International Journal of Web of Multidisciplinary Studies

( Peer-Reviewed, Open Access, Fully Refereed International Journal )

website: [Link] E-ISSN : 3049-2424
Volume 3 Issue 3 , March 2026
DOI: [Link]

AI –DRIVEN PUBLIC HEALTH CHATBOT FOR DISEASE AWARENESS

S.KAVIMANI1, K.BHARATHI2, [Link] MADHAVAN3, K.BHUVANESH4, P.YUHAPRASANTH5
1
Assistant Professor, Department of Information Technology , M P Nachimuthu M Jaganthan Engineering College
2,3,4,5
Final Year [Link] (IT), Department of Information Technology , M P Nachimuthu M Jaganthan Engineering
College.

Article Info Abstract:

The rapid growth of digital technologies has significantly transformed the healthcare
Article History: sector, enabling innovative solutions for disease awareness and preventive care.
Despite these advancements, a large portion of the population still lacks access to
reliable, real-time healthcare information, particularly in rural and underserved
Published:23 March 2026 regions. Traditional awareness systems rely on static communication channels such
Publication Issue: as websites, brochures, and campaigns, which often fail to provide personalized and
Volume 3, Issue 3 interactive guidance. This research proposes an AI-driven public health chatbot
March-2026 designed to deliver accurate, real-time, and user-specific health information. The
Page Number: system leverages Natural Language Processing (NLP), machine learning algorithms,
440-447 and a structured medical knowledge base to understand user queries and generate
context-aware responses. The chatbot integrates verified health data sources and
Corresponding Author: continuously updates its knowledge base to ensure reliability and accuracy. It also
[Link]
incorporates feedback mechanisms to improve performance over time.
Keywords: Artificial Intelligence, Public Health, Chatbot, Disease Awareness,
Natural Language Processing (NLP), Machine Learning, Health Informatics,
Preventive Healthcare, Digital Health Systems, Real-Time Health Monitoring

1. INTRODUCTION
Public health awareness is a fundamental component of a well-functioning healthcare system. It plays a crucial
role in preventing diseases, reducing healthcare costs, and improving the overall quality of life. However, access
to timely and accurate health information remains a major challenge, particularly in developing regions where
healthcare infrastructure is limited. Traditional methods of disseminating health information—such as awareness
campaigns, printed materials, and static online resources—are often insufficient in addressing the dynamic and
personalized needs of individuals. These methods lack interactivity, real-time responsiveness, and adaptability
to user- specific conditions . With the emergence of Artificial Intelligence (AI), intelligent chatbot systems have
gained popularity as effective tools for healthcare communication.. AI-driven chatbots have the potential to
bridge the gap between healthcare providers and the general public by offering accessible and reliable
[Link] research aims to design and develop an AI-based public health chatbot that enhances disease
awareness through intelligent interaction. The primary objectives include:

• Delivering accurate and real-time health information

IJWOS | Volume 3 Issue 3, March 2026 | [Link] 440

Copyright © 2025 The Author(s) : This is an open access article under the CC BY license ([Link]
 International Journal of Web of Multidisciplinary Studies
E-ISSN: 3049-2424

• Providing personalized responses based on user queries

• Reducing misinformation and panic during health crises
• Promoting preventive healthcare practices
• Enhancing accessibility in remote and rural areas
The proposed system represents a significant step toward digital healthcare transformation and supports the
broader vision of smart and inclusive healthcare systems.

2. LITERATURE REVIEW
The integration of AI in healthcare has been extensively explored in recent years, particularly in the domain of
conversational agents and chatbots. NLP- based chatbots are widely used for applications such as symptom
checking, mental health counseling, appointment scheduling, and patient engagement.

Early chatbot systems were primarily rule- based, relying on predefined patterns and responses. While these
systems were simple to implement, they lacked flexibility and could not handle complex queries. The
introduction of machine learning techniques significantly improved chatbot performance by enabling systems to
learn from data and adapt to user [Link] advancements in deep learning and NLP have further
enhanced chatbot capabilities. Techniques such as intent classification, entity recognition, and sentiment analysis
allow chatbots to understand user queries more accurately. Additionally, the use of probabilistic models and
neural networks has improved response [Link] healthcare organizations have deployed chatbots
during global health crises to provide real-time information and reduce the burden on healthcare systems.
However, existing systems face challenges such as limited contextual understanding, data privacy concerns, and
lack of scalability. The proposed system addresses these limitations by integrating advanced NLP techniques,
secure data handling mechanisms, and scalable architecture to ensure efficient and reliable performance

3. PROPOSED SYSTEM
System Overview
The proposed AI-driven chatbot is designed as an intelligent, interactive system capable of providing disease-
related information through natural language communication. It integrates multiple modules that work together
to process user queries and generate appropriate responses.

