Banks Fight Back: Fraud Database Allows Bankers to Cut Down on Losses
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
Skimming, pharming, phishing, check-kiting ... bankers have seen them all. And they all cost money, both to bankers and their customers. According to the 2004 ABA Deposit Account Fraud Survey Report, attempted check fraud alone cost the nation's banks $5.5 billion in 2003. With individuals becoming more techno-savvy, banks are searching for ways to stay one step ahead of the fraudsters. Fortunately, bankers are not without resources to help them combat the growing threat. Fraud-Net, a comprehensive database started in April 2002 by the Florida Bankers Association, allows bankers and law enforcement agencies to post information regarding crimes and fraudulent customers. The FBA holds the patent on Fraud-Net, and subscribers to the service can research local, statewide, and national postings and have new postings sent to their e-mail accounts. Postings can include anything from a suspect's aliases and Social Security number to photographs and information on other banks that have been hit by the same individuals or similar crimes. Thomas Kerr, senior vice-president and CFO for the FBA says the program grew out of a non-credit-loss task force meeting in 2000 dealing with fraud and identity theft. In Florida, banks write off an estimated $100 million in fraud related losses each year, and the FBA wanted to help prevent losses from occurring. According to Kerr, in 2002, 600 banks in Florida signed up for the program. Today that number has grown to 1,500 in Florida alone. The program became so successful in Florida that other states began to request access to the database. We never intended this to be a national database but that in fact has been created, says Kerr. Now 22 states are involved. The program is sponsored by state bankers associations, who often allow their members to access the information for free. Using Fraud-Net, the Secret Service, FBI, and local law enforcement agencies have all worked together to help bankers track down individuals, particularly those who commit a string of crimes. One of the advantages of Fraud-Net is that it detects patterns of check fraud. One case of a bad check may not garner much attention from law enforcement, but a string of bad checks passed at several institutions by the same person or groups of people around the state enables law enforcement to link cases together that before may have seemed unrelated. The real benefit for law enforcement agencies, says Kerr, is that they can take cold cases and turn them into workable cases again. Each bank has different pieces of information, says Kerr. Collectively we have enough information to give law enforcement something to work with. David Taxdal, assistant vice-president and security representative for Florida at Gold Bank in Bradenton, can vouch for this assessment. In August of 2004 he discovered through a Fraud-Net posting that several banks in Arkansas received counterfeit Gold Bank checks. Taxdal contacted the banks, and after working with Immigrations Customs Enforcement, discovered that the checks were coming out of Canada as part of a Nigerian lottery seam. Such collaborative efforts also helped Melody Shimmel, certified fraud examiner in charge of risk management and fraud at Century Bank in Sarasota, Fla. Shimmel, who checks Fraud-Net five to six times a day, says that the service is a boon to community banks. It really has cut down on our losses, she says. The program has enabled Shimmel and local law enforcement to solve a variety of cases affecting the ten-branch, $600 million assets bank. In August 2005 a posting appeared on Fraud-Net stating that a mail theft ring was stealing outgoing mail from residents' mailboxes in Manatee and Sarasota Counties. The individuals would take the checks they found in the mail, pretend to be bank customers and then cash the checks. Shimmel sent out a notice to her employees, and soon after a teller from the Bradenton branch called alerting her that the individuals were in the drive-up line of the bank. …
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.004 | 0.005 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it