Due to drunk-driving fatality is the most focal issue in the U.S. traffic accident, this paper applies panel data model to analyze the influence of beer tax and other drunk-driving related laws on the U.S. drunk-driving fatality rate from 1982-2006. Different from former references, this paper investigate if the drunk-driving fatality rate declines by the time and by region. The result shows that drunk-driving fatality rate has declined by the time, especially, in 1987; the drunk-driving fatality rate statistically significant drops. The posibility is that U.S. Secretary of Transportation, Elizabeth, ordered the automobile manufactures to set seat belt or air bag with the cars from 1987. However, the drunk-driving fatality rate doesn¡¦t show different significantly by region. This may suggest that regions have similar laws and cultural norms, which leads to similar drunk-driving fatality rate. Finally, the survey indicates the higher beer tax, BAC 0.08 Law, and Zero Tolerance Law are effective policies to reduce drunk-driving.

The amount of information on the internet is increasing every day and it is very difficult to find exact information from big list of search results. Users typically generate query specifying keywords and the search engine displays documents, what contains keywords what were set by user. Search within unstructured data starts from data processing: first of all documents are braked into sentences, then words and word groups are analyzed semantically and syntactically, to obtain facts. Facts describe objects of real world. Facts obtained from unstructured data will be stored into database, and unstructured data will be transformed into structured data. Afterwards it will be easy to analyze, conjunct, filter or do other manipulations with structured data.

Paleo data-model comparison is the process of comparing output from model simulations of past periods with paleoenvironmental data. It enables us to understand both the paleoclimate mechanism and responses of the earth environment to the climate and to evaluate how models work. This dissertation has two parts that each involve the development and application of approaches for data-model comparisons. In part 1, which is focused on the understanding of both past and future climatic changes/variations, I compare paleoclimate and historical simulations with future climate projections exploiting the fact that climate-model configurations are exactly the same in the paleo and future simulations in the Coupled Model Intercomparison Project Phase 5. In practice, I investigated large-scale temperature responses (land-ocean contrast, high-latitude amplification, and change in temperature seasonality) in paleo and future simulations, found broadly consistent relationships across the climate states, and validated the responses using modern observations and paleoclimate reconstructions. Furthermore, I examined the possibility that a small set of common mechanisms controls the large-scale temperature responses using a simple energy-balance model to decompose the temperature changes shown in warm and cold climate simulations and found that the clear-sky longwave downward radiation is a key control of the robust responses. In part 2, I applied the equilibrium terrestrial biosphere models, BIOME4 and BIOME5 (developed from BIOME4 herein), for reconstructing paleoclimate. I applied inverse modeling through the iterative forward-modeling (IMIFM) approach that uses the North American vegetation data to infer the mid-Holocene (MH, 6000 years ago) and the Last Glacial Maximum (LGM, 21,000 years ago) climates that control vegetation distributions. The IMIFM approach has the potential to provide more accurate quantitative climate estimates from pollen records than statistical approaches. Reconstructed North American MH and LGM climate anomaly patterns are coherent and consistent between variables and between BIOME4 and BIOME5, and these patterns are also consistent with previous data synthesis. This dissertation includes previously published and unpublished coauthored material.

Thesis (Ph.D.)--Boston University / PLEASE NOTE: Boston University Libraries did not receive an Authorization To Manage form for this thesis or dissertation. It is therefore not openly accessible, though it may be available by request. If you are the author or principal advisor of this work and would like to request open access for it, please contact us at open-help@bu.edu. Thank you. / This thesis describes a modularized universal relation data model, called URMSC (Universal Relation Data Model with Semantic Constructs). This model incorporates concepts from the semantic data model, such as the notions of entities, aggregation abstractions and generalization/specialization abstractions, into the universal relation data model. These semantic constructs allow us to reduce multiple paths among relation and allow us to loosen some of the constraints that the URM imposes. Consequently, a larger class of queries can be interpreted unambiguously. Modules are defined by aggregation hierarchies. Each module must satisfy the local unique name and attr£bute correlat£on assumptions. The former requires the uniqueness of attribute names on each individual module instead of the entire database scheme, to reduce the number of attribute renamings. The latter is used to disambiguate a query interpretation when there are duplicate attribute names in a database. In a universal relation system, data retrieval queries are expressed by mentioning only the attribute names of interest and their values. We extend this idea to the notion of database update. By taking advantage of the semantic constructs of the model, we provide criteria that guarantee unambiguous interpretation of update requests. / 2031-01-01

