This thesis analyses the monetary policy of the Fed in the period before the financial and economic crisis 2007. The aim of this analysis is to investigate whether the monetary policy significantly deviated from the preceding one which, according to John Taylor, adhered to the Taylor rule. This analysis is performed using both revised and real-time data. I compare the monetary policy with various specifications of the Taylor rule. I also estimate a forward-looking Taylor rule. The results confirm the deviation from the Taylor rule. However, deviations were usual in the whole sample period. Estimated models do not indicate that the monetary policy before the crisis was different from the rest of the analyzed period.

As more of the modern world is connected to the Internet, threats can reach further than ever before. Attacks happen all the time and many have serious consequences that disrupts the daily processes of people and companies, possibly causing lasting damage. To fight back, defensive tools are used to find and counter attacks. One of these tools is Snort. Snort finds malicious data packets and warns the user and counters the found attack. Snort relies on a list of signatures of different attacks, called a rule set, to know what is malicious. Many rule sets are available as paid subscriptions, but there are free alternatives. But how well can Snort defend a network using these free rule sets? By designing a network for experimentation and populating it with realistic background traffic, a group of rule sets are evaluated using a set of common attacks and tools. The performance hit when defending in a high speed, high bandwidth environment is evaluated as well. The results favour the Emerging Threats rule set. As for performance, Snort could not handle the most extreme amounts of traffic, with the rate of dropped packets making security dubious, but that occurred at the absolute peak of what consumer hardware can provide.

Bergmann’s and Allen’s rule suggest that for the same species individuals found further north, tend to have larger body sizes and smaller appendages compared to individuals further south, respectively, due to constraints for thermal regulation. This has shown especially true for birds and mammals.  In this paper, I test to see if global warming has led to a change in morphology for birds, i.e. body weight and wing size. We use year as a proxy for temperature as global warming has on average lead to an increase in the earth’s surface temperature over the last century with most of the change occurring since the 1980’s.  To test the hypothesis, I gathered data of male, young of the year birds during autumn migration for a select number of species with different wintering strategies, i.e. resident to long-distance migration. This data came from two Swedish bird observatories where standardized wing and weight measurements have been collected since 1986, coincident with the observed climate warming. The results of this study show that during this period, changes in body mass and wing length did not support the hypothesis that global warming has had a broad impact on the morphology of birds. Given that these results contradict that of other studies, analysis from species across a much wider latitudinal breadth of Europe, including, additional species and different age and sex classes should be investigated.

No description available.

No description available.

Elections are a democratic process intended to create peaceful and legitimate transfers of power. However, since the second and third waves of democratisation, policymakers and researchers have observed that electoral processes are transformed into arenas for political gain through violence. To counteract this destructive phenomenon, research needs to uncover why, where, how, by or on whom electoral violence occurs. A common denominator for countries that face election violence is colonial legacies. Therefore, this study attempts to explain the occurrence of electoral violence in postcolonial states through variance in colonial governance type, as they produce diverging societal aspects that can create conditions for electoral violence. The study applies an established theoretical framework from postcolonial theory on a regression testing the effect of indirect colonial rule, contra direct colonial rule on the occurrence of electoral violence. The regression results support the theory until colonial rulers and geographical location are controlled for, ultimately going against the proposed theory. However, the multivariate regression models are statistically insignificant, which makes the result of the study inconclusive. Robustness tests support the inconclusive result of the primary model and indicate the importance of segregating electoral violence based on perpetrator.

No description available.

No description available.

A Web Semântica tem como meta fazer com que os conteúdos disponibilizados na Web tenham significado não apenas para pessoas, mas também que possam ser processados por máquinas. Essa meta está sendo realizada com o desenvolvimento e uso de ontologias para criar dados anotados semanticamente. Entre as distintas formas de anotação semântica, a Semantic Web Rule Language (SWRL) torna possível criar anotações no formato de regras que combinam regras com conceitos definidos em ontologias, especificadas em Web Ontology Language (OWL), para representar conhecimento sobre dados por meio de afirmações condicionais. Todavia, à medida que o número dessas regras crescem, seus desenvolvedores podem enfrentar dificuldades para gerenciá-las adequadamente. Um grande conjunto de regras torna-se difícil de entender e propício a erros, principalmente quando usado e mantido de forma colaborativa. Neste trabalho é apresentado um conjunto de soluções para aprimorar o uso e gerenciamento de regras SWRL, que compreendem o desenvolvimento de novas representações visuais, técnicas de classificação de regras e ferramenta de detecção de erros. Essas soluções resultaram no SWRL Editor, uma ferramenta Web de visualização e composição de regras que roda como um plug-in para o Web Protégé. Como estudo de caso, foi utilizada a Autism Phenologue Rules, uma ontologia para caracterizar fenótipos de autismo, para exemplificar um conjunto grande e complexo de regras SWRL. A partir desse estudo, uma nova representação visual específica para as regras dessa ontologia foi elaborada, permitindo que um especialista em autismo, sem grandes conhecimentos computacionais, seja capaz de ver e editar regras sem ter de se preocupar com a sintaxe da linguagem SWRL. Os resultados obtidos indicam que o SWRL Editor é uma ferramenta clara e intuitiva, contribuindo para um melhor entendimento, criação e gerenciamento de regras SWRL. / The Semantic Web aims to make web content available not only to people but also to computers using machine-readable formats. This goal is being realized with the development and use of ontologies to create semantically annotated data. Among the different ways to annotate data, the Semantic Web Rule Language (SWRL) enables rule-based annotation that combines rules with ontology concepts, defined using the Web Ontology Language (OWL), to represent knowledge about data as conditional assertions. However, as the number of these rule-base annotations grows, developers face problems when trying to manage them. A large rule set becomes difficult to understand and prone to errors, especially when it is collaboratively maintained. This work presents solutions to improve SWRL rule use and management that include the development of new visual representations, classification techniques and error detection tools. These solutions resulted in the SWRL Editor, a webbased visualization and composition tool for SWRL rules that runs as a Web Protégé plug-in. As a case study, we used the Autism Phenologue Rules, an ontology to characterize autism phenotypes, to exemplify a large and complex SWRL rule set. From this study, a new visual representation, specific for this ontologys rules, has been developed, allowing an expert in autism, without a lot of computational knowledge, to be able to view and edit the rules without having to worry about SWRL syntax. The results obtained indicate that the SWRL Editor is a clear and intuitive tool, contributing for a better understanding and easing the creation and management of SWRL rule sets

In this thesis, we discuss different techniques to bridge the gap between two different approaches to artificial intelligence: the symbolic and the connectionist paradigm. Both approaches have quite contrasting advantages and disadvantages. Research in the area of neural-symbolic integration aims at bridging the gap between them. Starting from a human readable logic program, we construct connectionist systems, which behave equivalently. Afterwards, those systems can be trained, and later the refined knowledge be extracted.