Authors:
Miguel Feria
1
;
Juan Paolo Balbin
2
and
Francis Michael Bautista
2
Affiliations:
1
Mathematics and Statistics Department, De La Salle University, Taft Avenue, Manila, Philippines, Indigo Research, Katipunan Avenue, Quezon City and Philippines
;
2
Indigo Research, Katipunan Avenue, Quezon City and Philippines
Keyword(s):
Named Entity Recognition, Natural Language Processing, Graphs, Word Networks, Semantic Networks, Information Extraction.
Related
Ontology
Subjects/Areas/Topics:
Context-Awareness
;
Mobile Information Systems
;
Social Media Analytics
;
Society, e-Business and e-Government
;
Web Information Systems and Technologies
Abstract:
In this paper, we discuss a method for identifying a seed word that would best represent a class of named entities in a graphical representation of words and their similarities. Word networks, or word graphs, are representations of vectorized text where nodes are the words encountered in a corpus, and the weighted edges incident on the nodes represent how similar the words are to each other. Word networks are then divided into communities using the Louvain Method for community detection, then betweenness centrality of each node in each community is computed. The most central node in each community represents the most ideal candidate for a seed word of a named entity group which represents the community. Our results from our bilingual data set show that words with similar lexical content, from either language, belong to the same community.