Browse > Article

Generating Test Data for Deep Neural Network Model using Synonym Replacement  

Lee, Min-soo (중앙대학교 소프트웨어학부)
Lee, Chan-gun (중앙대학교 소프트웨어학부)
Publication Information
Journal of Software Engineering Society / v.28, no.1, 2019 , pp. 23-28 More about this Journal
Abstract
Recently, in order to effectively test deep neural network model for image processing application, researches have actively conducted to automatically generate data in corner-case that is not correctly predicted by the model. This paper proposes test data generation method that selects arbitrary words from input of system and transforms them into synonyms in order to test the bug reporter automatic assignment system based on sentence classification deep neural network model. In addition, we compare and evaluate the case of using proposed test data generation and the case of using existing difference-inducing test data generations based on various neuron coverages.
Keywords
Deep Neural Network; testing; corner-case; test data generation;
Citations & Related Records
연도 인용수 순위
  • Reference
1 K. Pei, Y. Cao, J. Yang, S. Jana, "DeepXplore : automated whitebox testing of deep learning systems," In Proc. of SOSP, ACM, pp. 1-18, Oct, 2017.
2 Lei Ma, Felix Juefei-Xu, Jiyuan Sun, Chunyang Chen, Ting Su, Fuyuan Zhang, Minhui Xue, Bo Li, Li Li, Yang Liu, Jianjun Zhao, and Yadong Wang, "Deepguage: Comprehensive and multi-granularity testing criteria for gauging the robustness of deep learning systems," In Proc. of ASE, pp. 120-13, 2018.
3 박해성, 김수빈, 이찬근, 채병훈, 딥러닝 모델 기반의 버그 담당자 자동 배정 시스템 성능 측정, 한국 소프트웨어공학 학술대회 2018
4 Houssem Ben Braiek, Foutse Khomh, "On Testing Machine Learning Programs,", CoRR, abs/1812.02257, 2018, URL: https://arxiv.org/abs/1812.02257
5 Francesco Leofante, Nina Narodytska, Luca Pulina, Armando Tacchella, "Automated Verification of Neural Networks: Advances, Challenges and Perspectives," CoRR, abs/1805.09938, 2018, URL: http://arxiv.org/abs/1805.09938.