<p>There are words and topics of conversation that are considered taboo and offensive in the English language. Offensive words can be divided into different categories, based on the way they are used and in what situation. Topics of conversation that are considered taboo are for example sexual activity and death.</p><p>Men and women are said to use language differently, in a number of areas. One of these areas concerns offensive language. It is considered masculine to swear and women are prone to use euphemisms more than men, i.e. the mildest form possible of an offensive word. Studies have shown that men and women feel more comfortable using taboo language with members of the same sex than with members of the opposite sex.</p><p>This paper aims to study the differences in language use between men and women and apply the findings on eight episodes of the American television series Sex and the City, with focus on offensive language. The purpose is to study whether or not the female characters of the show use a typically male language and if they do, if it could be a reason for them being perceived as strong women.</p>

Slurs are offensive words used about people on account of them belonging to certain groups,for example based on gender, sexual orientation or ethnicity. The topic of slurs has interestedphilosophers of language recently, and there are several theories trying to explain them. Mostassume that slurs have Neutral Counterparts, non-offensive words that can be used more or lesssynonymously with certain slurs. In this thesis I attempt to construct a theory of slurs withoutneutral counterparts. First, I give a background where I explain extension, content and neutralcounterparts. Then I describe popular theories about how slurs offend. After that I summarizeAlice Damirjian’s critique against neutral counterparts in her article “Rethinking slurs”. Againstthis background I explain my theory of slurs; slurs lack neutral counterparts, and they refer toentirely different classes than so called “neutral counterparts” do. Instead, slurs refer nonphysical sociocultural constructions, while “Neutral counterparts” refer to physical people.While the relation between slurs and “neutral counterparts” is non-semantic, it is pragmatic;there is a widespread misconception that slurs and “neutral counterparts” are more or lesssynonymous. Slurs offend since their sociocultural constructions are associated withstereotypes that are attributed to the people that their so called “neutral counterparts” refer to.After I explain my theory I respond to counterarguments, and the thesis is concluded with asummary.

<p>All languages have words that are considered taboo – words that are not supposed to be said or used. Taboo words, or swearwords, can be used in many different ways and they can have different meanings depending on what context they appear in. Another aspect of taboo words is the euphemisms that are used in order to avoid obscene speech. This paper will focus on x-words, words like the f-word or the c-word, which replace the words fuck or cunt, but as the study will show they also have other meanings and usages.</p><p>The purpose of this paper is also to investigate the significance of taboo words and their usage in English, as well as research on how they are used, or not used, in media. The aim is to examine how x-words are used in the British newspapers the Guardian and the Observer by using corpus searches.</p><p>The results show that there are several ways of using x-words, and that using them in order to show that a word either is taboo, or has become taboo in a certain context, is the most common way. It is also clear that x-words can represent many different words, and not only words that are generally considered taboo.</p>

There are words and topics of conversation that are considered taboo and offensive in the English language. Offensive words can be divided into different categories, based on the way they are used and in what situation. Topics of conversation that are considered taboo are for example sexual activity and death. Men and women are said to use language differently, in a number of areas. One of these areas concerns offensive language. It is considered masculine to swear and women are prone to use euphemisms more than men, i.e. the mildest form possible of an offensive word. Studies have shown that men and women feel more comfortable using taboo language with members of the same sex than with members of the opposite sex. This paper aims to study the differences in language use between men and women and apply the findings on eight episodes of the American television series Sex and the City, with focus on offensive language. The purpose is to study whether or not the female characters of the show use a typically male language and if they do, if it could be a reason for them being perceived as strong women.

