Le point de vue des étudiants et les minorités linguistiques sont peu présents dans la recherche sur la transition scolaire et l’expérience étudiante à l’université (Lamoureux, 2007; Lamoureux et coll., 2013; Tinto, 2017). Notre recherche porte sur le sentiment d’appartenance chez des étudiants diplômés d’écoles secondaires de langue française de l’Ontario lors de leur première année universitaire. Dans la présente étude, nous avons fait une utilisation secondaire de données provenant de 17 groupes de discussion réalisés auprès de 88 étudiants. Une analyse de contenu a été effectuée selon trois concepts, soit l’appartenance à l’institution, les relations sociales et le métier d’étudiant. Les résultats montrent qu’au premier trimestre, les étudiants sont en mode de survie et ont de la difficulté à trouver l’équilibre entre leur vie sociale et académique. Au deuxième trimestre, des moments de prise de conscience ont permis à plusieurs participants de comprendre ce sur quoi ils devaient travailler pour se sentir plus à l’aise à l’université. Cette étude contribue à la recherche sur le sentiment d’appartenance à l’université chez une population francophone minoritaire dans la mesure où nous avons montré de quelle façon celui-ci se développe lors de la première année d’études universitaires.

Consumer's opinions and sentiments on products can reflect the performance of products in general or in various aspects. Analyzing these data is becoming feasible, considering the availability of immense data and the power of natural language processing. However, retailers have not taken full advantage of online comments. This work is dedicated to a solution for automatically analyzing and summarizing these valuable data at both product and category levels. In this research, a system was developed to retrieve and analyze extensive data from public online resources. A parallel framework was created to make this system extensible and efficient. In this framework, a star topological network was adopted in which each computing unit was assigned to retrieve a fraction of data and to assess sentiment. Finally, the preprocessed data were collected and summarized by the central machine which generates the final result that can be rendered through a web interface. The system was designed to have sound performance, robustness, manageability, extensibility, and accuracy.

Cette thèse présente une contribution relative à l'étude de l'impact du sentiment de l'investisseur sur la dynamique boursière dans les marchés développés de G7. A cette fin, deux spécifications économétriques ont été proposées pour modéliser la dynamique des marchés boursiers tenant compte de l’hétérogénéité comportementale des investisseurs. Cette thèse est structurée autour de trois chapitres. Le premier chapitre est d’ordre théorique. Après avoir rappelé l'hypothèse de l'efficience informationnelle, ce chapitre introduit les concepts de base ( rationalité, finance comportementale, sentiment de l'investisseur, etc.). Les deux autres chapitres proposent deux essais empiriques sur l’effet du sentiment de l'investisseur sur la dynamique des rentabilités boursières. Dans le premier chapitre empirique, le modèle Smooth Transition Regression "STR" est utilisé afin de mieux reproduire les différents régimes des rentabilités boursières activés par l’effet du sentiment de l'investisseur. Le second chapitre empirique vise à tenir en compte simultanément la non linéarité et l'hétérogénéité comportementale à travers le double usage des dimensions temporelle et individuelle en employant le modèle Panel Smooth Transition Regression "PSTR". Nos résultats empiriques impliquent le rejet de l’hypothèse d’efficience ainsi que celle d’agent représentatif, en suggérant que la dynamique des rentabilités boursières exhibe de la non-linéarité. Nous montrons également l’existence d’effets de seuil significatifs permettant de distinguer différents régimes, suggérant que l'exposition des rentabilités au risque sentiment varie non-linéairement dans le temps et par régime. / This thesis presents a contribution to the study of the impact of investor sentiment upon stock market dynamics in the developed markets of the G7 countries. For this purpose, two econometric methods are proposed for modeling stock market dynamics that take the behavioral heterogeneity of investors into account. This thesis includes three chapters. The first chapter is theoretical. After recalling the hypothesis of informational efficiency, it introduces the basic concepts (rationality, behavioral finance, investor sentiment, etc.). The following two chapters offer two empirical tests for the impact of investor sentiment on the dynamics of stock market returns.In the first, the Smooth Transition Regression "STR" model is used to provide a better reproduction of the different regimes for market returns triggered by investor sentiment. The second empirical chapter aims to take the nonlinearity and the heterogeneity into account simultaneously, through the dual use of temporal and individual dimensions and by applying the Panel Smooth Transition Regression "PSTR" model. Our empirical results imply a rejection of the efficiency hypothesis and a rejection of the hypothesis of a representative agent, suggesting the nonlinearity of stock market returns. We also show that there are different transition speeds between different regimes, suggesting that the exposure of stock market returns to risk sentiment varies nonlinearly over time and by regime.

