360

Define the terms false match rate and false non-match rate, and explain the use of a threshold in relationship to these two rates. Justify your answers. Provide examples to support your response. Specifically, think of and give a real-life scenario portraying the following concepts:

1. False match rate
2. False non-match-rate

Be sure to post an initial, substantive response by Thursday at 11:59 p.m. MST and respond to 2 or more peers with substantive responses by Sunday at 11:59 p.m. MST. Be sure you write a substantive initial post, answer the question presented completely and/or ask a thoughtful question pertaining to the topic. Also ensure your substantive peer responses has a thoughtful question pertaining to the topic and/or has answers to a question (in detail) posted by another student or the instructor.

Please just answer responses

Greetings class,

A false match rate measures the percent of invalid inputs which are accepted when they should not be, while a false non-match rate measures the percent of inputs that were valid but were supposed to be rejected and were not. I never learned how to do this from a software perspective, but I am aware that adjusting the threshold may alter the likelihood that the false match rate will decrease and the false non-match rate may increase.

When it comes to biometric scanning, it isn’t 100% accurate all of the time. IT is a fickle creature in general, but biometric scans are a bit unique. In my experience, I think modern biometrics work well for the most part. These days we have devices like smart tablets and our technology is typically tuned well out of the box. My experiences with biometric collection aren’t limited to what we have now though.

When I was in the Army, one of the areas of information collection I had extensive training in was biometric data collection. This training didn’t involve coding or really trying to understand the underlying concepts, but rather tactical data collection as the data collector on a given site. This involved using a SEEK II built by a company called CrossMatch, first introduced in 2007. The SEEK II has a optical scanner, fingerprint reader, facial scanner/camera combo, and proprietary personnel enrollment system (think of a personal profile). From my understanding, SEEK II’s weren’t changed all that much from their original model, which by the time I was using them in Afghanistan in 2015, they were far out of date. The SEEK II was definitely a product of its time, and I remember it had somewhere around 30 gig of storage space in an SSD but ran on Windows XP and had maybe a single gig of RAM. The whole device was slow to boot and could have errors trying to collect data, especially trying to collect finger prints. I’d say about 80% of the time the SEEK II captured biometric data correctly and as intended, but I definitely remember times where the data would just not be recognized.

In 2007, we still had Windows Vista and about midway in the year the first generation of Apple’s iPhone was released to the general public. So by 2007 standards, SEEK II devices were behind the curve technologically but were also a niche device; they were clunky and slow compared to other technology, but they got the job done. I think the biggest flaw with the SEEK II was being ran on Windows XP. Hindsight is super vision, and I just don’t think that version of Windows would have been best. I think a Linux OS would have been a better pick, especially with how many distros were keeping up with changes in IT and XP wasn’t as powerful in 2007 as some of its contemporaries (despite still being widely used, I was using Windows XP on my family computer well into 2009).

response 2

Hello Class,

When it comes to any form of Biometric security there are two possibilities that could happen other than it working as it should. You have a False match rate which means there is a chance that two different people’s data could be confused for being the same by the service allowing them access as the correct person. Then there is also a false non-match-rate which equates to the service seeing data from the same person but do to day to day changes a person can go through it may flag as incorrect credentials.

Though I do not have a lot of experience with biometrics when it comes to professional life like many others, I use biometrics for my daily use. I do feel that fingerprint sensors are secure which is in line with companies such as Google and apple. This is due to technology using multiple forms of optical technology to check fingerprints, but I feel like facial has false and is more susceptible to these forms of issues. This is due to cameras sometimes having issues decerning features in low light levels. This could lead to false non-matches since the device is not able to