Esta pesquisa tem por finalidade analisar as construções sentenciais de tópico-comentário, de tópico marcado e de tópico não marcado, em produções textuais de estudantes do sexto ao nono anos do segundo segmento do ensino fundamental de escola pública localizada em comunidade da Zona Oeste do Rio de Janeiro, tendo sido analisadas 316 produções de um universo de 1040. O aporte teórico é o Funcionalismo, com ênfase na teoria dos três diferentes tipos de sujeito Sujeito Lógico, Sujeito Psicológico e Sujeito gramatical de Halliday (1994) e na teoria da Estrutura Informacional (Information Structure) de Lambrecht (1994). Partindo de análise quantitativa e qualitativa de corpus, investigou-se, de um lado, o percentual de aparecimento de construções de tópico marcado e não marcado e, de outro, os diferentes tipos de tópico não marcado e as motivações sintáticas e pragmático-discursivas de sua utilização. A análise foi feita em duas etapas. Primeiramente, para a identificação e a quantificação das construções sentenciais de tópico marcado, procedeu-se à investigação das seguintes variáveis: 1) tipo de tópico marcado mais recorrente, segundo categorização específica referente ao tópico e suas relações argumentais com o verbo 2) Posição mais frequente, considerando a estrutura textual e a ambiência sintática de favorecimento do uso do pronome correferente (presença/ausência de sintagma ou oração intercalada entre tópico sentencial e pronome correferente). A fim de investigar os contextos favorecedores de uso dos tópicos não marcados sujeito pronominal pleno de 3 pessoa e sujeito pronominal nulo de 3 pessoa, consideraram-se as seguintes variáveis: 1) tipo de oração e 2) conexão discursiva cf. Paredes Silva (2003). Os resultados obtidos permitem sistematizar as condições de uso das estruturas estudadas. Com relação às construções sentenciais de tópico marcado, verificou-se que se constitui em estratégia de ativar no discurso o elemento sobre o qual informações novas serão transmitidas, marcando-o como o tópico sentencial. Relativamente aos tópicos não marcados, as estratégias de uso estão relacionadas ao tipo de oração e ao grau de conexão discursiva da sentença. O uso do sujeito nulo está diretamente relacionado à conexão discursiva de grau 1, denominado ótimo. Uma breve análise de tais construções em autores consagrados da literatura brasileira e demais autores brasileiros contemporâneos permite afirmar que as construções de tópico, principalmente as que contêm tópico marcado, merecem uma atenção maior no ensino de Língua Portuguesa, uma vez que constituem uma estratégia pragmático-discursiva de uso corrente e de função relevante nos processos interacionais. O estudo das construções de tópico, portanto, merece uma atenção especial, uma vez que refletem a riqueza de nosso idioma, ao representar uma estratégia pragmático-discursiva que, se bem desenvolvida, constituir-se-á em uma ferramenta a mais para o usuário da língua no que concerne ao uso consciente e competente do idioma materno / The present study aims at analyzing sentence constructions of commentary topic, marked-topic and unmarked topic in textual productions of students attending the last four years of elementary education in a public school located in a poor community in the West zone of Rio de Janeiro. 316 out of 1040 productions have been analyzed. Functionalism has been used as the theoretical background with special emphasis on three different types of subject the Logical Subject, the Psychological Subject and the Grammatical Subject - Halliday (1994) and on Lambrechts information structure (1994). Based on the quantitative and qualitative analysis of the corpus, we have investigated, on the one hand, the percentage of marked and non-marked topic construction appearances and, on the other hand, the different types of non-marked topics and the syntactic and pragmatic discursive motivations of their usage. The analysis has been done in two stages. Firstly, for the identification and quantification of sentence constructions of marked topics, the investigation of the following variables has been made: 1st) most recurrent marked topic, according to specific categorization referring to the topic and its argumentative relationship with the verb, 2nd) most frequent position, taking into account the textual structure and the syntactic ambience of favoring the use of the coreferent pronoun (presence / absence of the syntagma or intercalated clause between sentence topic and coreferent pronoun). In order to investigate favoring contexts of the use of non-marked topics of 3rd person full pronominal subject and 3rd person null pronominal subject, the following variables have been considered: 1st) type of clause and 2nd) discursive connection, Paredes Silva (2003). The results obtained allow us to systematize the use conditions of the studied structures. When it comes to the sentence constructions of marked topic, we could perceive that it is a strategy of activating in the discourse the element about which the new information will be transmitted, marking it as the sentence topic. When it comes to non-marked topics, the strategies of use are related to the type of clause and to the degree of discursive connection of the sentence. The use of the null subject is directly related to the discursive connection degree 1, which is considered great. A brief analysis of such constructions in works of renowned authors in the field of Brazilian literature and other contemporary Brazilian authors allow us to say that topic constructions, especially the ones that contain marked topics, deserve greater attention in the education of Portuguese language, since they represent a pragmatic discursive strategy which is relevant and commonly used in interactional processes. Therefore, the study of topic construction deserves special attention since it reflects the richness of our language when it represents a pragmatic discursive strategy that, if well developed, can be one more tool for the language user when it comes to the conscientious and competent use of the mother tongue

