In recent years, web has experienced a tremendous growth concerning users and content. As a result information overload problem has always been always one of the main discussion topics. The aim has always been to find the most desired solution in order to help users when they find it increasingly difficult to locate the accurate information at the right time. Recommender systems developed to address this need by helping users to find relevant information among huge amounts of data and they have now become a ubiquitous attribute to many websites. A recommender system guides users in their decisions by predicting their preferences while they are searching, shopping or generally surfing, based on their preferences collected from past as well as the preferences of other users. Until now, recommender systems has been vastly used in almost all professional e-commerce websites, selling or offering different variety of items from movies and music to clothes and foods. This thesis will present and explore different recommender system algorithms such as User-User Collaborative and Item-Item Collaborative filtering using open source library Apache mahout. Algorithms will be developed in order to evaluate the performance of these collaborative filtering algorithms. They will be compared and their performance will be measured in detail by using evaluation metrics such as RMSE and MAE and similarity algorithms such as Pearson and Loglikelihood.

Les champs électromagnétiques sont couramment utilisés pour stimuler de manière non-invasive le cerveau humain soit à des fins thérapeutiques ou dans un contexte de recherche. Les effets de la stimulation magnétique varient en fonction de la fréquence et de l'intensité du champ magnétique. Les mécanismes mis en jeu restent inconnus, d'autant plus lors de stimulations à faible intensité. Dans cette thèse, nous avons évalué les effets de stimulations magnétiques répétées à différentes fréquences appliqués à faible intensité (10-13 mT ; Low Intensity Repetitive Magnetic Stimulation : LI-rMS) in vitro, sur des cultures corticales primaires et sur des modèles de réparation neuronale. De plus, nous décrivons une méthodologie pour la construction d'un dispositif instrumental fait sur mesure pour stimuler des cultures cellulaires.Les résultats montrent des effets dépendant de la fréquence sur la libération du calcium des stocks intracellulaires, sur la mort cellulaire, sur la croissance des neurites, sur la réparation neuronale, sur l'activation des neurones et sur l'expression de gènes impliqués. En conclusion, nous avons montré pour la première fois un nouveau mécanisme d'activation cellulaire par les champs magnétiques à faible intensité. Cette activation se fait en l'absence d'induction de potentiels d'action. Les résultats soulignent l'importance biologique de la LI-rMS par elle-même mais aussi en association avec les effets de la rTMS à haute intensité. Une meilleure compréhension des effets fondamentaux de la LI-rMS sur les tissus biologiques est nécessaire afin de mettre au point des applications thérapeutiques efficaces pour le traitement des conditions neurologiques. / Electromagnetic fields are widely used to non-invasively stimulate the human brain in clinical treatment and research. This thesis investigates the effects of different low intensity (mT) repetitive magnetic stimulation (LI-rMS) parameters on single neurons and neural networks and describes key aspects of custom tailored LI-rMS delivery in vitro. Our results show stimulation specific effects of LI-rMS on cell survival, neuronal morphology, neural circuit repair and gene expression. We show novel mechanisms underlying cellular responses to stimulation below neuronal firing threshold, extending our understanding of the fundamental effects of LI-rMS on biological tissue which is essential to better tailor therapeutic applications.

Yes / The paper contributes to development of RMS through linkage with external stakeholders such as customers and suppliers of parts/raw materials to handle demand fluctuations that necessitate information sharing across the supply chain tiers. RMS is developed as an integrated supply chain hub for adjusting production capacity using a hybrid methodology of decision trees and Markov analysis. The proposed Markov Chain model contributes to evaluate and monitor system reconfigurations required due to changes of product families with consideration of the product life cycles. The simulation findings indicate that system productivity and financial performance in terms of the profit contribution of product-process allocation will vary over configuration stages. The capacity of an RMS with limited product families and/or limited model variants becomes gradually inoperative whilst approaching upcoming configuration stages due to the end of product life cycles. As a result, reconfiguration preparation is suggested quite before ending life cycle of an existing product in process, for switching from a product family to a new/another product family in the production range, subject to its present demand. The proposed model is illustrated through a simplified case study with given product families and transition probabilities.