All languages have words that are considered taboo – words that are not supposed to be said or used. Taboo words, or swearwords, can be used in many different ways and they can have different meanings depending on what context they appear in. Another aspect of taboo words is the euphemisms that are used in order to avoid obscene speech. This paper will focus on x-words, words like the f-word or the c-word, which replace the words fuck or cunt, but as the study will show they also have other meanings and usages. The purpose of this paper is also to investigate the significance of taboo words and their usage in English, as well as research on how they are used, or not used, in media. The aim is to examine how x-words are used in the British newspapers the Guardian and the Observer by using corpus searches. The results show that there are several ways of using x-words, and that using them in order to show that a word either is taboo, or has become taboo in a certain context, is the most common way. It is also clear that x-words can represent many different words, and not only words that are generally considered taboo.

In today’s society there is a large number of social media users that are free to express their opinion on shared platforms. The socio-cultural differences between the people behind those accounts (in terms of ethnicity, gender, sexual orientation, religion, politics, . . . ) give rise to an important percentage of online discussions that make use of offensive language, which often affects in a negative way the psychological well-being of the victims. In order to address the problem, the endless stream of user-generated content engenders a need to find an accurate and scalable solution to detect offensive language using automated methods. This thesis explores different approaches to the offensiveness detection task focusing on five different languages: Arabic, Danish, English, Greek and Turkish. The results obtained using Support Vector Machines (SVM), Convolutional Neural Networks (CNN) and the Bidirectional Encoder Representations from Transformers (BERT) are compared, achieving state-of-the-art results with some of the methods tested. The effect of the embeddings used, the dataset size, the class imbalance percentage and the addition of sentiment features are studied and analysed, as well as the cross-lingual capabilities of pre-trained multilingual models.

Resumen El uso de palabrotas en el habla coloquial, más bien conocido como el uso de tacos, ha aumentado considerablemente entre mujeres durante las últimas décadas. En esta tesina se han analizado las diferencias de su uso entre géneros y en qué consisten, el tema presentado cabe dentro de la diciplina sociolingüística, la cual estudia los aspectos comunicativos entre individuos. Inicialmente se destacan las definiciones y se explican los diferentes usos de las palabrotas dentro del habla coloquial.  El análisis se ha realizado a través de un método mixto. La primera parte consiste en un análisis del corpus COLAM (2015) del cual el material ha sido tratado de modo cuantitativo para poder averiguar la diferencia entre frecuencia y uso entre géneros.  Las frecuencias están presentadas en diagramas para poder probar la supuesta hipótesis de que existen diferencias entre los géneros al usar tacos en el habla coloquial.  En la segunda parte del análisis se ha aplicado un método cualitativo en el que el uso de tacos es ejemplificados y explicados con ayuda del corpus PRESEEA (2014).             El resultado de la segunda parte del análisis nos muestra las diferentes funciones y usos sociolingüísticos dentro del habla coloquial independiente de género.  Los resultados de este estudio muestran que existen diferencias entre géneros al usar tacos, mayormente en frecuencia y temas de las cuáles los chicos predominan el uso con temas sexuales. El presupuesto aumento entre chicas es probablemente debido a un cambio de actitud sobre lo considerado ofensivo durante las últimas décadas. / Abstract The use of curses in everyday language, better known as the use of” tacos” in Spanish has increased considerably among women the last decades. In this thesis gender diferencials have been analized and explained, the theme presented is within the subject of area of sociolinguistics which studies communicative aspects between individuals. Inlitially the defenitions are determined and different uses of curses in everyday language are explained.       The analysis is implemented through a mixed method, the first part consists of an analysis of the COLAM (2015) corpus in wich the material has been treated in a quantitative fashioning order to determine the differences in frequency and use among opposing genders. The frequencies are presented in charts to enable the test of the hypothesis that diferences between genders exist in the use of cursing in everyday language. In the second part of the analysis a qualitative method has been applied in which the use of ”tacos” is exemplified and explained using the corpus of PRESEEA (2014). The focus here is to show the diferent functions ans sociolinguistic uses whithin everyday language independent of genders.                                                                              The results of this study shows that gender diferences in the use of ”tacos” exists, and that the diferances mainly consists of the frequency and theme of which boys predominatly uses sexual themes. The presumed increase among girls is probobly due to a change in attitude towards whats considered to be ofensive use of  language during the last decades.

