The Korean Journal of Applied Statistics (응용통계연구)

The Korean Statistical Society (한국통계학회)
권호 표지
DOI QR코드

DOI QR Code

Comparison of the performance of classification algorithms using cytotoxicity data

세포독성 자료를 이용한 분류 알고리즘 성능 비교

  • Yoon, Yeochang (Department of Information Security, Woosuk University) ;
  • Jeung, Eui Bae (Department of Veterinary Medicine, Chungbuk National University) ;
  • Jo, Na Rae (Department of Statistics, Chungbuk National University) ;
  • Ju, Su In (Department of Statistics, Chungbuk National University) ;
  • Lee, Sung Duck (Department of Statistics, Chungbuk National University)
  • 윤여창 (우석대학교 정보보안학과) ;
  • 정의배 (충북대학교 수의학과) ;
  • 조나래 (충북대학교 정보통계학과) ;
  • 주수인 (충북대학교 정보통계학과) ;
  • 이성덕 (충북대학교 정보통계학과)
  • Received : 2018.05.28
  • Accepted : 2018.06.08
  • Published : 2018.06.30
Download PDF
⟨ Previous Next ⟩

Abstract

An alternative developmental toxicity test using mouse embryonic stem cell derived embryoid bodies has been developed. This alternative method is not to administer chemicals to animals, but to treat chemicals with cells. This study suggests the use of Discriminant Analysis, Support Vector Machine, Artificial Neural Network and k-Nearest Neighbor. Algorithm performance was compared with accuracy and a weighted Cohen's kappa coefficient. In application, various classification techniques were applied to cytotoxicity data to classify drug toxicity and compare the results.

최근 동물실험의 대체방법 중 하나로 쥐의 줄기세포 유래 배상체를 이용하여 독성을 시험하는 방법이 개발되었다. 이는 동물에 직접 약물을 주입하는 것이 아닌 배상체 세포에 약물을 투입하여 세포의 변화에 따른 측정값들을 얻는 방법이다. 본 연구에서는 다범주 세포독성 자료를 이용해 통계적 기법인 판별분석(discriminant analysis)과 머신러닝 기법인 서포트 벡터 머신(support vector machine), 인공신경망(artificial neural network), k-인접이웃분류(k-nearest neighbor)의 성능을 비교하였다. 알고리즘의 성능은 분류 정확도(accuracy)와 가중카파계수(weighted Cohen's kappa coefficient)로 비교하였다.

Keywords

References

  1. Genschow, E., Scholz, G., Brown, N., et al. (2000). Development of prediction models for three in vitro embryotoxicity tests in an ECVAM validation study, ALTA, 13, 51-66.
  2. Kang, H. Y., Choi, Y. K., Jo, N. R., et al. (2017). Advanced developmental toxicity test method based on embryoid body's area, Reproductive Toxicology, 72, 74-85. https://doi.org/10.1016/j.reprotox.2017.06.185
  3. Scholkopf, B. and Smola, A. J. (2002). Learning with Kernels, Lon Don, The MIT Press.
  4. Seiler, A., Visan. A., Buesen, R., Genschow, E., and Spielmann, H. (2004). Improvement of an in vitro stem cell assay for developmental toxicity, Reproductive Toxicology, 18, 231-240. https://doi.org/10.1016/j.reprotox.2003.10.015
  5. Vapnik, V. N. (1998). Statistical Learning Theory, New York, Wiley.
  6. Yu, S. G. (2010). The Comparison of Classification Algorithms for Micro array data (Graduate School master's thesis), Chungbuk National University.