Abstract: Systems and methods are provided for detecting when a user uses one or both hands to interact with a device having a physical or virtual keyboard. In one implementation, one-handed typing is determined by analyzing intragroup and intergroup keyflight timing distributions between different groups of clustered keys, each cluster having distinctly different proximal properties. A statistical measure made between the intragroup and intergroup keyflight timing distributions may be performed to determine one-handed or two-handed typing input, which may be used to enhance a user experience by adapting a viewing or input element to an appropriate hand setting. In another implementation, the detection may be used as input to determine a likelihood of suspected coaching fraud for online banking applications. In yet another implementation, the detection of one-handed typing is used as an input to bot detection algorithms to reduce false positives from other modalities.

Abstract: Systems and methods are provided for improving behavioral biometrics scoring accuracy. A method includes receiving streamed behavioral biometrics data and associated contextual data, determining a first trust score by evaluating the streamed behavioral biometric data, and if the first trust score is below a first predetermined value, separating the streamed behavioral biometrics data into contextualized bins corresponding to the contextual data to produce binned data. The contextualized bins can include a target identifier such as a data field; or a data type field, an area of a browser window, a task, and/or a physical activity. Responsive to a first threshold amount of accrued binned data, the method can include evaluating the binned data to determine a second trust score that is context-aware and sending an alert to the enterprise server responsive to the second trust score being below a second predetermined value.

Abstract: Systems and methods are provided for automatically generating unique identifiers for text fields and determining the type of information of a specific text field in an online user journey. The method identifies a data type identifier for each text field by training on historical sessions from many user interactions to group text fields, even with randomly generated IDs, and based on the metadata associated with the text fields. The method can predict the data type for new data and can enable the automatic selection and application of a correct keystroke dynamics algorithm for authentication and fraud detection.

Abstract: Systems and methods are provided for detecting when a user uses one or both hands to interact with a device having a physical or virtual keyboard. In one implementation, one-handed typing is determined by analyzing intragroup and intergroup keyflight timing distributions between different groups of clustered keys, each cluster having distinctly different proximal properties. A statistical measure made between the intragroup and intergroup keyflight timing distributions may be performed to determine one-handed or two-handed typing input, which may be used to enhance a user experience by adapting a viewing or input element to an appropriate hand setting. In another implementation, the detection may be used as input to determine a likelihood of suspected coaching fraud for online banking applications. In yet another implementation, the detection of one-handed typing is used as an input to bot detection algorithms to reduce false positives from other modalities.

Abstract: A method for denying or nullifying a specific online transaction carried out by a specific user using a computing device associated with at least one input interface, while the specific user was coached by a fraudster. The method includes collecting a specific set of behavioral data relating to the behavior of the specific user during a specific online transaction, and using a multi-dimensional classification module to determine a probability that the specific user was coached during collection of the set of behavioral data. In response to the probability being greater than a predefined threshold, the specific transaction is denied or nullified.

Abstract: A method for denying or nullifying a specific online transaction carried out by a specific user using a computing device associated with at least one input interface, while the specific user was coached by a fraudster. The method includes collecting a specific set of behavioral data relating to the behavior of the specific user during a specific online transaction, and using a multi-dimensional classification module to determine a probability that the specific user was coached during collection of the set of behavioral data. In response to the probability being greater than a predefined threshold, the specific transaction is denied or nullified.

Abstract: A device identifying a coached fraudulent transaction carried out by a specific user using a computing device. A storage medium of the device has stored a training phase module, including instructions to receive a plurality of training sets of behavioral data and corresponding classifications indicating whether that training set was generated when the user was coached during the online transaction. The training phase module further includes instructions to generate a multi-dimensional classification model for classification of a set of behavioral data. The storage medium has stored an operational phase module, which includes instructions to receive, from the computing device via the network interface, a specific set of behavioral data relating to the behavior of the specific user during a specific online transaction, and instructions to determine, using the multi-dimensional classification model, a likelihood that the specific user was coached during the specific online transaction.