Functional Non-Functional Classification Dataset (FNFC):

Overview:
The Functional Non-Functional Classification Dataset (FNFC) contains 7,060 software requirements categorized into 14 distinct classes. These classes represent various functional and non-functional requirements, annotated by five experienced software engineering experts to ensure high-quality labeling.

Dataset Details:
Total Records: 7,060

Classes:
Availability (A)
Automation (AU)
Functional Requirements (FR)
Fault Tolerance (FT)
Look and Feel (LF)
Legal (LL)
Maintainability (M)
Operational (O)
Portability (P)
Performance (PE)
Reliability (R)
Scalability (SC)
Security (SE)
Usability (US)
Annotation Quality: Each requirement was reviewed and labeled by a team of 5 software engineering experts.

File Format:
The dataset is provided in .csv format with the following columns:

Requirement_ID: Unique identifier for each requirement.
Requirement_Text: Text of the requirement.
Category: The assigned class label.

Usage:
This dataset is designed for:
Classification of functional and non-functional requirements.
Research in requirements engineering and software fault analysis.
Development and testing of machine learning and deep learning models.

Contributors
Younes Abdeahad
Esmaeil Kheirkhah
Mahdi Kabootari
Yalda Kheirkhah
For inquiries, contact us at [[email protected]].