Understanding and managing change is a strategic objective for many organisations to successfully compete in a market place; as a result, organisations are leveraging their data asset and implementing data warehouses to gain business intelligence necessary to improve their businesses. Data warehouses are expensive initiatives, one-half to two-thirds of most data warehousing efforts end in failure. In the absence of well-formalised design methodology in the industry and in the context of the debate on data architecture in data warehousing, this thesis examines why multidimensional and relational data models define the data architecture landscape in the industry. The study develops a number of propositions from the literature and empirical data to understand the factors impacting the choice of logical data model in data warehousing. Using a comparative case study method as the mean of collecting empirical data from the case organisations, the research proposes a conceptual model for logical data model adoption. The model provides a framework that guides decision making for adopting a logical data model for a data warehouse. The research conceptual model identifies the characteristics of business requirements and decision pathways for multidimensional and relational data warehouses. The conceptual model adds value by identifying the business requirements which a multidimensional and relational logical data model is empirically applicable.

Esta tese envolve duas áreas de conhecimento, nomeadamente a de modelagem de dados para Sistemas de Gerecnciamento de Bases de Dados, e a de desenvolvimento de Máquinas de Bases de Dados. Devido a isso, esta tese apresenta-se também dividida em duas partes. Na primeira parte analisam-se os modelos já existentes e a partir das deficiências que apresentam para aplicações como Base de Dados para Engenharia, define-se o Modelo de Representação de Objetos. Na segunda parte são analisados arquiteturas de Máquinas de Base de Dados existentes e faz-se a proposta de uma nova arquitetura dedicada, para suportar uma implementação capaz de aproveitar o paralelismo que o modelo apresentado permite. Nas duas partes faz-se um levantamento de trabalhos relevantes que existem nas respectivas áreas, e mostra-se como as soluções apresentadas satisfazem as necessidades inerentes de cada parte / This work deals with two research areas: the data modeling for Database Management Systems; and the development of Database Machines. This work is divided in two parts. The first one analyzes already existent data models, and based on the characteristics required from engineering database applications, the Object Representation Model is defined. The second part analyzes existing database machines architectures and proposes a new one intended to support the intrinsic parallelism of the algorithms developed to implement the presented data model. A survey of relevant results obtained in both areas is included and a thorough discussion concludes the work

Esta tese envolve duas áreas de conhecimento, nomeadamente a de modelagem de dados para Sistemas de Gerecnciamento de Bases de Dados, e a de desenvolvimento de Máquinas de Bases de Dados. Devido a isso, esta tese apresenta-se também dividida em duas partes. Na primeira parte analisam-se os modelos já existentes e a partir das deficiências que apresentam para aplicações como Base de Dados para Engenharia, define-se o Modelo de Representação de Objetos. Na segunda parte são analisados arquiteturas de Máquinas de Base de Dados existentes e faz-se a proposta de uma nova arquitetura dedicada, para suportar uma implementação capaz de aproveitar o paralelismo que o modelo apresentado permite. Nas duas partes faz-se um levantamento de trabalhos relevantes que existem nas respectivas áreas, e mostra-se como as soluções apresentadas satisfazem as necessidades inerentes de cada parte / This work deals with two research areas: the data modeling for Database Management Systems; and the development of Database Machines. This work is divided in two parts. The first one analyzes already existent data models, and based on the characteristics required from engineering database applications, the Object Representation Model is defined. The second part analyzes existing database machines architectures and proposes a new one intended to support the intrinsic parallelism of the algorithms developed to implement the presented data model. A survey of relevant results obtained in both areas is included and a thorough discussion concludes the work

No description available.

