Made available in DSpace on 2016-08-17T14:52:40Z (GMT). No. of bitstreams: 1 Dissertacao_MilsonLouseiroLima.pdf: 4206975 bytes, checksum: 68293f1f1c80ce84d0573111677ff097 (MD5) Previous issue date: 2016-05-06 / FUNDAÇÃO DE AMPARO À PESQUISA E AO DESENVOLVIMENTO CIENTIFICO E TECNOLÓGICO DO MARANHÃO / Predicting the behavior of stocks in the stock market is a challenging task, a lot of times related to unknown factors or influenced by very distinct natures of variables, which can range from high-profile news to the collective sentiment, expressed in publications on social networks. Such market volatility may represent considerable financial losses for investors. In order to forestall such variations other mechanisms to predict the behavior of assets in the stock market have been proposed, based on pre-existing indicator data. Such mechanisms only analyze statistical data, not considering the collective human sentiment. This work aims to develop a model to predict the stock market, based on analysis of sentiment and it will make use of techniques of artificial intelligence as natural language processing (PLN) and Support Vector Machines (SVM) to predict the active behavior. However, it should be emphasized that this model is intended to be an aid tool in the decision-making process that involves buying and selling shares on the stock market. / Predizer o comportamento das ações na bolsa de valores é uma tarefa desafiadora, muita vezes relacionada a fatores desconhecidos ou influenciados por variáveis de naturezas bem distintas, que podem ir desde notícias de grande repercussão até o sentimento coletivo, expresso em publicações de redes sociais. Tal volatilidade do mercado pode representar perdas financeiras consideráveis para os investidores. No intuito de se antecipar a tais variações já foram propostos outros mecanismos para predizer o comportamento de ativos na bolsa de valores, baseados em dados de indicadores pré-existentes. Tais mecanismos analisam apenas dados estatísticos, não considerando o sentimento humano coletivo. Este trabalho tem como finalidade desenvolver um modelo para predição da bolsa de valores, baseado na mineração de opinião e, para isso, fará uso de técnicas de Inteligência artificial como processamento de linguagem natural(PLN) e Máquinas de Vetor de Suporte(SVM) para predizer o comportamento do ativo. No entanto, convém ressaltar que o referido modelo tem como finalidade ser uma ferramenta de auxílio no processo de tomada de decisão que envolve a compra e venda de ações na bolsa de valores.

Submitted by Georgia Vaz (georgia.vaz@mackenzie.br) on 2016-07-06T19:42:24Z No. of bitstreams: 2 license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) JOSE AHYRTON BATISTA LOPES FILHO.pdf: 1482320 bytes, checksum: 2162e0cdfb92a9b596af601d0f4c4ed1 (MD5) / Made available in DSpace on 2016-07-06T19:42:24Z (GMT). No. of bitstreams: 2 license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) JOSE AHYRTON BATISTA LOPES FILHO.pdf: 1482320 bytes, checksum: 2162e0cdfb92a9b596af601d0f4c4ed1 (MD5) Previous issue date: 2016-02-17 / With the growth of social media in recent years, there has been an increase on the interest in the automatic characterization of users based on the informal content they generate. In this context, the labeling of users in demographic categories, such as age, ethnicity, origin and race,and the investigation of other attributes inherent to users, such as political preferences, personality and gender expression, has received a great deal of attention, especially based on Twitter data. The present work focuses on the task of gender classification by using 65 textual meta-attributes, commonly used in text attribution tasks, for the extraction of gender expression linguistic cues in tweets written in Portuguese.The work takes into account characters, syntax, words, structure and morphology, as well as selected psycolinguistic cues of short length, multi-genre, content free texts posted on Twitter to classify author's gender via four different machine-learning algorithms. The proposed meta-attributes in this process are also evaluated. / Com o crescimento das mídias sociais nos últimos anos tem havido um aumento de interesse na caracterização automática dos usuários com base no conteúdo informal que eles geram. Neste contexto, a rotulação dos usuários em categorias demográficas tais como idade, etnia, origem e raça, bem como a investigação de outros atributos inerentes aos usuários, como preferências políticas, personalidade e expressão de gênero, tem recebido grande atenção, especialmente com base em dados do Twitter. O presente trabalho é centrado na tarefa de classificação de gênero, propondo 65 meta-atributos textuais, comumente usados em tarefas de atribuição de texto, para a extração de características linguísticas quanto à expressão de gênero em tweets escritos em Português. São considerados caracteres, sintaxe, palavras, estrutura e morfologia, além de determinados atributos psicolinguísticos, dos textos de comprimento curto, multi-gênero e de livre conteúdo postados no Twitter para a classificação de gênero do autor por meio de quatro algoritmos de aprendizado de máquina diferentes. Também é avaliada a influência dos meta-atributos propostos para este processo.

