Advertisement

Intelligent Call Triage System with Algorithm Combining Decision-Tree and SVM

Part of the Advances in Intelligent Systems and Computing book series (AISC, volume 273)

Abstract

In this paper, we propose a new online learning algorithm that constructs a subspace with a decision tree for a call triage support system. A call triage is an operation to determine the level of first-aid service quality on the basis of the severity and urgency of a victim. The call triage support system used in Yokohama is operated smoothly. However, further improvement in the accuracy detection is an issue because of the high percentage of misclassifications. This issue has been discussed in our prior implementation of a Bayesian network call triage support system. However, results indicate that the accuracy is decreased by over-training on the unlearned data. In this paper, we propose an algorithm to build a subspace with a decision tree for learning an increasing number of call triage records online. The evaluation experiment uses the past call triage records. Its results show that the proposed method can judge a call triage efficiently.

Keywords

Call Triage Decision tree Support vector machine 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Fire and Disaster Management Agency, Fire fighting White Paper 1999 edition, GYOUSEI (2000)Google Scholar
  2. 2.
    Fire and Disaster Management Agency, Fire fighting White Paper 2012 edition, Katsumi Print (2013)Google Scholar
  3. 3.
    Yunoki, S., Hamagami, T., Oshige, K., Kawakami, C., Suzuki, N.: High-accuracy Decision of Call-triage by using Bayesian Network. The Transactions of the Institute of Electrical Engineers of Japan. C, A Publication of Electronics., Information and Systems Society 132(1), 61–67 (2012)Google Scholar
  4. 4.
    Oshige, K., Kawakami, C., Mizusima, S., Moriaki, Y., Suzuki, N.: Evaluation of an algorithm for estimating a patient’s life threat risk from an ambulance call. BMC Emergency Medicine (2009)Google Scholar

Copyright information

© Springer International Publishing Switzerland 2014

Authors and Affiliations

  1. 1.Graduate School of EngineeringYokohama National UniversityHodogayakuJapan

Personalised recommendations