Neural
network technology has been successfully applied to a
wide range of real-world applications, such as
- Cost Management
- Quality Control
- Signal Processing
|
- Portfolio Management
- Targeted Marketing
- and many more.....
|
Below you will find brief descriptions
of a few of these successful neural network applications.
These summaries are intended to give you a feel for the
types of problems that are being solved by various
leading-edge companies.
Signal Processing
Yamaha Corp. has developed a more versatile
sound-synthesis system by tapping neural network
technology. Unlike conventional digital signal processing
methods, the new system can predict a signal's waveform
even before its full cycle of data is known. This
capability is expected to accelerate the
information-processing speed of guitar waveform-synthesis
components. Yamaha's G50 pitch detector for electric
guitars uses neural learning algorithms to reliably
predict the pitch of a signal from the characteristic
shape of its beginning transient. When the neural network
was trained on thousands of guitar sound samples, it
learned to detect frequency within a half cycle four
times faster than conventional systems, which need two
full cycles to reliably detect pitch.
Portfolio Management
Several well-known financial organizations have
disclosed where they are applying neural computing
effectively. For example, Deere & Company's pension
fund has been managing a portfolio of $100 million since
December 1993 using neural networks. The fund monitors a
pool of 1,000 US stocks on a weekly basis. For each of
these stocks there is a neural network which models the
future performance of the stock as a function of the
stock's exposure to 40 fundamental and technical factors,
and gives an estimate of its weekly price change. The
company then selects a portfolio of the top 100 stocks
and allocates the fund proportionately to predicted
returns. Its annual return has been well in excess of
industry benchmarks.
Targeted Marketing
Neural networks are increasingly being used in
the marketing profession to enhance bottom line sales
figures. A marketing manager first takes the demographic
data from a previous campaign to train the neural
network. Then the data for a new set of prospective
customers is fed into the network. Those customers
exhibiting the attributes that the network learned to
associate with purchasing earn higher scores. The highest
scoring prospective consumers will be the first ones
targeted.
Economic Forecasting
Neural networks are being used by various
economists to forecast economic indicators, such as the
U.S. Index of Industrial Production. In many cases, these
forecasts are then used as indicators for predicting
future stock market performance.
Credit Rating
Neural networks have been used to analyze the
credit worthiness of loan applicants.
Speech Recognition
Several of the products available for
translating speech to text use a neural network as the
classification engine.
Medical Diagnosis
The use of neural networks to detect cancerous
cells from digitized images of tumor slides is a very
active research topic.
Fraud Detection
Many banks have used neural network technology
to detect fraudulent credit card applications and
transactions.
Optical Character Recognition
(OCR)
The OmniPage ProTM7.0 OCR software product by
Caere Corp. uses a neural network to dramatically
increase its recognition accuracy and performance.
Target Recognition
The U.S. Military has funded various research
efforts to apply neural networks towards the automatic
recognition of enemy targets from image data.
Sales Forecasting
One of the UK's leading soft drink manufacturers
(Brivtic) used a neural network to forecast sales in an
effort to prevent either overproduction or
underproduction.
Control
Neural control systems have been developed to do
everything from driving trucks to flying planes.
Machine Diagnostics
Systems have been developed that detect abnormal
vibrations in rotating machinery so that the machines
will be shut down before damage can occur.
Resource Allocation
Neural networks have been used to determine the
lead time for manufacturing products as well as to
schedule plants to minimize the lead time.
Intelligent Searching
A few WWW search engines include neural network
sub-systems, which provide the most relevant content and
banner ads based on the users' past behavior.
Image Processing
A security system has been developed that uses
neural technology to recognize a person's face to grant
them access to a secured area.
Other Customer Applications
|
