| Name: | Description: | Size: | Format: | |
|---|---|---|---|---|
| 7.73 MB | Adobe PDF |
Authors
Advisor(s)
Abstract(s)
Esta dissertação aborda a necessidade de melhorar as capacidades analíticas da Direção de curso da Licenciatura de Engenharia de Informática (LEI), do Instituto Superior de Engenharia do Porto (ISEP), sendo neste documento identificado pelo Cliente. Para realizar essa melhoria foi decidido conceber uma solução de Business Intelligence à medida, por forma a responder a essas necessidades.
Ao longo do documento são abordados vários conceitos teóricos considerados importantes para a realização da solução, bem como várias ferramentas de ETL(extração, transformação e carregamento), ferramentas de análise de dados e alternativas à solução elaborada. É de salientar que nas ferramentas de ETL é feita uma comparação entre si, utilizando a mesma metodologia nas ferramentas de análises de dados e soluções alternativas.
Após a apresentação do estado da arte, é exposto o desenho e análise da estrutura do Armazém de Dados por forma a dar resposta às várias análises pretendidas pelo cliente e, também, outras que até aqui eram complicadas de realizar. De seguida é descrito todo o detalhe de implementação, orientado às ferramentas utilizadas.
Por fim são demonstrados os resultados obtidos, efetuando uma comparação entre as análises disponibilizadas pelo cliente e as obtidas no armazém de dados, permitindo assim demonstrar a veracidade dos dados obtidos nesta nova solução. Como se pôde constatar, as discrepâncias obtidas foram significativas, levando a uma análise exaustiva para comprovar que a nova solução contém os dados corretos.
Não obstante, esta dissertação tem como principal contributo a melhoria contínua na obtenção de análises que demonstram a performance do curso na instituição. Também, contribui para garantir a credibilidade e uniformização de dados bem como diminuir a margem de erro no tratamento dos mesmos, pois não requer intervenção humana.
This master thesis addresses the need to improve the analytical capabilities of the current Direction of Computer Engineering Degree (LEI), in the Instituto Superior de Engenharia do Porto (ISEP), that is identified by the client in this document. To achieve this improvement, it was decided to design a customized Business Intelligence solution to meet those needs. In this document are presented several theoretical concepts that were considered importante for the realization of the solution, as well as several ETL tools (extraction, treatment, and loading), data analysis tools and alternatives to this new solution. It should be noted that in ETL tools is made a comparison between them, using the same methodology in data analysis tools and alternative solutions. After the state of the art presentation, is exposed the design and analysis of the structure of the Data Warehouse to respond to the various analyzes intended by the client and others that until now were complicated to perform. Next, we describe the entire implementation detail, oriented to the tools used. Finally, the results obtained are shown, making a comparison between the analyzes made available by the client and those obtained in the data warehouse, thus allowing to demonstrate the veracity of the data obtained in this new solution. As it turned out, the discrepancies obtained were significant, leading to an exhaustive analysis to prove that the new solution contains the correct data. Nevertheless, it´s important to refer that main contribution of this master thesis, is the continuous improvement in the process of obtaining analysis, that demonstrate the performance of the course, in the institution. It also contributes to ensure the credibility and standardization of data in the data warehouse, removing the human intervention in the process of data handling.
This master thesis addresses the need to improve the analytical capabilities of the current Direction of Computer Engineering Degree (LEI), in the Instituto Superior de Engenharia do Porto (ISEP), that is identified by the client in this document. To achieve this improvement, it was decided to design a customized Business Intelligence solution to meet those needs. In this document are presented several theoretical concepts that were considered importante for the realization of the solution, as well as several ETL tools (extraction, treatment, and loading), data analysis tools and alternatives to this new solution. It should be noted that in ETL tools is made a comparison between them, using the same methodology in data analysis tools and alternative solutions. After the state of the art presentation, is exposed the design and analysis of the structure of the Data Warehouse to respond to the various analyzes intended by the client and others that until now were complicated to perform. Next, we describe the entire implementation detail, oriented to the tools used. Finally, the results obtained are shown, making a comparison between the analyzes made available by the client and those obtained in the data warehouse, thus allowing to demonstrate the veracity of the data obtained in this new solution. As it turned out, the discrepancies obtained were significant, leading to an exhaustive analysis to prove that the new solution contains the correct data. Nevertheless, it´s important to refer that main contribution of this master thesis, is the continuous improvement in the process of obtaining analysis, that demonstrate the performance of the course, in the institution. It also contributes to ensure the credibility and standardization of data in the data warehouse, removing the human intervention in the process of data handling.
Description
Keywords
Business Intelligence Armazém de Dados Análises Cliente Data Warehouse Analysis Client