Submitted by Renata Lopes (renatasil82@gmail.com) on 2016-06-30T18:01:53Z No. of bitstreams: 1 flaviavaleriolopes.pdf: 1120082 bytes, checksum: 242ab8638f47e6e042567ad17fd704ec (MD5) / Approved for entry into archive by Adriana Oliveira (adriana.oliveira@ufjf.edu.br) on 2016-07-01T12:31:57Z (GMT) No. of bitstreams: 1 flaviavaleriolopes.pdf: 1120082 bytes, checksum: 242ab8638f47e6e042567ad17fd704ec (MD5) / Approved for entry into archive by Adriana Oliveira (adriana.oliveira@ufjf.edu.br) on 2016-07-13T15:58:51Z (GMT) No. of bitstreams: 1 flaviavaleriolopes.pdf: 1120082 bytes, checksum: 242ab8638f47e6e042567ad17fd704ec (MD5) / Made available in DSpace on 2016-07-13T15:58:51Z (GMT). No. of bitstreams: 1 flaviavaleriolopes.pdf: 1120082 bytes, checksum: 242ab8638f47e6e042567ad17fd704ec (MD5) Previous issue date: 2012-03-30 / As redes sociais introduziram no atual cenário novas vozes, que passaram a participar do debate público a partir dos movimentos iniciados nos territórios do ambiente sociotécnico. Por meio dessas ferramentas, sujeitos evidenciam sua função de autoria, sem intermediários, e acabam sendo “convocados” a participar das discussões. Possibilitadas pelo avanço das tecnologias da sociedade em rede, as mídias sociais transformaram-se em um novo espaço público de discussão. Neste trabalho iremos investigar a reconfiguração da relação de interlocução entre imprensa e sociedade, com a chegada desses novos atores. Para isso, analisamos, discursivamente, a repercussão do trabalho de um grupo de jovens durante a ocupação/invasão da polícia no Complexo do Alemão, no Rio de Janeiro, pelo perfil @vozdacomunidade no Twitter. / Social networks have introduced in the current scenario new voices that have become involved in the public debate from the movements initiated in the domains of the sociotechnical settings. Through these tools, subjects identify their role of authorship, without intermediaries, and they end up being “summoned” to participate in discussions. Enabled by advancing technologies from the network society, social media have transformed themselves into a new public space for discussion. In this context, it is convenient to investigate the reconfiguration of the interlocution relation between press and society with the arrival of these new interlocutors. For that matter, it will be considered the repercussion of the work of a young people’s group during the police occupation-invasion of Complexo do Alemão in Rio de Janeiro through the profile @vozdacomunidade in Twitter.