The work deals with a research of essay writing lessons conditions plus most common methods used at elementary school. The work suggests new solutions to making the lessons more entertaining. It also says how to draw pupils´ attention to the subject. Finally, it focuses on choosing the most suitable topics and genres with concrete examples. It is attached with the main theoretical terms.

This paper explores topic modeling through the example text of Alice in Wonderland. It explores both singular value decomposition as well as non-­‐‑negative matrix factorization as methods for feature extraction. The paper goes on to explore methods for partially supervised implementation of topic modeling through introducing themes. A large portion of the paper also focuses on implementation of these techniques in python as well as visualizations of the results which use a combination of python, html and java script along with the d3 framework. The paper concludes by presenting a mixture of SVD, NMF and partially-­‐‑supervised NMF as a possible way to improve topic modeling.

Now, it is the world that high and new technology industries dominate. With the expansion of different industry chains and the trend of economic globalization, thousands of innovative solutions enter into the life of people. However, with the appearance of these ideas, initiators always face the challenge of presenting them. An intuitive and absorbing presentation plays the most important role of making the ideas accepted. As the saying goes, “No matter how good a movie is, it is boring without voice”. No matter how superior the products are, they will not be understood without perfect technical documents. This thesis presents the latest information typing architecture – DITA (Darwin Information Typing Architecture) which gives you an overview of technical presentations. As a newly developing XML-based (Extensible Markup Language) standard, DITA is not as popular as it ought to be. Seldom people even technical writers know about DITA and its characteristics. So far, the lack of practical materials leads to the difficulty for beginners to study DITA. The aim of this thesis is mainly to introduce DITA and make a technical tutorial. In order to strengthen the comprehension about DITA, the thesis makes a comparison with DocBook, another popular XML-based standard for technical writing and the most competitive standard of DITA. In the thesis, we collect a large number of materials of DITA and DocBook, and refine them for their comparison. For the practical tutorial, we make some simple examples in a DITA project and implement them on Serna Free, an XML editor. As a result, the thesis presents the detailed comparison between DITA and Docbook, and the tutorial includes the basic and vital part of DITA features.   Key words: DITA map, topic, Syntext Serna Free, tutorial

