This paper presents the development and evaluation of a machine learning-based system for automating the classification and data extraction of academic documents, such as high school transcripts and recommendation letters. The proposed system combines traditional machine learning models, including Logistic Regression, Random Forest, and Support Vector Machines (SVM), with advanced Natural Language Processing (NLP) techniques such as spaCy and BERT. Using real-world academic data, our study demonstrates how these models can significantly improve the accuracy of both document classification and key information extraction, such as student names, dates of birth, and school names. The results show a marked improvement in efficiency compared to manual methods, offering scalable solutions for academic document management.