BushraDBR: An Automatic Approach to Retrieving Duplicate Bug Reports

BushraDBR: An Automatic Approach
to Retrieving Duplicate Bug Reports
Ra'Fat Al-Msie'deen
Department of Software Engineering, Faculty of IT, Mutah University, Mutah 61710, Karak,
Jordan
E-mail address: rafatalmsiedeen@mutah.edu.jo
https://rafat66.github.io/Al-Msie-Deen/

➢ To cite this version:
R. Al-Msie’deen, “BushraDBR: An Automatic Approach to Retrieving
Duplicate Bug Reports,” International Journal of Computing and Digital
Systems, vol. 15, no. 1, pp. 221-238, 2024.
• URL: https://journal.uob.edu.bh/handle/123456789/5343

Keywords
Keywords: Software engineering, Software maintenance, Duplicate bug
reports, Formal concept analysis, Latent semantic indexing, Bug
tracking system, Bug report.

Abstract
A Bug Tracking System (BTS), such as Bugzilla,
is generally utilized to track submitted Bug
Reports (BRs) for a particular software system.
Duplicate Bug Report (DBR) retrieval is the
process of obtaining a DBR in the BTS. This
process is important to avoid needless work
from engineers on DBRs. To prevent wasting
engineer resources, such as effort and time, on
previously submitted (or duplicate) BRs, it is
essential to find and retrieve DBRs as soon as
they are submitted by software users.

Abstract …
Thus, this paper proposes an automatic approach
(called BushraDBR) that aims to assist an
engineer (called a triager) to retrieve DBRs and
stop the duplicates before they start. Where
BushraDBR stands for Bushra Duplicate Bug
Reports retrieval process. Therefore, when a new
BR is sent to the Bug Repository (BRE), an
engineer checks whether it is a duplicate of an
existing BR in BRE or not via BushraDBR approach.
If it is, the engineer marks it as DBR, and the BR is
excluded from consideration for any additional
work; otherwise, the BR is added to the BRE.
BushraDBR approach relies on Textual Similarity
(TS) between the newly submitted BR and the rest
of the BRs in BRE to retrieve DBRs.

Abstract …
BushraDBR exploits unstructured data from BRs
to apply Information Retrieval (IR) methods in
an efficient way. BushraDBR approach uses two
techniques to retrieve DBRs: Latent Semantic
Indexing (LSI) and Formal Concept Analysis
(FCA). The originality of BushraDBR is to stop
DBRs before they occur by comparing the newly
reported BR with the rest of the BRs in the BTS,
thus saving time and effort during the Software
Maintenance (SM) process. BushraDBR also
uniquely retrieves DBR through the use of LSI and
FCA techniques. BushraDBR approach had been
validated and evaluated on several publicly
available data sets from Bugzilla. Experiments
show the ability of BushraDBR approach to
retrieve DBRs in an efficient and accurate
manner.

List of Abbreviations
• Bug Tracking System (BTS)
• Bug Reports (BRs)
• Duplicate Bug Report (DBR)
• Bug Repository (BRE)
• Information Retrieval (IR)
• Latent Semantic Indexing (LSI)
• Software Maintenance (SM)
• Formal Concept Analysis (FCA)
• Master BRs set (MRs)
Vector Space Model (VSM)
Deep Learning (DL)
Machine learning (ML)
Convolutional Neural Network (CNN)
New BR (Nr)
Natural Language Processing (NLP)
Quality Assurance (QA)
Drawing Shapes Application (DSA)
Bushra Duplicate Bug Reports retrieval process (BushraDBR)
Cosine Similarity Matrix (CSM)
Cosine Similarity (CS)
Term-Query Matrix (TQM)
Term Document Matrix (TDM)
Textual Similarity (TS)
List of DBRs (LDRs)
Binary Formal Context (BFC)
Singular Value Decomposition (SVD)

Introduction
✓ In this work, reports that are textually similar to each other are called DBRs.
✓ TS between BRs is good evidence that they describe the same (or similar) issue.
✓ Frequently, DBRs are reported by multiple software users.
✓ These users come from different backgrounds and use different vocabularies to describe
the same (or similar) software bug.
✓ The important role of the triager is to check the reported BRs for any possible duplicates
before sending them to the BRE.
✓ A manual check of submitted BRs is a difficult task due to the huge number of BRs
reported every day.
✓ On the other hand, retrieving DBR after storing it in the BRE is a tedious and expensive task
for software developers.
✓ So, the process of stopping DBRs before they start is important and very useful.

