An AI-Driven Deep Learning Framework for Personalized Career Recommendation Using Multidimensional Student Profiling

Abstract-Career selection is a critical decision thatdirectly impacts a student’s professional growth andsatisfaction. Conventional guidance methods often rely onlimited assessments and fail to capture a student’scomplete potential. This paper presents a deep learning–based career recommendation model that providespersonalized guidance using multidimensional studentprofiles.The proposed framework integrates academicperformance, technical skills, behavioral traits,personality attributes, and domain interests through ahybrid feature representation. Numerical features arenormalized, while categorical features are modeled usingembedding layers to capture semantic relationships. Amulti-input deep neural network is trained using adaptivelearning and class-balanced optimization.Experimentalevaluation demonstrates high classification accuracy andreliable Top-3 and Top-5 career recommendations. Thesystem also supports skill-gap analysis, modulardeployment, and real-time career data updates,demonstrating the effectiveness of deep learning inenhancing traditional career counseling systems.