Graph Based KNN for Text Segmentation
- Authors
- Jo, Taeho; T.
- Issue Date
- 2017
- Publisher
- IEEE
- Citation
- Proceedings - 2017 International Conference on Computational Science and Computational Intelligence, CSCI 2017, pp.322 - 327
- Journal Title
- Proceedings - 2017 International Conference on Computational Science and Computational Intelligence, CSCI 2017
- Start Page
- 322
- End Page
- 327
- URI
- https://scholarworks.bwise.kr/hongik/handle/2020.sw.hongik/13076
- DOI
- 10.1109/CSCI.2017.54
- Abstract
- In this research, we propose that the graph based KNN should be applied to the text segmentation task. as well as other tasks of text mining. The text segmentation may be interpreted into the binary task of texts where each pair of adjacent sentences or paragraphs is classified into whether we put the boundary between topics, or not, and the ontology which has been used as the popular and standard knowledge representation is given as a graph. In this research, we encode the adjacent sentence or paragraph pairs into graphs, and use the graph based K Nearest Neighbor for the text segmentation task. As benefits from this research, we may expect the more graphical symbolic, and compact representations of texts as well as the improved performance. Therefore, the goal of this research is to implement the text segmentation system with the benefits.
- Files in This Item
- There are no files associated with this item.
- Appears in
Collections - School of Games > Game Software Major > 1. Journal Articles
![qrcode](https://api.qrserver.com/v1/create-qr-code/?size=55x55&data=https://scholarworks.bwise.kr/hongik/handle/2020.sw.hongik/13076)
Items in ScholarWorks are protected by copyright, with all rights reserved, unless otherwise indicated.