Genre and domain influence an author\'s style of writing and therefore a text\'s characteristics. Natural language processing is prone to such variations in textual characteristics: it is said to be genre and domain dependent. This thesis investigates genre and domain dependencies in sentiment analysis. Its goal is to support the development of robust sentiment analysis approaches that work well and in a predictable manner under different conditions, i.e. for different genres and domains. Initially, we show that a prototypical approach to sentiment analysis -- viz. a supervised machine learning model based on word n-gram features -- performs differently on gold standards that originate from differing genres and domains, but performs similarly on gold standards that originate from resembling genres and domains. We show that these gold standards differ in certain textual characteristics, viz. their domain complexity. We find a strong linear relation between our approach\'s accuracy on a particular gold standard and its domain complexity, which we then use to estimate our approach\'s accuracy. Subsequently, we use certain textual characteristics -- viz. domain complexity, domain similarity, and readability -- in a variety of applications. Domain complexity and domain similarity measures are used to determine parameter settings in two tasks. Domain complexity guides us in model selection for in-domain polarity classification, viz. in decisions regarding word n-gram model order and word n-gram feature selection. Domain complexity and domain similarity guide us in domain adaptation. We propose a novel domain adaptation scheme and apply it to cross-domain polarity classification in semi- and unsupervised domain adaptation scenarios. Readability is used for feature engineering. We propose to adopt readability gradings, readability indicators as well as word and syntax distributions as features for subjectivity classification. Moreover, we generalize a framework for modeling and representing negation in machine learning-based sentiment analysis. This framework is applied to in-domain and cross-domain polarity classification. We investigate the relation between implicit and explicit negation modeling, the influence of negation scope detection methods, and the efficiency of the framework in different domains. Finally, we carry out a case study in which we transfer the core methods of our thesis -- viz. domain complexity-based accuracy estimation, domain complexity-based model selection, and negation modeling -- to a gold standard that originates from a genre and domain hitherto not used in this thesis.

Sentiment analysis is a challenging task today for computational linguistics. Because of the rise of the social Web, both the research and the industry are interested in automatic processing of opinions in text. In this work, we assume a multilingual and multidomain environment and aim at automatic and adaptive polarity classification.We propose a method for automatic construction of multilingual affective lexicons from microblogging to cover the lack of lexical resources. To test our method, we have collected over 2 million messages from Twitter, the largest microblogging platform, and have constructed affective resources in English, French, Spanish, and Chinese.We propose a text representation model based on dependency parse trees to replace a traditional n-grams model. In our model, we use dependency triples to form n-gram like features. We believe this representation covers the loss of information when assuming independence of words in the bag-of-words approach.Finally, we investigate the impact of entity-specific features on classification of minor opinions and propose normalization schemes for improving polarity classification. The proposed normalization schemes gives more weight to terms expressing sentiments and lower the importance of noisy features.The effectiveness of our approach has been proved in experimental evaluations that we have performed across multiple domains (movies, product reviews, news, blog posts) and multiple languages (English, French, Russian, Spanish, Chinese) including official participation in several international evaluation campaigns (SemEval'10, ROMIP'11, I2B2'11).