El texto completo de este trabajo no está disponible en el Repositorio Académico UPC por restricciones de la casa editorial donde ha sido publicado. / In recent years, the increased use of digital tools within the Peruvian tourism industry has created a corresponding increase in revenues. However, both factors have caused increased competition in the sector that in turn puts pressure on small and medium enterprises’ (SME) revenues and profitability. This study aims to apply neural network based sentiment analysis on social networks to generate a new information search channel that provides a global understanding of user trends and preferences in the tourism sector. A working data-analysis framework will be developed and integrated with tools from the cloud to allow a visual assessment of high probability outcomes based on historical data, to help SMEs estimate the number of tourists arriving and places they want to visit, so that they can generate desirable travel packages in advance, reduce logistics costs, increase sales, and ultimately improve both quality and precision of customer service.

No description available.

Sentiment analysis is a burgeoning field in natural language processing used to extract and categorize opinion in evaluative documents. We look at recommendation letters, which pose unique challenges to standard sentiment analysis systems. Our dataset is eighteen letters from applications to UMass Worcester Memorial Medical Center’s residency program in Obstetrics and Gynecology. Given a small dataset, we develop a method intended for use by domain experts to systematically explore their intuitions about the topical make-up of documents on which they make critical decisions. By leveraging WordNet and the WordNet Propagation algorithm, the method allows a user to develop topic seed sets from real data and propagate them into robust lexicons for use on new data. We show how one pass through the method yields useful feedback to our beliefs about the make-up of recommendation letters. At the end, future directions are outlined which assume a fuller dataset.

Every week there are over a billion new posts to Twitter services and many of those messages contain feedback to companies about their services. One company that has recognized this unused source of information is Netflix. That is why Netflix initiated the development of a system that will let them respond to the millions of Twitter and Netflix users that are acting as sensors and reporting all types of user visible outages. This system will enhance the feedback loop between Netflix and its customers by increasing the amount of customer feedback that is being received by Netflix and reducing the time it takes for Netflix to receive the reports and respond to them. The goal of the SPOONS (Swift Perceptions of Online Negative Situations) system is to use Twitter posts to determine when Netflix users are reporting a problem with any of the Netflix services. This work covers a subset of the meth- ods implemented in the SPOONS system. The volume methods detect outages through time series analysis of the volume of a subset of the tweets that contain the word “netflix”. The sentiment methods first process the tweets and extract a sentiment rating which is then used to create a time series. Both time series are monitored for significant increases in volume or negative sentiment which indicates that there is currently an outage in a Netflix service. This work contributes: the implementation and evaluation of 8 outage detection methods; 256 sentiment estimation procedures and an evaluation of each; and evaluations and discussions of the real time applicability of the system. It also provides explanations for each aspect of the implementation, evaluations, and conclusions so future companies and researchers will be able to more quickly create detection systems that are applicable to their specific needs.