Core Modules
1. User Interaction Module: This module serves as the interface between the user and the
chatbot. It supports both text and voice inputs, enabling users to interact with the system easily. The interface
is designed to be user-friendly and accessible across multiple platforms such as web and mobile applications.
2. NLP Processing Module: This module processes user input using Natural Language

IJWOS | Volume 3 Issue 3, March 2026 | [Link] 441

 International Journal of Web of Multidisciplinary Studies
E-ISSN: 3049-2424

Processing techniques. It includes tokenization, stemming, and stop-word removal to convert raw text into
structured data suitable for analysis.
3. Intent Recognition Module: This module identifies the purpose of the user’s query. For example,
whether the user is asking about symptoms, treatment, prevention, or general information about a disease.
4. Knowledge Base Module: The knowledge base contains structured health- related
information, including diseases, symptoms, causes, treatments, and preventive measures. It is regularly updated
using verified sources.
5. Response Generation Module: This module generates responses based on the identified intent and
retrieved data. It ensures that responses are accurate, relevant, and easy to understand.

6. Feedback and Learning Module: This module collects user feedback and interaction data to
continuously improve the chatbot’s performance through machine learning.

4. METHODOLOGY
Data Collection
Data is collected from trusted sources such as medical databases, government health portals, and research
publications. The dataset includes disease-related information such as symptoms, causes, risk factors, and
treatment options.

4.1 Data Preprocessing

Data preprocessing ensures that the collected data is clean, consistent, and suitable for analysis.

𝑥 − 𝑥𝑚𝑖𝑛
𝑥′ =
𝑥𝑚𝑎𝑥 − 𝑥𝑚𝑎𝑥

Steps involved:
• Removing duplicate entries
• Handling missing values
4.4 Intent Classification

1
𝑃 = 𝑦=1∣𝑥

A logistic regression model is used to classify user queries into predefined categories. This probabilistic
approach allows the system to handle uncertainty and improve decision- making.

IJWOS | Volume 3 Issue 3, March 2026 | [Link] 442

 International Journal of Web of Multidisciplinary Studies
E-ISSN: 3049-2424

4.5 Response Generation

The chatbot retrieves relevant information from the knowledge base and generates responses using predefined
templates and dynamic content generation techniques.

5. ALGORITHM (Expanded)
AI Chatbot Algorithm (Detailed)
The proposed AI-driven public health chatbot follows a systematic sequence of steps to process user queries
and generate accurate, context-aware responses.

Step 1: Accept User Query The system receives input from the user in the form of text
or voice through the chatbot

• Normalizing numerical data

• Standardizing text formats
Step 2: Convert Input into Machine- Readable Format Relevant features such
as keywords, symptoms, and disease categories are extracted from the dataset. These features play a critical
role in improving the accuracy of classification models.

If the input is voice-based, it is converted into text using speech recognition techniques. The input is then
encoded into a format suitable for computational processing.

Step 3: Text Preprocessing The input text undergoes preprocessing to improve quality
and consistency:

• Tokenization (splitting text into words or phrases)

• Removal of stop words (e.g., “is”, “the”, “and”)
• Stemming or lemmatization (reducing words to root form)
• Noise removal (punctuation, special characters)
Step 4: Feature Extraction Relevant features such as keywords, symptoms, and intent-related terms
are extracted using NLP techniques. These features help in understanding the context of the query.

Step 5: Intent Classification Using ML Model

A machine learning model (e.g., logistic regression or classification model) analyzes the extracted features to
determine the user’s intent, such as:

• Symptom inquiry
• Disease information
• Prevention methods

IJWOS | Volume 3 Issue 3, March 2026 | [Link] 443

 International Journal of Web of Multidisciplinary Studies
E-ISSN: 3049-2424

• Treatment guidance
Step 6: Query Knowledge Base Based on the identified intent, the system searches the structured medical
knowledge base to retrieve accurate and relevant information.

Step7:GenerateContextAwareResponse The chatbot constructs a response using predefined templates and

dynamic content generation, ensuring clarity, relevance, and user-friendly language.

Step 8: Display Response to User The generated response is delivered to the user through the
interface in real time, ensuring quick and interactive communication.

Step 9: Store Interaction for Future Learning

User queries and chatbot responses are stored in a database for analysis and future improvements. This
helps in identifying common queries and system performance.

Step 10: Update Model Using Feedback Feedback collected from users is used to retrain and optimize
the machine learning model, improving accuracy, adaptability, and overall system performance over time.

6. SYSTEM ARCHITECTURE
The system follows a layered architecture:
• User Layer: Handles interaction via UI
• Application Layer: Manages chatbot logic and workflows
• Processing Layer: Performs NLP and machine learning operations
• Database Layer: Stores structured health data
This architecture ensures modularity, scalability, and efficient data processing

7. PERFORMANCE EVALUATION
The system is evaluated using the following metrics:

Accuracy: Correctness of responses Precision & Recall: Effectiveness of classification

Response Time: Time taken to respond

User Satisfaction: Feedback score

The evaluation demonstrates that the system performs efficiently under various conditions.

