AI-Driven Dimensionality Reduction for Visualizing Graduate Employment Outcomes

Abstract

This paper addresses the challenges faced by graduates in career choices and the limitations of traditional counseling services. It reviews the application of Artificial Intelligence (AI), particularly dimensionality reduction and visualization, in intelligent decision-support systems. Dimensionality reduction techniques extract essential features from high-dimensional data, while visualization provides intuitive insights. A multidimensional decision model is proposed, encompassing personal traits, socio-economic background, external environment, and school support. The paper discusses dimensionality reduction methods such as PCA and t−SNE , along with various visualization techniques. Challenges such as data privacy, algorithm fairness, and model interpretability are explored. The paper also looks ahead to the integration of multimodal data, human-AI collaboration, and other future directions, aiming to promote the responsible application of AI in graduate career planning.