Enhancing verification in mobile devices using model based on user interaction history

Patent metadata

Field	Value
Publication number	US-11880439-B2
Application number	US-202117304184-A
Country	US
Kind code	B2
Filing date	Jun 16, 2021
Priority date	Jun 16, 2021
Publication date	Jan 23, 2024
Grant date	Jan 23, 2024

Abstract

User verification using a mobile interaction can include receiving interaction data associated with a user of an application operating on a mobile device, the interaction data is generated during an interactive session of the user with the application on the mobile device. A plurality of features are extracted from the interaction data and one or more feature vectors are generated from the plurality of features. The plurality of features are aggregated to the one or more feature vectors and embedded within each feature vector. The embedded plurality of features are then projected to a global feature space by comparing a history of interactive sessions associated with the user and a history of interactive sessions associated with a plurality of different users for generating a global training dataset. Finally, a verification model is generated as a global binary classification model using the global training dataset.

First claim

What is claimed is:

1. A computer-implemented method for user verification, comprising:
receiving, by one or more processors, interaction data associated with a user of an application operating on a mobile device, the interaction data being generated during an interactive session of the user with the application on the mobile device;
extracting, by the one or more processors, a plurality of features from the interaction data;
generating, by the one or more processors, one or more feature vectors;
aggregating, by the one or more processors, the plurality of features of the one or more feature vectors;
embedding, by the one or more processors, the plurality of features within each feature vector and projecting the embedded plurality of features within each feature vector to a global feature space by comparing a history of interactive sessions associated with the user and a history of interactive sessions associated with a plurality of different users for generating a global training dataset; and
generating, by the one or more processors, a verification model as a global binary classification model using the global training dataset.

2. The computer-implemented method of claim 1 , further comprising:
continuously verifying, by the one or more processors, an identity of one or more users using the verification model.

3. The computer-implemented method of claim 1 , wherein embedding the plurality of features within each feature vector further comprises:
comparing, by the one or more processors, the interaction data associated with the interactive session of the user against a database of historical interactive sessions associated with the user; and
projecting, by the one or more processors, the interaction data associated with the interactive session of the user into the global feature space with positive labels.

4. The computer-implemented method of claim 3 , further comprising:
comparing, by the one or more processors, the interaction data associated with the interactive session of the user against a database of historical interactive sessions associated with the plurality of different users; and
projecting, by the one or more processors, data from each interactive session associated with the user together with the database of historical sessions associated with the plurality of different users into the global feature space with negative labels, wherein additional aggregative calculations can be performed on the global feature space to generate a transformed feature space.

5. The computer-implemented method of claim 1 , wherein the interaction data comprises a plurality of touch events including at least one of a swipe movement and a press movement performed by the user on the mobile device.

6. The computer-implemented method of claim 5 , further comprising:
for each of the plurality of touch events, extracting, by the one or more processors, at least one of raw X, Y coordinates together with a timestamp, a size of a user's finger, and a pressure exerted by the user's finger on the mobile device.

7. The computer-implemented method of claim 1 , wherein the plurality of features extracted from the interaction data comprises: a swipe direction, a gradient, an Euclidean distance, an average velocity, a maximum acceleration, and a curvature.

8. The computer-implemented method of claim 1 , further comprising:
dividing, by the one or more processors, the verification model into a finite number of models based on additional system related characteristics.

9. The computer-implemented method of claim 1 , further comprising:
dividing, by the one or more processors, the verification model into a plurality of semi-global verification models by:
clustering a plurality of users into k clusters based on behavior characteristics using a clustering technique; and
building a semi-global verification model for each of the k clusters.

10. A computer system for user verification, comprising:
one or more processors, one or more computer-readable memories, one or more computer-readable tangible storage devices, and program instructions stored on at least one of the one or more storage devices for execution by at least one of the one or more processors via at least one of the one or more memories, wherein the computer system is capable of performing a method comprising:
receiving, by one or more processors, interaction data associated with a user of an application operating on a mobile device, the interaction data being generated during an interactive session of the user with the application on the mobile device;
extracting, by the one or more processors, a plurality of features from the interaction data;
generating, by the one or more processors, one or more feature vectors;

aggregating, by the one or more processors, the plurality of features of the one or more feature vectors;
embedding, by the one or more processors, the plurality of features within each feature vector and projecting the embedded plurality of features within each feature vector to a global feature space by comparing a history of interactive sessions associated with the user and a history of interactive sessions associated with a plurality of different users for generating a global training dataset; and
generating, by the one or more processors, a verification model as a global binary classification model using the global training dataset.

11. The computer system of claim 10 , further comprising:
continuously verifying, by the one or more processors, an identity of one or more users using the trained verification model.

12. The computer system of claim 10 , wherein embedding the plurality of features within each feature vector further comprises:
comparing, by the one or more processors, the interaction data associated with the interactive session of the user against a database of historical interactive sessions associated with the user; and
projecting, by the one or more processors, the interaction data associated with the interactive session of the user into the global feature space with positive labels.

13. The computer system of claim 12 , further comprising:
comparing, by the one or more processors, the interaction data associated with the interactive session of the user against a database of historical interactive sessions associated with the plurality of different users; and
projecting, by the one or more processors, data from each interactive session associated with the user together with the database of historical sessions associated with the plurality of different users into the global feature space with negative labels, wherein additional aggregative calculations can be performed on the global feature space to generate a transformed feature space.

14. The computer system of claim 10 , wherein the interaction data comprises a plurality of touch events including at least one of a swipe movement and a press movement performed by the user on the mobile device.

15. The computer system of claim 14 , further comprising:
for each of the plurality of touch events, extracting, by the one or more processors, at least one of raw X, Y coordinates together with a timestamp, a size of a user's finger, and a pressure exerted by the user's finger on the mobile device.

16. The computer system of claim 10 , wherein the plurality of features extracted from the interaction data comprises: a swipe direction, a gradient, an Euclidean distance, an average velocity, a maximum acceleration, and a curvature.

17. The computer system of claim 10 , further comprising:
dividing, by the one or more processors, the v

Assignees

• IBM

Inventors

• Levi Matan
• Hazan Itay

Classifications

• G06F21/316 Primary

  by observing the pattern of computer usage, e.g. typical user behaviour · CPC title

• G06F16/285

  Clustering or classification · CPC title

• G06F2221/2139

  Recurrent verification · CPC title