Socure has filed a patent for a methodology and system that counters fraudulent document and image use for authentication purposes. The method involves receiving a selfie or document expression from an individual, evaluating the authenticity of the provided proof of identity, and determining the probability of fraud in the transaction. Machine learning is used to adapt the methodology, and the system can be tuned based on ranking insufficiencies to improve authentication results. GlobalData’s report on Socure gives a 360-degree view of the company including its patenting strategy. Buy the report here.
Smarter leaders trust GlobalData
Access deeper industry intelligence
Experience unmatched clarity with a single platform that combines unique data, AI, and human expertise.
According to GlobalData’s company profile on Socure, Content access control was a key innovation area identified from patents. Socure's grant share as of September 2023 was 52%. Grant share is based on the ratio of number of grants to total number of patents.
Method and system for countering fraudulent document and image use
A recently filed patent (Publication Number: US20230230088A1) describes a method and computing system for verifying the identity of an individual during a transaction. The method involves receiving a selfie of the individual or a document expression containing a headshot and identity information. An identity verifier then evaluates the authenticity of the selfie and/or document expression by cross-checking it against a known standard for the presented document. Based on this evaluation, an authentication result is determined, indicating the probability of fraud in the transaction.
To achieve this, a machine learning model called the Identity Verification Predictor (IVP) is trained on a first training set consisting of training selfies and training document expressions. Each element in the training set is initially paired with a predetermined fraud weighting, indicating the probability of fraud associated with its use. The evaluation of the selfie and/or document expression is converted into input for the IVP, which then produces the authentication result based on rankings and fraud risk weightings assigned to the elements.
If the IVP identifies a ranking insufficiency among certain elements, the system re-trains the IVP by tuning the assigned fraud risk weightings. The authentication result is then verified based on this retraining, resulting in an adjusted ordering of the elements in the hierarchy.
The evaluation of the selfie includes determining liveness and/or spoofing, age verification, and checking if the image appears in data stores associated with fraudulent activity. The evaluation of the presented document involves cross-checking the identity information with data stores to verify its accuracy and comparing the headshot with mathematical representations in the data stores.
The computing system for implementing this method includes one or more processors and memories storing instructions. The system receives the selfie or document expression, performs the evaluation, and identifies the authentication result using the IVP. It also re-trains the IVP and verifies the authentication result based on the retraining.
Overall, this patent describes a method and computing system that utilize machine learning and cross-checking techniques to verify the identity of an individual during a transaction, reducing the risk of fraudulent activity.
To know more about GlobalData’s detailed insights on Socure, buy the report here.
Data Insights
From
The gold standard of business intelligence.
Blending expert knowledge with cutting-edge technology, GlobalData’s unrivalled proprietary data will enable you to decode what’s happening in your market. You can make better informed decisions and gain a future-proof advantage over your competitors.
GlobalData, the leading provider of industry intelligence, provided the underlying data, research, and analysis used to produce this article.
GlobalData Patent Analytics tracks bibliographic data, legal events data, point in time patent ownerships, and backward and forward citations from global patenting offices. Textual analysis and official patent classifications are used to group patents into key thematic areas and link them to specific companies across the world’s largest industries.
