• Excerpt of the graph representation of the words associated with “jew” on /pol/. We extract the graph by finding the most similar words ( cutoff at 0.4 cosine distance value), and then we take the 2-hop ego network around “jew. In this graph the size of a node is proportional to its degree (i.e., the number of other nodes it is directly connected to); the color of a node is based on the community it is a member of; and the entire graph is visualized using a layout algorithm that takes edge weights into account (i.e., nodes with similar words will be closer in the visualization).
The Network Contagion Research Institute (NCRI) is a 501(c) organization (nonprofit) that deploys machine learning tools to expose hate on digital social networks within a cross-platform, public-minded, and global framework.

We are a multidisciplinary group of scientists and engineers who apply our technical skills to further public insight into the problem of online hate.

Learn more about us


Funding goes to paying for compute power, grad students, researchers, and engineers to help understand and stop hate.