Leveraging Name Entity Recognition and Text Classification for Talent Mapping

The paper presents a dual-model approach that can be used for categorizing the work experience data from resumes or other textual copra. The application tracking system (ATS), which is the main tool used to screen resumes, presents operational issues in the quickly changing software sector. Although this technique matches resumes with job descriptions effectively, it requires separate scans for every job posting, which could result in errors and inefficiencies. This initiative is motivated by the need to expedite the hiring process for tech companies and potential employees. The suggested approach presents a cutting-edge tool to improve and expedite job matching and resume analysis. The work attempts to improve the software industry's hiring process efficiency by overcoming present system restrictions. This study presents a novel solution—a text classification model intended for resume analysis and shortlisting to address the operational difficulties with the modern resume screening process. Even though the Application Tracking System (ATS) is currently used, it requires separate scans for every job posting, which could lead to inefficiencies. We propose dual-model architecture which includes our custom Name Entity Recognition (NER) Model which is trained on annotated work experience data and a Multi-Layer Perceptron (MLP) model for multi label classification, aiming to streamline the resume screening process. This dual-model approach enhances accuracy and efficiency in resume classification, benefiting both aspirants and tech companies. This research is a pivotal step in optimizing recruitment in the dynamic software industry. The complete work can be found on GitHub 1