Answers are grounded in this paper and cite their sources.
Demonstrates improved admissions processing in academic programs, suggesting enhanced efficiency through automation.
The manual screening of thousands of applications for Master’s degree programs at Hassan II University of Casablanca represents a highly time-consuming and labor-intensive process. To address this challenge, this study proposes an intelligent automated framework based on Natural Language Processing (NLP) for large-scale CV summarization and candidate ranking. The proposed system integrates pre-trained spaCy and Hugging Face Transformer-based Named Entity Recognition (NER) models to extract critical candidate information, including educational background, professional experience, and technical skills. The framework combines both extractive and abstractive summarization techniques to generate concise and meaningful candidate profiles. Extractive summarization is performed using BERT-based models to identify the most informative sentences within each CV, while abstractive summarization employs advanced Large Language Models (LLMs), particularly LLaMA, to generate coherent and contextually refined summaries. In addition, the system incorporates a semantic candidate-ranking module designed to evaluate applicant suitability according to program-specific requirements. To validate the effectiveness of the proposed framework, a case study was conducted using 2,325 CVs submitted to the Master’s program in Big Data and Data Science. Experimental evaluation demonstrated strong performance, achieving ROUGE-1 Recall of 72.67%, ROUGE-2 Recall of 74.32%, ROUGE-1 Precision of 73.15%, ROUGE-2 Precision of 57.28%, and Named Entity Recognition (NER) Precision of 82%. The system processed each CV in an average time of 3.84 seconds, demonstrating its suitability for large-scale admissions environments. Furthermore, a conversational AI assistant (chatbot) was integrated into the framework, enabling admissions evaluators to interactively query uploaded CVs in real time. This functionality significantly improves decision-making efficiency, enhances candidate exploration, and reduces administrative workload. Overall, the findings demonstrate that NLP-driven automation provides a scalable, efficient, and intelligent solution for modern academic admissions and candidate screening systems.
No takes yet. Share an insight, caveat, or question.
Abid et al. (2026) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: