Sales Prediction using Marketing Spends (KNIME + ML)

This project demonstrates how machine learning techniques can be applied to predict sales based on marketing expenditures across various channels like TV, radio, and newspapers. The workflow is built using KNIME Analytics Platform.

🎯 Objective

To develop a machine learning model that can accurately predict product sales based on the amount spent on different advertising media.

🧠 Tools & Technologies

• Platform: KNIME Analytics Platform
• Language: Visual Workflow (no-code/low-code via KNIME)
• Algorithms Used:
  • Linear Regression
  • Decision Tree Regressor (optional)
• Visualization: KNIME's built-in chart and plot nodes

📊 Dataset

• Source: Advertising Dataset
• Features:
  • TV Spend
  • Radio Spend
  • Newspaper Spend
• Target:
  • Sales

🔁 Workflow Overview

1. Data Import & Preprocessing
2. Exploratory Data Analysis
3. Model Training (Regression)
4. Model Evaluation (RMSE, R²)
5. Prediction & Visualization

📁 Files Included

• KNIME workflow file (.knwf)
• Dataset (CSV format)
• Screenshots of workflow and results
• Model performance summary

📎 Applications

• ROI estimation for marketing campaigns
• Budget allocation optimization
• Business forecasting

👤 Author

Swaroop Shinde
B.Tech ECE | AIML & Data Analytics Enthusiast

---

Let me know if you'd like to include a step-by-step tutorial or export the model for real-time predictions.