Submitted by Renata Lopes (renatasil82@gmail.com) on 2017-04-10T12:03:09Z No. of bitstreams: 1 daianamariaveigasigiliano.pdf: 4382171 bytes, checksum: 00c98a07bfce59fb7b6c48d4e03cbc50 (MD5) / Approved for entry into archive by Adriana Oliveira (adriana.oliveira@ufjf.edu.br) on 2017-04-11T11:34:28Z (GMT) No. of bitstreams: 1 daianamariaveigasigiliano.pdf: 4382171 bytes, checksum: 00c98a07bfce59fb7b6c48d4e03cbc50 (MD5) / Made available in DSpace on 2017-04-11T11:34:28Z (GMT). No. of bitstreams: 1 daianamariaveigasigiliano.pdf: 4382171 bytes, checksum: 00c98a07bfce59fb7b6c48d4e03cbc50 (MD5) Previous issue date: 2017-02-22 / CAPES - Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / O ambiente de convergência alterou de forma decisiva os modos de pensar, produzir, distribuir e consumir televisão. É a partir deste contexto que a social TV se configura, o fenômeno é caracterizado pelo compartilhamento de conteúdos (comentários, memes, vídeos, fotos, montagens, etc.) feito através das redes sociais (Twitter, Facebook, Snapchat, etc.) e dos aplicativos de segunda tela (TVShow Time, TV Tag, Viggle, etc.) de maneira síncrona à exibição da grade de programação.Ao estabelecer uma relação simbiótica entre a televisão e o ciberespaço, o fenômeno potencializa e reconfigura a conversação em torno do conteúdo televisivo e a experiência coletiva. Nesse sentido, esta dissertação tem como objetivo discutir de que forma a social TV reforça o conceito de laço social de Wolton (1996). Isto é, se o autor propunha o entendimento do laço social enquanto uma experiência compartilhada que aproxima os telespectadores e as diferentes comunidades que constituem uma sociedade, estabelecendo assim um conhecimento comum entre eles, o fenômeno não só potencializa este aspecto, como também propicia a formação de comunidades momentâneas e teias colaborativas no backchannel. Nesse contexto, analisamos as ações de social TV adotadas pelo canal estadunidense Fox para engajar os telespectadores e a forma como as publicações dialogam com o universo ficcional da série The X-Files. Criada por Chris Carter, a tramaé protagonizada pelos agentes do FBI Fox Mulder (David Duchovny) e Dana Scully (Gillian Anderson). Mulder e Scully são investigadores de casos não solucionados envolvendo fenômenos paranormais, conhecidos como x-files. Sendo assim, realizamos um estudo sobre as estratégias promovidas pela emissora estadunidense para os episódios da décima temporada de The X-Files e analisamos os conteúdos (comentários, memes, fotos, montagens, etc.) compartilhados pelos telespectadores interagentes no Twitter durante a exibição da atração. Desta forma, concluímos que, ao entrelaçar o fluxo televisivo e a temporalidade always on do Twitter, a social TV reforça o lado social de Wolton (1996) e ressignifica a experiência televisiva, possibilitando novas formas de participação, colaboração e expansão do universo ficcional. / The convergence environment has decisively changed the ways of thinking, producing, distributing and consuming television. It is from this context that thesocial TV gets configured; the phenomenon is characterized by the content sharing (comments, memes, videos, photos, montages, etc.) made through the social networks (Twitter, Facebook, Snapchat, etc.) and the second screen applications (TVShow Time, TV Tag, Viggle, etc.) synchronously to the programming griddisplay. By establishing a symbiotic relationship between the television and the cyberspace, the phenomenon potentializes and reconfigures the conversation around the television content and the collective experience. In this sense, this master thesis aims to discuss how the social TV reinforces the concept of social bond developed by Wolton (1996). In other words, if the author proposed the understanding of the social bond as a shared experience that approachesthe viewers and the different communities that constitute a society, establishing then a common knowledge among them, the phenomenon not only enhances this aspect, but also provides the formation of momentary communities and collaborative webs in the backchannel.In this context, the social TV actions adopted by the american channel Fox to engage the viewers and the way the publications dialog with the fictional universe of the series The X-Files were analyzed. Created by Chris Carter, the plot is starred by the FBI agents Fox Mulder (David Duchovny) and Dana Scully (Gillian Anderson). Mulder and Scully are investigators of unsolved cases involving paranormal phenomena, known as x-files. Therefore, we conducted a study on the strategies promoted by the american broadcaster for the episodes of the The X-Files tenth season and analyzed the content (comments, memes, photos, montages, etc.) shared by the interacting viewers on Twitter during the exhibition of the attraction. In this way, we concluded that, by interlacing the TV stream and the always on temporality of the Twitter, the social TV reinforces the social side of Wolton (1996) and re-signifies the television experience, allowing new forms of participation, collaboration and expansion of the fictional universe.

