Inspiration In today's world, data is abundant but often overwhelming. Sifting through heavy datasets can be daunting, especially for those without technical expertise. Inspired by this challenge, we aimed to democratize data analysis by developing a tool that simplifies data into understandable insights through AI-powered summaries and visualizations. Our goal is to make data more accessible, encourage exploratory learning, and possibly spark interest in data science among users from all backgrounds.
What it does ChartRAG empowers users to upload various data files, after which our generative AI takes over to automatically analyze the data. It provides a concise summary and creates a corresponding graph to visualize the findings. Users can further engage with the data by asking AI-driven questions, allowing for a deeper exploration of the data points and trends highlighted in the graph.
How we built it We built ChartRAG using a combination of cutting-edge technologies and frameworks. The frontend was developed with React, styled using CSS and Tailwind for a responsive design. The backend, orchestrated with Flask, handles data processing and integration with the OpenAI API. JavaScript facilitates the dynamic aspects of our application, while Python, paired with Plotly, generates the data visualizations. This diverse tech stack enabled us to create a seamless and interactive user experience.
Challenges we ran into We ran into a lot of challenges. For three of the four members, this was our first hackathon, and during this event we were introduced to a lot of frameworks and concepts that we were unfamiliar with. Because of this, we weren't able to work as efficiently as we wanted to, and we weren't able to accomplish everything that we were aiming for.
Accomplishments that we're proud of We have a lot of accomplishments that we are proud of. We worked tirelessly and learned and learned to our absolute limit, and we are not only proud of what we've made but of how far we've come this weekend. We've created a functional program that manipulates OpenAI and delves into generative AI, and we've studied and used more than five (new!) frameworks in this weekend. We've developed skills like teamwork, skills in GitHub, and skills in system organization.
What we learned First and foremost, we have studied and worked for this project diligently and tirelessly. We've learned a lot about AI, in OpenAI, how we were able to manipulate and use it for our own purposes; in GitHub Copilot and other AI coding assistants, where we learned to work more efficiently and quickly; and, of course, we learned a lot about ChatGPT, specifically how it works and how it can most benefit us.
What's next for ChartRAG Looking ahead, we plan to enhance ChartRAG by incorporating a feature to log the history of users' charts and interactions with the AI. This will allow users to track their inquiries and data explorations over time. Additionally, we aim to expand the customization options for graphs, enabling users to tailor visualizations to their specific preferences and needs.
Log in or sign up for Devpost to join the conversation.