Data-Driven Requirements Elicitation from App Reviews Framework Based on BERT
Source abstract
Abstract
Market-Driven Requirements Engineering (MDRE) integrates traditional Requirements Engineering (RE) practices, such as Requirements Elicitation and requirements prioritization, with market analysis. Offering software products to an open market has become a trend, yet it has many challenges. In MDRE, there are diverse sources for requirements including support teams, subcontractors, sales, and marketing teams. So, the MDRE process must provide ongoing requirements gathering techniques to ensure no crucial requirements are overlooked. It is generally possible for users to search and download software applications through app stores (such as the Google Play Store and Apple App Store) for various purposes. Users are allowed to express their opinions about the software applications by writing text messages which are widely known as “app reviews”. Utilizing these app reviews as a source of requirements while planning to develop a similar software application may have promising results. Therefore, the concept of “App Reviews Utilization” has emerged and can be applied for various purposes. This research utilizes app reviews in Requirements Elicitation while developing a software product in the market-driven development context. Furthermore, these reviews may be noisy and informally expressed. This paper proposes a framework, Automatic Requirements Elicitation from App Reviews (AREAR), that integrates Natural Language Processing (NLP) techniques with pre-trained Language Models to automatically elicit requirements from available app reviews while developing a market-driven software product. AREAR employed the Bidirectional Encoder Representation from the Transformers (BERT) Language Model. The proposed framework achieved an improved Accuracy and F1 score as compared to previous research.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Software Engineering Techniques and Practices · Software Engineering Research · Persona Design and Applications
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
Information Systems Department, Faculty of Computers and Artificial Intelligence, Damietta University, Damietta 34511, Egypt
Information Systems Department, Faculty of Computers and Artificial Intelligence, Cairo University, Cairo 12613, Egypt
Heikal Department of Management, Onsi Sawiris School of Business, American University in Cairo, Cairo 11835, Egypt
References (54)
Show source-provided references
1. Regnell, B., and Brinkkemper, S. (2005). Market-driven requirements engineering for software products. Engineering and Managing Software Requirements, Springer.10.1007/3-540-28244-0_13 ↗
2. Release planning in market-driven software product development: Provoking an understanding10.1007/s007660200010 ↗
3. Crowder, J.A., and Hoff, C.W. (2022). MDRE: The Requirements Engineering Process. Requirements Engineering: Laying a Firm Foundation, Springer.10.1007/978-3-030-91077-8 ↗
4. Data-driven requirements elicitation: A systematic literature review10.1007/s42979-020-00416-4 ↗
5. Ceci, L., and Statista (2024, June 26). Annual Number of Global Mobile App Downloads 2016–2023. Available online: https://www.statista.com/statistics/271644/worldwide-free-and-paid-mobile-app-store-downloads/.
6. Requirements elicitation techniques: A systematic literature review based on the maturity of the techniques10.1049/iet-sen.2017.0144 ↗
7. Ghosh, T.K., Pargaonkar, A., and Eisty, N.U. (2024). Exploring Requirements Elicitation from App Store User Reviews Using Large Language Models. arXiv.
8. Triantafyllou, I., Drivas, I.C., and Giannakopoulos, G. (2020). How to utilize my app reviews? A novel topics extraction machine learning schema for strategic business purposes. Entropy, 22.10.3390/e22111310 ↗
9. Kengphanphanit, N., and Muenchaisri, P. (2020, January 1–3). Automatic requirements elicitation from social media (ARESM). Proceedings of the 2020 International Conference on Computer Communication and Information Systems 2020, Ho Chi Minh City, Vietnam.10.1145/3418994.3419004 ↗
10. App store effects on software engineering practices
11. de Araújo, A.F., and Marcacini, R.M. (2021, January 22–26). RE-BERT: Automatic extraction of software requirements from app reviews using BERT language model. Proceedings of the 36th Annual ACM Symposium on Applied Computing (SAC 2021), ACM, Gwangju, Republic of Korea.10.1145/3412841.3442006 ↗
12. Dalpiaz, F., and Parente, M. (2019). RE-SWOT: From user feedback to requirements via competitor analysis. International Working Conference on Requirements Engineering: Foundation for Software Quality, Springer.10.1007/978-3-030-15538-4_4 ↗
13. On the automatic classification of app reviews10.1007/s00766-016-0251-9 ↗
14. Lu, M., and Liang, P. (2017, January 15–16). Automatic classification of non-functional requirements from augmented app user reviews. Proceedings of the 21st International Conference on Evaluation and Assessment in Software Engineering, Karlskrona, Sweden.10.1145/3084226.3084241 ↗
15. Extracting useful software development information from mobile application reviews: A survey of intelligent mining techniques and tools10.1016/j.eswa.2018.05.037 ↗
16. QUARE: Towards a question-answering model for requirements elicitation10.1007/s10515-023-00386-w ↗
17. The application of AI techniques in requirements classification: A systematic mapping10.1007/s10462-023-10667-1 ↗
