Prior-Knowledge-Embedded LDA with Word2vec - for Detecting Specific Topics in Documents
Prior-Knowledge-Embedded LDA with Word2vec - for Detecting Specific Topics in Documents
Hiroshi Uehara,A. Ito,Yutaka Saito,K. Yoshida
2019 · DOI: 10.1007/978-3-030-30639-7_10
Pacific Rim Knowledge Acquisition Workshop · 3 Citations
TLDR
The approach uses word2vec to acquire linkages between words related to specific topics to apply prior knowledge about topics of interest to Latent Dirichlet Allocation (LDA) and enables to detect very specific topics complying with users’ interests.