Sentiment analysis is a field of computational linguistics involving identification, extraction, and classification of opinions, sentiments, and emotions expressed in natural language. Sentiment classification algorithms aim to identify whether the author of a text has a positive or a negative opinion about a topic. One of the main indicators which help to detect the opinion are the words used in the texts. Needless to say, the sentiments expressed in the texts also depend on the syntactic structure and the discourse context. Supervised machine learning approaches to sentiment classification were shown to achieve good results. Classifying texts by emotions requires finer-grained analysis than sentiment classification. In this thesis, we explore the task of emotion and mood classification for blog postings. We propose a novel approach that uses the hierarchy of possible moods to achieve better results than a standard flat classification approach. We also show that using sentiment orientation features improves the performance of classification. We used the LiveJournal blog corpus as a dataset to train and evaluate our method. Another contribution of this work is extracting paraphrases for emotion terms based on the six basics emotions proposed by Ekman (\textit{happiness, anger, sadness, disgust, surprise, fear}). Paraphrases are different ways to express the same information. Algorithms to extract and automatically identify paraphrases are of interest from both linguistic and practical points of view. Our paraphrase extraction method is based on a bootstrapping algorithms that starts with seed words. Unlike in previous work, our algorithm does not need a parallel corpus. In Natural Language Generation (NLG), paraphrasing is employed to create more varied and natural text. In our research, we extract paraphrases for emotions, with the goal of using them to automatically generate emotional texts (such as friendly or hostile texts) for conversations between intelligent agents and characters in educational games. Nowadays, online services are popular in many disciplines such as: e-learning, interactive games, educational games, stock market, chat rooms and so on. NLG methods can be used in order to generate more interesting and normal texts for such applications. Generating text with emotions is one of the contributions of our work. In the last part of this thesis, we give an overview of NLG from an applied system's points of view. We discuss when NLG techniques can be used; we explained the requirements analysis and specification of NLG systems. We also, describe the main NLG tasks of content determination, discourse planning, sentence aggregation, lexicalization, referring expression generation, and linguistic realisation. Moreover, we describe our Authoring Tool that we developed in order to allow writers without programming skills to automatically generate texts for educational games. We develop an NLG system that can generate text with different emotions. To do this, we introduce our pattern-based model for generation. We show our model starts with initial patterns, then constructs extended patterns from which we choose ``final'' patterns that are suitable for generating emotion sentences. A user can generate sentences to express the desired emotions by using our patterns. Alternatively, the user can use our Authoring Tool to generate sentences with emotions. Our acquired paraphrases will be employed by the tool in order to generate more varied outputs.

À partir de 45 profils Facebook de jeunes franco-ontariens du Nord qui ont quitté leur région d'origine pour s'installer à Ottawa, la thèse s’intéresse au sens qu'ils donnent à ces deux milieux. Nous cherchions à mesurer la force du sentiment d’appartenance de ces migrants à leur milieu d'origine, tout en étudiant celui qu’ils développent à leur milieu d’accueil. Le rôle joué par les réseaux sociaux virtuels dans cette appartenance primait dans l'analyse. La recherche montre en fait que les jeunes ont des appartenances aussi fortes au nord de l'Ontario qu’à Ottawa, se situant ainsi dans l’entre-deux territorial. Notre analyse révèle que ce sont des dimensions différentes du milieu qui jouent dans l’appartenance aux milieux d’origine et d’accueil. Alors que les réseaux sociaux virtuels lient le migrant, maintenant dans le Sud, à sa région d'origine, l'imaginaire et les activités pratiquées dans les deux milieux servent eux aussi à maintenir cet attachement.

This multi-disciplinary study examines the linguistic characteristics which influence communication and social interaction in computer-mediated communication (CMC). We begin by conducting a qualitative data analysis on a group of graduate students taking online courses. Through this, we look more closely at their perception of social interaction in their online learning environment (Knowledge eCommons). We then take individual student notes and analyze their linguistic characteristics. We look at the emotional cues in notes, the use of factual, objective language and other linguistic features. We study these notes through the use of sentiment analysis methodologies – which will be explained in detail in the first and second chapter. We have proposed a method for deducing note objectivity and have computed reliability testing of this method. Our analyses show that there is a high correlation between the use of objective language in a note and the value that students place on that note.

This multi-disciplinary study examines the linguistic characteristics which influence communication and social interaction in computer-mediated communication (CMC). We begin by conducting a qualitative data analysis on a group of graduate students taking online courses. Through this, we look more closely at their perception of social interaction in their online learning environment (Knowledge eCommons). We then take individual student notes and analyze their linguistic characteristics. We look at the emotional cues in notes, the use of factual, objective language and other linguistic features. We study these notes through the use of sentiment analysis methodologies – which will be explained in detail in the first and second chapter. We have proposed a method for deducing note objectivity and have computed reliability testing of this method. Our analyses show that there is a high correlation between the use of objective language in a note and the value that students place on that note.

