Inspiration
As data scientists ourselves, we spend a lot of time finding insights from different datasets. We realized that many of these processes could be automated with AI, enabling anyone, regardless of their technical expertise, to have an AI-powered data scientist at their disposal to help extract insights from their data. Canva seemed like the perfect platform for this, as it’s the go-to tool for creating reports and presentations, and integrating our solution directly within Canva would streamline the entire process for users.
What it does
Plotit AI automates the extraction of insights from any tabular data and creates visualizations that can be customized to fit brand aesthetics. Users simply upload their data, and the AI identifies key trends, patterns, and insights, generating a range of charts and graphs that can be easily edited and incorporated into their Canva designs.
How we built it
We automated many of the processes we previously handled manually using Python. This includes loading, cleaning, and preprocessing datasets, conducting exploratory data analysis (EDA), and identifying interesting insights to plot. We leveraged machine learning and data processing libraries to streamline these tasks, ensuring that the AI could handle a wide variety of datasets and consistently generate meaningful visualizations.
Challenges we ran into
Handling corrupted data: Ensuring the AI could manage and clean corrupted or incomplete datasets was a significant challenge, requiring robust error handling and preprocessing techniques. Plotting only interesting insights: We had to develop algorithms that could identify truly valuable insights, filtering out noise and irrelevant data to avoid overwhelming users with unnecessary visuals. Making plots fully customizable: Achieving a balance between automation and customization was tricky, as we wanted to provide users with flexibility in how they present their data while still benefiting from the AI’s automated insights.
Accomplishments that we're proud of
Automating complex data analysis: We successfully built an AI that can replicate the tasks of an experienced data scientist, from data cleaning to generating insightful visualizations. Seamless integration with Canva: We managed to integrate Plotit AI with Canva, allowing users to effortlessly create and customize visualizations within their favorite design platform. User-friendly interface: Despite the complexity of the underlying processes, we created an intuitive interface that makes it easy for users of all skill levels to extract and visualize data insights.
What we learned
Throughout this project, we learned the importance of balancing automation with user control. While AI can automate many tasks, users still value the ability to customize and tweak the output to fit their specific needs. We also deepened our understanding of the challenges involved in making data analysis accessible to non-technical users, particularly in terms of simplifying the user experience without sacrificing functionality.
What's next for Plotit AI
Providing insights along with plots: We plan to enhance Plotit AI by offering not just visualizations but also narrative insights that help users understand the story behind their data. Direct integration with HubSpot: Expanding our integrations to include HubSpot, allowing users to analyze and visualize marketing data directly within Canva. Finding insights across multiple apps: We aim to enable cross-platform data analysis, allowing users to combine datasets from multiple apps and uncover deeper, more complex insights.
Log in or sign up for Devpost to join the conversation.