Evolving Knowledge Extraction from Online Resources

In this paper, we present an evolving knowledge
extraction system named AKEOS (Automatic Knowledge Extraction
from Online Sources). AKEOS consists of two modules, including
a one-time learning module and an evolving learning module.
The one-time learning module takes in user input query, and
automatically harvests knowledge from online unstructured resources
in an unsupervised way. The output of the one-time learning is a
structured vector representing the harvested knowledge. The evolving
learning module automatically schedules and performs repeated
one-time learning to extract the newest information and track the
development of an event. In addition, the evolving learning module
summarizes the knowledge learned at different time points to produce
a final knowledge vector about the event. With the evolving learning,
we are able to visualize the key information of the event, discover
the trends, and track the development of an event.

Authors:

• Zhibo Xiao
• Tharini Nayanika de Silva
• Kezhi Mao

Keywords:

• Evolving learning
• knowledge extraction
• knowledge graph
• text mining.

References:

[1] J. Piskorski and R. Yangarber, Information Extraction: Past, Present
and Future. Berlin, Heidelberg: Springer Berlin Heidelberg, 2013, pp.
23–49.
[2] O. Etzioni, M. Cafarella, D. Downey, A.-M. Popescu, T. Shaked,
S. Soderland, D. S. Weld, and A. Yates, “Unsupervised named-entity
extraction from the web: An experimental study,” Artificial intelligence,
vol. 165, no. 1, pp. 91–134, 2005.
[3] O. Etzioni, M. Banko, S. Soderland, and D. S. Weld, “Open information
extraction from the web,” Communications of the ACM, vol. 51, no. 12,
pp. 68–74, 2008.
[4] S. S. Tan, T. Y. Lim, L.-K. Soon, and E. K. Tang, “Learning to extract
domain-specific relations from complex sentences,” Expert Systems with
Applications, vol. 60, pp. 107 – 117, 2016.
[5] L. Shou, Z. Wang, K. Chen, and G. Chen, “Sumblr: Continuous
summarization of evolving tweet streams,” in Proceedings of the 36th
International ACM SIGIR Conference on Research and Development in
Information Retrieval, ser. SIGIR ’13. New York, NY, USA: ACM,
2013, pp. 533–542.
[6] S. P. Kasiviswanathan, P. Melville, A. Banerjee, and V. Sindhwani,
“Emerging topic detection using dictionary learning,” in Proceedings of
the 20th ACM International Conference on Information and Knowledge
Management, ser. CIKM ’11. New York, NY, USA: ACM, 2011, pp.
745–754.
[7] Z. Jiang, X. Liu, and L. Gao, “Dynamic topic/citation influence modeling
for chronological citation recommendation,” in Proceedings of the
5th International Workshop on Web-scale Knowledge Representation
Retrieval & Reasoning, ser. Web-KR ’14. New York, NY, USA: ACM,
2014, pp. 15–18.
[8] Z. Jiang, “Chronological scientific information recommendation via
supervised dynamic topic modeling,” in Proceedings of the Eighth ACM
International Conference on Web Search and Data Mining, ser. WSDM
’15. New York, NY, USA: ACM, 2015, pp. 453–458.
[9] Z. Jiang, X. Liu, and L. Gao, “Chronological citation recommendation
with information-need shifting,” in Proceedings of the 24th ACM
International on Conference on Information and Knowledge
Management, ser. CIKM ’15. New York, NY, USA: ACM,
2015, pp. 1291–1300.
[10] S. Teller, Data Visualization with D3.Js. Packt Publishing, 2013.
[11] L. Tan, “Pywsd: Python implementations of word sense disambiguation
(wsd) technologies (software),” https://github.com/alvations/pywsd.