Cette étude qualitative multicas est ancrée dans l'approche sociale-cognitive de la théorie de l'autoefficacité de Bandura (1977). Elle s’intéresse à quatre stages à l’enseignement qui se sont déroulés au Musée canadien de la nature en 2009. L’utilisation de données secondaires issues du questionnaire STEBI-B traduit et modifié (Dionne et Couture, 2010) ainsi que des entrevues semi-dirigées ont permis une analyse du changement d’autoefficacité en sciences chez les stagiaires. Les éléments les plus intéressants de cette recherche sont l’apprentissage vicariant et la possibilité de répétition qui favorise une meilleure connaissance de soi et une pratique réflexive. Les résultats, dans l’ensemble positifs, illustrent bien le potentiel d’un tel stage afin de rehausser le sentiment d’autoefficacité en sciences chez des stagiaires en enseignement, particulièrement chez ceux qui se destinent à enseigner à l’élémentaire puisque ceux-ci ont souvent une formation académique sans science.

We examine how investor sentiment affects the changes in implied volatility, and discover investor sentiment has impact on the size of the changes in implied volatility through returns, especially when returns are negative. We examine the short-tern relation between the S&P 500 index returns and the changes of VIX from January 1990 to January 2011, and between the NASDAQ-100 index returns and the changes of VXN from February 2001 to January 2011 with proxy for beginning-of-period investor sentiment at both the daily and weekly level. We find that during high sentiment periods, the negative and asymmetric relation of return to changes in implied volatility can be mitigated significantly. When returns are segregated into positive and negative returns, investor sentiment has different impact on the size of changes in implied volatility. In negative returns, investors are more panic than in positive returns, but the panic can be mitigated significantly when investors are in high sentiment. Thus, sentiment can alter the risk attitude of investors and reduce their panic in the future, especially when market has negative performance.

Thèse 3 cycle--Histoire--Montpellier III, 1975. / Contient le texte traduit de l'arabe des discours du colonel Boumediene non recueillis dans l'édition française. Bibliogr. p. 351-371. Index.

隨著金融市場的全球化，自一九〇〇年代起各國進行跨國上市的企業逐年增長，而該現象也引起學者對於可能造成跨國上市之因與其中之利弊進行進一步的思考與研究，從而發展出許多假說與相關實證結果。過去的傳統假說以市場分割假說、流動性假說以及投資人認知假說等對跨國上市進行解釋，且多以各國至美國跨國上市作為實證，由於上述假說經實證仍留有無法解釋的部分，進而發展出綁定假說，但無論是傳統或是新興的理論，都留有空間讓我們透過全新的角度去賦予見解，因此本論文期以透過行為財務學的觀點，以投資人情緒來解讀公司進行跨國上市的決策制定。 不同於以往，我們以美國作為實證，檢視全球投資人情緒對於美國公司至全球進行跨國上市決策是否有所影響，樣本期間取自二〇〇三年至二〇一四年，完整樣本數共4,955家企業進行跨國上市，而透過參考文獻我們在考量了公司、交易所與國家三個層級的控制變數後進行相關實證。 實證結果顯示全球投資人情緒確實影響公司進行跨國上市的決策，當全球投資人情緒越高漲，公司進行跨國上市的可能性則提高，而反之亦然。本論文提供已經過長時間假說與實證的跨國上市領域一個新的思考方向，全球投資人情緒的波動將會是一個影響企業至海外進行權益融資的指標之一。 / With the globalization of financial markets, boundaries between countries are getting vague. Since the 〖20〗^(th) century, the amount of firms having their stocks cross-listed oversea is increasing each year, hypotheses and empirical test have long been formed and conducted to figure out the cause and effect of such phenomenon. As for the conventional wisdom, market segmentation, liquidity and investor recognition hypotheses are constructed but still left puzzle unexplained. Bonding theory then been brought up after. But no matter how the conventional wisdom or new research initiatives are trying to interpret, behavioral finance can always bring up a brand new aspect and a whole new explanation. Our paper use global sentiment as a determinant to demonstrate the cross-listing decision-making of a firm. Firms in the United States are using as our samples to test our hypothesis, which is expressed that the higher the global sentiment is, the more possible that a firm would have its stock cross-listed. Our sample period is from 2003 to 2014 and the amount of firms cross-listed in the sample period is 4,955. Familiar with the previous studies we have our control variables divided into three levels, firm, exchange and country. The empirical result indicates that while the sentiment of the globe is high, firms in the United States have the intention to have their stock cross-listed oversea, and vice versa. Our main contribution of this study is that though research in cross-listing has long been studied, we provide a new viewpoint that we confirm the connection between global sentiment and cross-listing decision of financing.