https://github.com/gurupatil0003/algorithm
Clustering techniques like K-Means, EM Algorithm, and Apriori are widely used in data analysis for various purposes. K-Means is commonly used for customer segmentation, where it groups customers with similar behaviors or characteristics.
https://github.com/gurupatil0003/algorithm
github ml mlops-workflow mongodb
Last synced: 5 months ago
JSON representation
Clustering techniques like K-Means, EM Algorithm, and Apriori are widely used in data analysis for various purposes. K-Means is commonly used for customer segmentation, where it groups customers with similar behaviors or characteristics.
- Host: GitHub
- URL: https://github.com/gurupatil0003/algorithm
- Owner: Gurupatil0003
- Created: 2025-04-23T08:40:00.000Z (about 1 year ago)
- Default Branch: master
- Last Pushed: 2025-04-24T10:56:25.000Z (about 1 year ago)
- Last Synced: 2025-07-02T15:51:05.178Z (12 months ago)
- Topics: github, ml, mlops-workflow, mongodb
- Homepage:
- Size: 172 KB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
```
https://in.images.search.yahoo.com/search/images;_ylt=Awrx_3uA5gloLgIAoTq7HAx.;_ylu=Y29sbwNzZzMEcG9zAzEEdnRpZAMEc2VjA3BpdnM-?p=aprior+algorithm&fr2=piv-web&type=E211IN714G0&fr=mcafee#id=355&iurl=https%3A%2F%2Frailsware.com%2Fblog%2Fwp-content%2Fuploads%2F2018%2F09%2Fillustration-9.jpg&action=click
```
# EM ALgorithm
``` python
1. **Guess initial values** for the hidden parts of your data (e.g., cluster centers or parameters).
↓
2. Repeat until things stop improving:
a. **Estimate**: Using your current guesses, figure out the missing information (like how likely each data point belongs to each group).
b. **Maximize**: Update your guesses based on the new information. Adjust the hidden parts of your data to better fit the observed data.
↓
3. You now have the best estimates for the hidden parts and the model.
```
# K means
```pytohn
1. Pick how many groups (K) you want.
↓
2. Randomly place K “center” points in your data space.
↓
3. Repeat until things stop changing much:
a. **Sort**: For each data item, find which center it’s closest to and put it in that group.
b. **Move**: For each group, shift its center to the middle of the items in that group.
↓
4. You now have K groups and their centers—clustering complete!
```python
# Apriori
```pytohn
1. **Start with individual items**: Look at the items in each transaction and see how often they appear together.
↓
2. **Find frequent item sets**: Identify which combinations of items appear frequently (above a certain threshold).
↓
3. **Generate larger item sets**: Combine the frequent item sets to make larger combinations and check if they appear often.
↓
4. **Repeat** until no more large item sets are frequent.
↓
5. **Find associations**: Based on frequent item sets, find rules like "If you buy X, you might also buy Y".
```
### Further Reading & Interactive Notebooks
| Topic | Medium Article | Google Colab |
|---------------|----------------------------------------------------------------------------------------------------------------------------------|------------------------------------------------------------------------------------------------------------------------------|
| K Means | [
](https://medium.com/@gurupatil327/k-means-clustering-algorithm-f642cae2f627) | [
](https://colab.research.google.com/drive/1dIdZpkMgT6XRfPMNmQ5MSfO3O_0a7m6O?usp=sharing) |
| EM Algorithm | [](https://medium.com/@gurupatil327/introduction-to-em-algorithm-in-ml-dfe2f1ea079c) | [](https://colab.research.google.com/drive/1TPiUsaPO-V-NBEqVx7L5oXTDIsNh3NIp?usp=sharing) |
| Apriori | [](https://medium.com/@gurupatil327/association-rule-mining-concept-and-implementation-0d2217fd1d1d) | [](https://colab.research.google.com/drive/1Mt71U44NfFJJo0YErwH2kvRtXqJOHzOi?usp=sharing) |