We present results from a project on sentiment analysis of drama texts, more concretely the plays of Gotthold Ephraim Lessing. We conducted an annotation study to create a gold standard for a systematic evaluation. The gold standard consists of 200 speeches of Lessing’s plays and was manually annotated with sentiment information by five annotators. We use the gold stand-ard data to evaluate the performance of different German sentiment lexicons and processing configurations like lemmatization, the extension of lexicons with historical linguistic variants, and stop words elimination, to explore the influence of these parameters and to find best prac-tices for our domain of application. The best performing configuration accomplishes an accu-racy of 70%. We discuss the problems and challenges for sentiment analysis in this area and describe our next steps toward further research

Social media has recently become an essential component of our daily modern life, with platforms like Facebook, YouTube, and Twitter serving as popular venues for people to share their opinions on various topics, including sustainable food. The interest in consumer sentiments towards sustainable practices has increased particularly after Covid-2019. This study investigates the attitudes and needs of Swedish consumers regarding sustainable food consumption as reflected in their social media interactions using 4588 comments from Facebook and YouTube. The methodology used are sentiment analysis and topic modelling with VADER and Latent Dirichlet Allocation (LDA) respectively. The results reveal a generally strong positive attitude toward sustainable food. However, the study observes further a decline in positive sentiments over time, indicating changing consumer opinions. The primary topic identified is market challenges, such as high pricing. Furthermore, health concerns and environmental considerations are identified both as important factors influencing the choice of sustainable food. The findings highlight the necessity for policy interventions to enhance the affordability and accessibility of sustainable food, as well as the effective use of social media for raising consumer awareness.

Customer reviews are becoming an increasingly important factor for the success of hotels. Many potential clients rely on reviews when making booking decisions, and negative reviews can seriously impact the hotel's reputation. The hotel business is constantly evolving under the influence of new trends, technologies, and customer expectations. Understanding which aspects of service cause the most dissatisfaction among customers allows hotels to adapt and effectively respond to changes. / Отзывы клиентов становятся все более важным фактором для успеха отелей. Многие потенциальные клиенты ориентируются на отзывы при принятии решения о бронировании, и негативные отзывы могут серьезно повлиять на репутацию отеля. Отельный бизнес постоянно меняется под воздействием новых тенденций, технологий и ожиданий клиентов. Понимание того, какие аспекты обслуживания вызывают наибольшее недовольство клиентов, позволяет отелям адаптироваться и эффективно реагировать на изменения.

本篇研究專注於首次公開發行公司上市櫃初始交易日之異常報酬與新聞情緒兩 者間之關係。本研究建立情緒字典以判別新聞之正負情緒,並過濾出與首次公開發 行有關之新聞,利用本研究建立之情緒字典以過濾出正負情緒之詞組。利用正負情 緒詞組數量計算出三種新聞情緒變數,並採實證研究方法檢測三種新聞情緒變數與 首次公開發行公司之初始交易日之異常報酬兩者間之關係。根據本研究之實證結果, 發現初始交易日之前的新聞能影響首次公開發行之異常報酬,而相關新聞之情緒語 調亦和異常報酬有關。此外,本研究亦檢測三種情緒變數和三種傳統變數之交乘項 對異常報酬之影響,發現公司規模大小與首日交易量與情緒變數之交乘項會對初始 交易日之異常報酬有影響。總言論之,本研究對新聞會影響首次公開發行初始交易 日之異常報酬提供了實證證據。 / This study focuses on the relation between IPOs’ abnormal returns on initial trading days and news sentiment. To identify the tone of news, sentiment dictionary was established for this study, and news regarding IPO firms was picked out to count positive and negative words and phrases based on the sentiment dictionary. Using quantities of positive and negative words and phrases, three news variables were adopted and calculated. And linear regression was utilized to investigate the relation between IPOs’ abnormal returns on initial trading days and news sentiment. According to empirical results, I find that news prior to the IPO’s initial trading day can affect IPOs’ abnormal returns. The number of negative words and phrases is negatively related to the abnormal returns; the tone of news is positively related to the abnormal returns. Furthermore, I also investigated whether interaction terms of news variables and three control variables are related to abnormal returns on IPOs’ initial trading days. I find that interaction terms of the natural logarithm of firm size and two news variables and interaction terms of the natural logarithm of first-day trading volume and two news variables are related to abnormal returns. Overall, there is evidence that news can influence IPOs’ abnormal returns on initial trading days.

