[Unsupervised ML] Implement Clustering Algorithm for Data Segmentation

[ravicoder01]

📌 Description:
As part of the unsupervised machine learning task, I would like to implement clustering algorithms to analyze and segment the given dataset.

🔧 Proposed Algorithms:

• K-Means Clustering
• DBSCAN (Density-Based)
• Hierarchical Clustering (optional)

🎯 Objective:

• Perform clustering on the dataset (if available)
• Visualize the clusters using matplotlib/seaborn
• Evaluate clusters using silhouette score or similar metric

🔗 Tools & Libraries:

• Python
• scikit-learn
• pandas
• matplotlib / seaborn

🧑‍💻 I’m a GSSoC’25 Contributor and would love to work on this under guidance. Kindly assign this issue to me if valid. 😊

[ravicoder01]

@manasvi-0 can you assign it to me ..?

[viv2005ek]

Hi @ravicoder01 👋
Thanks for your interest!

To maintain consistency across our project structure, you can follow this architecture:

Steps:

1. Modularize your algorithm- Inside unsupervised_module.py, write separate functions for each algorithm.
   2.Integrate Streamlit UI-
   Add relevant Streamlit widgets such as:
   Sliders for n_clusters, eps, min_samples, etc.
   Dropdowns for selecting distance metrics or linkage types
   Checkboxes/toggles if needed for additional control
   3.Add Visualizations (using Matplotlib / Seaborn)
   4.Each algorithm function should return Streamlit components, so they can be directly rendered from app.py
   5.Create a docs/ folder (if not present) and add a short markdown file for your algorithm(s) – explaining logic, parameters used, and visual outputs.

try to complete this task within 5-6days.

[ravicoder01]

Thank you so much for assigning me this task @viv2005ek

I’ll start working on it as per the structure you've shared.
I’ll modularize the unsupervised algorithms and integrate them with Streamlit UI and visualizations.
I’ll also make sure to include a markdown file in the docs/ folder as instructed.

If I face any blockers, I’ll make sure to reach out.

[manasvi-0]

🔧 While building the unsupervised learning workflow, please keep the following in mind:

1. Add a select box to let users choose between Clustering and Dimensionality Reduction.
2. Create a reusable dataset generation function for both clustering and dimensionality reduction.
3. Integrate appropriate visualizations based on the selected algorithm.
4. Include relevant evaluation metrics (e.g., silhouette score, explained variance).
5. Add a video-based visual demo to show how parameters affect results — especially helpful for clustering visuals.

[esharatnaparkhi]

Hello, Can you assign me this task

[viv2005ek]

@ravicoder01
Please try to complete the task as soon as possible. Before creating a PR, make sure to pull the latest changes from the main repo into your forked repo to avoid any conflicts and ensure code efficiency.

[manasvi-0]

@viv2005ek i g @moah0911 has done all the unsupervised algos now we dont need any other contributor for this

[ravicoder01]

hello @viv2005ek @manasvi-0 i have done all the tasks of the unsupervised algorithm as per your instructions. As I am new to this, I don't know how pr works exactly that much, but I am learning all the things.
I have done with my program, changes, and documents as per instructed and all in hierarchical manner. Please guide me with how to resolve this conflict and merge pr. I've spent my enough time to complete this repo , guide me the next part if possible.

[ravicoder01]

this task was assigned to me, and the deadline isn't over yet. How can somebody else contribute to it.

[manasvi-0]

okay complete it and i will review your code and add it to the main