Social media has become an important place for modern people to conveniently share and exchange their ideas and opinions. However, not all content on the social media have positive impact. Hate speech is one kind of harmful content that people use abusive speech attacking or promoting hate towards a specific group or an individual. With online hate speech on the rise these day, people have explored ways to automatically recognize the hate speech, and among the ways people have studied, the Bert-based approach is promising and thus dominates SemEval-2019 Task 6, a hate speech detection competition. In this work, the method of fusion of sentiment features and Bert-based approach is proposed. The classic Bert architecture for hate speech detection is modified to fuse with additional sentiment features, provided by an extractor pre-trained on Sentiment140. The proposed model is compared with top-3 models in SemEval-2019 Task 6 Subtask A and achieves 83.1% F1 score that better than the models in the competition. Also, to see if additional sentiment features benefit the detectoin of hate speech, the features are fused with three kind of deep learning architectures respectively. The results show that the models with sentiment features perform better than those models without sentiment features. / Master of Science / Social media has become an important place for modern people to conveniently share and exchange their ideas and opinions. However, not all content on the social media have positive impact. Hate speech is one kind of harmful content that people use abusive speech attacking or promoting hate towards a specific group or an individual. With online hate speech on the rise these day, people have explored ways to automatically recognize the hate speech, and among the ways people have studied, Bert is one of promising approach for automatic hate speech recognition. Bert is a kind of deep learning model for natural language processing (NLP) that originated from Transformer developed by Google in 2017. The Bert has applied to many NLP tasks and achieved astonished results such as text classification, semantic similarity between pairs of sentences, question answering with given paragraph, and text summarization. So in this study, Bert will be adopted to learn the meaning of given text and distinguish the hate speech from tons of tweets automatically. In order to let Bert better capture hate speech, the approach in this work modifies Bert to take additional source of sentiment-related features for learning the pattern of hate speech, given that the emotion will be negative when people trying to put out abusive speech. For evaluation of the approach, our model is compared against those in SemEval-2019 Task 6, a famous hate speech detection competition, and the results show that the proposed model achieves 83.1\% F1 score better than the models in the competition. Also, to see if additional sentiment features benefit the detection of hate speech, the features are fused with three different kinds of deep learning architectures respectively, and the results show that the models with sentiment features perform better than those without sentiment features.

The automatic analysis of emotions is a well-established area in the natural language processing ( NLP ) research field. It has shown valuable and relevant applications in a wide array of domains such as health and well-being, empathetic conversational agents, author profiling, consumer analysis, and security. Most emotion analysis research till now has focused on sources such as news documents and product reviews. In these cases, the NLP task is the classification into predefined closed-set emotion categories (e.g. happy, sad), or alternatively labels (positive, negative). A deep and fine-grained emotion analysis would require explanations of the trigger events that may have led to a user state. This type of analysis is still in its infancy. In this work, we introduce the concept of Emotion Carriers (EC) as the speech or text segments that may include persons, objects, events, or actions that manifest and explain the emotions felt by the narrator during the recollection. In order to investigate this emotion concept, we analyze Personal Narratives (PN) - recollection of events, facts, or thoughts from one’s own experience, - which are rich in emotional information and are less explored in emotion analysis research. PNs are widely used in psychotherapy and thus also in mental well-being applications. The use of PNs in psychotherapy is rooted in the association between mood and recollection of episodic memories. We find that ECs capture implicit emotion information through entities and events whereas the valence prediction relies on explicit emotion words such as happy, cried, and angry. The cues for identifying the ECs and their valence are different and complementary. We propose fine-grained emotion analysis using valence and ECs. We collect and annotate spoken and written PNs, propose text-based and speech-based annotation schemes for valence and EC from PNs, conduct annotation experiments, and train systems for the automatic identification of ECs and their valence.

Yes / While research on factors driving corporate investment decisions has blossomed, knowledge related to the Chief Executive Officer’s (CEO’s) market sentiment on investment decision outcomes is lacking. In this study, we extend the existing corporate finance literature by examining the underexplored issue of how CEOs’ market sentiment drives firms’ investment decisions. Capitalising on a large sample of US firms for the period 2004-2014, we uncovered some crucial observations. First, we found empirical support for our theoretical contention that market sentiment drives corporate investment decisions. Second, we established that, while financial flexibility induces managers to overinvest, the expectation of future profitability leads firms to underinvest during high sentiment periods. In addition, we uncovered that the 2007/08 financial crisis significantly impacted firm behaviour and realigned managerial decision-making. Thus, the sentiment-investment relationship is more pronounced during the crisis and the post-crisis periods. Our results are robust after accounting for the possibility of endogeneity and using alternative measures of both CEOs’ market sentiment and firm investment.