Simplistic Collection and Labeling Practices Limit the Utility of Benchmark Datasets for Twitter Bot Detection

Open in new window