El presente trabajo abarca el estudio de la variación en las construcciones adverbiales locativas. Esta variación se manifiesta en dos niveles: (1) la alternancia entre la construcción estándar ADVERBIO + preposición de + PRONOMBRE PERSONAL (por ejemplo, delante de mí), y su variante normativamente no recomendada ADVERBIO + PRONOMBRE POSESIVO (por ejemplo, delante mío); y (2) la alternancia entre la construcción posesiva con el sufijo -o (por ejemplo, delante mío), y la construcción posesiva con el sufijo -a (por ejemplo, delante mía), variante normativamente aún más disuadida. El objetivo fue estudiar estas dos variables lingüísticas y sus posibles predictores sociolingüísticos. Con esta finalidad, se han analizado datos provenientes de Twitter. La recolección de datos se llevó a cabo en febrero del 2017. Se recogieron 2357 observaciones de construcciones adverbiales locativas de 21 países de habla hispana (incluyendo los Estados Unidos), de las cuales 1006 eran de la variante posesiva. En cuanto a la variación entre construcciones preposicionales y posesivas, se clasificaron los datos según cuatro variables independientes: el género (o sexo) del tuitero, la procedencia geográfica del tuit, el tipo de adverbio precedente y el número gramatical del referente. Con relación al uso del sufijo -a del posesivo, las observaciones fueron clasificadas según dos variables adicionales: la vocal final del adverbio y el género del referente gramatical. Se ha construido un modelo estadístico de regresión logística que mide el efecto de las variables independientes investigadas sobre el uso de (1) la construcción con el posesivo y (2) la construcción con el posesivo del sufijo -a respectivamente. Referente a (1), la alternancia en los complementos, el análisis revela que la zona dialectal del tuitero, el tipo del adverbio precedente y el número gramatical son los factores estadísticamente significativos para predecir el uso de la construcción posesiva. En lo que concierne a (2), la alternancia en los sufijos del posesivo, los datos apuntan a que tanto la zona dialectal del tuitero como el género del referente gramatical condicionan el uso del posesivo con la desinencia -a. / The present thesis studies the morphosyntactic variation in adverbial locative phrases in Spanish. This variation occurs at two levels: (1) the alternation between the standard construction ADVERB + the preposition de + PERSONAL PRONOUN (as in delante de mí) and the normatively nonstandard construction ADVERB + POSSESSIVE PRONOUN (as in delante mío); (2) the alternation between the possessive construction with the suffix -o (as in delante mío), and the possessive construction with the suffix -a (as in delante mía), the last variant being normatively highly discouraged. The goal of this study was to investigate these two linguistic variables and their possible sociolinguistic predictors. The data was obtained by API searches of Twitter data in February of 2017 and yielded material from 21 Spanish-speaking countries (including the US). 2357 tweets containing the constructions of interest were gathered, of which 1006 were possessive constructions. Concerning the alternation between constructions with prepositions and possessives, the data was coded according to four independent variables: the gender (or sex) of the Twitter user, the location of the user, the type of adverb used in the phrase, and the grammatical number of the referent. As to the use of the suffix -a with possessives, two additional variables were coded for, the final vowel of the preceding adverb and grammatical gender of the referent. A statistical model was constructed in order to measure the effect of the independent variables on the use of (1) the possessive construction, and (2) the possessive construction with the suffix -a, respectively. The results of the first analysis indicate that the geographical region of the Twitter user, as well as the preceding adverb and the grammatical number of the referent, were statistically significant in predicting the use of the possessive construction. Furthermore, with regards to the variation in the suffix of the possessive pronoun, the analysis supports the view that the geographical region of the user and the grammatical gender of the referent are predictors that condition the use of the possessive pronoun with the suffix -a.

