4 ms·
We've got huge quantities of multilingual news data with social metrics (e.g. newsapi.ai), classification models, research APIs for Twitter. But Nature's runni
by specproc 4y ago
We've got huge quantities of multilingual news data with social metrics (e.g. newsapi.ai), classification models, research APIs for Twitter.
But Nature's running Upworthy and dictionary-based classification? It's almost low-N by today's standards. This is the sort of paper that could have been written a couple of decades ago.
- h2odragon 4y agoThere was no need to reach for more sophisticated tools; these were sufficient to give them the results they wanted. I suspect you wanted research where people look at the world and discover what it means. This is the other kind, where people decide what they want the world to look like, then paint their data to reflect that.