How Does the TripAdvisor Review Tracking System Work?
Here we explain how the technology behind our review tracking system works, and how we use it to catch fake reviews and prevent them from reaching the site...
When someone submits a review on TripAdvisor, it goes through a series of checks before it is posted to the site. These checks are done automatically by our review tracking system, which analyzes hundreds of pieces of information.
The tracking system is our first line of defense against fake reviews, and it means we can analyze every review prior to it being posted on the site. Such a task would be impossible if we relied only on human moderation, because the volume of review submissions from the global travel community is so high — TripAdvisor receives hundreds of contributions every minute.
What pieces of information does the tracking system check?
Because reviews are submitted online, there are hundreds of pieces of electronic information we can analyze to help us understand more about every review’s origin and circumstance. For example, we can see the location of the device that was used to submit a review, as well as some details about the specifications of that device. Even if someone is trying to hide information about their device location, which fraudsters often do, we can spot that too.
That is just one example — there are lots of other pieces of information that we look out for. They act like pieces of a puzzle that our tracking system is designed to join together, creating a complete picture of each review.
Can TripAdvisor give other examples of what the tracking system checks?
In the same way a bank cannot share too much information about the security systems it uses, unfortunately we cannot share a full list of everything our system tracks. We would love to show off all of the things our technology can do, but we also have to be cautious not to provide any information that fraudsters might try to use to get around our systems.
What is the tracking system looking for?
Our tracking system is designed to do two things — block reviews that are clearly in breach of our guidelines, such as those that include offensive language or plagiarized content, and also spot unusual patterns that might indicate a review is biased or fake.
When our tracking system identifies a review that is clearly in breach of our guidelines, it rejects that review and blocks it from being posted to the site in the first place.
When our tracking system identifies an unusual pattern of review activity, it triggers an investigation by our team of moderators. While the investigation takes place, the reviews associated with the unusual activity are blocked from being posted to the site. You can learn more about the types of reviews that might be flagged for investigation here.
What does TripAdvisor mean by ‘unusual patterns’?
Our tracking system has been analyzing hundreds of millions of reviews for more than a decade. With all of this data, we know how most reviewers behave, and can spot patterns that don’t conform to that behavior.
The techniques we use are similar to how a bank catches credit card fraud. Banks use technology to track the transactions taking place on individual accounts. By analyzing these transactions, the technology can determine whether spending on the account seems normal, or whether there are transactions that appear suspicious.
Our system works in a similar way — we see the reviews that are coming in for any given property, compare those to the historical patterns we’ve observed for that property, and identify any anomalies in those patterns that look suspicious. We also analyze the patterns of individual reviewers to spot any anomalies in their behavior too — such as the location of the places they are reviewing, how frequently they are reviewing, and so on.
So even if a review might look legitimate on the surface, our system can spot certain characteristics that are suspicious and trigger an investigation by our expert team.
To learn more about the tools we use to spot unusual patterns, watch this video.
You can also read more about the different types of fake review and how we catch them here.
Last Updated: July 13, 2018