kadie@m.cs.uiuc.edu (03/10/89)
In article joe@amos.ling.ucsd.edu (Fellow Sufferer) writes: > > Hecht-Nielsen Corp of San Diego, Ca is doing just such research. > > It seems they've had some success, too. The neural net was quite good at > predicting loan reliabilty. > > Their real problem was explaining why an applicant was refused: evidently > there is a law which requires that the institution explain exatactly what > was wrong with an application. > > That's not quite as easy as it sounds. > There is another potential problem, even if an explanation is found, it may be illegal. For example, the ANN may be very sensitive to zipcode. This is called redlining; in many places it is illegal. - Carl
c289-cf@seymour.Berkeley.EDU (Mr. AI) (03/12/89)
In article <37300001@m.cs.uiuc.edu> kadie@m.cs.uiuc.edu writes: >In article joe@amos.ling.ucsd.edu (Fellow Sufferer) writes: >> Their real problem was explaining why an applicant was refused: evidently >> there is a law which requires that the institution explain exatactly what >> was wrong with an application. That's not quite as easy as it sounds. > > There is another potential problem, even if an explanation is found, it may > be illegal. > For example, the ANN may be very sensitive to zipcode. This is called > redlining; in many places it is illegal. I don't see any problem with using a NN to judge loans. Can't factors that may be illegal to consider just not be used as input to the network? If zip codes cannot be used don't make it part of the input. The same for race or sex or whatever. Also, if it is required that an explaination be gives as to why the individual was turned doewn, I still see no trouble. Loan application along with most decision processes require complex weighing of factors. Therefore, I see nothing wrong with supplying a list of the most harmful factors to his loan acceptance. People may want a simple answer to why their loan was rejected, but there probably isn't one, because several factors of varying importance were the causes. Maybe a list giving the factors that if changed would result in a loan acceptance, such as: Loan Would Be Accepted If.. 20% higher income or 7% higher asset values or 10% higher income and 5% high asset values ... ... These nubers could be obtained by tweeking the input values. ______________ __ ____ _ ____ ___ _______________ =========/ /_||___//_\\___ \___\ \=========== =========\___/ || \/ \___\ ___\\__\========== =====Roger David Carasso. Computer Science.=====