Scribd
Upload
3 views2 pages

Document 4

Read 4

Uploaded by

danielfaloye8
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
3 views2 pages

Document 4

Read 4

Uploaded by

danielfaloye8
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 2

NAME: EKEOGU CHUKWUEBUKA PROSPER

MATRIC NO.: F/HD/23/3210154

COURSE: ARTIFICIAL INTELLIGENCE ASSIGNMENT

CODE: COM327

UNSUPERVISED LEARNING: A SHORT NOTE

Unsupervised learning is a type of machine learning where the algorithm analyzes


unlabeled data to find patterns or groupings. Unlike supervised learning, where the data
has predefined categories or outcomes, unsupervised learning algorithms discover these
patterns and structures on their own.

KEY CHARACTERISTICS:

• Unlabeled Data: The algorithm works with data that doesn't have any predefined
labels or categories.

• Pattern Discovery: The goal is to find inherent structures, clusters, or relationships


within the data.

• No Explicit Guidance: The algorithm learns without any human supervision or


guidance.

COMMON TECHNIQUES:

1. Clustering: This involves grouping similar data points together. Some common
clustering algorithms include:

a. K-means: Partitions data into K clusters based on their distance to cluster


centroids.

b. Hierarchical clustering: Creates a hierarchy of clusters, starting with


individual data points and merging them into larger clusters.

c. DBSCAN: Identifies clusters based on density, grouping closely packed data


points together.
2. Dimensionality Reduction: This reduces the number of features in a dataset while
preserving as much information as possible. This is useful for visualizing high-
dimensional data or speeding up other machine learning algorithms. Common
techniques include:
a. Principal Component Analysis (PCA): Finds the principal components that
explain most of the variance in the data.

b. t-SNE: A non-linear technique that maps high-dimensional data to a lower-


dimensional space while preserving local similarities.

3. Anomaly Detection: This identifies unusual data points that deviate from the norm.
This is useful for fraud detection, network intrusion detection, and fault diagnosis.
Common techniques include:

a. Isolation Forest: Isolates anomalies by randomly partitioning the data.

b. One-Class SVM: Learns a boundary around normal data points, identifying


anomalies outside the boundary.

APPLICATIONS:

• Customer Segmentation: Grouping customers with similar purchasing behavior.

• Image and Video Analysis: Identifying objects or patterns in images or videos.

• Anomaly Detection: Detecting fraudulent transactions or network intrusions.

• Recommendation Systems: Recommending products or services based on user


preferences.

• Drug Discovery: Identifying new drug compounds with potential therapeutic


properties.

Unsupervised learning is a powerful tool for discovering hidden patterns and


structures in data. It can be used to gain insights into data that would be difficult or
impossible to obtain through human analysis.

You might also like