I argue that political communication plays a key role in how parties try to maintain a strong individual profile. I have defined three main strategies that parties can use to achieve this. Based on a dataset that contains over 35'000 press releases by German and Dutch coalition parties, I analyze what influences the usage of the three strategies in question. I find that the usage personalization, which is defined as focusing on individual politicians, is affected by the ideological distance between coalition partners. I use a supervised classiffication algorithm to classify the press releases issued by German coalition parties into issue areas to analyze whether parties use election campaigns to emphasize issues that are already important to them. I find that this is not the case, and that parties actually de-emphasize these issues as an election approaches, regardless of ideological distance between them and their partners. Lastly, I analyze the presence of negative campaigning in German press releases and in a selection of episodes from a political talk show. In the case of press releases, I find that parties are more likely to specifically mention hostile politicians as an election approaches, and that these mentions correlate with a more negative sentiment of the press releases in question.

Since software and hardware is well available for automated text analysis and since a large data that describes real projects submitted to grant program is opened up, there is a possibility to follow phenomena of behavioral economics and psycholinguistics which evidence particularities in textual descriptions may be statistically associated with a reader's behavior or with a reader's decision-taking, which, in this case, involves an influence on final allocation of grant funds. The thesis uses forenamed areas as a starting-point and also employs quantitative indicators from the field of forensic linguistics in order to perform a computer-aided quantitative text analysis. The main goal is to evaluate from correlation perspective, if there in real operational programmes were present any associable relationships between the quantitative features of a proposed project's textual description and the amount of grant allocated to a project. The thesis is divided into four chapters, where it introduces basis, describes analyzed data and used methods, comments on made analyses and found relations, and all the performed research is summarized and evaluated in the last chapter.