Introduction …
• The novelty of this paper is that it proposes a textual-based
approach (i.e., an IR-based solution) to stopping DBRs before
they start.
• BushraDBR prevents DBRs by continuously checking the
recently reported or submitted BR against the BRs stored in the
BTS.
• In the event that the submitted BR is textually similar to any of
the BRs inside the BTS, the triager will exclude this BR from any
further work and not include it in the BTS.

Figure 1. An
example of
a BR from
drawing
shapes
application.
“Bug details”
Bug ID: 000001.
Product: Drawing shapes App.
Submitted on: Sept. 12, 2022.
Summary: The PDF viewer is very slow for line drawings.
Component: Layout.
Type: defect.
Severity: S4.
Priority: P5.
Status: Assigned.
“Bug description”
Actual results:
PDF is very slow to load. The text parts of the PDF appear fast, but
every line drawing is very slow.
Expected results:
PDFs with line drawings should show up a lot faster without any delay.

TABLE I. A pair
of DBRs from
Bugzilla (i.e.,
core product).

Figure 2. The workflow
for retrieving DBRs via
BushraDBR approach.
BR_001: …
Bug repository
BushraDBR
approach
Similarity
BR documents
BR_002: …
BR_003: …
…
BR_n: …
BR_001: 0.02
BR_002: 0.58
BR_003: 0.98
…
BR_n: 0.45
Duplicate
Nr_Bug-ID
BR _003
Duplicate or
non-duplicate
New bug report (Nr)
Nr_Bug-ID
Query
LSI
Test duplicate bug reports (DBRs)
✓
Nr
FCA

Figure 3. The key
elements of
BurshraDBR
approach.
New bug report (Nr)
BR documents from BTS
Duplicate BR
Non-duplicate BR
Threshold = 80
LSI
FCA
OUTPUT
DATA SETS
EVALUATION METRICS Precision
F-measure
Recall
WebPayments UI
Firefox
Symbolication & General
Eliot
DOM: Editor
Audio/Video
Core
Drawing Shapes App
TECHNIQUES
BurshraDBR
approach

TABLE II. Structured
and unstructured
information that is
leveraged by the
selected approaches
(i.e., survey).

TABLE III. Data
sets that are
utilized by the
selected
approaches
(i.e., survey).

TABLE IV. Review and classification of current studies relevant for DBR retrieval approaches.

Figure 4. DBRs describe the
same software failure and use
similar vocabularies.
Bug Report X
Bug Report Y
Bug X Bug Tracking
System (BTS)
Triager or
developer
BushraDBR
approach
Non-duplicate
Duplicate
Master (or original) bug report
Software
user
Submit
Software
failure X
Nr
Similar
vocabularies

Figure 5. The DBR
retrieval process -
BushraDBR approach.
Master bug reports
set (MRs)
List of duplicate bug reports (LDRs)
Repository
BushraDBR approach
Non-duplicate
Duplicate
No
Yes
BR documents
Similarity
value >=
threshold
(Thr)
Insert into
MRs
Insert into
LDRs
New bug
report (Nr)
Query (Nr)
document
Inputs
Latent Semantic Indexing
Formal Concept Analysis
Preprocessing

Figure 6. An example of a bug
document generated by BushraDBR
approach.
Bug Document
BushraDBR
Bug ID: 000007
Incapable of using the
mouse wheel to scroll on an
art site.
Scrolling the contents of
the art site by clicking
the scroll wheel of the
mouse device does not
always work well.

Algorithm 1:
Preprocessing
of BRs.

Figure 7. An
example of a
pre-processed
bug
document.
Bug Document
BushraDBR
Bug ID: 000007
incapable
use
mouse
wheel
scroll
art
site
scroll
content
art
site
click
scroll
wheel
mouse
device
always
work
well
1. Extracting
bug content
2. Splitting
words or
tokenization
3. Removing
stop words
4. Stemming
Pre-processed document

TABLE V. Bug reports from the drawing shapes application data set.

TABLE VI. BushraDBR preprocessing steps, explanation, and a real-world example
from the DSA data set.

Figure 8. Retrieving DBRs
using LSI and FCA
techniques — the
BushraDBR approach.
BushraDBR approach
2
3
Extracting
Similarity matrix (SM)
SM as binary formal context
Nr
BR-1 BR-2 BR-N
0.01 0.93 0.07
Concept_0
BR-2
Nr
Nr
BRs
Nr
BR-1 BR-2 BR-N
0 1 0
Nr
BR-1 BR-2 BR-N
x
The AOC-poset
Formal context
DBRs
Formal
concept
..., of the myOval()
and … displayed, …
…, of, the, my, oval,
and, ..., displayed, ..
.., oval, .., displayed,
…
…, oval, …, display, ..
Splitting
Preprocessing
1
The intent
of concept
The extent of concept
Stop words Stemming

Algorithm 2:
Measuring
TS between
BRs via LSI.