There has been much hype in vocational and academic circles surrounding the emergence of web 2.0 or social media; however, relatively little work was dedicated to substantiating the actual concept of web 2.0. Many have dismissed it as not deserving of this new title, since the term web 2.0 assumes a certain interpretation of web history, including enough progress in certain direction to trigger a succession [i.e. web 1.0 → web 2.0]. Others provided arguments in support of this development, and there has been a considerable amount of enthusiasm in the literature. Much research has been busy evaluating current use of web 2.0, and analysis of the user generated content, but an objective and thorough assessment of what web 2.0 really stands for has been to a large extent overlooked. More recently the idea of collective intelligence facilitated via web 2.0, and its potential applications have raised interest with researchers, yet a more unified approach and work in the area of collective intelligence is needed. This thesis identifies and critically evaluates a wider context for the web 2.0 environment, and what caused it to emerge; providing a rich literature review on the topic, a review of existing taxonomies, a quantitative and qualitative evaluation of the concept itself, an investigation of the collective intelligence potential that emerges from application usage. Finally, a framework for harnessing collective intelligence in a more systematic manner is proposed. In addition to the presented results, novel methodologies are also introduced throughout this work. In order to provide interesting insight but also to illustrate analysis, a case study of the recent financial crisis is considered. Some interesting results relating to the crisis are revealed within user generated content data, and relevant issues are discussed where appropriate.

Topic modeling refers to the process of algorithmically sorting documents into categories based on some common relationship between the documents. This common relationship between the documents is considered the “topic” of the documents. Sentiment analysis refers to the process of algorithmically sorting a document into a positive or negative category depending whether this document expresses a positive or negative opinion on its respective topic. In this paper, I consider the open problem of document classification into a topic category, as well as a sentiment category. This has a direct application to the retail industry where companies may want to scour the web in order to find documents (blogs, Amazon reviews, etc.) which both speak about their product, and give an opinion on their product (positive, negative or neutral). My solution to this problem uses a Non-negative Matrix Factorization (NMF) technique in order to determine the topic classifications of a document set, and further factors the matrix in order to discover the sentiment behind this category of product.

