SRI SAI RANGANATHAN ENGINEERING COLLEGE
THEME BASED TASK ASSIGNMENT ON JIRA USING TOPIC
MODELLING
PRESENTED BY:
HEMAMALINI.P
KISHORE KUMAR.S
PRAVINKUMAR.P
CONTENTS:
✓ INTRODUCTION
✓ ABSTRACT
✓ LITERATURE SURVEY
✓ PROBLEM STATEMENT
✓ EXISTING SYSTEM
1. EXISTING ARCHITECTURE
2. DISADVANTAGES
✓ PROPOSED SYSTEM
1. PROPOSED ARCHITECTURE
2. ADVANTAGES
✓ MODULE DESCRIPTION
✓ REQUIREMENTS
✓ CONCLUSION
✓ FUTURE WORK
INTRODUCTION
• JIRA lets you prioritize, track, report, manually assign and audit your issues from software bugs and helpdesk tickets to
project tasks and change requests
• More than just an issue tracker, JIRA is an extensible platform that you can customize to match your business processes
• JIRA improves productivity by cutting down on time wated on tracking issues and coordination
• JIRA improves quality by ensuring all tasks are recorded down with all the details and followed up till completion
ABSTRACT:
More and more software teams are relying on online issue tracking systems to work together on software projects. Properly
assigning new issues to the right team members can greatly impact the project's success. To streamline this process, many
research efforts have focused on automating it to save time and effort. However, existing methods primarily target software
bugs and use models that don't fully consider the meaning and additional information of issues. They may also require manual
adjustments. This project introduces a methodology that extracts both textual and non-textual features from various types of
issues, including bugs, new features, documentation issues, patches, etc. Additionally, it employs a topic modeling technique
to effectively capture the meaning of issue text, which is then optimized based on the assignment results. Finally, the
methodology combines probabilities from multiple models to make the final assignment decision. The effectiveness of this
approach was evaluated using a dataset of Jira issues for Theme based task assignment. The results suggest that it can indeed
be successful for this purpose.
LITERATURE SURVEY
TITLE AUTHOR YEAR ABSTRACT
In this paper, we propose to apply machine learning
techniques to assist in bug triage by using text
Automatic bug triage using text Murphy, G., Cubranic, D 2004 categorization to predict the developer that should
categorization work on the bug based on the bug’s description.
In this paper, we propose a semi-supervised text
classification approach for bug triage to avoid the
deficiency of labeled bug reports in existing
supervised approaches. This new approach
Automatic bug triage using Xuan, J., et al. 2010 combines naive Bayes classifier and expectation-
semi‐supervised text classification maximization to take advantage of both labeled and
unlabeled bug reports.
In this paper, we have evaluated the performance of
different machine learning techniques namely
Support Vector Machine (SVM), Naive Bayes (NB),
Predicting the priority of a Sharma, M., et al. 2012 K-Nearest Neighbors (KNN) and Neural Network
reported bug using machine (NNet) in predicting the priority of the newly
learning techniques and cross coming reports on the basis of different performance
project validation. measures.
TITLE AUTHOR YEAR ABSTRACT
Finding the appropriate developer for a bug
report, so called ‘Bug Triage’, is one of the
bottlenecks in the bug resolution process The
results of the experiments applied to both
industrial and open source projects reveal
Applying deep learning based Lee, S.R., et al 2017 benefits of the automatic approach and suggest
automatic bug triager to industrial co-operation of human and automatic triage.
projects Our experience in integrating and operating the
proposed system in an industrial development
environment is also reported
The full integration of online repositories in the
contemporary software development process
promotes remote work and remote
collaboration. In this context, we have analyzed
Employing contribution and Diamantopoulos, T., et al 2020 the 3000 most popular Java GitHub projects
quality metrics for quantifying the and provide the data to the community
software development process
Usually, managers or technical leaders in
software projects assign issues manually. As
main contributions, we present a literature
review, an industrial report comparing different
Issue auto‐assignment in software Oliveira, P., et al 2021 algorithms, and lessons learned during the
projects with machine learning project
techniques
PROBLEM STATEMENT
The current task assignment process in JIRA lacks a systematic approach to align tasks with project themes, leading to
potential inefficiencies and misalignment of efforts. Without a method to harness project insights effectively, teams face
challenges in prioritizing tasks and maintaining project cohesion. Therefore, there is a pressing need to implement a theme-
based task assignment system using topic modeling on JIRA to ensure tasks are aligned with project themes, enhancing
efficiency and cohesion among team member.
EXISTING SYSTEM
❖ JIRA is a widely used project management tool developed by Atlassian. It is designed to help teams plan, track, and
manage their work efficient
❖ One of the core features of JIRA is its robust issue tracking system. Users can create, prioritize, and assign tasks, bugs, or
user stories within the platform. Each issue can be categorized, labeled, and linked to other related items.
❖ JIRA enables teams to create custom workflows to reflect their specific processes. Workflows define the lifecycle of an
issue, including status transitions, approvals, and notifications. This helps streamline the work process and ensures
consistency.
❖ In the existing system NLP technique named topic modelling and a classifier named SVM & Naïve bayes is used to attain
60% accuracy rate.
❖ JIRA allows users to define custom workflows to match specific assignment processes. This means you can create
workflows tailored to the steps involved in manual assignments, such as assignment creation, review, approval, and
completion.
