If we are to understand how we can build machines capable of broadpurpose learning and reasoning, we must first aim to build systemsthat can represent, acquire, and reason about the kinds of commonsenseknowledge that we humans have about the world. This endeavor suggestssteps such as identifying the kinds of knowledge people commonly haveabout the world, constructing suitable knowledge representations, andexploring the mechanisms that people use to make judgments about theeveryday world. In this work, I contribute to these goals by proposingan architecture for a system that can learn commonsense knowledgeabout the properties and behavior of objects in the world. Thearchitecture described here augments previous machine learning systemsin four ways: (1) it relies on a seven dimensional notion of context,built from information recently given to the system, to learn andreason about objects' properties; (2) it has multiple methods that itcan use to reason about objects, so that when one method fails, it canfall back on others; (3) it illustrates the usefulness of reasoningabout objects by thinking about their similarity to other, betterknown objects, and by inferring properties of objects from thecategories that they belong to; and (4) it represents an attempt tobuild an autonomous learner and reasoner, that sets its own goals forlearning about the world and deduces new facts by reflecting on itsacquired knowledge. This thesis describes this architecture, as wellas a first implementation, that can learn from sentences such as ``Ablue bird flew to the tree'' and ``The small bird flew to the cage''that birds can fly. One of the main contributions of thiswork lies in suggesting a further set of salient ideas about how wecan build broader purpose commonsense artificial learners andreasoners.

We seek to both detect and segment objects in images. To exploit both local image data as well as contextual information, we introduce Boosted Random Fields (BRFs), which uses Boosting to learn the graph structure and local evidence of a conditional random field (CRF). The graph structure is learned by assembling graph fragments in an additive model. The connections between individual pixels are not very informative, but by using dense graphs, we can pool information from large regions of the image; dense models also support efficient inference. We show how contextual information from other objects can improve detection performance, both in terms of accuracy and speed, by using a computational cascade. We apply our system to detect stuff and things in office and street scenes.

Context-free tree languages play an important role in algebraic semantics and are applied in mathematical linguistics. In this thesis, we present some new results on context-free tree languages.

Scholars have researched how voters make decisions for well over a half a century, but these studies are limited in what they are able to say about how voters make decisions because they have focused on the choice rather than the process. Most of these studies have focused on the choice that voters reach or the way their memories are structured, overlooking the importance of the search and acquisition of information. Specifically, scholars in political science have paid little attention to how contextual variations in the information environment affect how voters make decisions. This dissertation investigate`s how changes in context affect how voters search for information. I explore three specific contexts: the number of offices on the ballot, the availability of partisan information about the candidates, and the presence or absence of campaign dialogue between two candidates. Indeed, one of the prominent features of American elections is the variation in the number of elections across jurisdictions, the availability of partisan information about candidates, and the amount of campaign dialogue between candidates--the three contexts that I examine in this study. I conduct three experiments that manipulate each of these contexts, using a dynamic information board that simulates the campaign environment and process tracing methods to track the information subjects chose to view and in what order they chose to view it. Results indicate that context shapes the way voters search for and acquire information. When faced with long ballots, subjects examined information less closely, they compared more information between candidates, and they searched for information less systematically. When subjects were unable to use the partisan cue, they compared less information between candidates and searched for information less systematically. Finally, when there was no dialogue between candidates, subjects searched for less information and had a less systematic search for it. These findings suggest that there are better ways to design elections.

This study explores the complexity of teachers’ attitudes towards promoting their pupils’ mental health in Kuwait middle schools, teachers’ perceptions of the contextual factors that shaped those attitudes, the barriers they identify that might hinder the implementation of the promotion process, and changes required to put such a promotion process into practice. A mixed-methodological research approach is adopted within a complementary research design, consisting of two stages. A total of 497 Kuwaiti middle school teachers completed a systematic survey and twelve teachers were subsequently chosen purposely to take part in semi-structured interviews. The study’s findings suggest that Kuwaiti middle school teachers tend to hold moderately favourable attitudes towards promoting pupils’ mental health. However, a variety of personal, interpersonal, socio-cultural, and structural-organizational barriers were perceived by teachers, which could undermine positive attitudes and impact on the implementation of promoting pupils’ mental health. The study also showed teachers’ attitudes and perceptions as markedly embedded within their socio-cultural and religious context.

The aim of this study was to describe the humor use by an 11-year-old adolescent with ASD. Through an iterative coding process to identify a successful, unsuccessful, and neutral rating of instances of humor, we describe how does an adolescent with ASD uses humor and the functions humor serves in his interactions with a friend. We describe the personal and environmental factors that support the successful use of humor. The adolescent used two main types of humor (self-initiated and environmentally-initiated), consisting of two forms (verbal and physical). We describe 12 main behavioral indicators to identify the instances of humor. Humor appeared to sever as a means of engaging his friend and he appeared to be most successful in using humor with his friend when in a familiar environment and engaging in a familiar activity. The implications for future research are discussed.