<p>In order to understand how waves influence the exchange of momentum, latent heat and other parameters, between the ocean surface and the atmosphere, one can use models. A coupling between a wave model and an atmospheric regional climate model, for the Baltic Sea, will be performed at the Meteorology Institute in Uppsala University. The wave model is a state of the art, third generation wave model called WAM.</p><p>The new version of the WAM model (cycle 4) needs to be validated. The aim of this thesis is to perform this validation and also to investigate what meteorological forcing one should use to achieve best results. Two different types of forcing are analyzed, ERA40 reanalysis and the RCA climate model. In order to do this, observations from six different buoys in the Baltic Sea will be compared with the model output from WAM. The parameters that will be compared in this study are significant wave height, direction and peak period.</p><p>A consistent phenomenon for all the buoys is a slightly overestimation by the model of what the rate of this increases with increasing wave height. If one compares the model output when WAM are forced with the RCA climate model and when it is forced with ERA40 reanalysis, the differences between them are notable but not large. ERA40 is slightly better.</p><p>Significant wave height is quite good and gives a reasonably result. Some buoys and periods are better and some are worse. There are some differences for the significant wave height between the east coast and the west coast of Sweden, when forcing the model with RCA. It is slightly better on the west coast. On the contrary, the results from ERA40 are very coherent. The quality of the hindcast for the direction and the peak period, in contrast to the significant wave height, is not that good. The results are not bad, but it only gives a rough picture of the sea state.</p>

Student Number : 9811923K - PhD thesis - School of Mechanical Engineering - Faculty of Engineering and the Built Environment / Methods are presented for obtaining models used for predicting welded sample resistance and effective weld current (RMS) for desired weld diameter (weld quality) in the resistance spot welding process. These models were used to design predictive controllers for the welding process. A suitable process model forms an important step in the development and design of process controllers for achieving good weld quality with good reproducibility. Effective current, dynamic resistance and applied electrode force are identified as important input parameters necessary to predict the output weld diameter. These input parameters are used for the process model and design of a predictive controller. A three parameter empirical model with dependent and independent variables was used for curve fitting the nonlinear halfwave dynamic resistance. The estimates of the parameters were used to develop charts for determining overall resistance of samples for any desired weld diameter. Estimating resistance for samples welded in the machines from which dataset obtained were used to plot the chart yielded accurate results. However using these charts to estimate sample resistance for new and unknown machines yielded high estimation error. To improve the prediction accuracy the same set of data generated from the model were used to train four different neural network types. These were the Generalised Feed Forward (GFF) neural network, Multilayer Perceptron (MLP) network, Radial Basis Function (RBF) and Recurrent neural network (RNN). Of the four network types trained, the MLP had the least mean square error for training and cross validation of 0.00037 and 0.00039 respectively with linear correlation coefficient in testing of 0.999 and maximum estimation error range from 0.1% to 3%. A prediction accuracy of about 97% to 99.9%. This model was selected for the design and implementation of the controller for predicting overall sample resistance. Using this predicted overall sample resistance, and applied electrode force, a second model was developed for predicting required effective weld current for any desired weld diameter. The prediction accuracy of this model was in the range of 94% to 99%. The neural network predictive controller was designed using the MLP neural network models. The controller outputs effective current for any desired weld diameter and is observed to track the desired output accurately with same prediction accuracy of the model used which was about 94% to 99%. The controller works by utilizing the neural network output embedded in Microsoft Excel as a digital link library and is able to generate outputs for given inputs on activating the process by the push of a command button.

In order to understand how waves influence the exchange of momentum, latent heat and other parameters, between the ocean surface and the atmosphere, one can use models. A coupling between a wave model and an atmospheric regional climate model, for the Baltic Sea, will be performed at the Meteorology Institute in Uppsala University. The wave model is a state of the art, third generation wave model called WAM. The new version of the WAM model (cycle 4) needs to be validated. The aim of this thesis is to perform this validation and also to investigate what meteorological forcing one should use to achieve best results. Two different types of forcing are analyzed, ERA40 reanalysis and the RCA climate model. In order to do this, observations from six different buoys in the Baltic Sea will be compared with the model output from WAM. The parameters that will be compared in this study are significant wave height, direction and peak period. A consistent phenomenon for all the buoys is a slightly overestimation by the model of what the rate of this increases with increasing wave height. If one compares the model output when WAM are forced with the RCA climate model and when it is forced with ERA40 reanalysis, the differences between them are notable but not large. ERA40 is slightly better. Significant wave height is quite good and gives a reasonably result. Some buoys and periods are better and some are worse. There are some differences for the significant wave height between the east coast and the west coast of Sweden, when forcing the model with RCA. It is slightly better on the west coast. On the contrary, the results from ERA40 are very coherent. The quality of the hindcast for the direction and the peak period, in contrast to the significant wave height, is not that good. The results are not bad, but it only gives a rough picture of the sea state.