Measuring TS between BRs via LSI
1
2
3
4

Figure 9. The TS
values between
Nr and BRs as a
directed graph.

Algorithm 3: Retrieving DBRs using FCA.

Retrieving DBRs using FCA
Figure 10. The AOC-poset for the
DSA data set.
1
2
3
4

Figure 10. The AOC-
poset for DSA data
set.

Concept_0
1819151
1819152
Drawing Shapes
Application
(DSA)
Symbolication
& General -
Eliot product
Concept_0
1490824
1510066
WebPayments UI
– Firefox product
(partial)
Concept_0
1545235
1545237
Audio/Video –
Core product
(partial)
DOM editor –
Core product
(partial)
Concept_0
176525
1841744
Concept_0
000006
000007
The extent of
Concept_0
The intent of
Concept_0
The
concept ID
Formal concept
Concept_0
000006
000007
Formal concept analysis

Master bug reports set (MRs)
List of duplicate
bug reports
(LDRs)
Repository
BushraDBR approach
Non-duplicate
Duplicate
No
Yes
Preprocessing
1
2
3
Extracting Splitting
Stop words Stemming
BR documents
Similarity
value >=
threshold
(Thr)
Nr
BR-1 BR-2 BR-N
0.01 0.93 0.07
Nr
BR-1 BR-2 BR-N
0 1 0
Insert into MRs
Insert into
LDRs
Concept_0
BR-2
Nr
New bug
report (Nr)
Query (Nr)
document

Figure 9. The TS values between
Nr and BRs as a directed graph.
000007
000002
000003
000004
000005
000001
000006

New bug report
(Nr)
List of duplicate
bug reports (LDRs)
BushraDBR approach Non-duplicate
Duplicate
No
Yes
Preprocessing
1
2
3
Stop words Stemming
Query (Nr) document
Similarity
value >=
threshold
(Thr)
Insert into LDRs
BushraDBR
approach

Experimentation
Master bug
reports set (MRs)
Repository
BR documents

Figure 10. The AOC-poset for each data set in
the experiments: (A) DSA, (B) Eliot, (C)
WebPayments UI [partial], (D) Audio / Video
[partial], and (E) DOM editor [partial].
Concept_0
1819151
1819152
Concept_1
1768863
1814509
1729698
1741434
1801212
1812345
1815981
1707879
1815982
1811299
1801169
1811236
1475334
1649535
1745533
Concept_1
000005
000003
000004
000001
000002
Concept_0
1490824
1510066
Concept_1
1476344
1498447
1446577
1498225
1438784
1494439
1501447
1499837
1507623
1464356
1494559
1470197
1495151
.
.
Concept_0
1545235
1545237
Concept_1
1582074
1673285
1532646
1732199
1816175
1776641
1680362
1743870
1676015
1691996
1545237
1757124
1799132
.
.
Concept_1
1088194
201410
1036856
1276391
1327934
1723853
1840784
1710784
458524
460903
377297
1462368
1567160
.
.
Concept_0
176525
1841744
Concept_0
000006
000007
(A) (B) (C) (D) (E)

Conclusion
• This paper has introduced a novel
approach called BushraDBR targeted
at automatically retrieving DBRs
using LSI and FCA. BushraDBR aimed
to prevent developers from wasting
their resources, such as effort and
time, on previously submitted BRs.
The novelty of BushraDBR is that it
exploits textual data in BRs to apply
LSI and FCA techniques in an efficient
way to retrieve DBRs.
Preprocessing
1
2
3
Stop words Stemming

Conclusion …
• BushraDBR prevents DBRs
before they occur by
comparing the newly reported
BR with the rest of the BRs in
the repository. The suggested
approach had been validated
and evaluated on different data
sets from Bugzilla. Experiments
show the capacity of
BushraDBR approach to
retrieve DBRs in an efficient
and accurate manner.

Future work
• Regarding BushraDBR's future work, the author
plans to extend the current approach by
developing an ML-based solution to retrieve DBRs
and prevent duplicates before they start. Also, he
plans to compare BushraDBR (i.e., an IR-based
approach) with current ML-based approaches.
Furthermore, additional empirical tests can be
conducted to verify BushraDBR approach using
open-source and industrial data sets. There is also
a necessary need to conduct a comprehensive
survey and make comparisons between all current
approaches relevant to DBR retrieval.
New bug
report (Nr)
List of duplicate
bug reports
(LDRs)
Non-duplicate
Duplicate
No
Yes
Query (Nr) document
Similarity value >= threshold (Thr)
Insert into
LDRs

