The Localised Chatbot for Hospital Customer Care project is aimed at developing an intelligent and interactive chatbot system tailored to enhance customer support services for users who communicate in the native Sri Lankan languages, such as Sinhala or Tamil. This innovative chatbot will serve as a virtual assistant, offering quick and accurate information, addressing medical inquiries, guiding users through hospital procedures, and ultimately enhancing the overall patient experience.
The primary objective of this project is to create a chatbot that can seamlessly understand and respond in both Sinhala and Tamil languages. This linguistic fluency will enable users to interact comfortably and receive accurate and contextually appropriate responses.
The chatbot will streamline the appointment booking process, allowing patients to schedule appointments conveniently through natural language interactions. Additionally, it will send automated reminders to patients about their upcoming appointments.
To provide a more personalized experience, the chatbot will be designed to recognize returning patients and access their medical history (with proper consent). This capability will enable the chatbot to offer tailored guidance, appointment reminders, and post-treatment follow-up instructions.
Data security and patient privacy are of utmost importance. The chatbot will be developed with robust encryption and compliance measures to protect sensitive medical information and adhere to healthcare regulations.
For more details about this project, please refer to the project documentation and source code. Your contributions and feedback are highly encouraged and appreciated.
This project is implemented using the following technologies and models:
-
Retrieval Augmented Generation: The chatbot utilizes retrieval augmented generation techniques to provide contextually relevant responses. It can retrieve information from a knowledge base and generate responses using a language model.
-
ChromaDb: ChromaDb is employed as a database system to store and manage the knowledge base of medical information and hospital procedures. It ensures quick and efficient access to relevant data.
-
Langchain: Langchain technology is used to facilitate language adaptability, enabling the chatbot to understand and respond fluently in both Sinhala and Tamil languages.
-
LLAMA2 Chat 7b LLM (Large Language Model): The chatbot leverages the power of the LLAMA2 Chat 7b Large Language Model to understand user queries, generate natural language responses, and provide personalized patient support.
