Assessing User Interactions with Medical Crowdfunding Apps in India: A Hybrid Approach of Topic and Sentiment Analysis

• Sahiba Khan,
• Ranjit Singh,
• Jot Singh Bindra,
• Dhrubajyoti Bordoloi,
• Ditalak Mpanme,

Vol. 19, No. 10, October 31, 2025
10.3837/tiis.2025.10.013, Download Paper (Free):

• Medical crowdfunding
• donation
• topic modeling
• sentiment analysis
• reviews

Abstract

This research uses topic modeling and sentiment analysis to examine the primary themes and user sentiments expressed in reviews of medical crowdfunding applications (apps) in India. The study focuses on two specific apps, Ketto and Milaap. Using Latent Dirichlet Allocation, seven key topics were identified and categorized as application, ads, limit, login, calls, service, and withdrawal. Sentiment analysis using VADER revealed that users generally hold positive sentiments toward these apps, particularly regarding their usefulness and customer service. A comparative study of the two apps, overall and within specific topics, showed that Ketto received the highest positive reviews. A detailed topic analysis highlights areas for improvement in both apps. Ketto performs well in limits and service, while Milaap excels in the application part. However, both apps face criticism regarding the frequency of ads and calls. Additionally, the study evaluated the predictive accuracy of four machine-learning (ML) models for sentiment and bias detection. The Support Vector Machine (SVM) model demonstrated high accuracy in sentiment prediction using three different feature extraction methods: Bag of Words (BoW), Term Frequency-Inverse Document Frequency (TF-IDF), and hashing. The Quadratic Discriminant Analysis (QDA) model outperformed the others for bias prediction. This research offers valuable guidance for users in selecting the most appropriate app and assists developers in addressing the identified issues. Along with this, platforms can utilize the recommended ML models to analyze and predict the sentiment of authentic reviews from large volumes of feedback, improving customer satisfaction and enhancing their services.

Statistics

Cite this article