K-Means Clustering Workflow in KNIME

📌 Project Overview

This project demonstrates how to use K-Means Clustering in KNIME Analytics Platform to group respondents/customers based on psychographic survey data.

The workflow:

• Reads survey data from Excel
• Performs clustering using K-Means
• Generates:
  • Customer cluster labels
  • Cluster centroid summaries
• Exports results to Excel files

This workflow is useful for:

• Customer segmentation
• Market research
• Consumer behavior analysis
• Target marketing strategies

---

🛠 Workflow Structure

---

📂 Workflow Nodes Explanation

---

1️⃣ Excel Reader (Node 1)

Purpose

Imports the psychographic survey dataset into KNIME.

Input File

https://1drv.ms/f/c/f483042b9735aab9/IgAor9UeCPV3RaXZHMC1fhD-AZESzdtcVALLU7APVUZlYuc?e=T1gbFb

What Happens

• Reads Excel data
• Each row = one respondent/customer
• Each column = survey variable/statement

Output

A clean dataset ready for clustering.

---

2️⃣ k-Means Node (Node 2)

Purpose

Performs clustering analysis using the K-Means algorithm.

What Happens

• Performs K-means clustering on psychographic variables.
• Excludes irrelevant columns like sr.no
• In this we applied 3 cluster solution, we will get 3 customer segments (Disciplined planners, Homebody values, Social trendsetters).

---

📤 Outputs from k-Means Node

The k-Means node produces two outputs.

---

Output 1 → Cluster Labels

Connected To

Excel Writer (Node 3)

Purpose

Seperate into clusters

Contains

• Original dataset
• Additional cluster column:
  • Cluster 0
  • Cluster 1
  • Cluster 2

Example

Respondent	Cluster
R1	Cluster 0
R2	Cluster 1
R3	Cluster 2

Use Cases

• Market segmentation
• Customer profiling
• Target marketing

Output File

https://1drv.ms/f/c/f483042b9735aab9/IgAor9UeCPV3RaXZHMC1fhD-AZESzdtcVALLU7APVUZlYuc?e=Us5jjf

---

Output 2 → Cluster Summary / Centroids

Connected To

Excel Writer (Node 4)

Purpose

Stores average values (centroids) for each cluster.

Contains

Mean values of all variables for every cluster.

Example

Variable	Cluster 0	Cluster 1	Cluster 2
Fashion Conscious	4.5	2.1	3.8
Health Conscious	1.8	4.6	3.2

Interpretation

Helps identify characteristics of each customer segment.

Example Segment Names

• Cluster 0 → Fashion-focused customers
• Cluster 1 → Health-conscious customers
• Cluster 2 → Balanced customers

Output File

https://1drv.ms/x/c/f483042b9735aab9/IQCwBL9fCuFYQKlFVWAmFx0FAXv_feIFh1X9w4ojhpQwsoM?e=3XiiGC

---

📊 Final Objective

The workflow helps:

• Identify similar customer groups
• Understand consumer behavior
• Perform psychographic segmentation
• Support marketing decision-making

---

✅ Conclusion

• Excel Reader imports the data
• k-Means creates customer clusters
• Node 3 stores customer cluster assignments
• Node 4 stores cluster centroids/summary
• The workflow identifies meaningful customer segments for analysis

---

🧰 Technologies Used

• KNIME Analytics Platform
• K-Means Clustering

---

🚀 How to Run

KNIME Workflow - https://1drv.ms/u/c/f483042b9735aab9/IQANv-neYWeHQqpT4cv8GDyjAYWNTEQof6MifmzlESVXnzo?e=gTsKKR

1. Open KNIME Analytics Platform
2. Import the workflow
3. Open the workflow
4. Configure file paths if needed
5. Execute all nodes
6. View exported cluster results

---

👩‍💻 Author

Dhathri Narne

---

⭐ If you found this project useful, consider giving it a star on GitHub!