Patient reviews analysis using machine learning
by Bijayalaxmi Panda; Chhabi Rani Panigrahi; Bibudhendu Pati
International Journal of Computational Science and Engineering (IJCSE), Vol. 26, No. 2, 2023

Abstract: In the present scenario, health-related tweets are available in several online forums for communication. Doctors as well as patients share their views in different discussion forums that help people seek similar information. In this work, an investigation was done by the authors on unstructured patient reviews collected from different forums regarding different diseases. The dataset was collected from Figshare which helps in identifying several features from the text provided by patients into numerical forms. Sentiment analysis (SA) was applied on the dataset to determine positive and negative tweets. Then, the bag-of-words model was used for disease detection. From the considered dataset, specific features were selected and machine learning classification algorithms such as support vector machine (SVM), Gaussian naive Bayes (NB), and random forest (RF) were applied to classify the features. Finally, the performance of classifiers was measured in terms of the parameters such as precision, recall, F1-score, support and accuracy. From the experimental results, it was found that RF results in higher accuracy of 98% as compared to SVM and Gaussian NB.

Online publication date: Wed, 22-Mar-2023

The full text of this article is only available to individual subscribers or to users at subscribing institutions.

 
Existing subscribers:
Go to Inderscience Online Journals to access the Full Text of this article.

Pay per view:
If you are not a subscriber and you just want to read the full contents of this article, buy online access here.

Complimentary Subscribers, Editors or Members of the Editorial Board of the International Journal of Computational Science and Engineering (IJCSE):
Login with your Inderscience username and password:

Username:        Password:          Forgotten your password?

Want to subscribe?
A subscription gives you complete access to all articles in the current issue, as well as to all articles in the previous three years (where applicable). See our Orders page to subscribe.

If you still need assistance, please email subs@inderscience.com