Clusterização de dados georreferenciados utilizando métodos de machine learning
Saved in:
| Main Author: | |
|---|---|
| Publication Date: | 2025 |
| Format: | Bachelor thesis |
| Language: | por |
| Source: | Repositório Institucional da UPF |
| Download full: | https://repositorio.upf.br/handle/123456789/9533 |
Summary: | This study investigates the application of machine learning techniques to the clustering of georeferenced data, focusing on a case study related to the management of Municipal Solid Waste (MSW) in the state of Rio Grande do Sul, Brazil. Data from different towns in the area was merged with details about waste disposal sites, and clustering methods like K-Means, K-Nearest Neighbors (KNN), DBSCAN, and Random Forest were used. The objective was to identify coherent territorial groupings that could support more effective public policies. The resulting regionalizations were compared with a baseline study for validation and comparative analysis of the models. The results show that spatial clustering methods perform satisfactorily, with K-Means and DBSCAN standing out, demonstrating the potential of machine learning techniques in spatial analysis applied to georeferenced data. |
Similar Items: Clusterização de dados georreferenciados utilizando métodos de machine learning
- Análise comparativa de métodos de clusterização em dados geoespaciais
- Detecção de tráfego anômalo de rede utilizando clusterização em Big Data
- Analise de técnicas de clusterização em MMO com dados restritos : o caso de Final Fantasy XIV
- Clusterização espacial e não espacial : um estudo aplicado à agropecuária brasileira
- Análise tensorial para prevenção de falsificações em sistemas de reconhecimento facial : uma proposta baseada em clusterização
- Insect pest image recognition : a few-shot machine learning approach including maturity stages classification