People talk about coherent episodes of their experience, leading to strong dependencies between words and the contexts in which they appear. Consequently, language within a context is more repetitive and more coherent than language sampled from across contexts. In this dissertation, I investigated how patterns in infant-directed speech differ under context-sensitive compared to context-independent analysis. In particular, I tested the hypothesis that cues to word boundaries may be clearer within contexts. Analyzing a large corpus of transcribed infant-directed speech, I implemented three different approaches to defining context: a top-down approach using the occurrence of key words from pre-determined context lists, a bottom-up approach using topic modeling, and a subjective coding approach where contexts were determined by open-ended, subjective judgments of coders reading sections of the transcripts. I found substantial agreement among the context codes from the three different approaches, but also important differences in the proportion of the corpus that was identified by context, the distribution of the contexts identified, and some characteristics of the utterances selected by each approach. I discuss implications for the use and interpretation of contexts defined in each of these three ways, and the value of a multiple-method approach in the exploration of context. To test the strength of statistical cues to word boundaries in context-specific sub-corpora relative to a context-independent analysis of cues to word boundaries, I used a resampling procedure to compare the segmentability of context sub-corpora defined by each of the three approaches to a distribution of random sub-corpora, matched for size for each context sub-corpus. Although my analyses confirmed that context-specific sub-corpora are indeed more repetitive, the data did not support the hypothesis that speech within contexts provides richer information about the statistical dependencies among phonemes than is available when analyzing the same statistical dependencies without respect to context. Alternative hypotheses and future directions to further elucidate this phenomenon are discussed. / 2019-02-17

This dissertation examines China’s market reforms over the last few decades, and their implications for (re)shaping socioeconomic inequalities in health. Specifically, I study the effect of marketization and related structural changes at community level on individual health outcomes. The first part of this dissertation revisits the market transition theory of Nee (1989), using individual health status as the outcome variable to assess Nee’s theory. Using multiple waves of a longitudinal survey from 1991 to 2006, I compare temporal changes in the role of human capital, political capital, and state policy in determining health under marketization. In partial support of the market transition theory, the empirical results show that the significance of human capital for health increases with marketization, while the return to political capital and one’s household registration status diminishes with a growing market. Additionally, I distinguish between marketization effects on community level, and different aspects of community context in shaping the SES-health link. I find that the level of urbanization and available resources within each community exert influences on self-rated health and change the relative importance of individual socioeconomic conditions in shaping health. Overall, this study provides new longitudinal evidence from China to support the notion that health is influenced by dynamic processes moderated by the structural changes as well as the social stratification system. I discuss the findings in the context of China’s market reform, fundamental causes theory, and socio-ecological perspectives, highlighting that health is determined by a nexus of life experiences and social environment that impact individuals at different levels. / 10000-01-01

This thesis looks at the role of Peter and Judas in the four canonical gospels, seeking to answer the question of why two such dramatic examples of failure in discipleship became, from so early on, an established and central part of the gospel narrative. The first chapter discusses the literary context of the gospels, considering issues such as the oral medium, the gospel communities, and the genre of the gospels. The second chapter examines the historical Peter and Judas, discussing the evidence in the rest of the New Testament and elsewhere in Christian tradition, followed by a range of parallel disciple figures in ancient literature, from the followers and debate partners of Plato’s Socrates to the students of the neo-Platonic philosopher Plotinus, making the case that a literary archetype for a disciple exists and is made use of in all of these texts. The third, fourth, fifth and sixth chapters look at each of the gospels in turn, discussing the way in which Peter and Judas correspond to the literary archetype and the ways that this archetype interacts both with the actual events of Jesus’ career and arrest, and potentially ongoing or recent events in each evangelist’s own community.

Purpose: The relationship between context and e-governance has been gaining a significant momentum in academic circles due its social and technical complexities. There are many challenges posed by the disparity between the context and the system when it comes to e-governance in developing countries. This research aims to reveal more successful adoption of e-governance initiatives and exposes factors that hinder its implementation. We develop a conceptual framework showing the reciprocity between the context and the system or what is termed “Context-System Gap”. Therefore, this research will study the appropriateness of the context and its influence on the system and the influence of the system on the context. The purpose of this research is to explore the factors that enable successful e-government adoption in Lebanon, where e-governance is still at its initial stage. Most empirical research and theories on the implementation of e-governance in developing countries remain at the macro-level and miss out on the complexities of the context of deployment and the role of the gap between the citizens and the government. The purpose of this thesis is to provide an empirical model differentiating between the electronic context and the electronic system and shed a light over a new gap, government-citizen gap, in the adoption of e-government. Design/methodology/approach: Following previous research on e-government services adoption, this study uses several technology use and acceptance models and literature to examine the elements behind the adoption and use of e-government services in Lebanon from citizen and government perspectives. The research strategy is a quantitative method approach employing questionnaire. Quantitative data will be collected from e-government users (citizens) and statistical tests will be conducted in order to examine the relation between variables. Practical implications: The findings are useful for policy-makers and decision-makers to develop a better understanding of citizens' needs. The proposed model can be used as a guideline for the implementation of e-government services in developing countries. Originality/value: This study is the only one to examine the dimensions influencing citizens’ adoption of e-government technologies in developing countries using a unified model merging context and system elements.