Semantic Word Clouds with Background Corpus Normalization and t-distributed Stochastic Neighbor Embedding
This addresses the need for more informative visual document summaries for users of text analysis tools, though it is an incremental improvement over existing word cloud methods.
The authors tackled the problem of generating semantically meaningful word clouds by modeling word significance and affinity relative to a background corpus, using a modified t-SNE for placement and a scalable sketch-based method for corpus statistics. They empirically validated their approach on a large Wikipedia corpus, demonstrating improved semantic organization compared to random layouts.
Many word clouds provide no semantics to the word placement, but use a random layout optimized solely for aesthetic purposes. We propose a novel approach to model word significance and word affinity within a document, and in comparison to a large background corpus. We demonstrate its usefulness for generating more meaningful word clouds as a visual summary of a given document. We then select keywords based on their significance and construct the word cloud based on the derived affinity. Based on a modified t-distributed stochastic neighbor embedding (t-SNE), we generate a semantic word placement. For words that cooccur significantly, we include edges, and cluster the words according to their cooccurrence. For this we designed a scalable and memory-efficient sketch-based approach usable on commodity hardware to aggregate the required corpus statistics needed for normalization, and for identifying keywords as well as significant cooccurences. We empirically validate our approch using a large Wikipedia corpus.