Skip to main content
Fig. 3 | BMC Bioinformatics

Fig. 3

From: Discovering themes in biomedical literature using a projection-based algorithm

Fig. 3

Similarity of document pairs within Themes and LDA-based clusters. The similarity between a pair of documents is computed as the dot product of two document vectors. These values are averaged over all within-theme document pairs and, further averaged over all themes of the same size. Same computation is applied to LDA-based clusters. Each point on the graph presents that average as a function of Themes / LDA-based cluster size

Back to article page