• Introduction
• MultiDimensional Data Model
• Association Analysis
• Data Warehouse Architecture
• Capacity Planning
• Q and A
• Search Engine
• Clustering_Code
• Practical Questions
• Outlier finding Clustering Algorithm
• k-means clustering algorithm
• k-medoids clustering algorithm
• k-means vs k-medoids
• DBSCAN
• Divisive Hierarchical Clustering
• Agglomerative hierarchical Clustering
• Limitations of Data Mining
• Predictive and Descriptive Data Mining
• Decision Tree
• Bayesian Classifier
• Association Analysis Algorithms
• Frequent Pattern Growth
• Star Schema
• Data Cube
• OLAP operations
• Linear Classification
• Reconstruction-based Outlier Detection
• Reconstruction-based Outlier Detection using Auto Encoder
• Q_and_A_2025_new
• Data warehouse - Example - i
• Simple - Apriori- First step
• Bayesian Classifier - Concept

Data Mining And Data Warehousing

DBSCAN

Density-Base Spatial Clustering of Applications with Noise (DBSCAN)

• Clusters are of arbitrary shapes
• Automatically detect the number of clusters
• Data contains noise or outliers

1. Core Point: A point with at least min_samples neighbors within a distance eps.
2. eps distance: minimum distance from core point to another point.
3. min points: number of points close to given point such that the point is core point
4. Density reachable and density connected: Density-reachable refers to a point being reached via a chain of core points within a specified radius, while density-connected means two points can be reached from a common core point, also within the specified radius.
5. Border Point: A point within eps of a core point, but not a core point itself.
6. Noise (Outlier): A point that is not a core point and not within eps of any core point.

• For each point, find its eps-neighborhood (nearby points within a radius eps).
• If it has greater than min_samples neighbors, it becomes a core point and starts forming a new cluster.
• Expand the cluster by recursively adding:
• Core points\' neighbors
• Border points
• Label points as noise that do not belong to any cluster.

• No need to specify the number of clusters (unlike K-Means).
• Can find arbitrarily shaped clusters.
• Automatically detects outliers.
• Works well for spatial and noisy datasets.

• Not good with varying densities (some where very dense data, somewhere low dense data) in the same dataset.
• Sensitive to parameter settings (eps, min_samples).
• Not ideal for high-dimensional data (distance becomes less meaningful).