A web-based platform for visualizing datasets and predicting outcomes using multiple machine learning models with automatic preprocessing and model evaluation.
This project allows users to upload datasets, perform real-time data visualization, analyze patterns, and generate predictions using various ML algorithms. The system preprocesses the dataset (handles nulls, types, duplicates), reduces features, trains multiple ML models, and shows the best model based on accuracy. It integrates an intuitive web UI with a powerful Django + Python backend.
- Clone the repository:
git clone https://github.com/YourUsername/Data-Visualization-and-Model-Prediction-for-Analytical-Data.git
cd Data-Visualization-and-Model-Prediction-for-Analytical-Data
pip install -r requirements.txt
python manage.py runserver
- Users upload .csv datasets through the UI.
- Handling null values
- Removing duplicates
- Converting datatypes
- Feature scaling (normalization/standardization)
- Feature selection & dimensionality reduction
The platform provides:
- Bar, Line, Scatter, Area, Pie, and Histogram plots
- Drag-and-drop column selectors for X, Y, and Z axes
- Real-time interactive charts using Chart.js
The system trains multiple machine learning models:
-
Logistic Regression
-
Random Forest
-
K-Nearest Neighbors
-
Support Vector Machine
-
Decision Tree
-
Naive Bayes
-
Gradient Boosting
Accuracy of each model is evaluated and the best-performing model is displayed to the user with metrics.
Great for beginners to learn how ML models work and how preprocessing impacts prediction.
Used in real-world projects to visualize datasets before building ML models.
Automates ML workflow from raw CSV to insights, including preprocessing and evaluation.
Upload company data to forecast sales, customer churn, or behavior with ease.
-
Preprocess, visualize, train, and evaluate models all in one place.
-
Eliminates the need to code ML pipelines manually.
- Simple UI to select graph dimensions.
- Auto-generates different graph types like bar, scatter, area, and more.
- Multiple Model Comparison
- Automatically trains multiple algorithms.
- Real-Time Visualization
- Interactive, responsive charts with HTML/CSS/JS and Chart.js.
- Supports 1D, 2D, and 3D graphs depending on column selection.
DataVisualizationandModelPre.../
├── MyApp/
│ ├── Code/
│ │ └── PythonCodes/
│ │ ├── __pycache__/
│ │ ├── classification.py
│ │ ├── clustering.py
│ │ ├── dataanalysis.py
│ │ ├── preprocess.py
│ │ └── regression.py
│ ├── Templates/
│ │ ├── data_visualization.html
│ │ ├── home.html
│ │ ├── index.html
│ │ ├── login2.html
│ │ ├── model_prediction.html
│ │ └── register2.html
│ ├── __pycache__/
│ ├── migrations/
│ │ └── __init__.py
│ ├── __init__.py
│ ├── admin.py
│ ├── apps.py
│ ├── models.py
│ ├── tests.py
│ ├── urls.py
│ └── views.py
├── OutputScreenshots/
├── LICENSE
├── README.md
├── db.sqlite3
└── manage.py
Preprocessing and ML model training using pandas, sklearn, numpy, etc.
Manages backend logic, API routes, and model evaluation pipeline.
Frontend interface for drag-and-drop, visualization, and interactions.
Renders dynamic charts in the browser.
This project is open-source and available under the MIT License.
© 2025 Sruthi Pulipati
Sruthi Pulipati
This project was built to make data science accessible by combining automation, visualization, and real-time model training in a single, easy-to-use web application as a part of MINI-PROJECT.