International Journal of All Research Education & Scientific Methods

An ISO Certified Peer-Reviewed Journal

ISSN: 2455-6211

Latest News

Visitor Counter
5854311643

Experience Aware Question Generation- A Perso...

You Are Here :
> > > >
Experience Aware Question Generation- A Perso...

Experience Aware Question Generation- A Personalized Approach to Adaptive Interview Question Generation

Author Name : Mallikarjuna G D, Dr. M. John Basha, Dr. A. Suresh Kumar, V G Likhith Gowda

DOI: https://doi.org/10.56025/IJARESM.2025.130725001

 

ABSTRACT The Experience-Aware Question Generation (EAQG) system introduces an adaptive approach to interview questioning by dynamically generating tailored questions based on a candidate’s experience level. Using Natural Language Processing (NLP) and machine learning, the system analyzes resume content to classify candidates as Beginner, Intermediate, or Expert and formulates questions aligned with Bloom’s Taxonomy to ensure cognitive appropriateness. Additionally, it generates possible answers to assist interviewers in evaluation. The framework enhances recruitment by improving relevance, difficulty adaptation, diversity, and fluency in question generation. To assess question quality, multiple NLP metrics were utilized, including BLEU, ROUGE-L, METEOR, and BERTScore for lexical and contextual similarity, TF-IDF Cosine Similarity for topic relevance, and Self-BLEU for diversity. Results indicate that intermediate-level questions scored highest in similarity, while expert-level questions exhibited greater conceptual depth and variation, highlighting the balance required between accuracy and diversity. The integration of experience-aware classification and Bloom's Taxonomy reduces human bias and enhances assessment consistency. Future expansions could focus on domain-specific question generation and real-time adaptation based on candidate feedback. By automating and personalizing question generation, EAQG can transform recruitment, ensuring fairer and more effective hiring practices