[3] N. Jalbert and W. Weimer, “Automated duplicate detection for bug tracking systems,” in The 38th Annual IEEE / IFIP
International Conference on Dependable Systems and Networks, DSN 2008, June 24-27, 2008, Anchorage, Alaska, USA, Proceedings.
IEEE Computer Society, 2008, pp. 52–61. [Online]. Available: https://doi.org/10.1109/DSN.2008.4630070
[4] A. Hindle and C. Onuczko, “Preventing duplicate bug reports by continuously querying bug reports,” Empir. Softw. Eng., vol.
24, no. 2, pp. 902–936, 2019. [Online].Available: https://doi.org/10.1007/s10664-018-9643-4
[11]A. Kukkar, R. Mohana, Y. Kumar, A. Nayyar, M. Bilal, and K. Kwak, “Duplicate bug report detection and classification system
based on deep learning technique,” IEEE Access, vol. 8, pp. 200 749–200 763, 2020. [Online]. Available:
https://doi.org/10.1109/ACCESS.2020.3033045
[17] X. Wang, L. Zhang, T. Xie, J. Anvik, and J. Sun, “An approach to detecting duplicate bug reports using natural language and
execution information,” in 30th International Conference on Software Engineering (ICSE 2008), Leipzig, Germany, May 10-18,
2008, W. Sch¨afer, M. B. Dwyer, and V. Gruhn, Eds. ACM, 2008, pp. 461–470. [Online]. Available:
https://doi.org/10.1145/1368088.1368151
[36] C. Sun, D. Lo, X. Wang, J. Jiang, and S. Khoo, “A discriminative model approach for accurate duplicate bug report
retrieval,” in Proceedings of the 32nd ACM / IEEE International Conference on Software Engineering - Volume 1, ICSE 2010, Cape
Town, South Africa, 1-8 May 2010, J. Kramer, J. Bishop, P. T. Devanbu, and S. Uchitel, Eds. ACM, 2010, pp. 45–54. [Online].
Available: https://doi.org/10.1145/1806799.1806811
[37] C. Sun, D. Lo, S. Khoo, and J. Jiang, “Towards more accurate retrieval of duplicate bug reports,” in 26th IEEE / ACM
International Conference on Automated Software Engineering (ASE 2011), Lawrence, KS, USA, November 6-10, 2011, P. Alexander,
C. S. Pasareanu, and J. G. Hosking, Eds. IEEE Computer Society, 2011, pp. 253–262. [Online]. Available:
https://doi.org/10.1109/ASE.2011.6100061

[38] F. Thung, P. S. Kochhar, and D. Lo, “Dupfinder: integrated tool support for duplicate bug report detection,” in ACM / IEEE
International Conference on Automated Software Engineering, ASE ’14, Vasteras, Sweden - September 15 - 19, 2014, I. Crnkovic, M.
Chechik, and P. Gr¨unbacher, Eds. ACM, 2014, pp. 871–874. [Online].Available: https://doi.org/10.1145/2642937.2648627
[39] J. He, L. Xu, M. Yan, X. Xia, and Y. Lei, “Duplicate bug report detection using dual-channel convolutional neural
networks,” in ICPC ’20: 28th International Conference on Program Comprehension, Seoul, Republic of Korea, July 13-15, 2020.
ACM, 2020, pp. 117–127. [Online].Available: https://doi.org/10.1145 /3387904.3389263
[40] P. Runeson, M. Alexandersson, and O. Nyholm, “Detection of duplicate defect reports using natural language processing,” in
29th International Conference on Software Engineering (ICSE 2007), Minneapolis, MN, USA, May 20-26, 2007. IEEE Computer
Society, 2007, pp. 499–510. [Online].Available: https://doi.org/10.1109/ICSE.2007.32
[44] M. S. Rakha, C. Bezemer, and A. E. Hassan, “Revisiting the performance evaluation of automated approaches for the
retrieval of duplicate issue reports,” IEEE Trans. Software Eng., vol. 44, no. 12, pp. 1245–1268, 2018. [Online]. Available:
https://doi.org/10.1109/TSE.2017.2755005
[47] B. S. Neysiani and S. Morteza Babamir, “Automatic duplicate bug report detection using information retrieval-based versus
machine learning-based approaches,” in 2020 6th International Conference on Web Research (ICWR), April 2020, pp. 288–293.

BushraDBR: An Automatic Approach to Retrieving Duplicate Bug
Reports

BushraDBR: An Automatic Approach to Retrieving Duplicate Bug Reports

More Related Content

More from Ra'Fat Al-Msie'deen (20)

Recently uploaded (20)

BushraDBR: An Automatic Approach to Retrieving Duplicate Bug Reports