Design and Development of a Computational Framework for Automated Document Classification of Mental Health Diagnosis Documents Using Machine Learning

Author(s): Loveneet Kumar and Rupali

Abstract:

Mental health diagnosis often relies on qualitative evaluations by clinicians, making the process subjective and time-consuming. With the increasing volume of digital medical records, automating diagnostic classification can enhance efficiency and consistency. This research presents a computational framework for classifying mental health diagnosis documents using advanced text preprocessing, feature extraction, and machine learning algorithms. A dataset of anonymized diagnostic notes was pre-processed using tokenization, lemmatization, and stop-word removal. Feature vectors were generated using TF-IDF and Word2Vec representations. Machine learning algorithms including Naïve Bayes, Support Vector Machine (SVM), Random Forest, and a Neural Network model were applied for classification. The SVM model achieved the highest accuracy (92.6%) and F1-score (0.91). The proposed framework demonstrates the potential of computational text classification in supporting preliminary mental health diagnosis and clinical decision-making.

keywords:

Pages: 146-149     |    3 View     |    0 Download