18. Utilizing BERT for Information Retrieval: Survey, Applications, Resources, and Challenges
19. Hidalgo, M., Yanine, F., Paredes, R., Frez, J., and Solar, M. (2024). What Is the Process? A Metamodel of the Requirements Elicitation Process Derived from a Systematic Literature Review. Processes, 13.10.3390/pr13010020 ↗
20. Grace, R., Burnham, K., and Na, H.S. (2023). How Much Context Do Users Provide in App Reviews? Implications for Requirements Elicitation. International Conference on Information, Springer.10.1007/978-3-031-28032-0_2 ↗
21. Automatic requirements elicitation from user-generated content: A review of data, methods, and representations10.1016/j.engappai.2025.111110 ↗
22. Big consumer opinion data understanding for Kano categorization in new product development10.1007/s12652-021-02985-5 ↗
23. What do mobile app users complain about?10.1109/MS.2014.50 ↗
24. Panichella, S., Di Sorbo, A., Guzman, E., Visaggio, C.A., Canfora, G., and Gall, H.C. (October, January 29). How can I improve my app? Classifying user reviews for software maintenance and evolution. Proceedings of the 2015 IEEE International Conference on Software Maintenance and Evolution (ICSME), Bremen, Germany.10.1109/ICSM.2015.7332474 ↗
25. Ciurumelea, A., Schaufelbühl, A., Panichella, S., and Gall, H.C. (2017, January 20–24). Analyzing reviews and code of mobile apps for better release planning. Proceedings of the 2017 IEEE 24th International Conference on Software Analysis, Evolution and Reengineering (SANER), Klagenfurt, Austria.10.1109/SANER.2017.7884612 ↗
26. Convolutional neural network based classification of app reviews10.1109/ACCESS.2020.3029634 ↗
27. Leveraging Large Language Model ChatGPT for enhanced understanding of end-user emotions in social media feedbacks10.1016/j.eswa.2024.125524 ↗
28. Deep neural networks in natural language processing for classifying requirements by origin and functionality: An application of bert in system requirements10.1115/1.4063764 ↗
29. Requirements engineering challenges in market-driven software development–An interview study with practitioners10.1016/j.infsof.2007.02.008 ↗
30. Shaheen, R., Ahsan, A., and Anwar, Z. (2018, January 3–4). Requirements management for market driven software products—Key issues. Proceedings of the 2018 International Conference on Computing, Mathematics and Engineering Technologies (iCoMET), Sukkur, Pakistan.10.1109/ICOMET.2018.8346345 ↗
31. Gharatkar, S., Ingle, A., Naik, T., and Save, A. (2017, January 17–18). Review preprocessing using data cleaning and stemming technique. Proceedings of the 2017 International Conference on Innovations in Information, Embedded and Communication Systems (ICIIECS), Coimbatore, India.10.1109/ICIIECS.2017.8276011 ↗
32. A survey of visual analytics techniques for machine learning10.1007/s41095-020-0191-7 ↗
33. Improving active learning by data balance to reduce annotation efforts
34. Importance-SMOTE: A synthetic minority oversampling method for noisy imbalanced data10.1007/s00500-021-06532-4 ↗
35. Devlin, J., Chang, M.W., Lee, K., and Toutanova, K. (2019, January 2–7). Bert: Pre-training of deep bidirectional transformers for language understanding. Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Minneapolis, MN, USA.
36. AugGPT: Leveraging ChatGPT for Text Data Augmentation10.1109/TBDATA.2025.3536934 ↗
37. Text Classification With Frequency-Based Text Vectorization Methods For Enhancing Call Center Efficiency
38. Cao, J., Su, Z., Yu, L., Chang, D., Li, X., and Ma, Z. (December, January 30). Softmax cross entropy loss with unbiased decision boundary for image classification. Proceedings of the 2018 Chinese Automation Congress (CAC), Xi’an, China.10.1109/CAC.2018.8623242 ↗
39. Adahessian: An adaptive second order optimizer for machine learning
40. A review on evaluation metrics for data classification evaluations10.5121/ijdkp.2015.5201 ↗
41. Sebastiano, P., and Ciurumelea, A. (2024, January 30). Replication Package for: ‘Analyzing Reviews and Code of Mobile Apps for a better Release Planning Organization’. Available online: https://github.com/panichella/UserReviewReference-Replication-Package/tree/URR-v1.0.
42. Attention Is All You Need
43. Exploring the limits of transfer learning with a unified text-to-text transformer
44. A survey of transfer learning10.1186/s40537-016-0043-6 ↗
45. Single document automatic text summarization using term frequency-inverse document frequency (TF-IDF)
46. Comparison of term frequency and document frequency based feature selection metrics in text categorization10.1016/j.eswa.2011.09.160 ↗
47. Text Data Augmentation Techniques for Word Embeddings in Fake News Classification10.1109/ACCESS.2024.3369918 ↗
48. Palanivinayagam, A., El-Bayeh, C.Z., and Damaševičius, R. (2023). Twenty Years of Machine-Learning-Based Text Classification: A Systematic Review. Algorithms, 16.10.3390/a16050236 ↗
49. Cloze Procedure: A New Tool for Measuring Readability
50. Deep contextualized word representations
51. Alec, R., Karthik, N., Tim, S., and Ilya, S. (2018). Improving Language Understanding with Unsupervised Learning: Technical Report. OpenAI Res., Available online: https://cdn.openai.com/research-covers/language-unsupervised/language_understanding_paper.pdf.
52. Comparing BERT Against Traditional Machine Learning Models in Text Classification
53. Confusion matrix
54. A Study on Text Classification in the Age of Large Language Models10.3390/make6040129 ↗