Issuu

International Research Journal of Engineering and Technology (IRJET) Volume: 12 Issue: 04 Apr2025

www.irjet.net

p-ISSN: 2395-0072

VIZARD: AI-Driven Data exploration & Visualization platform using Streamlit Dr.Gulab Singh Chauhan*, S Milan Kumar2, Ruthvik N P°, Prajwal Y S4, Punya K S5 1 Professor, ISE, Acharya Institute Of Technology, Karnataka, India

B.E Student, ISE, Acharya Institute Of Technology, Karnataka, India

B.E Student, ISE, Acharya Institute Of Technology, Karnataka, India B.E Student, ISE, Acharya Institute Of Technology, Karnataka, India

--------------------------------------------------------------------------***---------------------------------------------------------------------------Abstract - With the increasing complexity of datasets, non-technical users struggle to extract meaningful insights. Traditional data querying methods require programming expertise, making data exploration inaccessible for many. This paper introduces “Vizard” an Al-powered conversational data analysis tool that allows users to interact with datasets using natural language queries. Built using Streamlit, Pandas, and PandasAI, this system enables seamless data exploration without requiring coding knowledge. The integration of LLM's ensures accurate query interpretation and dynamic data visualization. Our results demonstrate that Al-assisted querying enhances efficiency and usability, making data analysis more accessible for nonprogrammers. Performance evaluations indicate a 92% user satisfaction rate, 94% accuracy in query responses, and a 4x reduction in query execution time compared to traditional data analysis methodscritical gaps in modern suweillance systems.

2. LITERATURE REVIEW Recent advancements in Conversational AI have led to the development of intelligent query systems that enable users to interact with structured and unstructured datasets. Works such as Smith et al. (2021) demonstrate how LLM-powered chat interfaces improve accessibility for non-technical users. Similarly, research by Brown et al. (2022) highlights the effectiveness of natural language processing (NLP) models in automating complex data retrieval tasks The emergence of LLMs such as GPT-4, Google Gemini and BERT has significantly influenced automated data processing. Studies by Zhou et al. (2023) compare various LLM architectures, showcasing how fine-tuned models improve the interpretability of dataset queries. Our work builds upon these findings by integrating Google Gemini with PandasAI for enhanced query precision. Several studies discuss no-code platforms that bridge the gap between technical and non-technical users. Research by Patel et al. (2020) reviews popular Bl tools such as Tableau, Power BI, and Google Data Studio, noting their limitations in handling complex analytical queries. Our project overcomes these issues by providing an AI-driven, conversational interface for dataset interactions. Prior research has examined the efficiency of SQL-based vs. Al-powered querying systems. A study by Kim et al. (2022) found that traditional SQL queries require explicit schema knowledge, whereas Ai-driven solutions offer more intuitive data retrieval. Our work expands upon this by implementing a conversational model capable of generating real-time insights without predefined query structures. Furthermore, the integration of conversafional AI with data analytics is reshaping how businesses and researchers interact with information. By reducing dependency on technical expertise, these advancements empower a broader audience to engage with data-driven insights. Our study aims to contribute to this evolving landscape by demonstrating how Al-enhanced querying can streamline decision-making across various domains.

Key Words: Conversational AI, Data Science, Natural Language Processing, PandasAI, Data Exploration.

Streamlit,

1. INTRODUCTION Data-driven decision-making is critical in various industries, yet many professionals lack the technical skills to analyze data effectively. Conventional tools like SQL or Pandas require knowledge of query languages, creating barriers for nontechnical users. Recent advancements in Conversational AI have enabled users to interact with data intuitively through natural language. Our project, “Vizard” leverages AI to bridge this gap, providing an accessible interface for dataset querying and visualization. This paper explores the development of a Streamlit-based interactive data exploration tool that integrates PandasAl for AI-driven query processing. The system allows users to upload datasets, ask questions in plain English, and receive structured outputs, including tables, statistics, and visualizations. By eliminating the need for complex coding, Vizard empowers professionals to make data-driven decisions effortlessly. The intuitive interface fosters accessibility, enabling users from diverse backgrounds to explore and analyze data with ease.

Impact Factor value: 8.315

ISO 9001:2008 Certified Journal

Page 1086