After Donald Trump won the presidency, his use of social media has gotten massive attention because of its incredibly inappropriate content. Many people question how appropriate his use of Twitter is, and in the linguistic field prominent voices point out that the former businessman does not have the skills to speak; he only talks (McWhorter, 2017). As America’s President with views of his own about worldly events, it is of special interest to see how people interact with him online. Because of the low 39% approval rating, one can assume that many disagree with him, possibly in a way involving irony. One sub-type of irony is described as “the use of words to express something other than and especially the opposite of the literal meaning” (Merriam-Webster, 2018), and it is reasonable to assume that some responses to Trump’s tweets use irony to present an alternative view of his actions or opinions. The study sets out to find replies to Trump’s tweets that are ironic and identify if they correspond to earlier developed theories of irony. The replies are examined to determine which of Grice’s maxims of conversation that is flouted the most. The data included 337 replies of which 78 (23%) were shown to follow an ironical pattern, mostly in order to ridicule and condemn Trump. 88 % of all comments disagree with Trump’s tweets. The most applicable theory of irony is the echoic interpretation theory by Wilson and Sperber, and the most flouted maxim is relevance. / Sedan Donald Trump ställde upp i presidentvalet och till slut vann valet, har hans användning av sociala medier fått mycket uppmärksamhet p.g.a dess olämpliga innehåll. Många har ifrågasatt lämpligheten i hans Twitter-användning, och lingvistiskt sett menar framstående röster att den tidigare företagsmannen inte äger förmågan att tala; han kan bara prata; och det gör han på Twitter som USA:s president och ger sin egen syn på världsliga händelser. Därför är det av intresse att titta på hur människor reagerar och tilltalar honom på nätet. Med tanke på hans låga popularitetssiffra på 39%, kan det förmodas att många inte håller med honom, och att ironi används som ett sätt att uttrycka en annan åsikt. Ironi beskrivs som “användningen av ord för att uttrycka någonting annat eller speciellt det motsatta av den bokstavliga meningen” (översatt från Merriam-Webster, 2018), vilket är hypotetiskt förväntat i svar till Trumps tweets. Denna studie syftar till att identifiera ironiska kommentarer på Trumps tweets och se om de överensstämmer med tidigare utvecklade teorier kring ironi. Även analys av vilken av Grices konversationsmaximer som bryts oftast görs i studien. Studiens data inkluderar 337 kommentarer, av vilka 78 (23%) är ironiska, oftast med intention att förlöjliga och fördöma Trump. 88% av alla kommentarer håller inte med om innehållet i Trumps tweets. Den mest applicerbara ironi-teorin var “echoic interpretation” teorin (Wilson & Sperber), medan den mest ignorerade konversationsmaximen var relevans.

Abstract Organizations continuously strive to engage customers in the services development process. The social web facilitates this process by enabling novel channels for voluntary feedback-sharing and collaborative interaction through social media and technologically advanced environments. The component parts of these environments are information systems that are linked with social media and designed for large displays to support interactivity. The work performed during this research involved the design and assessment of operational software features for encouraging user engagement through publicly displayed information systems. Drawing upon socio-psychological theories and interconnecting them to the seven social influence design principles listed in the Persuasive Systems Design model, this dissertation examines the role of persuasive software features in altering human behavior with respect to engagement in feedback-sharing and collaborative interaction. The dissertation consists of five quantitative studies, including one Twitter survey (involving 403 respondents) and four experiments