A Mineração de Opiniões é um processo que tem por objetivo extrair as opiniões e suas polaridades de sentimentos expressas em textos em língua natural. Essa área de pesquisa tem ganhado destaque devido ao volume de opiniões que os usuários compartilham na Internet, como revisões em sites de e-commerce, rede sociais e tweets. A Mineração de Opiniões baseada em Aspectos é uma alternativa promissora para analisar a polaridade do sentimento em um maior nível de detalhes. Os métodos tradicionais para extração de aspectos e classificação de sentimentos exigem a participação de especialistas de domínio para criar léxicos ou definir regras de extração para diferentes idiomas e domínios. Além disso, tais métodos usualmente exploram algoritmos de aprendizado supervisionado, porém exigem um grande conjunto de dados rotulados para induzir um modelo de classificação. Os desafios desta tese de doutorado estão relacionados a como diminuir a necessidade de grande esforço humano tanto para rotular dados, quanto para tratar a dependência de domínio para as tarefas de extração de aspectos e classificação de sentimentos dos aspectos para Mineração de Opiniões. Para reduzir a necessidade de grande quantidade de exemplos rotulados foi proposta uma abordagem semissupervisionada, denominada por Aspect-based Sentiment Propagation on Heterogeneous Networks (ASPHN) em que são propostas representações de textos nas quais os atributos linguísticos, os aspectos candidatos e os rótulos de sentimentos são modelados por meio de redes heterogêneas. Para redução dos esforços para construir recursos específicos de domínio foi proposta uma abordagem baseada em aprendizado por transferência entre domínios denominada Cross-Domain Aspect Label Propagation through Heterogeneous Networks (CD-ALPHN) que utiliza dados rotulados de outros domínios para suportar tarefas de aprendizado em domínios sem dados rotulados. Nessa abordagem são propostos uma representação em uma rede heterogênea e um método de propagação de rótulos. Os vértices da rede são os aspectos rotulados do domínio de origem, os atributos linguísticos e os candidatos a aspectos do domínio alvo. Além disso, foram analisados métodos de extração de aspectos e propostas algumas variações para considerar cenários nãosupervisionados e independentes de domínio. As soluções propostas nesta tese de doutorado foram avaliadas e comparadas as do estado-da-arte utilizando coleções de revisões de diferentes produtos e serviços. Os resultados obtidos nas avaliações experimentais são competitivos e demonstram que as soluções propostas são promissoras. / Opinion Mining is a process that aims to extract opinions and their sentiment polarities expressed in natural language texts. This area of research has been in the highlight because of the volume of opinions that users share on the available visualization means on the Internet (reviews on e-commerce sites, social networks, tweets, others). Aspect-based Opinion Mining is a promising alternative for analyzing the sentiment polarity on a high level of detail. The traditional methods for aspect extraction and sentiment classification require the participation of domain experts to create lexicons or define extraction rules for different languages and domains. In addition, such methods usually exploit supervised machine learning algorithms, but require a large set of labeled data to induce a classification model. The challenges of this doctoral thesis are related on to how to reduce the need for great human effort both: (i) to label data; and (ii) to treat domain dependency for the tasks of aspect extraction and aspect sentiment classification for Opinion Mining. In order to reduce the need for a large number of labeled examples, a semi-supervised approach was proposed, called Aspect-based Sentiment Propagation on Heterogeneous Networks (ASPHN). In this approach, text representations are proposed in which linguistic attributes, candidate aspects and sentiment labels are modeled by heterogeneous networks. Also, a cross-domain learning approach called Cross-Domain Aspect Label Propagation through Heterogeneous Networks (CD-ALPHN) is proposed in order to reduce efforts to build domain-specific resources, This approach uses labeled data from other domains to support learning tasks in domains without labeled data. A representation in a heterogeneous network and a label propagation method are proposed in this cross-domain learning approach. The vertices of the network are the labeled aspects of the source domain, the linguistic attributes, and the candidate aspects of the target domain. In addition, aspect extraction methods were analyzed and some variations were proposed to consider unsupervised and domain independent scenarios. The solutions proposed in this doctoral thesis were evaluated and compared to the state-of-the-art solutions using collections of different product and service reviews. The results obtained in the experimental evaluations are competitive and demonstrate that the proposed solutions are promising.

Sistemas de recomendação surgiram da necessidade de selecionar e apresentar conteúdo relevante a usuários de acordo com suas preferências. Dentre os diversos métodos existentes, aqueles baseados em conteúdo faz em uso exclusivo da informação inerente aos itens. Estas informações podem ser criadas a partir de técnicas de indexação automática e manual. Enquanto que as abordagens automáticas necessitam de maiores recursos computacionais e são limitadas á tarefa específica que desempenham, os métodos manuais são caros e propensos a erros. Por outro lado, com a expansão da Web e a possibilidade de usuários comuns criarem novos conteúdos e anotações sobre diferentes itens e produtos, uma alternativa é obter esses metadados criados colaborativamente pelos próprios usuários. Entretanto, essas informações, em especial revisões e comentários, podem conter ruídos, além de estarem em uma forma desestruturada. Deste modo, este trabalho1 tem como objetivo desenvolver métodos de construção de representações de itens baseados em descrições colaborativas para um sistema de recomendação. Objetiva-se analisar o impacto que diferentes técnicas de extração de características, aliadas à análise de sentimento, causam na precisão da geração de sugestões, avaliando-se os resultados em dois cenários de recomendação: predição de notas e geração de ranques. Dentre as técnicas analisadas, observa-se que a melhor apresenta um ganho no poder descritivo dos itens, ocasionando uma melhora no sistema de recomendação. / Recommender systems arose from the need to select and present relevant content to users according to their preferences. Among several existent methods, those based on content make exclusive use of information inherent to the items. This information can be created through automatic and manual indexing techniques. While automa-tic approaches require greater computing resources and are limited to the specific task they perform, manual methods are expensive and prone to errors. On the other hand, with the expansion of theWeb and the possibility of common users to create new content and descriptions about different items and products, an alternative is to get these metadata created collaboratively by the users. However, this information, especially reviews and comments, may contain noise, be- sides being in a unstructured fashion. Thus, this study aims to develop methods for the construction of items representations based on collaborative descriptions for a recommender system. This study aims to analyze the impact that different feature extraction techniques, combined with sentiment analysis, caused in the accuracy of the generated suggestions, evaluating the results in both recommendations cenarios: rating prediction and ranking generation. Among the analyzed techniques, it is observed that the best is able to describe items in a more effcient manner, resulting in an improvement in the recommendation system.

