Skip to main content
. 2021 Jun 21;23(6):e25482. doi: 10.2196/25482

Figure 2.

Figure 2

Frequently occurring obsessional words and their clustering, based on semantic relationships. The word embedding was trained on the entirety of the data set and clustered using k-means with k=3 clusters. The font is scaled according to the frequency of occurrence of each word. For reference, bolded words are those that also appear in the Yale-Brown Obsessive-Compulsive Scale Symptom Checklist.