EXISTING SYSTEM:
EXISTING ARCHITECTURE:
ACCURACY
Task assignment accuracy, F‐measure, and Area Under the Curve for different classifier
Task assignment accuracy and F measure for different number of Assignees and LDA
technique
EXISTING SYSTEM:
DISADVANTAGES:
1. Complexity
2. Cost
3. Customization overload
4. Learning curve
5. Performance issues
6. Dependency on third-party integrations
7. Limited support for non-software projects
PROPOSED SYSTEM
❖ Automated issue assignment using topic modeling in JIRA involves utilizing natural language processing techniques to
analyze the content of incoming issues or tickets.
❖ By employing topic modeling algorithms, such as Latent Dirichlet Allocation (LDA) or Non-negative Matrix Factorization
(NMF), the text of the issues can be categorized into different topics or themes. These topics represent the underlying
concepts or subjects discussed in the issues.
❖ Once the topics are identified, rules or algorithms can be set up to automatically assign the incoming issues to the
appropriate team or individual based on the topics extracted from the text.
❖ For example, if an issue primarily discusses "user interface design," it could be automatically assigned to the UI/UX team.
❖ In the proposed system , to enhance and speed up the performance , a NLP technique named Topic modelling and Classifier
named Random forest is used to attain the 70% accuracy.
PROPOSED SYSTEM:
PROPOSED ARCHITECTURE
ACCURACY
Task assignment accuracy ,F measure and area under curve for RF classifier
Task assignment accuracy and F-measure for different number of assignees and LDA technique
PROPOSED SYSTEM
ADVANTAGES:
➢ Better handling of Non-linear Data
➢ Robustness to overfitting
➢ Efficient handling of large datasets
➢ Ensemble learning
➢ Reduced sensitivity to data scaling
MODULE DESCRIPTION:
The proposed system consists of the following modules,
❖ Natural Language Processing
✓ Topic modelling
❖ Machine learning module
✓ Random Forest Algorithm
REQUIREMENTS
• Software Requirements:
1. Python: Latest version installed on your system.
2. Flask: Python web framework for building the backend.
3. MongoDB: NoSQL database for storing project data.
4. PyMongo: Python driver for MongoDB, to interact with the database.
5. Frontend Technologies: HTML, CSS, JavaScript for building the user interface.
• Hardware Requirements:
1. Processor: A modern multi-core processor for handling concurrent requests efficiently.
2. RAM: Sufficient RAM to handle database operations and concurrent user requests.
3. Storage: Adequate storage space for storing project data in MongoDB.
4. Network: Stable internet connectivity if you're planning to deploy the application online.
CONCLUSION
In this paper, we have focused on the challenge of automated issue assignment with the aim of saving considerable time and
effort during the software development process. Our methodology uses diverse data from a Jira installation and builds a set of
features that capture the semantics of different types of issues. Moreover, we employ a topic modelling technique that is
optimized taking into account the final assignment, thus mitigating the need for manual tuning, while achieving significant
improvement over simple vector space modelling. Finally, as supported also by our evaluation, we see that employing
non‐textual features further improves the results and produces effective recommendations
FUTURE WORK
1. Enhanced Integration
2. AI and Automation
3. Performance Optimization
4. Mobile Optimization
5. Customization and Flexibility
6. Security and Compliance
7. Collaboration Features
8. Community Engagement
Above are the features that needs to be enhanced in future for a best work experience.
REFERENCE
❖ Murphy, G., Cubranic , D.: Automatic bug triage using text categorization. In: Proceedings of the 16th International
Conference on Software Engineering and Knowledge Engineering SEKE ’04, pp. 92–97. Knowledge Systems Institute,
USA (2004)
❖ Xuan, J., et al.: Automatic bug triage using semi‐supervised text classification. In: Proceedings of the 22nd International
Conference on Software Engineering and Knowledge Engineering SEKE ’10, pp. 209–214. Knowledge Systems Institute,
USA (2010)
❖ Lee, S.R., et al.: Applying deep learning based automatic bug triager to industrial projects. In: Proceedings of the 2017 11th
Joint Meeting on Foundations of Software Engineering ESEC/FSE 2017, pp. 926–931. Association for Computing
Machinery, New York (2017)
❖ Yang, C.Z., et al.: An empirical study on improving severity prediction of defect reports using feature selection. In:
Proceedings of the 2012 19th Asia‐Pacific Software Engineering Conference ‐ Volume 01 APSEC ’12, pp. 240–249. IEEE
Computer Society, USA (2012)
REFERENCES
❖ Diamantopoulos, T., et al.: Employing contribution and quality metrics for quantifying the software development process. In:
Proceedings of the IEEE/ACM 17th International Conference on Mining Software Repositories MSR ’20, pp. 558–562. ACM,
Seoul (2020)
❖ Oliveira, P., et al.: Issue auto‐assignment in software projects with machine learning techniques. In: 2021 IEEE/ACM 8th
International Workshop on Software Engineering Research and Industrial Practice (SER&IP), pp. 65–72 (2021)
❖ Sia, S., Dalmia, A., Mielke, S.J.: Tired of topic models? Clusters of pretrained word embeddings make for fast and good topics
too. In: Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP) Association for
Computational Linguistics, pp. 1728–1736 (2020)
❖ Mohsin, H., et al.: SPAN: a self‐paced association augmentation and node embedding‐based model for software bug
classification and assignment. Know‐Based Syst 236(C), 107711 (2022). [Link] knosys.2021.107711
JIRA TOOL
JIRA TOOL
JIRA TOOL
DATA
DATA
DATA
RESULT DATA
RESULT DATA
RESULT DATA
THANK YOU