Le discours haineux est un problème sérieux sur les média sociaux. Dans ce mémoire, nous étudions le problème de détection automatique du langage haineux sur réseaux sociaux. Nous traitons ce problème comme un problème de classification de textes. La classification de textes a fait un grand progrès ces dernières années grâce aux techniques d’apprentissage profond. En particulier, les modèles utilisant un mécanisme d’attention tel que BERT se sont révélés capables de capturer les informations contextuelles contenues dans une phrase ou un texte. Cependant, leur capacité à saisir l’information globale sur le vocabulaire d’une langue dans une application spécifique est plus limitée. Récemment, un nouveau type de réseau de neurones, appelé Graph Convolutional Network (GCN), émerge. Il intègre les informations des voisins en manipulant un graphique global pour prendre en compte les informations globales, et il a obtenu de bons résultats dans de nombreuses tâches, y compris la classification de textes. Par conséquent, notre motivation dans ce mémoire est de concevoir une méthode qui peut combiner à la fois les avantages du modèle BERT, qui excelle en capturant des informations locales, et le modèle GCN, qui fournit les informations globale du langage. Néanmoins, le GCN traditionnel est un modèle d'apprentissage transductif, qui effectue une opération convolutionnelle sur un graphe composé d'éléments à traiter dans les tâches (c'est-à-dire un graphe de documents) et ne peut pas être appliqué à un nouveau document qui ne fait pas partie du graphe pendant l'entraînement. Dans ce mémoire, nous proposons d'abord un nouveau modèle GCN de vocabulaire (VGCN), qui transforme la convolution au niveau du document du modèle GCN traditionnel en convolution au niveau du mot en utilisant les co-occurrences de mots. En ce faisant, nous transformons le mode d'apprentissage transductif en mode inductif, qui peut être appliqué à un nouveau document. Ensuite, nous proposons le modèle Interactive-VGCN-BERT qui combine notre modèle VGCN avec BERT. Dans ce modèle, les informations locales captées par BERT sont combinées avec les informations globales captées par VGCN. De plus, les informations locales et les informations globales interagissent à travers différentes couches de BERT, ce qui leur permet d'influencer mutuellement et de construire ensemble une représentation finale pour la classification. Via ces interactions, les informations de langue globales peuvent aider à distinguer des mots ambigus ou à comprendre des expressions peu claires, améliorant ainsi les performances des tâches de classification de textes. Pour évaluer l'efficacité de notre modèle Interactive-VGCN-BERT, nous menons des expériences sur plusieurs ensembles de données de différents types -- non seulement sur le langage haineux, mais aussi sur la détection de grammaticalité et les commentaires sur les films. Les résultats expérimentaux montrent que le modèle Interactive-VGCN-BERT surpasse tous les autres modèles tels que Vanilla-VGCN-BERT, BERT, Bi-LSTM, MLP, GCN et ainsi de suite. En particulier, nous observons que VGCN peut effectivement fournir des informations utiles pour aider à comprendre un texte haiteux implicit quand il est intégré avec BERT, ce qui vérifie notre intuition au début de cette étude. / Hate speech is a serious problem on social media. In this thesis, we investigate the problem of automatic detection of hate speech on social media. We cast it as a text classification problem. With the development of deep learning, text classification has made great progress in recent years. In particular, models using attention mechanism such as BERT have shown great capability of capturing the local contextual information within a sentence or document. Although local connections between words in the sentence can be captured, their ability of capturing certain application-dependent global information and long-range semantic dependency is limited. Recently, a new type of neural network, called the Graph Convolutional Network (GCN), has attracted much attention. It provides an effective mechanism to take into account the global information via the convolutional operation on a global graph and has achieved good results in many tasks including text classification. In this thesis, we propose a method that can combine both advantages of BERT model, which is excellent at exploiting the local information from a text, and the GCN model, which provides the application-dependent global language information. However, the traditional GCN is a transductive learning model, which performs a convolutional operation on a graph composed of task entities (i.e. documents graph) and cannot be applied directly to a new document. In this thesis, we first propose a novel Vocabulary GCN model (VGCN), which transforms the document-level convolution of the traditional GCN model to word-level convolution using a word graph created from word co-occurrences. In this way, we change the training method of GCN, from the transductive learning mode to the inductive learning mode, that can be applied to new documents. Secondly, we propose an Interactive-VGCN-BERT model that combines our VGCN model with BERT. In this model, local information including dependencies between words in a sentence, can be captured by BERT, while the global information reflecting the relations between words in a language (e.g. related words) can be captured by VGCN. In addition, local information and global information can interact through different layers of BERT, allowing them to influence mutually and to build together a final representation for classification. In so doing, the global language information can help distinguish ambiguous words or understand unclear expressions, thereby improving the performance of text classification tasks. To evaluate the effectiveness of our Interactive-VGCN-BERT model, we conduct experiments on several datasets of different types -- hate language detection, as well as movie review and grammaticality, and compare them with several state-of-the-art baseline models. Experimental results show that our Interactive-VGCN-BERT outperforms all other models such as Vanilla-VGCN-BERT, BERT, Bi-LSTM, MLP, GCN, and so on. In particular, we have found that VGCN can indeed help understand a text when it is integrated with BERT, confirming our intuition to combine the two mechanisms.

