AI-Powered Sponsorship Recommendation System for Students using TF-IDF, SpaCy NLP, and Rule-Based Skill Mapping

This paper presents an intelligent and interpretable recommendation system designed to effectively match students with potential sponsors, fostering a mutually beneficial relationship between emerging talent and organizations seeking innovative contributors. The platform utilizes a hybrid approach combining natural language processing (NLP) techniques and vector space models to analyze and compare student resumes with job or sponsorship descriptions. Core components of the system include text preprocessing using SpaCy—where stopwords, punctuation, and irrelevant content are removed and words are lemmatized to their base form—followed by the transformation of text into TF-IDF vectors to capture the relative importance of terms. Cosine similarity is then used to compute the degree of alignment between resumes and sponsor-defined criteria. Additionally, the system features a rule-based skill extraction module that identifies technical proficiencies from resumes using a curated skill list, and a recommendation engine that matches these extracted skills against predefined job roles through intersection logic. This combined methodology ensures accurate and explainable results, helping organizations identify suitable candidates while empowering students to access meaningful opportunities based on merit. The platform serves as a step forward in democratizing access to sponsorships and enhancing visibility for skilled students across diverse domains.