• 

  Upload page

• 

  Front page after upload

• 

  2nd page for EDA perform

• 

  Third Page for ML features

About the Project

Inspiration:
The idea behind Steemit came from the need to simplify data analysis for everyone. Whether you’re a student, analyst, or developer, uploading and understanding datasets quickly can be a challenge—especially when dealing with messy or incomplete data. I wanted to build a tool that not only evaluates your data quality but also performs exploratory data analysis (EDA) and applies machine learning techniques, all in one place.

What I Learned:
This project helped me improve my skills in data preprocessing, visualization, and machine learning. I gained hands-on experience with Python libraries like Pandas for data manipulation, Seaborn and Plotly for creating insightful visualizations, and scikit-learn for applying algorithms like KNN and KMeans. Additionally, I learned how to build interactive web apps using Streamlit that make these complex processes accessible to non-experts.

How I Built It:
I developed Steemit using Python as the core language. Users can upload datasets in multiple formats such as CSV, Excel, or JSON. The app evaluates the dataset for missing values and suggests improvements to make the data cleaner and more usable. It then performs EDA with visualizations including heatmaps, box plots, and scatter plots. For machine learning, the app runs KNN classification (providing accuracy) or KMeans clustering (showing clusters graphically), depending on the data type. Streamlit was used to create a responsive and user-friendly web interface.

Challenges Faced:
One of the main challenges was handling different types and formats of datasets smoothly, especially ensuring that the ML algorithms adapt correctly to categorical and continuous variables. Time constraints also pushed me to optimize the app’s performance while maintaining a clean and intuitive UI. Balancing comprehensive functionality with simplicity was tough but rewarding.

Built With

• fpdf
• kmeans
• knn
• matplotlib
• pandas
• plotly
• prophet
• scikit-learn
• streamlit
• xlsxwriter