Artificial Neural Networks and Their Applications

1
Artificial Neural Networks and Their
Applications

• Prof. Les Sztandera

2
Artificial Neural Networks

• Artificial neural networks (ANNs) are programs
  designed to simulate the way a simple biological
  nervous system is believed to operate. These
  networks have the capacity to learn, memorize and
  create relationships amongst data
• The most widely used ANN is known as the back
  propagation ANN. This type of ANN is excellent at
  prediction and classification tasks
• Another is the Kohonen or self organizing map
  which is excellent at finding relationships
  amongst complex sets of data.

3
Who Needs ANNs?

• People that have to work with or analyze data of
  any kind.
• People in business, finance, industry, education
  and science whose problems are complex,
  laborious, fuzzy or simply un-resolvable using
  present methods.
• People who want better solutions and wish to gain
  a competitive edge.

4
Why Are ANNs Better?

• 1. They deal with the non-linearity in the world
  in which we live.
• 2. They handle noisy or missing data.
• 3. They create their own relationship amongst
  information - no equations!
• 4. They can work with large numbers of variables
  or parameters.
• 5. They provide general solutions with good
  predictive accuracy.

5
What Are ANNs Used For ?

• Their applications are almost limitless, but
  fall into a few simple categories
• Classification
• Forecasting
• Modeling

6
Classification

• Customer/market profiles, medical diagnosis,
  signature verification, loan risk evaluation,
  voice recognition, image recognition, spectra
  identification, property valuation,
  classification of cell types, microbes,
  materials, samples.

7
Forecasting

• Future sales, production requirements, market
  performance, economic indicators, energy
  requirements, medical outcomes, chemical reaction
  products, weather, crop forecasts, environmental
  risk, horse races, jury panels.

8
Modeling

• Process control, systems control, chemical
  structures, dynamic systems, signal compression,
  plastics molding, welding control, robot control,
  and many more.

9
Neural Networks at work an example

• This simulation will take you inside a neural
  network, so you can get a good overview of how
  neural networks are constructed internally