Method, device, and system of differentiating among users based on platform configurations
Devices, systems, and methods of detecting user identity, differentiating between users of a computerized service, and detecting a cyber-attacker. An end-user device interacts and communicates with a server of a computerized server (a banking website, an electronic commerce website, or the like). The interactions are monitored, tracked and logged. User Interface (UI) interferences or irregularities are introduced; and the server tracks the response or the reaction of the end-user to such interferences. The system determines whether the user is a legitimate user, or a cyber-attacker or automated script posing as the legitimate user. The system utilizes classification of users into classes or groups, to deduce or predict how a group-member would behave when accessing the service through a different type of device. The system identifies user-specific traits that are platform-independent and thus can be further monitored when the user switches from a first platform to a second platform.
1. A method comprising:
(A) in a first usage session of a user that utilizes an electronic device to interact with a computerized service:
(A1) monitoring user interactions with an input unit of said electronic device;
(A2) extracting from said user interactions, a first user-specific characteristic;
determining that said first user-specific characteristic is a cross-platform user-specific characteristic that is expected to characterize said user if said user switches to utilizing a subsequent, different, electronic device to access said computerized service; and
assigning a first weight value to the first user-specific characteristic;
(A3) extracting from said user interactions, a second user-specific characteristic;
determining that said first user-specific characteristic is not a cross-platform user-specific characteristic that is expected to characterize said user if said user switches to utilizing a subsequent, different, electronic device to access said computerized service; and
assigning a second, smaller, weight value to the second user-specific characteristic;
(B) in a second, subsequent, usage session of said user that utilizes said subsequent electronic device to interact with computerized service:
(B1) monitoring user interactions with an input unit of said subsequent electronic device;
(B2) differentiating between: (i) said user, and (ii) other users of said computerized service, by analyzing said user interactions with the input unit of the subsequent electronic device through a weighted formula that takes into account, at least, said first weight value for the first user-specific characteristic and said second weight value for the second user-specific characteristic.
2. The method of claim 1 , wherein the checking comprises:
if it is determined that said user is utilizing in said subsequent session, a computing device that belongs to said first type of computing device, then: determining whether user interactions of said user during the subsequent usage session of said user, match the first cross-platform user-specific trait.
3. The method of claim 1 , wherein the checking comprises:
if it is determined that said user is utilizing in said subsequent session, a computing device that belongs to said first type of computing device, then:
comparing user interactions of said user during said subsequent usage session, against a group of user-specific traits that includes the first cross-platform user-specific trait and that excludes the second cross-platform user-specific trait.
4. The method of claim 1 , wherein the method comprises:
identifying the first cross-platform user-specific trait as a first trait which characterizes usage sessions in which said user interacted with the computerized service via any portable computing device selected from the group consisting of smartphone and tablet;
identifying the second cross-platform user-specific trait as a second trait which characterizes usage sessions in which said user interacted with the computerized service via any non-portable computing device;
in a subsequent usage session that is assumed to belong to said user, checking whether user interactions during said subsequent usage session match only one of said first or second traits, based on a determination of whether said user utilizes either a portable or a non-portable computing device to access the computerized service during the subsequent usage session.
5. The method of claim 1 , wherein the method comprises:
identifying the first cross-platform user-specific trait as a first trait which characterizes usage sessions in which said user interacted with the computerized service via any non-portable computing device selected from the group consisting of laptop computer and desktop computer;
identifying the second cross-platform user-specific trait as a second trait which characterizes usage sessions in which said user interacted with the computerized service via any portable computing device;
in a subsequent usage session that is assumed to belong to said user, checking whether user interactions during said subsequent usage session match only one of said first or second traits, based on a determination of whether said user utilizes either a portable or a non-portable computing device to access the computerized service during the subsequent usage session.
6. The method of claim 1 , wherein the method comprises:
identifying the first cross-platform user-specific trait as a first trait which characterizes usage sessions in which said user interacted with the computerized service via any computing device having a touch-screen;
identifying the second cross-platform user-specific trait as a second trait which characterizes usage sessions in which said user interacted with the computerized service via any computing device having a non-touch screen;
in a subsequent usage session that is assumed to belong to said user, checking whether user interactions during said subsequent usage session match only one of said first or second traits, based on a determination of whether said user utilizes either a touch-screen computing device or a non-touch-screen computing device to access the computerized service during the subsequent usage session.
7. The method of claim 1 , wherein the method comprises:
identifying the first cross-platform user-specific trait as a first trait which characterizes usage sessions in which said user interacted with the computerized service via any computing device through any Web browser;
identifying the second cross-platform user-specific trait as a second trait which characterizes usage sessions in which said user interacted with the computerized service via any computing device through a non-browser native application of said computerized service;
in a subsequent usage session that is assumed to belong to said user, checking whether user interactions during said subsequent usage session match only one of said first or second traits, based on a determination of whether said user utilizes either a Web browser or a native application to access the computerized service during the subsequent usage session.
8. The method of claim 1 , comprising:
identifying a first cross-platform user-specific trait that characterizes usage sessions in which said user interacted with the computerized service via a first type of software application;
identifying a second cross-platform user-specific trait that characterizes usage sessions in which said user interacted with the computerized service via a second type of software application;
in a subsequent usage session that is assumed to belong to said user, checking whether user interactions during said subsequent usage session match only one of said first or second cross-platform user-specific traits, based on a type of software application that is actually utilized by said user in said subsequent usage to access the computerized service.
9. The method of claim 1 , wherein the method comprises:
identifying the first cross-platform user-specific trait as a first trait which characterizes usage sessions in which said user interacted with the computerized service via any computing device through any Web browser;
identifying the second cross-platform user-specific trait as a second trait which characterizes usage sessions in which said user interacted with the computerized service via any computing device through a non-browser native application of said computerized service;
in a subsequent usage session that is assumed to belong to said user, checking whether user interactions during said subsequent usage session match only one of said first or second traits, based on a determination of whether said user utilizes either a Web browser or a native application to access the computerized service during the subsequent usage session.