8. RESULTS
The experimental results indicate significant improvements over traditional systems

IJWOS | Volume 3 Issue 3, March 2026 | [Link] 444

 International Journal of Web of Multidisciplinary Studies
E-ISSN: 3049-2424

Paramete r Existing Proposed

System System
Accuracy 60–70% 90–95%

Response
High Low
Time
User
Engagem ent Moderate High

Fraud/Mi
sinformat High Low
ion

The chatbot effectively provides accurate, fast, and reliable health information

9. DISCUSSION
The proposed system highlights the significant potential of Artificial Intelligence in transforming public health
communication by enabling fast , interactive, and reliable information delivery. By automating responses to
common health-related queries, the chatbot reduces the dependency on healthcare professionals for basic
consultations, allowing medical experts to focus on more critical cases. This contributes to improved efficiency
within healthcare systems, especially in regions with limited medical resources . Furthermore, the system
enhances transparency and user trust by relying on verified and structured health data sources. The use of
standardized datasets and continuous updates ensures that users receive accurate and up-to-date information,
which is crucial in preventing the spread of misinformation. The interactive nature of the chatbot also improves
user engagement and encourages individuals to seek health information [Link] these advantages,
certain challenges remain. Data privacy and security are critical concerns, as the system handles sensitive user
information. Proper encryption and secure data management techniques are necessary to protect user data.
Additionally, language barriers may limit accessibility, particularly in multilingual regions, highlighting the need
for multi- language [Link] limitation is the chatbot’s ability to handle complex medical conditions or
emergency situations, where human expertise is essential. While the system is effective for general awareness
and basic guidance, it cannot replace professional medical [Link] enhancements may include the
integration of advanced AI models, multilingual capabilities, voice-based interaction, and stronger data security
mechanisms. These improvements can further increase the system’s reliability, accessibility, and overall impact
on public health awareness.

IJWOS | Volume 3 Issue 3, March 2026 | [Link] 445

 International Journal of Web of Multidisciplinary Studies
E-ISSN: 3049-2424

10. CONCLUSION
This research presents an AI-driven public health chatbot designed to enhance disease awareness and improve
accessibility to reliable healthcare information. By integrating Natural Language Processing (NLP), machine
learning techniques, and a structured health knowledge base, the system is capable of delivering accurate, real-
time, and user-specific responses. The chatbot effectively bridges the gap between healthcare information and
the general public, particularly in regions with limited access to medical resources. It also improves user
engagement through interactive communication and supports preventive healthcare practices by encouraging
early awareness and informed decision-making. Additionally, the automation of basic health queries helps reduce
the workload on healthcare professionals, contributing to a more efficient healthcare ecosystem.

Future enhancements may include the incorporation of multilingual support to improve accessibility across
diverse populations, voice-based interaction for better usability, and integration with wearable health devices
for real-time health monitoring. Overall, the proposed chatbot provides a practical, efficient, and scalable
solution for promoting public health awareness and supporting the digital transformation of healthcare services.

References
1. Jurafsky, D., & Martin, J. H. (2020). Speech and Language Processing (3rd ed.). Pearson.
2. World Health Organization (WHO). (2021). Global Strategy on Digital Health 2020–2025.
3. Bickmore, T., & Giorgino, T. (2006). Health dialog systems for patients and consumers. Journal of
Biomedical Informatics, 39(5), 556–571.
4. IBM Watson Health. (2020). Artificial Intelligence in Healthcare: Applications and Benefits.
5. Government of India. (2020). National Digital Health Mission (NDHM).
6. Ministry of Health and Family Welfare (MoHFW). (2021). Public Health Initiatives in India.
7. NITI Aayog. (2018). Strategy for New India @75: Health Sector.
8. Young, T., et al. (2018). Recent trends in deep learning based natural language processing. IEEE Computational
Intelligence Magazine, 13(3), 55–75.
9. Devlin, J., Chang, M. W., Lee, K., & Toutanova, K. (2019). BERT: Pre-training of deep bidirectional
transformers for language understanding.
10. NAACL-HLT Huang, J., Zhou, M., & Yang, D. (2007). Extractin chatbot knowledge from online discussion
forums. IJCAI.

IJWOS | Volume 3 Issue 3, March 2026 | [Link] 446

 International Journal of Web of Multidisciplinary Studies
E-ISSN: 3049-2424

8 Miner, A. S., et al. (2016). Talking to machines about personal mental health problems. JAMA, 316(22),
2367–2368.
9 .Topol, E. (2019). Deep Medicine: How Artificial Intelligence Can Make Healthcare Human Again. Basic
Books.
10 Kaur, H., & Kumari, V. (2020). Chatbots in healthcare: A systematic review. International Journal of
Advanced Computer Science and Applications, 11(4).
11 National Informatics Centre (NIC). (2020). E-Governance Infrastructure in India.
12 Reserve Bank of India (RBI). (2021). Digital Systems and Data Security in India.

IJWOS | Volume 3 Issue 3, March 2026 | [Link] 447