How to prevent fraud?

[hadoob024]

I just joined this startup venture. It's basically a company that's trying to create a general ratings system for products (similar to Amazon, Epinions, etc.). You see a product online, highlight it, right-click and then you can rate the product or search on the product's ratings. They then plan on charging sites to display these ratings (among other revenue avenues).

The two issues I can see are:

1) How to make sure that ratings posted on a website are paid for by the website? If a company wants to post our ratings on their site (for marketing/bragging purposes), they must pay to do so. How can I make sure that they just don't use the information?


2) How to prevent marketing firms from posting fraudulent ratings for hire? If Company A's product is rated lowly by our users, how can we stop them from hiring someone (or some company) to make fraudulent ratings to boost the overall image of their product?

For the second problem, I was thinking of tying the ratings to IP addresses, maybe? Or maybe the MAC address instead? Anyone have any thoughts on either of these two issues?

[qjensen]

For issue #1, there is very little you can do. You can do some basic searching and look for sites that use your name or image and catch some of those using your information without permission, but you won't put much of a dent in it. Reputable companies won't do that, so you hope to come out on top in the long run.

For issue #2, you can do a variety of things. The IP address/time between votes is a good start. You can also use cookies, although tech savvy users can find a way around that. Having more than one method of handling this will make it harder to abuse. Use a captcha, rate limit by IP, use cookies and allow registered users with accounts to vote more frequently to encourage them to get an account and add another barrier to abuse. I believe Google even uses geotargeting to limit ad sense abuse.

Neither of these problems has an easy solution. That is why the big G is always changing its ranking algorithms and finding new ways to eliminate click fraud. Technology makes it easier for us to develop new ideas, but it also makes it easier for the bad guys to abuse the system.

Quint Jensen
Web Developer
Scaled Solutions (http://www.scaled-solutions.com)