language:
- en
library_name: transformers
pipeline_tag: text-classification
base_model: distilbert/distilbert-base-uncased
base_model_relation: finetune
datasets:
- devleoespinosa/qus-user-story-quality-refined
tags:
- distilbert
- transformers
- text-classification
- requirements-engineering
- software-engineering
- user-stories
- qus
- quality-assessment
- natural-language-processing
Model Description
This repository contains a fine-tuned DistilBERT model for the automated assessment of the Conceptually Sound criterion of the Quality User Story (QUS) framework.
The model performs binary text classification to determine whether the components of a user story are conceptually consistent. Under QUS, the means of the story should describe a feature or desired functionality, while the optional ends should express the rationale or benefit associated with that functionality.
A violation may occur when these components do not fulfill their intended semantic roles or when the relationship between the requested functionality and its rationale is conceptually inconsistent.
The model was developed as part of the study "Fine-Tuned DistilBERT for Automated User Story Quality Assessment".
Classification task
- Input: A user story written in natural language.
- Output: Binary classification indicating compliance with the Conceptually Sound criterion.
- Correct: The requested functionality and its rationale fulfill their expected semantic roles and form a conceptually coherent requirement.
- Incorrect: The story contains a conceptual inconsistency between its components or uses them for inappropriate semantic purposes.
This model is one of eight criterion-specific DistilBERT models developed for the individual quality criteria of the QUS framework.