This thesis undertakes to research the development of domestic legal protection for privacy rights within employment. Unusually for a current work, it does this by returning to the genesis of the protections. The work challenges pervasive arguments that the notion of confidence did not protect privacy and exceptionally argues that the earliest incarnation of the notion of confidence was well equipped to protect private and personal information against disclosure. The thesis puts forward the novel view that the problems that have arisen in providing domestic protection for such information, in the absence of a right to privacy, are the result of an unfortunate narrowing of the original notion. This position is underpinned by the contention that a number of cases misinterpreted the nature of the original tort and did not fully recognise the requirement to provide protection against the acquisition of personal information. The historical analysis contributes to current knowledge by providing an alternative interpretation of the historical legal framework. This analysis provides an unorthodox assessment of the opportunities provided to the courts by the notion of confidence, to enhance both the theory and practical impact of the protection of privacy rights within employment Furthermore the thesis evaluates a broad range of case law from the European Court of Human Rights. These include general privacy cases, those involving wrongdoers as well as general employment cases. These cases are used to identify any consistent themes or conflicts in the application of the right to privacy. The evaluation produces a highly developed analysis and uncovers the latent significance of employment policies in both the protection of and the intrusion into, an employee's privacy. Most notably and distinctively the thesis identifies the important role that a well-crafted policy can have in augmenting an insubstantial legislative framework, provided that some legal basis provides the foundation for the policy. This evaluation also exposes the implications of any policy, which provides the basis for an interference with an employee's privacy and unconventionally highlights that the mere existence of such a policy can of itself amount to an intrusion, whether or not it is put into practice. Moreover, the thesis considers whether the incorporation of the right to privacy into domestic law has any impact upon private sector employers and employees. It heightens knowledge of the positive obligations placed upon the state and the courts to protect the right to privacy of all individuals against intrusions by the state and significantly against intrusions by other individuals or private sector organisations. The thesis therefore provides a valuable addition to current understanding of the interventionist and rigorous protections for privacy rights within employment, provided by the Strasbourg Court. This in turn provides the foundation for the unique evaluation of how effectively the right to privacy is incorporated into current domestic law. The thesis has taken the valuable opportunity provided by the tenth anniversary of the publication of the Human Rights Act 1998 to consider its impact upon privacy rights within employment. It takes advantage of the occasion to re-evaluate the categories of confidence and privacy and to analyse the principles underpinning the notions within domestic courts. The work exceptionally compares the development of domestic employment law with the development of general civil and criminal cases. This is an effective structure, which facilitates the development of arguments outlining how privacy rights within employment can be more effectively recognised and protected. The thesis does not shirk the challenges posed by the complex and difficult piece of legislation known as the Regulation of Investigatory Powers Act 2000, a particularly puzzling statute'. 1 It singularly and painstakingly reviews its provisions and questions whether accepted interpretations are correct or workable. The work offers an independent analysis of the rationale and application of its terms and brings to light the finding that despite its name Part II of the Act does not provide any powers and does not regulate the use of any existing powers but merely provides a framework, by which, intrusions may be undertaken by public authorities 'in accordance with the law' where there is no other statutory basis for the interference. The thesis makes an interesting case that other than where it creates criminal offences and civil liabilities for the interception of communications2, it has little if any effect upon the monitoring of employees, whether in the public or private sectors. The thesis also evaluates the Data Protection Act 1998 and The Employment Practices Code. It originally asserts that the Act and the Code provide the statutory basis for employers to intrude into the private lives of employees and prospective employees 'in accordance with the law' where it is necessary and proportionate to so; providing the employer has established and published policies that make the intrusion foreseeable in the circumstances. The fact that these arguments need to be evaluated and explained, naturally leads to the explicit conclusion that the incorporation of the rights protected by Article 8 in the statutory framework have failed to supply the necessary clarity to provide forseeability or to give, strengthen or explicitly restrict privacy rights within employment. Additionally, the creative comparison with the general privacy and criminal cases brings to light the evident disparity in the development of the case law in these areas compared with the development within employment cases. The thesis sets out the &adual but definite maturing of the protection noticeable in general privacy and criminal cases, particularly in v W [2003] EWCA Crim 1632 p. 98 2 Which themselves are modified by the Telecommunication (Lawful Business Practice) (Interception of Communications) Regulations, 2000 L1 relation to wrongdoing and rehabilitation, arguing that this has yet to filter into the employment cases. This leads to proposals for reform to remove the evident confusion for employers, employees, legal advisors, those providing oversight and commentators. The proposals encourage the judiciary to embrace the challenges and possibilities provided by the Human Rights Act 1998 to provide appropriate protections for privacy rights within employment. The thesis provides a platform for further research within this area and makes recommendations as to how the findings could be developed by both empirical research or by further comparative studies. The unusual approach to the research, the original nature of the findings and proposals for reform provide a valuable contribution to knowledge of the domestic legal framework, both statutory and common law suggesting both how it may be more effectively applied and how it could profitably be developed and clarified for both employers and employees. The thesis has thereby moved the debate to a different theoretical place from the established view of the ability of domestic law to effectively protect privacy rights within employment.

