US 12,406,280 B2
Method and system for hardware and software based user identification for advertisement fraud detection
Anuj Khanna Sohum, Singapore (SG); Charles Yong Jien Foong, Singapore (SG); and Anurag Singh, Gurgaon (IN)
Assigned to Affle (India) Limited, Mumbai (IN)
Filed by Affle (India) Limited, Mumbai (IN)
Filed on Apr. 30, 2019, as Appl. No. 16/399,747.
Claims priority of application No. 201821016233 (IN), filed on Apr. 30, 2018.
Prior Publication US 2019/0333102 A1, Oct. 31, 2019
Int. Cl. G06Q 30/0241 (2023.01); G06N 7/01 (2023.01)
CPC G06Q 30/0248 (2013.01) [G06N 7/01 (2023.01); G06Q 30/0277 (2013.01)] 16 Claims
OG exemplary drawing
 
1. A fraud-protection computer system comprising:
a plurality of devices, wherein each device among the plurality of devices comprises:
a plurality of components, wherein the plurality of components comprises a plurality of device sensors, a plurality of biometrics, and a plurality of device connective components;
a plurality of third party devices, wherein the plurality of third party devices is connected with each other and the plurality of devices; and
a fraud detection system comprising:
one or more processors; and
a memory coupled to the one or more processors, the memory for storing instructions which, when executed by the one or more processors, cause the one or more processors to perform a method for detection of online advertisement fraud and commercial fraud comprising:
collecting a first set of data from the plurality of components associated with each device of the plurality of devices in real-time;
receiving a second set of data associated with each device of a plurality of third party devices in real-time;
assigning each device of the plurality of third party devices with a unique identity;
creating a unique device profile for each device of the plurality of devices, wherein the unique device profile stores the corresponding first set of data collected from each device of the plurality of devices;
analyzing the first set of data and the second set of data to:
determine that a time period in which the first set of data collected from the plurality of device sensors is zero is longer than a predetermined time period;
determine a behavior of a plurality of users with a similar finger size, wherein the plurality of users are associated with the plurality of devices, wherein a touch sensor determines the finger size by determining the hardness of a press by a finger and the size of the area pressed, and wherein the behavior of the plurality of users is determined before and after installation of an application, and
wherein the analyzing is done using a correlation data, the correlation data comprising a positive correlation data and a negative correlation data, wherein the positive correlation data is based on pre-event and post-event data, and wherein the negative correlation data is based on non-human behavior data; and
detecting the online advertisement fraud and the commerce fraud based on a deviation in the determined behavior of the plurality of users before and after the installation of the application and based on the determination that the time period in which the first set of data collected from the plurality of device sensors is zero is longer than the threshold time period.