10. The method of claim 1 , wherein the method comprises:
identifying the first cross-platform user-specific trait as a first trait which characterizes usage sessions in which said user interacted with the computerized service via both (A1) a first type of hardware configuration, and (B1) a first type of software configuration;
identifying the second cross-platform user-specific trait as a second trait which characterizes usage sessions in which said user interacted with the computerized service via both (A2) a second type of hardware configuration, and (B2) a second type of software configuration;
in a subsequent usage session that is assumed to belong to said user, checking whether user interactions during said subsequent usage session match only one of said first or second traits, based on a determination of whether said user utilizes either: (x) the first type of hardware configuration and also the first type of software configuration, or (y) the second type of hardware configuration and also the second type of software configuration.
11. The method of claim 1 , wherein the method comprises:
identifying the first cross-platform user-specific trait as a first trait which characterizes usage sessions in which said user interacted with the computerized service via both (A1) any non-portable computing device, and (B1) any Web browser;
identifying the second cross-platform user-specific trait as a second trait which characterizes usage sessions in which said user interacted with the computerized service via both (A2) any portable computing device, and (B2) a non-browser native application of said computerized service;
in a subsequent usage session that is assumed to belong to said user, checking whether user interactions during said subsequent usage session match only one of said first or second traits, based on a determination of whether said user utilizes either: (x) a non-portable computing device and also a Web browser, or (y) a portable computing device and also a non-browser native application of said computerized service.
12. The method of claim 1 , wherein the method comprises:
identifying the first cross-platform user-specific trait as a first trait which characterizes usage sessions in which said user interacted with the computerized service via both (A1) any non-portable computing device, and (B1) any Web browser;
identifying the second cross-platform user-specific trait as a second trait which characterizes usage sessions in which said user interacted with the computerized service via both (A2) any portable computing device, and (B2) any Web browser;
in a subsequent usage session that is assumed to belong to said user, checking whether user interactions during said subsequent usage session match only one of said first or second traits, based on a determination of whether said user utilizes either: (x) a non-portable computing device and also any Web browser, or (y) a portable computing device and also any Web browser; wherein said checking excludes checking of user interactions against the first trait and further excludes checking of user interactions against the second traits if it is determined that said user utilizes both a non-portable computing device and a non-browser native application to access the computerized service during said subsequent session.
13. The method of claim 1 , wherein the method comprises:
identifying user-specific behavioral characteristics of said user based on analysis of user interactions;
classifying said user-specific behavioral characteristics based on their dependency on the a combination of hardware configuration and software configuration that said user utilizes to access the computerized service.
14. The method of claim 1 , wherein the method comprises:
identifying user-specific cognitive characteristics of said user based on analysis of user interactions that indicate a cognitive trait of user-specific typical response time;
classifying said user-specific cognitive characteristics based on their dependency on the a combination of hardware configuration and software configuration that said user utilizes to access the computerized service.
15. The method of claim 1 , wherein the method comprises:
identifying user-specific cognitive characteristics of said user based on analysis of user interactions that indicate a cognitive trait of user-specific typical adaptation time to input/output anomalies;
classifying said user-specific cognitive characteristics based on their dependency on the a combination of hardware configuration and software configuration that said user utilizes to access the computerized service.
16. The method of claim 1 , wherein the method comprises:
identifying user-specific physiological characteristics of said user based on analysis of user interactions that indicate a physiological trait of the user that corresponds to whether said use utilizes right-hand or left-hand to operate the computing device when accessing the computerized service;
classifying said user-specific physiological characteristics based on their dependency on the a combination of hardware configuration and software configuration that said user utilizes to access the computerized service.
17. The method of claim 1 , wherein the method comprises:
identifying user-specific physiological characteristics of said user based on analysis of user interactions that indicate a physiological trait of the user that corresponds to curvature of an on-screen gesture that said user performs to operate the computing device when accessing the computerized service;
classifying said user-specific physiological characteristics based on their dependency on the a combination of hardware configuration and software configuration that said user utilizes to access the computerized service.
18. The method of claim 1 , wherein the method comprises:
applying a signal processing algorithm to extract from user interactions user-specific behavioral features that are not platform-dependent;
applying a feature selection algorithm to classify user-specific behavioral features, that are identified across multiple hardware configurations associated with said user, as belonging to a particular same type of computing platforms as a differentiator among users if it is determined that said user utilizes, in a subsequent usage session, a platform of said particular same type of computing platforms.
19. The method of claim 1 , wherein the method comprises:
applying a signal processing algorithm to extract from user interactions user-specific behavioral features that are not platform-dependent;
applying a feature selection algorithm to classify in a same group multiple user-specific behavioral features, which belong to a particular same type of computing platforms, if said multiple user-specific behavioral features are identified across multiple hardware configurations associated with said user;
applying a user differentiation algorithm that is based on multiple user-specific behavioral features that belong to said same group and that exclude other user-specific behavioral features that are platform-dependent.
20. The method of claim 1 , wherein the method comprises:
applying a signal processing algorithm to extract from user interactions user-specific behavioral features that are not platform-dependent;
applying a feature selection algorithm to classify in a same group multiple user-specific behavioral features, which were identified across both a tablet of said user and a smartphone of said user;
if it is determined that said user accesses the computerized device, in a subsequent usage session, via either a tablet or a smartphone, then applying a user differentiation algorithm that is based on multiple user-specific behavioral features that belong to said same group and that exclude other user-specific behavioral features that are platform-dependent.