This doctoral thesis explores a range of issues within the development of the offence of incitement to genocide. It examines how the 'notorious Jew-baiter' Julius Streicher, was prosecuted in 1945/46 before the International Military Tribunal (IMT) for 'the incitement of the persecution of the Jews'. The newly defined category of 'crimes against humanity' under Article 6( c) of the Nuremberg Charter, classified 'persecutions on political, racial, or religious grounds' as a new criminal offence. In 1945, the prosecution alleged that Streicher's anti­Semitic propaganda had paved the way for Hitler's extermination policy, leading to the mass murder of approximately six million Jews. On October I, 1946, the Tribunal held that Streicher's actions of 'incitement to murder and extermination' were classified as 'persecution' and found him guilty of crimes against humanity. This thesis asks the question how was it that 'words used as persecution' became recognised as an international criminal offence as a sub set of crimes against humanity, without being directly or expressly criminalised by the Charter? In order to provide an answer to this question, this thesis conducts the first in-depth analysis and comprehensive reconstruction of how the prosecution's case against Streicher developed during the various phases of the pre-trial and trial process. This detailed reconstruction, based on archival sources not previously discussed in the academic literature on Nuremberg, forms the most original element of the thesis. The present study critically examines the factual evidence raised and dropped during different stages in the development of Streicher' s case. It explores the various strategies and tactics deployed by the prosecution, and evaluates the success and weaknesses, along with the counterarguments submitted by Streicher' s defence counsel. The aim in explicating and discussing these issues and conflicts of strategic interpretations and reinterpretations is to provide an original perspective that effectively explores the process 'behind the scenes'. This study suggests that it was this process of selective interpretation and decision-making that resulted in the emergence, or 'birth', of a new offence, 'incitement to genocide' that today would be recognised as 'direct and public incitement to commit genocide', under Article III( c) of the 1948 Convention on the Prevention and Punishment of the Crime of Genocide.

Structured probabilistic inference has shown to be useful in modeling complex latent structures of data. One successful way in which this technique has been applied is in the discovery of latent topical structures of text data, which is usually referred to as topic modeling. With the recent popularity of mobile devices and social networking, we can now easily acquire text data attached to meta information, such as geo-spatial coordinates and time stamps. This metadata can provide rich and accurate information that is helpful in answering many research questions related to spatial and temporal reasoning. However, such data must be treated differently from text data. For example, spatial data is usually organized in terms of a two dimensional region while temporal information can exhibit periodicities. While some work existing in the topic modeling community that utilizes some of the meta information, these models largely focused on incorporating metadata into text analysis, rather than providing models that make full use of the joint distribution of metainformation and text. In this thesis, I propose the event detection problem, which is a multidimensional latent clustering problem on spatial, temporal and topical data. I start with a simple parametric model to discover independent events using geo-tagged Twitter data. The model is then improved toward two directions. First, I augmented the model using Recurrent Chinese Restaurant Process (RCRP) to discover events that are dynamic in nature. Second, I studied a model that can detect events using data from multiple media sources. I studied the characteristics of different media in terms of reported event times and linguistic patterns. The approaches studied in this thesis are largely based on Bayesian nonparametric methods to deal with steaming data and unpredictable number of clusters. The research will not only serve the event detection problem itself but also shed light into a more general structured clustering problem in spatial, temporal and textual data.

