Using virtual edges to extract keywords from texts modeled as complex networks
This work addresses keyword extraction for text mining applications, representing an incremental improvement by integrating embeddings into graph-based methods.
The study tackled keyword extraction by modeling texts as co-occurrence networks and adding virtual edges based on word embeddings from Word2vec and BERT, finding that this approach improved discriminability, with best results at low percentages of added edges and using metrics like degree, PageRank, and accessibility.
Detecting keywords in texts is important for many text mining applications. Graph-based methods have been commonly used to automatically find the key concepts in texts, however, relevant information provided by embeddings has not been widely used to enrich the graph structure. Here we modeled texts co-occurrence networks, where nodes are words and edges are established either by contextual or semantical similarity. We compared two embedding approaches -- Word2vec and BERT -- to check whether edges created via word embeddings can improve the quality of the keyword extraction method. We found that, in fact, the use of virtual edges can improve the discriminability of co-occurrence networks. The best performance was obtained when we considered low percentages of addition of virtual (embedding) edges. A comparative analysis of structural and dynamical network metrics revealed the degree, PageRank, and accessibility are the metrics displaying the best performance in the model enriched with virtual edges.