The impact of the law on the lives of homeless people in Queensland has, to date, remained largely unexplored by legal academics and researchers. This is despite the fact that homeless people experience a number of legal difficulties that seriously affect their lives. This thesis by published papers aims to make a significant and original contribution to filling this gap in the research evidence by presenting the results of analyses of the legal, theoretical and practical issues that arise in the context of homeless persons' interactions with the legal system in Queensland. Most notably, it is comprised of three pieces of empirical research which identify those areas of law that impact most on homeless people in Queensland and explore the consequences of the operation of these laws on their lives. In sum, this thesis examines the extent of the law's influence on the lives of homeless people in Queensland, and finds that the consequences of the law's operation on homeless people in Queensland are serious. The thesis first examines the effect on Queensland's homeless people of laws which regulate behaviour conducted in public space. The criminal offences of vagrancy, begging and public nuisance are analysed; their historical origins, the reasons for their retention on modern statute books, and arguments in favour of their repeal are discussed. The impact of 'public space law' on homeless people in Queensland is also explored through a survey of 30 homeless people residing in inner-city Brisbane. This part of the thesis concludes that public space law in Queensland results in breaches of homeless persons' human rights, as well as the contravention of rule of law principles. The thesis then explores the impact of the law on homeless persons' experiences of citizenship. Empirical research and theoretical analysis demonstrate that the application of various laws, particularly public space laws, social security laws and electoral laws, encroaches on homeless persons' citizenship rights. The thesis then reports on the results of a unique survey of Queensland's homelessness service providers. This survey is the most extensive piece of empirical research ever conducted on the extent to which various laws impact on homeless people. Respondents were asked to indicate which areas of law impact most adversely on their homeless clients. Based on the research findings outlined above, the hypothesis was that criminal law issues, particularly public space offences, would be proven to impact particularly adversely on homeless people in Queensland. Somewhat unexpectedly, the findings of the survey indicated that fines law, debt law and family law difficulties are those legal difficulties most often encountered by homeless people in Queensland. Difficulties produced by criminal laws, social security laws and electoral laws, while still generally relevant, rated less highly. However, the survey did demonstrate that experiences differ between sub-groups within the homeless population, for example Indigenous homeless people were reported to be most affected by criminal law issues, while young homeless people were reported to be most affected by social security law issues. Together, the five papers which comprise this thesis make an original and substantial contribution to knowledge by identifying empirically for the first time the various laws that have a significant impact on the lives of homeless people in Queensland, and analysing the consequences of this in terms of their effect on homeless persons' citizenship rights, human rights and rule of law entitlements.