Data Type and Data Sources for Agricultural Big Data and Machine Learning

Sustainable agriculture is currently being challenged under climate change scenarios since extreme environmental processes disrupt and diminish global food production. For example, drought-induced increases in plant diseases rainfall caused a decrease Machine Learning Agricultural Big Data are high-performance computing technologies that allow analyzing large amount of data to understand agricultural the processing analysis amounts heterogeneous for which intelligent IT high-resolution remote sensing techniques required. However, selection ML algorithms depends on types be used. Therefore, scientists need sources from they derived. These can structured, such as temperature humidity data, usually numerical (e.g., float); semi-structured, those spreadsheets information repositories, these not previously defined stored No-SQL databases; unstructured, files PDF, TIFF, satellite images, have been processed therefore any database but repositories Hadoop). This study provides insight into used along with their main challenges trends. It analyzes 43 papers selected through protocol proposed by Kitchenham Charters validated PRISMA criteria. was found primary Databases, Sensors, Cameras, GPS, Remote Sensing, capture Platforms Hadoop, Cloud Computing, Google Earth Engine. In future, Lakes will integration across different platforms, provide representation models other relationships between them, improving quality integrated.