The model improvescoherence by exploiting neural word embeddings through a graph-based clusteringmethod . Unlike typical topic models, this approach works without knowing the true number of topics . Experimental results on the real-life multimodal dataset MuSe-CaR demonstrates that our approach extracts coherent and meaningfultopics, outperforming baseline methods . Furthermore, we successfullydemonstrate the generalisability of our approach on a pure text review dataset . We successfullydemonstrated the generalIsability of the approach on the pure textreview dataset .

Author(s) : Lukas Stappen, Gerhard Hagerer, Björn W. Schuller, Georg Groh

Links : PDF - Abstract

Code :
Coursera

Keywords : approach - dataset - pure - generalisability - graph -

Leave a Reply

Your email address will not be published. Required fields are marked *