Le développement de technologies comme les systèmes de positionnement par satellites (GNSS), les communications sans fil, les systèmes de radio-identification (RFID) et des capteurs a augmenté la disponibilité de données spatio-temporelles décrivant des trajectoires d’objets mobiles. Des bases de données relationnelles peuvent être utilisées pour stocker et questionner les données capturées. Des applications récentes montrent l’intérêt d’une approche intégrant des trajectoires « sémantiques » pour intégrer des connaissances sur les comportements d’objets mobiles. Dans cette thèse, nous proposons une approche basée sur des ontologies. Nous présentons une ontologie pour les trajectoires. Nous appliquons notre approche à l’étude des trajectoires de mammifères marins. Pour permettre l’exploitation de nos connaissances sur les trajectoires, nous considérons l’objet mobile, des relations temporelles et spatiales dans notre ontologie. Nous avons évalué la complexité du mécanisme d’inférence et nous proposons des optimisations, comme l’utilisation d’un voisinage temporel et spatial. Nous proposons également une optimisation liée à notre application. Finalement, nous évaluons notre contribution et les résultats montrent l’impact positif de la réduction de la complexité du mécanisme d’inférence. Ces améliorations réduisent de moitié le temps de calcul et permettent de manipuler des données de plus grande dimension. / Spatio-temporal data describing trajectories of moving objects has increased as a consequence of the larger availability of such data due to current sensors techniques. These devices use different technologies like global navigation satellite system (GNSS), wireless communication, radio-frequency identification (RFID), and sensors techniques. Although capturing technologies differ, the captured data has common spatial and temporal features. Thus, relational database management systems (RDBMS) can be used to store and query the captured data. RDBMS define spatial data types and spatial operations. Recent applications show that the solutions based on traditional data models are not sufficient to consider complex use cases that require advanced data models. A complex use case refers not only to data, but also to the domain expert knowledge and others. An inference mechanism enriches semantic trajectories with this knowledge. Temporal and spatial reasoning are fundamental for the inference mechanism on semantic trajectories. Several research fields are currently focusing on semantic trajectories to discover more information about mobile object behavior. In this thesis, we propose a modeling approach based on ontologies. We introduce a high-level trajectory ontology. The temporal and spatial parts form an implicit background of the trajectory model. So, we choose temporal and spatial models to be integrated with our trajectory model. We apply our modeling approach to a particular domain application : marine mammal trajectories. Therefore, we model this application and integrate it with our ontology. We implement our approach using RDF. Technically, we use Oracle Semantic Data Technologies. To accomplish reasoning over trajectories, we consider mobile objects, temporal and spatial knowledge in our ontology. Our approach demonstrates how temporal and spatial relationships that are common in natural language expressions (i.e., relations between time intervals like ”before”, ”after”, etc.) are represented in the ontology as user-defined rules. To annotate data with this kind of rules, we need an inference mechanism over trajectory ontology. Experiments over our model using the temporal and spatial reasoning address an inference computation complexity. This complexity is indicated in term of time computations and space storage. In order to reduce the inference complexity, we propose optimizations, such as domain constraints, temporal and spatial neighbor refinements. Moreover, controlling the repetition of the inference computation is also proposed. Even more, we define a refinement specifically for the application domain. Finally, we evaluate our contribution. Results show their positive impact on reducing the complexity of the inference mechanism. These refinements reduce half of the time computation and allow considering bigger size of the data.

Mining association rules is an important technique for discovering meaningful patterns in transaction databases. In the current literature, the properties of algorithms to mine associations are discussed in great detail. In this paper we investigate properties of transaction data sets from a probabilistic point of view. We present a simple probabilistic framework for transaction data and its implementation using the R statistical computing environment. The framework can be used to simulate transaction data when no associations are present. We use such data to explore the ability to filter noise of confidence and lift, two popular interest measures used for rule mining. Based on the framework we develop the measure hyperlift and we compare this new measure to lift using simulated data and a real-world grocery database. / Series: Research Report Series / Department of Statistics and Mathematics