Nowadays, a growing volume of data is generated from various sources, such as sensors and social media, making data storage and management increasingly challenging. Extracting valuable insights from this raw data to support decision-making requires data mining methods, especially when combined with machine learning (ML) algorithms to uncover hidden patterns. This work presents the development of an integrated platform that combines a scientific data repository with ML tools, data science methods, and resources for geographic data visualization. The platform enables users to implement complete pipelines—from data storage and management to ML model training and deployment—streamlining workflows and making data science more accessible to a broader audience. Key features include a dataset repository, classic ML algorithms for tabular data, and new deep learning algorithms for image classification. The new model-sharing module facilitates collaboration by allowing users to share and manage trained models within the platform.

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Data Symbion EI: Expanding Horizons with Deep Learning and Model Sharing

  • Amanda Mendes,
  • Rafael Huszcza,
  • Jeferson Lopes,
  • Eduardo N. Borges,
  • Giancarlo Lucca,
  • Pablo D. B. Guilherme,
  • Leandro A. Pereira

摘要

Nowadays, a growing volume of data is generated from various sources, such as sensors and social media, making data storage and management increasingly challenging. Extracting valuable insights from this raw data to support decision-making requires data mining methods, especially when combined with machine learning (ML) algorithms to uncover hidden patterns. This work presents the development of an integrated platform that combines a scientific data repository with ML tools, data science methods, and resources for geographic data visualization. The platform enables users to implement complete pipelines—from data storage and management to ML model training and deployment—streamlining workflows and making data science more accessible to a broader audience. Key features include a dataset repository, classic ML algorithms for tabular data, and new deep learning algorithms for image classification. The new model-sharing module facilitates collaboration by allowing users to share and manage trained models within the platform.