Graph Based AHC Algorithm for Text Clustering
Taeho Jo · 2017
In this research, we propose the graph based AHC algorithm where a graph is given as the input, as the approach to the text clustering tasks. The ontology is the popular and standard representations of knowledge as a graph, so it may be more natural to encode texts into graphs rather than numerical vectors. In this research, we encode texts into graphs, define the similarity measure between graphs, and modify the AHC algorithm into the graph based version. As the benefits from this research, we may expect the better performance and more graphical representations of data items, by proposing that texts should be encoded so. Therefore, the goal of this research is to implement the text clustering system which has the benefits.