Interview Question Analyzer: A Framework for Bias-Free Question Distribution and Evaluation
Ms. Surabhi KS1, Varadha K2
1Assistant professor, Department of Computer Applications, Nehru College of Management, Coimbatore, Tamil Nadu, India.
ksurabhi454@gmail.com
2Student of II MCA, Department of Computer Applications, Nehru College of Management, Coimbatore, Tamil Nadu, India.
varadhadas78@gmail.com
Abstract
Recruitment processes often rely on question banks that are manually distributed among candidates, which can unintentionally introduce bias and inconsistency in evaluation. This paper presents Interview Question Analyzer, a framework designed to ensure fairness in question distribution and candidate assessment. The system integrates a secure authentication module, automated difficulty classification, and equity-driven allocation algorithms to generate balanced interview papers. Question pools are ingested from CSV or PDF formats, with difficulty levels determined through heuristic keyword analysis and contextual length evaluation. A fairness audit mechanism highlights disparities in distribution, while a corrective equity algorithm enforces structural parity across candidates. Additionally, relevancy scoring is performed using TF-IDF vectorization against job descriptions to align questions with role-specific competencies. The framework incorporates archival and reporting features, enabling consolidated candidate assessments in PDF/CSV formats and secure storage in a relational database. An interactive dashboard built with Streamlit and Plotly provides HR professionals with real-time analytics, fairness notifications, and thematic visualizations. By combining automated distribution, bias detection, and equity enforcement, the proposed system contributes to transparent and standardized interview evaluation practices, reducing subjectivity and enhancing reliability in candidate assessment.
Keywords:- Interview systems, bias-free evaluation, fairness algorithm, question distribution, TF-IDF relevancy, equity enforcement, recruitment analytics.
