Scribd
Upload
4 views9 pages

Name - Sameer Ali PPT of Machine Learning

Uploaded by

Shubham Singh Rajput
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
4 views9 pages

Name - Sameer Ali PPT of Machine Learning

Uploaded by

Shubham Singh Rajput
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 9

NAME – SAMEER ALI CLUSTURING AND

INCREMENTAL
CLUSTURING
SLIDE 1: •Definition: Clustering is a technique in
machine learning and data analysis that

INTRODUCTION TO
involves grouping similar data points together.
•Purpose: Discover hidden patterns,
relationships, and structures within a dataset.
•Common Applications: Image segmentation,

CLUSTERING
customer segmentation, anomaly detection.
SLIDE 2: TYPES OF
CLUSTERING
1.Hierarchical Clustering
1. Explains the hierarchical structure of clusters.
2. Dendrogram illustration.

2.Partitioning Clustering
1. Divides data into non-overlapping subsets.
2. Example: K-means algorithm.

3.Density-Based Clustering
1. Identifies regions of high data point density.
2. Example: DBSCAN (Density-Based Spatial Clustering of Applications with
Noise).
SLIDE 3: INCREMENTAL
CLUSTERING OVERVIEW
•Definition: Incremental clustering is an approach that updates
existing clusters when new data points are introduced.
•Benefits: Efficient handling of evolving datasets, real-time analysis,
reduced computational costs.
SLIDE 4: ADVANTAGES OF
INCREMENTAL
CLUSTERING
1.Real-Time Adaptability
1. Easily adapts to changing data in real-time.
2. Suitable for streaming data applications.

2.Efficient Resource Utilization


1. Reduces the need to reprocess the entire dataset.
2. Saves computational resources.

3.Scalability
1. Scales well with large and dynamic datasets.
2. Handles continuous data influx.
SLIDE 5: INCREMENTAL
CLUSTERING
ALGORITHMS
1.BIRCH (Balanced Iterative Reducing and Clustering using
Hierarchies)
1. Overview of the algorithm.
2. Suitable for large datasets and stream data.

2.Online K-means
1. Modification of traditional K-means for incremental updates.
2. Explanation of how it works.
SLIDE 6: CHALLENGES IN INCREMENTAL
CLUSTERING
CONCEPT DRIFT

1.Concept Drift
1. Addressing changes in the underlying data distribution.
2. Methods to handle concept drift.

2.Maintaining Cluster Quality


1. Ensuring clusters remain representative and meaningful.
2. Continuous evaluation and adjustment.
SLIDE 7: BEST PRACTICES
•Provide tips and best practices for implementing incremental
clustering effectively.
•Emphasize the importance of monitoring and updating clustering
models.
SLIDE 8: CONCLUSION
•Summarize key points about clustering and incremental clustering.
•Reinforce the benefits and practical applications.

You might also like