Conversational AI Platform
on Call Centre Data
CLIENT & PROBLEM STATEMENT
- A Large call centre Company needed an efficient way to interact with data stored in SQL database .
- The client desires a conversational interaction where the solution is capable of recalling previous conversations and providing responses that are contextually awares.
APPROACH
- Integrated Client’s SQL database with G-Square ask narrator functionality of narrator model to handle big data queries.
- A LLama-based model is used to convert the user’s natural language questions into an appropriate SQL query.
- The generated SQL query is executed to fetch the required data from the underlying database, ensuring the query matches the user’s request accurately and efficiently .
- The retrieved data is passed back to the LLama model, which transforms the raw data into a natural language explanation.
- The solution is develops such that LLama model maintains contextual memory across conversations, allowing the system to recall previous user inputs and queries.
SOLUTION & OUTPUT
- The output section is designed to provide the SQL query generated by the LLama model based on the user’s input ie displayed in a clear and readable format.
- The data retrieved from the database as a result of the SQL query is presented in a structured and easy-to-read format.
- The system offers a natural language explanation of the fetched data.
- A query editing feature is provided, allowing the user to manually modify the SQL query if desired.
- The system may offer suggestions or improvements to the query or data explanation based on previous conversations.
