NLP-Based Multi-Class Disease Classification Using Patient Symptom Text
Research Digest on Engineering Management and Social Innovations · 04.04.2026 · 2 (4)
DOI: 10.46647/icetetas094 ↗Kurz gefasst
Fachartikel in der Zeitschrift Research Digest on Engineering Management and Social Innovations, online erschienen am 04.04.2026. Verfasst von P. Nikhitha und 1 weiteren Autor:innen.
Abstract (Originalsprache)
Healthcare systems produce large amounts of unstructured patient data in the form of patient symptoms and notes, which pose a challenge in the efficient diagnosis of patients. The current paper aims to develop a model of disease classification using NLP techniques on a synthetic dataset of 25,000 patient records and 30 different categories of diseases. The proposed model applies a text pre-processing technique to classify the diseases using three different classification algorithms to determine the efficiency of the model in terms of accuracy. The text pre-processing technique employed in the proposed model is Bag of Words and Term Frequency-Inverse Document Frequency. The three different classification algorithms employed in the proposed model are Multinomial Naive Bayes, Logistic Regression, and Support Vector Machines. The classification algorithm that achieved the highest accuracy of 94.2%, precision of 93.8%, recall of 93.5%, and F1-score of 93.6% in the proposed model is Support Vector Machines. A simple interface has been employed to fetch the symptoms of the patient to classify the disease by probability-based classification. The proposed model has proved that NLP can be employed as a viable and efficient tool for decision-making.
Zitieren
APA
Nikhitha, P., & Dhanya, G. D. (2026). NLP-Based Multi-Class Disease Classification Using Patient Symptom Text. Research Digest on Engineering Management and Social Innovations, 2(4). https://doi.org/10.46647/icetetas094
Vancouver
Nikhitha P, Dhanya GD. NLP-Based Multi-Class Disease Classification Using Patient Symptom Text. Research Digest on Engineering Management and Social Innovations. 2026;2(4). doi:10.46647/icetetas094
BibTeX
@article{Nikhitha2026,
author = {Nikhitha, P. and Dhanya, G. Deena},
title = {NLP-Based Multi-Class Disease Classification Using Patient Symptom Text},
journal = {Research Digest on Engineering Management and Social Innovations},
year = {2026},
volume = {2},
number = {4},
doi = {10.46647/icetetas094}
}
EUVIMED