In the era of "big data", scalable statistical inference is necessary to learn from new and growing sources of quantitative information. However, many commercial and scientific applications also require models to be interpretable to end users in order to generate actionable insights about quantities of interest. We present three case studies of Bayesian hierarchical models that improve the interpretability of existing models while also maintaining or improving the efficiency of inference. The first paper is an application to online advertising that presents an augmented regression model interpretable in terms of the amount of revenue a customer is expected to generate over his or her entire relationship with the company---even if complete histories are never observed. The resulting Poisson Process Regression employs a marginal inference strategy that avoids specifying customer-level latent variables used in previous work that complicate inference and interpretability. The second and third papers are applications to the analysis of text data that propose improved summaries of topic components discovered by these mixture models. While the current practice is to summarize topics in terms of their most frequent words, we show significantly greater interpretability in online experiments with human evaluators by using words that are also relatively exclusive to the topic of interest. In the process we develop a new class of topic models that directly regularize the differential usage of words across topics in order to produce stable estimates of the combined frequency-exclusivity metric as well as proposing efficient and parallelizable MCMC inference strategies. / Statistics

In this dissertation, Noto-ya Masogana is described as a moral story. The moral story can be distinguished from the didactic story and from the picaresque. The picaresque deals with a picaro who is the only main character. He is not presented as being trustworthy and responsible. As a result of this, this type of narrative is known as a 'Schelmroman' in German. The development of the story is simple and deals with the escapades of the picaro who goes on a journey. The didactic story presents the reader with a lesson, while the moral story indicates that a life well led will be rewarded. A life not well led will, in turn, be punished. In this narrative, Lesibana pays for his unfaithfulness to Mamahlo. He repents and the narrative has a happy ending. The method used in this investigation is based on an adjusted narratological model. Only the content and structure levels feature. The topic is identified as an important concept with regard to level one, the reason for this being that it influences the four different elements of the narrative, namely: characters, events time and place. Two main characters are distinguished, and the mutual relationship between them and less important characters receives attention. Events are divided into three categories according to importance. Events are thus called either essential, appropriate or coincidental. Two issues are relevant with regard to time; namely the moment at which an event takes place and the time which is occupied by an event. Place has a bearing on the natural as well as the sociocultural place. The concepts theme and narrating strategies play an important role on a structural level. In the discussion of the structure of Noto-ya-Masogana attention is paid to the exposition, the development, the climax and the resolution. The following techniques are relevant in this regard: motif, foreshadowing, retardation and acceleration of time, contrast, complication, focus and cycle. The study is conducted by comparing this novelette to the story of the Prodical Son and to the picaresque. Attention is also drawn to the author's use of the 'Makgoweng' motif. / Dissertation (MA)--University of Pretoria, 2012. / Philosophy / unrestricted

Human-machine teaming aims to meld human cognitive strengths and the unique capabilities of smart machines to create intelligent teams adaptive to rapidly changing circumstances. One major contributor to the problem of human-machine teaming is a lack of communication skills on the part of the machine. The primary objective of this research is focused on a machine’s interruption timings or when a machine should share and communicate information with human teammates within human-machine teaming interactions. Previous work addresses interruption timings from the perspective of single human, multitasking and multiple human, single task interactions. The primary aim of this dissertation is to augment this area by approaching the same problem from the perspective of a multiple human, multitasking interaction. The proposed machine is the Collaborative Communication Interruption Management System (C-CIMS) which is tasked with leveraging speech information from a human-human task and making inferences on when to interrupt with information related to an orthogonal human-machine task. This study and previous literature both suggest monitoring task boundaries and engagement as candidate moments of interruptibility within multiple human, multitasking interactions. The goal then becomes designing an intermediate step between human teammate communication and points of interruptibility within these interactions. The proposed intermediate step is the mapping of low-level speech information such as prosodic and lexical information onto higher constructs indicative of interruptibility. C-CIMS is composed of a Task Boundary Prosody Model, a Task Boundary Topic Model, and finally a Task Engagement Topic Model. Each of these components are evaluated separately in terms of how they perform within two different simulated human-machine teaming scenarios and the speed vs. accuracy tradeoffs as well as other limitations of each module. Overall the Task Boundary Prosody Model is tractable within a real-time system because of the low-latency in processing prosodic information, but is less accurate at predicting task boundaries even within human-machine interactions with simple dialogue. Conversely, the Task Boundary and Task Engagement Topic Models do well inferring task boundaries and engagement respectively, but are intractable in a real-time system because of the bottleneck in producing automatic speech recognition transcriptions to make interruption decisions. The overall contribution of this work is a novel approach to predicting interruptibility within human-machine teams by modeling higher constructs indicative of interruptibility using low-level speech information.