Presentation 2022-01-12
Execution-trace embedding using word-proximity metric for a method to automatically classify test results
Takuma Ikeda, Kozo Okano, Shinpei Ogata, Shin Nakajima,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) The problem to solve automatically classifying the results of test executions is called the test oracle problem. This is one of the most important problems in test automation. In this paper, we propose a method for convolution of execution traces in machine learning models that aims to solve the test oracle problem. In the proposed method, we focus on the fact that the execution trace is a sequence of method invocations, and obtain a variance vector for each method invocation information using Word2vec. The variance vector of the execution trace is obtained by giving the obtained variance vector to a Long Short-Term Memory (LSTM) and encoding it. Using the learned NN model, we automatically classify the test execution results by grading them as pass or fail. We confirmed the effectiveness of the proposed method by using the execution traces of tests for various programs.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Test Oracle Problem / Supervised Learning / Execution Trace / Cosine Similarity
Paper # MSS2021-46,SS2021-33
Date of Issue 2022-01-04 (MSS, SS)

Conference Information
Committee SS / MSS
Conference Date 2022/1/11(2days)
Place (in Japanese) (See Japanese page)
Place (in English) Nagasakiken-Kensetsu-Sogo-Kaikan Bldg.
Topics (in Japanese) (See Japanese page)
Topics (in English) Mathematical Systems Science and its Applications, Software Science, etc.
Chair Takashi Kobayashi(Tokyo Inst. of Tech.) / Atsuo Ozaki(Osaka Inst. of Tech.)
Vice Chair Kozo Okano(Shinshu Univ.) / Shingo Yamaguchi(Yamaguchi Univ.)
Secretary Kozo Okano(Hiroshima City Univ.) / Shingo Yamaguchi(Tokyo Inst. of Tech.)
Assistant Shinpei Ogata(Shinshu Univ.) / Masato Shirai(Shimane Univ.)

Paper Information
Registration To Technical Committee on Software Science / Technical Committee on Mathematical Systems Science and its Applications
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Execution-trace embedding using word-proximity metric for a method to automatically classify test results
Sub Title (in English)
Keyword(1) Test Oracle Problem
Keyword(2) Supervised Learning
Keyword(3) Execution Trace
Keyword(4) Cosine Similarity
1st Author's Name Takuma Ikeda
1st Author's Affiliation Shinshu University(Shinshu Univ.)
2nd Author's Name Kozo Okano
2nd Author's Affiliation Shinshu University(Shinshu Univ.)
3rd Author's Name Shinpei Ogata
3rd Author's Affiliation Shinshu University(Shinshu Univ.)
4th Author's Name Shin Nakajima
4th Author's Affiliation National Institute of Informatics(NII)
Date 2022-01-12
Paper # MSS2021-46,SS2021-33
Volume (vol) vol.121
Number (no) MSS-317,SS-318
Page pp.pp.83-88(MSS), pp.83-88(SS),
#Pages 6
Date of Issue 2022-01-04 (MSS, SS)