Download as ppt, pdf, or txt
Download as ppt, pdf, or txt
You are on page 1of 9

Modeling and Synthesis of Principle Component

Analysis (PCA) and Partial Least Squares Regression


(PLSR) using Neurocomputing

Presented by Under The Guidance Of


SUNIL PRASAD.G Co-Supervisor
Roll No:00900ph428 Dr.K.RAMANAIDU
Under The Guidance Of
Supervisor
Dr.K.E.SREENIVASA MURTHY
Objective of Study of PCA

1.      It ia way of identifying patterns in data and expressing the


data in such a way as to highlight the differences and
similarities.

2.      In order to reduce the high-dimensional space to lower-


dimensional space
 
3.      In order to this PCA & PLSR are the tools which will reduce
  the higher dimensional space to lower dimensional space
and this should be done by retaining variation in original data

as much as possible
 
a) Classical statistical Method
b) Artificial Neural Networks

There are many ways of in which PCA and PLSR have been
implemented but very less work has been carried using
Artificial Neural Networks.
Artificial Neural Network are the model that attempt to
achieve performance via dense inter connection of simple
Computational elements.
Advantages of ANN

The most important property of neural network is the ability


to learn from its environment.

Fault tolerance is an inbuilt option that is available with


 neural networks.

As Ann's mature into commercial systems, they are likely to


be implemented in hardware. Their fault tolerance and
reliability are therefore vital to the functioning of the system
in which they are embedded.
Pattern Recognition with Neural Networks
• A central issue of Pattern Recognition is that of
generalization i.e the performance of the trained model when
applied to previously unseen data.

• ANN posses characteristics that make them a useful tool for


Pattern Recognition. Important features such as
generalization , tolerance to noise and graceful degradation
makes a robust learning paradigm/ model. Their performance
strongly lies on large amount of data for training.

• ANN with their remarkable ability to derive meanings from


complicated/ imprecise data can be used to extract patterns
and detect trends that are too complex to be noticed by either
human or other computer related techniques.
PCA has been called one of the most valuable
results from applied linear algebra.
PCA is abundantly used in all forms of analysis
from neuroscience to computer graphics.
Feature extraction from a data or a pattern is a
necessary step in pattern recognition, among many
feature extraction methods PCA is also one of the
most popular, relatively effective and simple.
Fault Tolerance in Artificial Neuron

• ANN are composed of large number of simple


computational units operating in parallel, they have a
potential to provide fault tolerance.

• Neural Networks offers a parallel distributed method of


processing information.

• Common assertion is that neural networks are inherently


fault tolerant.
Why Fault Tolerance in Neural Networks

•Neural Computation is an emerging technology with variety of potential


applications such as classification, identification etc.

•The fault tolerance of internal components is highly desirable since it


directly reduces maintenance costs and increases overall productivity.

•In near future, when ANN are developed into commercial system they are
likely to be implemented in hardware. The fault tolerance capability is
vital in functioning of systems embedded in hardware.

•The design of NEURO MICROPROCESOR is targeted for a


peak performance of 1011 connections per sec for a network up to
million units and a billion connections.
Main Areas of Fault to be Explored

• The capability of neural networks to withstand failure of internal


components.

• Producing satisfactory results in presence of noise in input data from


external environment.

• Being able to tolerate noise or imprecision between its internal sub-


systems.

You might also like