Loading presentation...

Present Remotely

Send the link below via email or IM


Present to your audience

Start remote presentation

  • Invited audience members will follow you as you navigate and present
  • People invited to a presentation do not need a Prezi account
  • This link expires 10 minutes after you close the presentation
  • A maximum of 30 users can follow your presentation
  • Learn more about this feature in our knowledge base article

Do you really want to delete this prezi?

Neither you, nor the coeditors you shared it with will be able to recover it again.


Using Neural Networks for data analysis

Case Study 2: Using Neural Networks for Chase Bank

Mukta Misal

on 12 November 2013

Comments (0)

Please log in to add your comment.

Report abuse

Transcript of Using Neural Networks for data analysis

What is Neural Network?
The Development of Neural Network
Artificial Neural Network (ANN) is a system loosely modeled on the human brain.

It is an attempt to simulate within specialized hardware or sophisticated software, the multiple layers of simple processing elements called neurons.

Each neuron is linked to certain of its neighbors with varying coefficients of connectivity that represent the strengths of these connections.

(Daniel Klerfors, November 1998)

McCulloch and Pitts introduced the first neural network computing model in the early 1940's.

In the 1950's, Rosenblatt's work resulted in a two-layer network, the perceptron, which was capable of learning certain classifications by adjusting connection weights.

In the early 1980's, researchers showed renewed interest in neural networks. Recent work includes Boltzmann machines, Hopfield nets, competitive learning models, multilayer networks, and adaptive resonance theory models.
(Wikipedia, 8 November 2013)
How Neural Netwok Works
(Oolution Technologies, 14 Jan 2013)
Use of Neural Network
Neural Network at Chase Manhattan Bank
A statistical-based hybrid neural network at Chase Manhattan Bank is one of the largest and most successful AI applications in the United States.

It addresses a critical success factor in the bank's strategic plan: reducing losses on loans made to public and private corporations.

Most of Chase's business for corporations involves assessing their creditworthiness.
(NIBS Inc., n.d.)
The Problem
In 1985 Chase began a search for new quantitative techniques to assist senior loan officers in forecasting the creditworthiness of corporate loan candidates.
Another problem was to figure out which Asian stock market has shown the least volatility.

Consequently, Chase established a 36-member internal consulting organization called Chase Financial Technologies to oversee the development of pattern-analysis network models for evaluating corporate loan risk.

The resulting models, called the Creditview system, perform three-year forecasts that indicate the likelihood of a company being assigned a Chase risk classification of good, criticized, or charged-off.
Benefits of Creditview
Creditview provides a detailed listing of the items that significantly contributed to the forecast, an expert-system-generated interpretation of those items, and several comparison reports.
Creditview models run on a Chase Financial Technologies host computer. A user system resides at each user's PC and communicates with the host through telephone lines.

Still neural networks have their flaws
For neural networks to perform most effectively, they must constantly be fed updated information so they can uncover new trends in the market.
There are some things, like emotions, that the systems can't predict.

John Raasch of Hamilton Investments said he got a buy signal when U.S. Surgical hit $46 recently. But the news the next day that the health-care company had poor sales pushed the stock down 33 percent in one day. "The institutions panicked," said Raasch. "There was nothing we could do."

(Bloomberg Business News, May 02 1993)

Neural Networks for Data Analysis
Future Trends
Neural networks will allow us to explore new realms of human capability realms previously available only with extensive training and personal discipline.
Neural Networks will fascinate user-specific systems for education, information processing, and entertainment.

(Marcus D. Odom, Ramesh Sharda, n.d.)
The computing world has a lot to gain from NN's. Their ability to learn by example makes them very flexible and powerful.

They are also very well suited for real time systems because of their fast response and computational times which are due to their parallel architecture.

Neural networks also contribute to other areas of research such as neurology and psychology.

Finally, I would like to state that neural networks have a huge potential, we will only get the best of them when they are integrated with Computing AI and related subjects.

Input Layer
Output Layer
Hidden Layers
1. Artificial Neural Networks, Daniel Klerfors (November 1998)
2. Wikipedia (8 November 2013)
Retrieved from http ://en.wikipedia.org/wiki/Artificial_neural_network
3. NewWave Intelligent Business Systems, NIBS Inc. (n.d.)
Retrieved from http:// Sharda (n.d.)ulcar.uml.edu/~iag/CS/intronn.htm
4. Bloomberg Business News (May 02, 1993)
Retrieved from http ://articles.chicagotribune.com/1993-05-02/business/9305040156_1_neural-networks-factors-and-stock-price-lbs-capital-management
5. A Neural Network Model, Marcus D. Odom, Ramesh
ANNs are used for interpretation, prediction, data analysis, data filtering, diagnosis and many more.
The most successful applications of neural networks are in categorization and pattern recognition (Wikipedia)
One of the best known applications is the bomb detector installed in some U.S. airports.
(Daniel Klerfors, November 1998)

Case Study 2 by Mukta Misal
"Compared to human traders, (neural networks) are not nearly as brilliant, but they are a lot more consistent," said Thomas Young, a currency trader with A.G. Edwards in Plano, Texas, who uses neural networks.
To solve these problems, Neural networks with Adaptive Decision Systems has been applied.
Chase located Inductive Inference Inc. (headed by Dr. David Rothenberg), a New York City company with a history of successfully applying neural-network technology to statistical pattern analysis.
A test model was built, evaluated, and independently audited.

Lessons Learned
Some of the recent improvements done in the field like : exploration of neuromodulators etc. helped in better understanding the application to data analysis.
The analysis of neural networks was restricted to text data but now a days with the emerge of big data, a combination of text, voice and image processing is being handled by neural networks the study for the same could have been done.
The large and diversified usage of neural networks discovered during the course of research was not anticipated and this gave the increase opportunity to learn.
Critical Review
Conventional financial statement analysis is also performed using Chase's Financial Reporting System, an independent financial spreading and analysis package designed for conventional financial statement analysis.

This helped Chase to save resources in terms of money and time.
Full transcript