What are the real world implications of this project?

Slides:



Advertisements
Similar presentations
Understand business credit and risk management.
Advertisements

Carl Johnson Financial Literacy Jenks High School.
Requirements for Opening a Checking Account
MLC Online Applications Adviser Guide. 2 Introduction MLC Online Applications have been designed to provide you with an improved means of submitting client.
1 Simulation Modeling and Analysis Verification and Validation.
Opening and Using a Bank Account December 16 th, 2008.
Check Writing All About Checks. Check Writing Checks include the following information:
zipLogix ® Digital Ink Training Sending files to be signed using the new zipForm ® Plus Go Beyond.
Method for verifying a credit card number: Example:
A Neural Network Approach to Predicting Stock Performance John Piefer ECE/CS 539 Project Presentation.
2015 ANNUAL TRAINING By: Denise Goff
APPLICATION PROCESS, CREDIT BUREAU, CREDIT DOCUMENTS.
Richard M. Alderman University of Houston Law Center
Test Review Banking. 1 List the guidelines for selecting a PIN number. Don’t pick a number that anyone else could figure out.
STEPS TO SUCCESS. STEPS TO COMPLETE A TAX RETURN Taxpayer enters location with necessary forms and information to file their return … Income forms (W-2,1099’s.
Chapter 9 CHECKING ACCOUNTS. Financial institutions (like banks and credit unions) offer a number of services. Can you name a service?
Check Writing All About Checks. Check Writing Checks include the following information:
1 Payroll Schemes Chapter 6. 2 List and understand the three main categories of payroll fraud. Understand the relative cost and frequency of payroll frauds.
Chapter 6 Payroll Schemes.
An Artificial Neural Network Approach to Surface Waviness Prediction in Surface Finishing Process by Chi Ngo ECE/ME 539 Class Project.
Checking and Banking
Artificial Neural Network System to Predict Golf Score on the PGA Tour ECE 539 – Fall 2003 Final Project Robert Steffes ID:
Today’s Agenda ML Development Workflow –Emacs –Using use –The REPL More ML –Shadowing Variables –Debugging Tips –Boolean Operations –Comparison Operations.
1 Identity Theft Prevention and the Red Flag Rules.
What are different ways you can save and invest your money? SS.8.FL.5.1 Describe the differences among the different types of financial assets, including.
Chapter 7, standard 065: Prepare all forms necessary for opening and maintaining a checking and savings account.
STEPS TO SUCCESS.
How much do you know about the most common form of bank account?
Freedom Independence Transition
Check Writing All About Checks.
Handling Checks Module 2.
CHAPTER 23 Creating a Negotiable Instrument
MODEL 0. The Simple Circular Flow with CASH KEYNESIAN THEORY (KT)
Digital Signatures A digital signature is a protocol that produces the same effect as a real signature: It is a mark that only the sender can make but.
STEPS TO SUCCESS.
Check Writing All About Checks.
Open, Manage, and Reconcile
Chapter 10 Verification and Validation of Simulation Models
Where should I put my money?
Students: Meiling He Advisor: Prof. Brain Armstrong
Chapter 16: Introduction
CSSE463: Image Recognition Day 11
Introduction to Business
Check Writing All About Checks.
CS539: Project 3 Zach Pardos.
Fraud Prevention.
What is BankMobile? A process to select how to receive student refunds and student payroll payments It is fast, secure, and convenient. Go to:
Checking Accounts Chapter 23.2
Chapter 11 Practical Methodology
Check Writing All About Checks.
The Six Steps in New Product Development
CSSE463: Image Recognition Day 18
Check Writing All About Checks.
Check Writing All About Checks.
Year 3 (National Numeracy Strategy) (Based on DFEE Sample Lessons)
Chapter 16: Check Digit Systems, Continued
Saving at a depository institution Choosing Savings Tools
Computer Hardware Optimization
Method for verifying a credit card number:
Check Writing All About Checks.
Check Writing All About Checks.
Evaluating Classifiers
Check Writing All About Checks.
Viking Independent bank Mrs. sorrell
Check Writing All About Checks.
Section 11.1: Significance Tests: Basics
Check Writing Simulation
Check Writing All About Checks.
High School Financial Planning Program Use a checking account
Check Writing All About Checks.
Presentation transcript:

What are the real world implications of this project? Signatures are used to verify a person is who they say they are. Things like credit cards, checks use signatures to verify id Things like contracts use a signature to the contract legally binding Finding forges could save banks and credit card companies lots of money

Gathering the Data First: Sign my name into a computer 100 times in a 2x6 inch image real Then: Have 2 other people sign their name once, sign my name 4 times and then try to forge my signature while looking at my signature as a guide real forged

Do Dimension Reduction on the Inputs This makes it so we have a reasonable amount of inputs

With these parameters the MLP converges in < 300 Epochs ! Run the Inputs on an MLP Found optimal parameters are Alpha = .01 Mom = .8 Epochs = 1000 Epoch size = 64 Checks every 10 Epochs With these parameters the MLP converges in < 300 Epochs !

The MLP has lots of problems with this forged signature Results Valid Signature If the data is trained with one persons forges and tested with the other persons forges worst case scenario it is classifies 22% of the samples incorrectly The MLP has lots of problems with this forged signature Forged Signature

Conclusion The MLP preformed quite well compared to both of the baseline models. If we look at real world scenarios, having absolutely no data on the forger, then the model was accurate 80-90% of the time. What makes these results useful is that the MLP was 100% accurate when claiming a signature was forged. This means that if someone were to implement this style in a bank to check for forged bank checks, if it ever flagged a signature as being forged, it would be a good idea to double check to see if is a valid transaction since the MLP appears not to give false positives in this scenario. The toughest part of this would be generating good forged signatures to test over. Other then having a person forge a signature several times for input, the MLP will not work. The good news is that if one can get this data, the MLP seems to work to detect forges signed by someone else.