Requirements are a key aspect of software development. Requirements are also related with other software artefacts including designs, test cases and documentation. These artefacts are often captured with specialized models. However, many tools lack support for traceability relationships between requirements artefacts and model artefacts, leading to analysis issues. To establish traceability between models and other types of requirements artefacts, this thesis proposes a new Domain-Specific Language (DSL) for describing the concepts of a modeling language that would be intended to be traced using a Requirements Management System (RMS), with tool support handling the evolution of models and of their traceability links. In the first part of this thesis, the syntax and metamodel of the Model Traceability DSL (MT-DSL) are defined, together with an editor implemented using Xtext. In the second part of the thesis, a library of import and maintenance functions is generated automatically (using Xtend) from model traceability descriptions written using MT-DSL. The target language for this library is the DOORS eXtension Language (DXL), the scripting language of a leading commercial RMS with traceability support, namely IBM Rational DOORS. The implementation has been tested successfully for importing and evolution scenarios with two different modeling languages (User Requirements Notation and Finite State Machines). This work hence contributes a reliable mechanism to define and support traceability between requirements and models.

p. 1 - 107 / Submitted by Santiago Fabio (fabio.ssantiago@hotmail.com) on 2013-02-18T18:16:31Z No. of bitstreams: 1 2222.pdf: 583246 bytes, checksum: 538e8f7a78a59116ab39517f159109f5 (MD5) / Approved for entry into archive by Fatima Cleômenis Botelho Maria (botelho@ufba.br) on 2013-02-19T16:39:06Z (GMT) No. of bitstreams: 1 2222.pdf: 583246 bytes, checksum: 538e8f7a78a59116ab39517f159109f5 (MD5) / Made available in DSpace on 2013-02-19T16:39:06Z (GMT). No. of bitstreams: 1 2222.pdf: 583246 bytes, checksum: 538e8f7a78a59116ab39517f159109f5 (MD5) Previous issue date: 2007 / Esta dissertação aborda os fatores condicionantes de sobrevivência das empresas participantes do Arranjo Produtivo Local de Tecnologia da Informação (APL-TI) da Região Metropolitana de Salvador e Feira de Santana, como também os pilares para seu crescimento e longevidade. O Relatório de Pesquisa do SEBRAE Nacional, publicado em Agosto de 2004 – Fatores Condicionantes e Taxas de Mortalidade de Empresas no Brasil - é utilizado como base nesse trabalho de pesquisa, no que tange aos fatores condicionantes de sobrevivência. A premissa é que a gestão é o diferencial de sobrevivência, que se não a garante, colabora decisivamente para isso, sendo pressuposto que esse diferencial é válido para pequenas e médias empresas de qualquer segmento de negócio, inclusive Tecnologia da Informação. Foi desenvolvida revisão do referencial teórico acerca do tema sobrevivência e fatores condicionantes relacionados à gestão. A pesquisa de campo, com aplicação de questionário, mapeou as empresas do APL-TI fornecendo dados necessários à identificação do perfil do empresário e da empresa. A análise das evidências derivadas da sistematização das respostas aos questionários, permite, ao final, oferecer sugestões para a melhoria do processo de consolidação desse segmento de negócio no Estado da Bahia. / Salvador

