Open Access
Subscription Access
Retail Insights Generator
Abstract
Retail businesses generate large volumes of data, but extracting meaningful insights often requires technical expertise in SQL and data analytics. This project, Retail Insights Generator, presents an AI-powered system that converts natural language queries into SQL queries using LangChain and large language models. The system supports multilingual input, voice interaction, and intelligent recommendation features such as restock alerts and discount suggestions. It leverages a dataset of over 73,100 retail records and integrates ChromaDB for few-shot learning to improve query accuracy. The solution simplifies data access for non-technical users, enhances decision-making, and improves operational efficiency. With its lightweight architecture and user-friendly interface, the system provides an effective tool for modern retail analytics.
References
Chen, T., Sun, X., & Li, J. (2020). Neural text-to-SQL generation: A survey. IEEE Transactions on Knowledge and Data Engineering.
Rajkumar, S., & Singh, P. (2019). Natural language interface to databases using deep learning. International Journal of Computer Applications.
LangChain. (n.d.). LangChain documentation. https://docs.langchain.com
Google. (n.d.). Generative AI (Gemini) documentation. https://aistudio.google.com
ChromaDB. (n.d.). ChromaDB documentation. https://www.trychroma.com
Kumar, A., & Sharma, R. (2021). Retail data analytics using machine learning techniques. International Journal of Data Science and Analytics.
Hugging Face. (n.d.). Transformers documentation. https://huggingface.co/docs
SQLite. (n.d.). SQLite documentation. https://www.sqlite.org/docs.html
Gupta, M., & Verma, S. (2022). AI-based decision support systems in retail. International Journal of Advanced Research in Computer Science.
Chauhan, A. (n.d.). Retail store inventory.
Refbacks
- There are currently no refbacks.