Abstract
Root cause analysis of failed projects and incidents is an important and necessary step to working out measures for preventing their recurrences. In this paper, to better analyze the causes of failed projects and incidents, we propose a novel topic-document-cause(TDC) model that reveals the corresponding relationships among topics, documents, and causes. We use the JST failure knowledge base to construct a TDC model with machine learning methods such as LDA and perceptron. The experimental results show that our approach performed better at discovering the causes of failures for projects and incidents.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Failure knowledge database (in Japanese), http://www.sozogaku.com/fkd/
Failure Knowledge Database of Startups (in Japanese), http://www.meti.go.jp/policy/newbusiness/kikidatabase/
David, M., Andrew, Y., Michael, I.: Latent Dirichlet Allocation. Journal of Machine Learning Research 3, 993–1022 (2003)
Ito, K., Horiguchi, Y., Nakanishi, H., Nakanishi, H.: Structural Analysis of Latent Similarity among Errors by Applying Text-Mining Methods to Documents of Failure Cases. In: 23rd Fuzzy System Symposium, pp. 527–532 (2007)
Aono, H., Ohta, M.: Construction of a Causal Network and Acquisition of Causal Knowledges by Searching Factors. In: The 2nd Forum on Data Engineering and Information Management B9-1 (2010) (in Japanese)
Nojima, Y., Ma, Q., Yoshikawa, M.: Causal Relation Mining by Utilizing Contents and Structural Features of News Articles. In: The 3rd Forum on Data Engineering and Information Management F4-4 (2011) (in Japanese)
Ishii, H., Ma, Q., Yoshikawa, M.: Incremental Construction of Causal Network from News Articles. Journal of Information Processing 20(1), 207–215 (2012)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2012 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Awano, Y., Ma, Q., Yoshikawa, M. (2012). Cause Analysis of New Incidents by Using Failure Knowledge Database. In: Liddle, S.W., Schewe, KD., Tjoa, A.M., Zhou, X. (eds) Database and Expert Systems Applications. DEXA 2012. Lecture Notes in Computer Science, vol 7447. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-32597-7_8
Download citation
DOI: https://doi.org/10.1007/978-3-642-32597-7_8
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-32596-0
Online ISBN: 978-3-642-32597-7
eBook Packages: Computer ScienceComputer Science (R0)