Nomad Analytix: Text-rich visual reasoning using vision models for insights and recommendations
Nomad Analytix is an innovative business intelligence tool that uses state-of-the-art vision models to transform data analysis. This software automates complex tasks traditionally handled by data analysts, empowering non-technical teams such as marketing and sales to access advanced data analysis easily. By using natural language prompts, users can interact with data intuitively and gain valuable insights without needing extensive technical expertise. A prototype of the software, built on the Streamlit platform, will showcase its ability to generate visualizations from various data sources, including CSV, JSON, SQLite, Excel, and databases, with potential extensions to data warehouses. The integration of Vision Language Models GPT 4 Omni and GPT 4 Turbo- with this framework provides a seamless interface for data querying, visualization creation, and recommendation generation. Nomad Analytix serves as an inclusive, intelligent, and intuitive solution, bridging the gap between data and decision-making across diverse industries. • Nomad Analytix is an innovative business intelligence tool designed to transform data analysis by leveraging state-of-the-art vision models. • This software is built on the Streamlit platform and can generate visualizations from various data sources. • The integration of Vision Language Models gpt4-Omni and gpt-4-turbo-2024-04-09 with this framework provides a seamless interface for data querying, visualization creation, and recommendation generation.
Authors
- Harini Sriraman (ORCID: https://orcid.org/0000-0002-2192-8153)
- Aswathy Ravikumar (ORCID: https://orcid.org/0000-0003-0897-6991)
- Sai Jeevan Puchakayala
Institutions
- Living Systems (United States) (US)
- Vellore Institute of Technology University (IN)
Publication Details
- Journal
- Software Impacts
- Published
- 2025-04-29
- DOI
- https://doi.org/10.1016/j.simpa.2025.100765
- Primary Topic
- Multimodal Machine Learning Applications
- Type
- article
- Field-Weighted Citation Impact
- 0.00