To provide affordable and quality software that can analyze the legitimacy of online reviews
We want to act as a beacon for businesses with real reviews by providing reassurances that they are a high-quality business worthy of their customer's hard-earned money.
We work hard to bring a new element of trust back to the online review marketplace.
Curtis Boyd is the founder of The Transparency Company. His expertise in Online Reputation Management gave The Transparency Company the direction for the technology.
Sr. DevOps Engineer
Roman is the right hand man behind the scenes, making the API's work and driving functionality to a very large data pipeline. He's more than a full-stack developer, he's family.
Rashid is one of the hardest working individuals, who refuses to say, " I can't do that." His commitment to his projects has been invaluable to our team.
What is our workflow?
Hybrid Data Mining/API’s
We collect all the information we can from: 1. Review Content 2. Reviewer Profiles 3. Business Profiles
The aggregated data is analyzed by our server and evaluated based on pre-trained AI Models.
We build profiles
Based on the unique scores of each reviewer, we profile them based on known fraudulent profiles.
Profile data includes the behavioural data of a reviewer. The businesses they have reviewed, the types of businesses, the types of reviews they leave.
Distance Matrix Analysis
We look at each reviewer individually and look at all the businesses they have reviewed, then calculate the distances between all of those businesses.This helps us to understand a group of reviewers of a particular business, particularly about where they live and write reviews.
Reviewer Pod Analysis
The review pod analysis measures if there are large groups of reviewers who have reviewed the same businesses. It looks a lot like a private reviewer circle. This is often a sign of someone being paid or incentivized to publish reviews.
Reviewer Image Analysis
Every profile belongs to a unique human being, with unique access to images and where they source their profile pictures. Often times, review spammers will steal photos from celebrities from instagram and facebook. This is why we check reviewer images to understand a bit about the profile.
Content Metrics are created from the content of the review itself. Not all reviews are created equal, and we look at them closely.
Natural Language Processing
As human beings we all have our own unique authorship style, the way we write sentences, ect. We use machine learning to identify unique authorship styles. This is helpful when looking for reviews written by the same author.
Fake reviews are generally higher in sentiment than real reviews, which is why we look at sentiment to make a prediction whether a review is real or fake. We also look at use of exclamation marks and other indications that show over the top positive sentiment.
We read review content and measure keyword density. This way we can understand if the business is stuffing keywords into the reviews in order to rank better for specific industry-related keywords. This is absolutely a sign of review fraud.
See our technology in action
Consumers are 15x more likely to have a negative experience from a company with fake positive reviews.