Yeah, or just find some grad student. Watching 1000 videos to get a quick sense of how it can be categorized as “justified” vs “not” vs “debatable” would take some time, but it wouldn’t be all that hard. Lots of research things take time. Requesting all the footage would be hard, dealing with all the holes in the database would be hard, basically the biggest of the underlying problems is that no one really cares enough to try to make any of this easy. But yes, having the reality to base the conversation on would be a very nice thing to have.
Yeah, or just find some grad student. Watching 1000 videos to get a quick sense of how it can be categorized as “justified” vs “not” vs “debatable” would take some time, but it wouldn’t be all that hard. Lots of research things take time. Requesting all the footage would be hard, dealing with all the holes in the database would be hard, basically the biggest of the underlying problems is that no one really cares enough to try to make any of this easy. But yes, having the reality to base the conversation on would be a very nice thing to have.
I don’t think a grad student could handle the volume here and still has some bias, even if the bias comes from “I didn’t get great sleep last night”.
Classification algorithms have been around for decades and are the perfect use of AI.