Robot Finder v2: Search and Ingestion of Educational Robotics Data from Youtube

• Gabriel Batistuta Urbano Lopes UERN
• Ryllari Raianne Marques de Santana Costa UERN / UFERSA
• Sebastião Emidio Alves Filho UERN / UFERSA

Resumo

Educational Robotics offers an interactive learning approach where students can turn abstract theories into concrete projects. With the exponential growth in content creation on internet platforms, finding specific, high-quality learning objects about robotics in education can be challenging. Platforms like YouTube, for example, have a vast collection of educational videos that cover various topics related to educational robotics. However, they are scattered among a massive amount of available data, making it difficult to have more assertive searches and identify the best resources. This work aims to present Robot Finder v2, a software that uses an automated approach to feed RepositORE, a repository of Learning Objects in Educational Robotics. The Robot Finder v2 searches and stores data from YouTube videos to make the search for this content simpler and more effective. The collected data feeds the RepostiORE, where these objects can be adapted and researched by users looking to acquire skills in educational robotics. For the objects to be found accurately, the software uses the YouTube Data API v3, which has a set of metadata in addition to the descriptions of the videos provided by their creators. The processing done in the Robot Finder optimizes the search and registration of information by mapping API endpoints and seeking correlations of metadata between RepositORE objects and YouTube videos.

Palavras-chave: Video on demand, Correlation, Education, Metadata, Search problems, Software, Web sites, Feeds, Robots, Videos, Learning Objects, Educational Robotics, YouTube API, Data Ingestion, System Interoperability, Data Integration, API Integration