Submitted by admin tede (tede@pucgoias.edu.br) on 2018-09-27T18:13:44Z No. of bitstreams: 1 ParcileneFernandesdeBrito.pdf: 2444083 bytes, checksum: eea4e1b897bdd9e57504c34888b57c01 (MD5) / Made available in DSpace on 2018-09-27T18:13:44Z (GMT). No. of bitstreams: 1 ParcileneFernandesdeBrito.pdf: 2444083 bytes, checksum: eea4e1b897bdd9e57504c34888b57c01 (MD5) Previous issue date: 2018-06-29 / The vast amount of information available on the Internet have enabled numerous multidisciplinar investigations aimed to understand nuances of human consumption behavior, especially to identify people's opinions about products and services. From Behavioral Perspective Model (BPM), consumer behavior analysis can be conducted focusing on antecedent variables (behavioral setting and consumer learning history) and consequences (reinforcement and punishment, utilitarian and informative) to the occurrence of behavior. The present thesis investigated consumption behavior in the context of tourism, with BPM as theoretical support for interpretations of verbal data extracted from comments available on the TripAdvisor®, a website about tourism. Verbal responses of tourism consumers, engaged in the process of online avaluation of components of tourism products (specifically, Accommodations [ACO], Restaurants [RES] and Attractions [ATR]), were analyzed. Research participants were the unknown individuals who, between the beginning of February and the end of March 2017, emitted 6.438.497 comments distributed among the 100 most evaluated Brazilian touristic destinations at TribAdvisor®. In two studies (Study 1 [E1] and Study 2 [E2]), the thesis research aimed at: a) extraction and analysis of tourist´s verbal information (commentaries) throught a Sentiment Analysis (SA) computational technique; extraction the number of touristic product component (ACO, RES e ATR) evaluative indications emitted by tourism consumers with different statuses as TripColaborators; extraction of the number of votes (Likes) to the comments; b) describe the polarized evaluative response attributed to the 100 evaluated touristic destinations and interpret such responding considering BPM concepts. E1 resulted in the successful development of the SentimentALL tool, focusing on the AS module, and the generation the primary variables explored in E2 (n = 197). In E2, data generated in E1 and derived measures were explored (described in rankings and correlation analyzes) and interpreted using the BPM conceptual framework. With a fundamental exploratory caracter, the interpretative effort suggested profitable research lines and utility of the computational and psychological knowledges integration. / A grande quantidade de informações disponíveis na internet tem viabilizado numerosas investigações de caráter multidisciplinar com o objetivo de entender nuances do comportamento de consumo humano, especialmente identificar as opiniões das pessoas sobre produtos e serviços. A partir do Behavioral Perspective Model (BPM), análises do comportamento do consumidor podem ser realizadas considerando variáveis antecedentes (cenário do comportamento e história de aprendizagem do consumidor) e consequências (reforços e punições, utilitários e informativos) à ocorrência do comportamento. A presente tese investigou o comportamento de consumo no contexto do turismo, com o BPM como suporte teórico para interpretações de dados verbais extraídos de comentários disponíveis no TripAdvisor®, website do setor. Para tanto, analisaram-se as respostas verbais de turistas-consumidores no processo de “opinar on-line” sobre componentes de produtos turísticos (especificamente, acomodações – ACO, restaurantes – RES e atrações – ATR). Os participantes da pesquisa foram os indivíduos (desconhecidos) que, entre o início de fevereiro e o final de março de 2017, emitiram, no TripAdvisor®, 6.438.497 comentários distribuídos entre os 100 destinos turísticos brasileiros mais avaliados. Descrita em dois estudos (Estudo 1 [E1] e Estudo 2 [E2]), a pesquisa de tese se propôs a: a) extração e análise de informações verbais (comentários) dos turistas com base na técnica computacional Análise de Sentimentos (AS); extração do número de indicações avaliativas dos componentes do produto turístico (ACO, RES e ATR) emitidas por turistas-consumidores com diferentes status como TripColaboradores; extração do número de votos úteis (Likes) nos comentários; b) descrever o responder avaliativo polarizado atribuído aos 100 destinos turísticos avaliados e analisar interpretativamente tal responder a partir do BPM. O E1 resultou no desenvolvimento da ferramenta SentimentALL, com foco no módulo de AS, e na geração das variáveis primárias exploradas no E2 (n = 197). No E2, dados gerados no E1 e medidas derivadas foram explorados (descritos em rankings e análises de correlação) e interpretados com recurso ao referencial conceitual do BPM. De caráter fundamentalmente exploratório, o esforço interpretativo sugeriu linhas profícuas de pesquisa e a utilidade da integração entre conhecimentos computacionais e psicológicos.

