Requirement or not, that is the question!

This repository contains the replication package for the paper "Requirement or not, that is the question: A case from the railway industry". Please cite the original paper.

Tool

A running tool with the best model---BERT uncased---is available here.

Replication Package

• REFSQ23_results.xlsx contains the obtained results of our experiments.
• models directory should contain all the trained ML models on five folds (both on data with and without pre-processing) of the Dronology Dataset. The instructions to download the models are in the following section.
• dronology.csv contains the original Dronology dataset in CSV format. IDs starting with RE are considered requirements and all other entries are considered as non-requirements.
• data contains the generated stratified five folds from the original Dronology dataset, both with and without pre-processing. These folds are used for cross-validation of all the pipelines.

How to run?

1. Clone the repository
2. Download (from here and here) and unzip the additional material in the models directory
3. Create a virtual environment with Python 3.8.10 and install the requirements.txt using pip install -r requirements.txt
4. Run the Jupyter Notebook RorNot_Pipelines.ipynb

Cite as

@InProceedings{10.1007/978-3-031-29786-1_8,
   author="Bashir, Sarmad
   and Abbas, Muhammad
   and Saadatmand, Mehrdad
   and Enoiu, Eduard Paul
   and Bohlin, Markus
   and Lindberg, Pernilla",
   editor="Ferrari, Alessio
   and Penzenstadler, Birgit",
   title="Requirement or Not, That is the Question: A Case from the Railway Industry",
   booktitle="Requirements Engineering: Foundation for Software Quality",
   year="2023",
   publisher="Springer Nature Switzerland",
   address="Cham",
   pages="105--121",
   isbn="978-3-031-29786-1"
}

OR

Bashir, S., Abbas, M., Saadatmand, M., Enoiu, E.P., Bohlin, M., Lindberg, P. (2023).
Requirement or Not, That is the Question: A Case from the Railway Industry.
In: Ferrari, A., Penzenstadler, B. (eds) Requirements Engineering: Foundation for Software Quality. REFSQ 2023.
Lecture Notes in Computer Science, vol 13975. Springer, Cham.
https://doi.org/10.1007/978-3-031-29786-1_8

Funding

This work is partially funded by the AIDOaRt (KDT) and SmartDelta (ITEA) projects.