Submitted by Automa??o e Estat?stica (sst@bczm.ufrn.br) on 2016-04-15T20:05:32Z No. of bitstreams: 1 SamaraDeCavalcantePaiva_DISSERT.pdf: 3765795 bytes, checksum: eed2c79e7ce6b6439acadf41bfab260e (MD5) / Approved for entry into archive by Arlan Eloi Leite Silva (eloihistoriador@yahoo.com.br) on 2016-04-19T23:21:12Z (GMT) No. of bitstreams: 1 SamaraDeCavalcantePaiva_DISSERT.pdf: 3765795 bytes, checksum: eed2c79e7ce6b6439acadf41bfab260e (MD5) / Made available in DSpace on 2016-04-19T23:21:12Z (GMT). No. of bitstreams: 1 SamaraDeCavalcantePaiva_DISSERT.pdf: 3765795 bytes, checksum: eed2c79e7ce6b6439acadf41bfab260e (MD5) Previous issue date: 2015-06-22 / Conselho Nacional de Desenvolvimento Cient?fico e Tecnol?gico (CNPq) / Um sistema el?trico de pot?ncia t?pico ? caracterizado pela centraliza??o da gera- ??o de energia. Entretanto, com a reestrutura??o do sistema el?trico, esta topologia vem modificando-se com a utiliza??o de geradores ao longo do sist ema de distribui??o (gera- ??o distribu?da) que proporciona diversos benef?cios por l ocalizar-se pr?ximo aos centros de consumo. Com isso, a inser??o de geradores distribu?dos, especialmente provenientes de fontes renov?veis, no sistema brasileiro tende a se torna r a cada ano mais comum. En- tretanto, esta nova configura??o do sistema traz novos desafi os concernentes ao controle, opera??o e prote??o. Um dos principais problemas da gera??o distribu?da nos sistemas de prote??o ? a forma??o de ilhamentos, que pode resultar em ris cos de seguran?a ?s pessoas e a rede de energia el?trica. Dentre as diversas t?cnicas de p rote??o de ilhamentos, as t?cnicas passivas destacam-se devido ao custo de implement a??o e simplicidade, sendo necess?rias apenas medi??es de tens?o e corrente para detec ??o de problemas no sistema. Prop?e-se neste trabalho um sistema de prote??o baseado na t ransformada wavelet com as fun??es de sobrecorrente, sub/sobretens?es e informa?? es dos transit?rios de falta para r?pida detec??o e identifica??o de situa??es de falta no sist ema. O esquema de prote??o proposto foi avaliado por meio de estudos de simula??o e expe rimental, apresentando de- sempenho similar aos m?todos de sobrecorrente e sub/sobret ens?o convencionais, por?m, com o adicional da detec??o do instante exato das faltas. / A typical electrical power system is characterized by centr alization of power gene- ration. However, with the restructuring of the electric sys tem, this topology is changing with the insertion of generators in parallel with the distri bution system (distributed gene- ration) that provides several benefits to be located near to e nergy consumers. Therefore, the integration of distributed generators, especially fro m renewable sources in the Brazi- lian system has been common every year. However, this new sys tem topology may result in new challenges in the field of the power system control, ope ration, and protection. One of the main problems related to the distributed generati on is the islanding formation, witch can result in safety risk to the people and to the power g rid. Among the several islanding protection techniques, passive techniques have low implementation cost and simplicity, requiring only voltage and current measuremen ts to detect system problems. This paper proposes a protection system based on the wavelet transform with overcur- rent and under/overvoltage functions as well as infomation of fault-induced transients in order to provide a fast detection and identification of fault s in the system. The propo- sed protection scheme was evaluated through simulation and experimental studies, with performance similar to the overcurrent and under/overvolt age conventional methods, but with the additional detection of the exact moment of the fault.

Requirements are a key aspect of software development. Requirements are also related with other software artefacts including designs, test cases and documentation. These artefacts are often captured with specialized models. However, many tools lack support for traceability relationships between requirements artefacts and model artefacts, leading to analysis issues. To establish traceability between models and other types of requirements artefacts, this thesis proposes a new Domain-Specific Language (DSL) for describing the concepts of a modeling language that would be intended to be traced using a Requirements Management System (RMS), with tool support handling the evolution of models and of their traceability links. In the first part of this thesis, the syntax and metamodel of the Model Traceability DSL (MT-DSL) are defined, together with an editor implemented using Xtext. In the second part of the thesis, a library of import and maintenance functions is generated automatically (using Xtend) from model traceability descriptions written using MT-DSL. The target language for this library is the DOORS eXtension Language (DXL), the scripting language of a leading commercial RMS with traceability support, namely IBM Rational DOORS. The implementation has been tested successfully for importing and evolution scenarios with two different modeling languages (User Requirements Notation and Finite State Machines). This work hence contributes a reliable mechanism to define and support traceability between requirements and models.