Submitted by PPG Ci?ncia da Computa??o (ppgcc@pucrs.br) on 2018-09-03T14:11:33Z No. of bitstreams: 1 WILLIAN EDUARDO BECKER_DIS.pdf: 2142751 bytes, checksum: e6501a586bb81f7cbad7fa5ef35d32f2 (MD5) / Approved for entry into archive by Sheila Dias (sheila.dias@pucrs.br) on 2018-09-04T14:43:25Z (GMT) No. of bitstreams: 1 WILLIAN EDUARDO BECKER_DIS.pdf: 2142751 bytes, checksum: e6501a586bb81f7cbad7fa5ef35d32f2 (MD5) / Made available in DSpace on 2018-09-04T14:57:29Z (GMT). No. of bitstreams: 1 WILLIAN EDUARDO BECKER_DIS.pdf: 2142751 bytes, checksum: e6501a586bb81f7cbad7fa5ef35d32f2 (MD5) Previous issue date: 2017-11-24 / Nowadays, the use of social media has become a daily activity of our society. The huge and uninterrupt flow of information in these spaces opens up the possibility of exploring this data in different ways. Sentiment Analysis (SA) is a task that aims to obtain knowledge about the polarity of a given text relying on several techniques of Natural Language Processing, with most of solutions dealing with only one language at a time. However, approaches that are not restricted to explore only one language are more related to extract the whole knowledge and possibilities of these data. Recent approaches based on Machine Learning propose to solve SA by using mainly Deep Learning Neural Networks have obtained good results in this task. In this work is proposed three Convolutional Neural Network architectures that deal with multilingual Twitter data of four languages. The first and second proposed models are characterized by the fact they require substantially less learnable parameters than other considered baselines while are more accurate than several other Deep Neural architectures. The third proposed model is able to perform a multitask classification by identifying the polarity of a given sentences and also its language. This model reaches an accuracy of 74.43% for SA and 98.40% for Language Identification in the four-language multilingual dataset. Results confirm that proposed model is the best choice for both sentiment and language classification by outperforming the considered baselines. / A utiliza??o de redes sociais tornou-se uma atividade cotidiana na sociedade atual. Com o enorme, e ininterrupto, fluxo de informa??es geradas nestes espa?os, abre-se a possibilidade de explorar estes dados de diversas formas. A An?lise de Sentimento (AS) ? uma tarefa que visa obter conhecimento sobre a polaridade das mensagens postadas, atrav?s de diversas t?cnicas de Processamento de Linguagem Natural, onde a maioria das solu??es lida com somente um idioma de cada vez. Entretanto, abordagens que n?o restringem se a explorar somente uma l?ngua, est?o mais pr?ximas de extra?rem todo o conhecimento e possibilidades destes dados. Abordagens recentes baseadas em Aprendizado de M?quina prop?em-se a resolver a AS apoiando-se principalmente nas Redes Neurais Profundas (Deep Learning), as quais obtiveram bons resultados nesta tarefa. Neste trabalho s?o propostas tr?s arquiteturas de Redes Neurais Convolucionais que lidam com dados multi-linguais extra?dos do Twitter contendo quatro l?nguas. Os dois primeiros modelos propostos caracterizam-se pelo fato de possu?rem um total de par?metros muito menor que os demais baselines considerados, e ainda assim, obt?m resultados superiores com uma boa margem de diferen?a. O ?ltimo modelo proposto ? capaz de realizar uma classifica??o multitarefa, identificando a polaridade das senten?as e tamb?m a l?ngua. Com este ?ltimo modelo obt?m-se uma acur?cia de 74.43% para AS e 98.40% para Identifica??o da L?ngua em um dataset com quatro l?nguas, mostrando-se a melhor escolha entre todos os baselines analisados.