with publicly displayed Twitter-based systems (involving 284 users), that successively complement each other to address the main research question: How can social influence design principles persuade people to engage with publicly displayed systems that are integrated with social media? The results of these studies reveal interplay between the design principles and indicate that they have the capacity to improve the persuasiveness of information systems and predict the behavioral intentions of users to engage with such systems in the future. Based on these findings, a framework for studying socially influencing systems (SIS) is proposed. This framework is potentially instrumental in achieving a richer understanding of how to effectively harness social influence for enhanced user engagement through socio-technical environments and for the future development of persuasive information systems. / Tiivistelmä Organisaatioiden palvelukehityksen tavoitteena on sitouttaa ja innostaa asiakkaita. Sosiaalinen verkko edistää tätä prosessia tarjoamalla uusia väyliä niin vapaaehtoiselle palautteenjaolle kuin sosiaalisessa mediassa ja edistyneissä teknologisissa ympäristöissä tapahtuvalle interaktiiviselle yhteistyöllekin. Nämä ympäristöt koostuvat tietojärjestelmistä, jotka on yhdistetty sosiaaliseen mediaan, ja jotka on suunniteltu interaktiota edistäville suurille näytöille. Tässä tutkimuksessa on suunniteltu ja arvioitu ohjelmisto-ominaisuuksia, joiden tarkoitus on suostutella ja osallistuttaa julkisesti esilläolevien tietojärjestelmien ja näyttöjen kautta. Tutkimus nojaa sosiopsykologiseen teoriapohjaan, ja yhdistää näitä seitsemään sosiaalisen vaikuttamisen suunnitteluperiaatteeseen, jotka on kuvattu vaikuttavien suunnittelumenetelmien mallissa. Tämä väitöstyö tutkii vaikuttavien ohjelmistollisten ominaisuuksien osuutta käyttäytymisen muuttamisessa kohdistuen erityisesti siihen, miten loppukäyttäjää voidaan suostutella jakamaan palautetta ja ottamaan osaa interaktiiviseen yhteistyöhön. Työ koostuu viidestä määrällisestä tutkimuksesta, jotka vastaavat tutkimuskysymykseen: kuinka sosiaalisen vaikuttamisen suunnitteluperiaatteet voivat suostutella ja vakuuttaa ihmiset käyttämään julkisesti esilläolevia järjestelmiä, jotka on yhdistetty sosiaaliseen mediaan? Tutkimuksista yksi on Twitter-kyselytutkimus (403 vastaajaa) ja neljä tutkimusta tehtiin julkisesti esilläolevalla Twitter-pohjaisella järjestelmällä (284 vastaajaa). Tutkimusten tulokset osoittavat mainittujen suunnitteluperiaatteiden keskinäisen vuorovaikutuksen, sekä kuinka niiden avulla voidaan parantaa tietojärjestelmien vaikuttavuutta ja ennustaa tulevaa käyttöä. Tutkimustulosten pohjalta työssä luodaan viitekehys sosiaalisesti vaikuttavien järjestelmien tutkimukseen. Tämä viitekehys on keskeisessä asemassa pyrittäessä saavuttamaan syvällisempi ymmärrys siitä, kuinka käyttäjiä voidaan suostutella hyödyntämällä sosiaalista vaikuttamista. Viitekehystä voidaan käyttää myös vaikuttavien järjestelmien kehittämisessä.

... Im Interesse eines lebendigen Dialogs mit ihrer weitaus größten Benutzergruppe hat sich die SLUB im Frühjahr 2009 entschlossen, die Klaviatur der sozialen Netzwerke, der Blogs und Mikro-Blogs, planmäßig zu nutzen. Erklärte Ziele sind die möglichst zielgruppengerechte Vermittlung und beschleunigte Optimierung der Bibliotheksdienstleistungen sowie eine hohe Identifikation der Studierenden mit „ihrer“ Bibliothek. Die SLUB zählt zu den ersten Bibliotheken in Deutschland, die die neuen Werkzeuge konsequent einsetzt, jedoch kann kein Zweifel bestehen, dass die Kommunikation via Blogs und Mikro-Blogs recht bald weithin zum bibliothekarischen Alltag gehören wird. ...

(in English): The diploma thesis deals with yet in czech language unprocessed complex issue of Twitterverse with a focus on community Twitter and third party applications. The thesis is mapping origin and development of Twitter and the related phenomenon Twitterverse using quantative research methods and case study. The work should contribute to a better and more detailed understanding of this issue.

No description available.