An Efficient Training-Free Resume Matching System with NLP-Based Extraction and Custom Scoring for Enhanced Candidate Selection

As the volume of job applications continues to grow, efficient resume-matching systems have become essential for organizations to streamline candidate selection. This paper presents a resume-matching system that utilizes natural language processing (NLP) and rule-based algorithms for efficient candidate screening without requiring a training phase. The system extracts critical attributes, including skills, age, and educational qualifications, from resumes using customized pattern files and regular expressions. By eliminating the need for training, the system reduces data dependency and enhances processing speed, making it suitable for real-time applications. Skills are matched using a semantic similarity function, while age compatibility is assessed through a non-linear scoring approach, and degrees are compared based on predefined educational levels. Experimental evaluations with ۲۰۰ manually reviewed resumes show high accuracy in attribute extraction and matching, demonstrating the system's advantages in speed and efficiency over training-dependent models.

Keywords:

resume matching , natural language processing , training-free system , candidate screening , semantic similarity