Publicação
Sistema para análise de transações bancárias realizadas em comerciantes utilizando Datascience
| Resumo: | With the increasing amount of data in the financial sector, data analysis has become more and more important. With such analysis, it is possible not only to benefit financial institutions but also to favor customers. The report describes the work developed in partnership with the company Innovation Makers, which falls in the area of payments made at Point of Sales (POS). Considering that the use of these machines has been increasing significantly, it is important for the owners to get information, in a simple and fast way. This way, it will be possible to manage it in the most efficient way. The goal is to create a score that dictates the performance of each POS. This score should be generated using the information coming from the terminals. It is also intended to predict the amount that passes through each terminal. For this, we will use data science and data mining methodologies, with the intention of building a relational database. For the scoring system, Data Envelopment Analysis methodology will be used, and for Time Series forecast, the technique that best fits the data is the Prophet model. Furthermore, it is intended to elaborate a Dashboard report that displays this information and other elements in a clear way, in order to assist decision-making for the management of POS machinery. |
|---|---|
| Autores principais: | Fonseca, Inês Filipa Mascarenhas |
| Assunto: | Terminais de Pagamento Automático Análise Envoltória de Dados Dashboards Séries temporais Aprendizagem automática Teses de mestrado - 2024 |
| Ano: | 2024 |
| País: | Portugal |
| Tipo de documento: | dissertação de mestrado |
| Tipo de acesso: | acesso aberto |
| Instituição associada: | Universidade de Lisboa |
| Idioma: | português |
| Origem: | Repositório da Universidade de Lisboa |
| Resumo: | With the increasing amount of data in the financial sector, data analysis has become more and more important. With such analysis, it is possible not only to benefit financial institutions but also to favor customers. The report describes the work developed in partnership with the company Innovation Makers, which falls in the area of payments made at Point of Sales (POS). Considering that the use of these machines has been increasing significantly, it is important for the owners to get information, in a simple and fast way. This way, it will be possible to manage it in the most efficient way. The goal is to create a score that dictates the performance of each POS. This score should be generated using the information coming from the terminals. It is also intended to predict the amount that passes through each terminal. For this, we will use data science and data mining methodologies, with the intention of building a relational database. For the scoring system, Data Envelopment Analysis methodology will be used, and for Time Series forecast, the technique that best fits the data is the Prophet model. Furthermore, it is intended to elaborate a Dashboard report that displays this information and other elements in a clear way, in order to assist decision-making for the management of POS machinery. |
|---|