Evolving Knowledge Extraction From Online Resources
Lists
Only you see this
Abstract
In this paper, we present an evolving knowledge<br> extraction system named AKEOS (Automatic Knowledge Extraction<br> from Online Sources). AKEOS consists of two modules, including<br> a one-time learning module and an evolving learning module.<br> The one-time learning module takes in user input query, and<br> automatically harvests knowledge from online unstructured resources<br> in an unsupervised way. The output of the one-time learning is a<br> structured vector representing the harvested knowledge. The evolving<br> learning module automatically schedules and performs repeated<br> one-time learning to extract the newest information and track the<br> development of an event. In addition, the evolving learning module<br> summarizes the knowledge learned at different time points to produce<br> a final knowledge vector about the event. With the evolving learning,<br> we are able to visualize the key information of the event, discover<br> the trends, and track the development of an event.
BibTeX
@inproceedings{xiao2017evolving,
author = {Zhibo Xiao and Tharini Nayanika de Silva and Kezhi Mao},
title = {{Evolving Knowledge Extraction From Online Resources}},
booktitle = {Zenodo},
year = {2017},
doi = {10.5281/zenodo.1130978},
}Loading related content...