Aiming　at　the　characteristics　of　human-thinking-centric　studies,　brain　informatics　(BI)　emphasizes　a　systematic　methodology,　in　which　systematic　brain　data　analysis　is　a　key　issue.　In　this　paper,　a　multiagent　system,　namely,　the　global　learning　scheme　for　BI,　is　proposed　to　perform　a　data-brain　driven　process　planning.　Utilizing　the　obtained　three　types　of　mining　workflows,　a　three-step　analytical　process　can　be　carried　out　for　multiaspect　brain　data　analysis.　The　proposed　work　presents　a　practical　approach　toward　the　data-brain　driven　systematic　brain　data　analysis.　A　scenario　is　provided　to　demonstrate　the　significance　and　usefulness　of　the　proposed　approach.　©　2012　IEEE.