Scribd
Upload
8 views4 pages

Unsupervised Learning Networks 1

Uploaded by

Priya Senthilkumar
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
8 views4 pages

Unsupervised Learning Networks 1

Uploaded by

Priya Senthilkumar
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 4

UNSUPERVISED LEARNING NETWORKS

 The unsupervised learning networks is


independent. As there exists no feedback
from the system or environment to indicate
the desired outputs of a network.

 The network by itself should discover any


relationships of interest, such as features,
patterns, contours, correlations or categories,
classifications in the input data, and thereby
translate the discovered relationships into
outputs. Such networks are also called self-
organizing networks.

 An unsupervised learning can judge how


similar a new input pattern is from patterns
already
seen, and the network gradually learns what
the
similarity is.
 The network may construct a set of axes
along which to measure similarity to previous
patterns, it can be performed by principal
component analysis(PCA), clustering,
adaptive vector quantization and feature
mapping.

 As unsupervised learning is the type of ANN


under unsupervised learning tasks, input
vectors of similar types are combined to form
clusters.

 To find patterns or regularity in the input


data by forming clusters, there exist several
neural networks that come under this
category.

Max net
Mexican hat
Hamming net
Kohonen self-organizing feature map
Counter propagation net
Learning vector quantization (LVQ)
Adaptive resonance theory (ART).

ADVANTAGES:
 No need for labelled datasets.
 Discovery of hidden patterns.
 Flexibility and Adaptability.
 Data Dimensionality reduction.
 Enhanced decision making.

CHALLENGES:
 Evaluation of results.
 Scalability.
 Interpretability.
 Data sparsity and noise.
 Complexity of model selection.
 Overfitting.
REAL TIME APPLICATIONS:

 Anomaly detection in cybersecurity:


Unsupervised learning networks like autoencoders and
SOMs are widely used to detect unusual patterns or anomalies in
network traffic, which may indicate cybersecurity threats such as
intrusions or fraud.

 Customer segmentation in marketing:

Clustering algorithms such as K-means and Fuzzy C-


Means are employed to segment customers into groups based on
purchasing behaviour, preferences, or demographics, enabling
more targeted marketing strategies.

 Image and video segmentation:

In medical imaging and computer vision, unsupervised


learning is used to segment images or video frames into
meaningful parts, such as separating tumours from healthy tissue
in medical scans.

 Financial market analysis:

Unsupervised learning is applied in finance for clustering


stocks based on historical price patterns, anomaly detection in
trading, and uncovering latent factors that drive market
movements.
 Recommendation systems:
Unsupervised learning techniques, especially RBMs and
DBNs, are used to build recommendation systems that suggest
products, movies, or content based on patterns in user behaviour
without requiring explicit ratings.

 Natural language processing:

In NLP, unsupervised learning models are used for tasks


like topic modeling, where the goal is to discover the underlying
themes in a large corpus of text, or for word embeddings, which
capture the relationships between words based on their context in
text data.

You might also like