New Google-sponsored GSRank algorithm highlights opinion spam in app stores

If you browse the App Store or Google Play for more than a few minutes, you'll likely come across some rather suspect reviews.

It's not unusual to see a bunch of five-star reviews that have clearly been left by people tied to the game in question, followed by scathing opinions from actual players. There's also the issue of for-pay opinion spamming. Basically, people being paid to leave favourable reviews.

Thankfully, a new Google-sponsored study, entitled 'Spotting Fake Reviewer Groups in Consumer Reviews', by University of Illinois researchers could help to spot opinion spam using a new relation-based algorithm called GSRank (Group Spam Rank).

The algorithm uses numerous bits of information, including how closely together members post reviews, similarities in review content, and observed relationships between group spam and products.

Apple in particular has had trouble with people manipulating its App Store - several companies reportedly offer services (for a price) that guarantee your app will make it into the Apple's top 25 iTunes charts.

These companies recruit people to download a specified app during a given period of time, which propels it up the charts and into a position that's likely to attract real downloaders.

It looks like Apple and Google - at the very least - are taking notice of methods that will help them to combat both forms of manipulation.