A Software Defect Prediction Method Based on Program Semantic Feature Mining

Round 1

Reviewer 1 Report

1. research gap is not properly identified.

2. some urls are given inside text, bring them to reference section with proper citation.

3. overall working  algorithm of the proposed system is missing

4. certain figs. have low quality, improve them

5. all tables and figs. should be cited before their occurrence.

6.  how the authors did compare with baseline works ?

7. some important references are missing:

    (a)  SePiCo: Semantic-Guided Pixel Contrast for Domain Adaptive Semantic Segmentation. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1-17. doi: 10.1109/TPAMI.2023.3237740

(b)A Robust Light Field Semantic Segmentation Network Combining Contextual and Geometric Features. Frontiers in Environmental Science, 1443. doi: 10.3389/fenvs.2022.996513

(c)   UrbanLF: A Comprehensive Light Field Dataset for Semantic Segmentation of Urban Scenes. IEEE Transactions on Circuits and Systems for Video Technology. doi: 10.1109/TCSVT.2022.3187664

 

 

 

 

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

The paper proposes a software defect prediction method based on program semantics feature mining (PSFM). The semantic information is extracted from grammatical structure information and code text information. The work is interesting, but the presentation is not sufficient.

1) The authors wrote: “In summary, the current research on software defect prediction methods ignore the impact of program semantics on defect performance.” This claim is not true since this is not the first approach which includes semantics for defect prediction.

2) The related work section must be improved.

3) How can the approach be generalized to be applicable to non-Java code? Why has any parser/compiler generator (e.g., ANTLR, LISA) not been used?

4) The following statement: “the vector representation of the code text information is finally output”, is not completed.

5) Figure 2 and 3 lack details and are currently not meaningful.

6) The proposed approach is not sufficiently described and lacks details.

7) Some acronyms are not defined (e.g., AUROC, KPCA).

8) References are not written with enough care. There is inconsistent style (“Liu Wenjie” vs “Radjenović, D.,”), as well as full of other typos.

9) Too many typos:

First This paper

->

First this paper

 

machine learning[4].

->

machine learning [4].

 

path set.Liu

->

path set. Liu

 

Shafiq et.al [1]proposed

->

Shafiq et.al [1] proposed

 

Tian et al.[3]propose

->

Tian et al. [3] propose

 

Shafiq et al.[18]

->

Shafiq et al. [18]

 

Tian et al.[19]propose

->

Tian et al. [19] propose

 

3.5. Combining Code Semantics.

->

3.5. Combining Code Semantics

 

this paper put

->

This paper put

 

References used in this review:

===============================

Shi et al.  2020: PathPair2Vec: An AST path pair-based code representation method for defect prediction. Journal of Computer Languages, Volume 59, August 2020, 100979

Khalilian et al. 2020: APRSuite: A suite of components and use cases based on categorical decomposition of automatic program repair techniques and tools. Journal of Computer Languages, Volume 57, April 2020, 100927

 

Fu and Menzies 2017: Revisiting unsupervised learning for defect prediction Proceedings of the 2017 11th Joint Meeting on Foundations of Software Engineering, ESEC/FSE 2017, ACM, New York, NY, USA (2017), pp. 72-83.

 

Kovačević et al. 2020: From Grammar Inference to Semantic Inference—An Evolutionary Approach. Mathematics 8, no. 5: 816

 

Kovačević et al. 2022: Automatic compiler/interpreter generation from programs for Domain-Specific Languages: Code bloat problem and performance improvement. Journal of Computer Languages, Volume 70, June 2022, 101105

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

all changes are done

 

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Reviewer 2 Report

My comments have been, more or less, addressed. I have only a few minor comments:

1) Two consecutive statements in the Abstract start with: “This is because … .” Please rewrite one statement.

2) Typos: below mistakes appears twice: in the text and in the title of Fig. 1

(a)code semantic

->

(a) code semantic

;(b)

->

; (b)

;(c)

->

; (c)

3) Additional typos:

queries.K

->

queries. K

4) References are again not adequately written. All authors should be credited. Hence, provide complete information for references with “et al.” abbreviation (e.g., “Pachouly, J., et al.”.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf