Abstract: A distributed feature collection and correlation engine is provided, Feature extraction comprises obtaining one or more data records; extracting information from the one or more data records based on domain knowledge; transforming the extracted information into a key/value pair comprised of a key K and a value V, wherein the key comprises a feature identifier; and storing the key/value pair in a feature store database if the key/value pair does not already exist in the feature store database using a de-duplication mechanism. Features extracted from data records can be queried by obtaining a feature store database comprised of the extracted features stored as a key/value pair comprised of a key K and a value V, wherein the key comprises a feature identifier; receiving a query comprised of at least one query key; retrieving values from the feature store database that match the query key; and returning one or more retrieved key/value pairs.

Abstract: A distributed feature collection and correlation engine is provided, Feature extraction comprises obtaining one or more data records; extracting information from the one or more data records based on domain knowledge; transforming the extracted information into a key/value pair comprised of a key K and a value V, wherein the key comprises a feature identifier; and storing the key/value pair in a feature store database if the key/value pair does not already exist in the feature store database using a de-duplication mechanism. Features extracted from data records can be queried by obtaining a feature store database comprised of the extracted features stored as a key/value pair comprised of a key K and a value V, wherein the key comprises a feature identifier; receiving a query comprised of at least one query key; retrieving values from the feature store database that match the query key; and returning one or more retrieved key/value pairs.

Abstract: A method for identifying an unknown user according to a plurality of facets of user activity in a plurality of contexts includes receiving a plurality of priors for the facets with respect to the contexts, receiving a plurality of footprints of known users, aggregating the footprints of the users to determine an ensemble prior, receiving a plurality of network traces relevant to an unknown user in a computer environment, matching the network traces against each of the footprints to determine a plurality of matches, aggregating the matches using the ensemble prior according to the facets and the contexts, and outputting a probable user identity for the unknown user.

Abstract: A method for identifying an unknown user according to a plurality of facets of user activity in a plurality of contexts includes receiving a plurality of priors for the facets with respect to the contexts, receiving a plurality of footprints of known users, aggregating the footprints of the users to determine an ensemble prior, receiving a plurality of network traces relevant to an unknown user in a computer environment, matching the network traces against each of the footprints to determine a plurality of matches, aggregating the matches using the ensemble prior according to the facets and the contexts, and outputting a probable user identity for the unknown user.

Abstract: Methods and apparatus are provided for detecting unauthorized bulk forwarding of sensitive data over a network. A bulk forwarding of email from a first network environment is automatically detected by determining an arrival rate for internal emails received from within the first network environment into one or more user accounts; determining a sending rate for external emails sent from the one or more user accounts to a second network environment; and detecting the bulk forwarding of email from a given user account by comparing the arrival rate for internal emails and the sending rate for external emails. The bulk forwarding of email from a given user account can be detected by determining whether statistical models of the arrival rate for internal emails and of the sending rate for external emails are correlated in time.

Abstract: Methods and apparatus are provided for detecting unauthorized bulk forwarding of sensitive data over a network. A bulk forwarding of email from a first network environment is automatically detected by determining an arrival rate for internal emails received from within the first network environment into one or more user accounts; determining a sending rate for external emails sent from the one or more user accounts to a second network environment; and detecting the bulk forwarding of email from a given user account by comparing the arrival rate for internal emails and the sending rate for external emails. The bulk forwarding of email from a given user account can be detected by determining whether statistical models of the arrival rate for internal emails and of the sending rate for external emails are correlated in time.

Abstract: Methods and apparatus are provided for automatic identification of affected network resources after a computer intrusion. The network resources affected by a computer intrusion can be identified by collecting information about an external system from an external source; deriving a list of one or more affected internal systems on an internal network by correlating the information with internal information about internal systems that interacted with the external system: and identifying one or more user accounts associated with the one or more affected internal systems. Data residing on systems accessible by the one or more user accounts can also optionally be identified. A list can optionally be presented of the network resources that may be affected by the computer intrusion. The affected network resources can be, for example, servers, services and/or client machines.

Abstract: Methods and apparatus are provided for automatic identification of affected network resources after a computer intrusion. The network resources affected by a computer intrusion can be identified by collecting information about an external system from an external source; deriving a list of one or more affected internal systems on an internal network by correlating the information with internal information about internal systems that interacted with the external system; and identifying one or more user accounts associated with the one or more affected internal systems. Data residing on systems accessible by the one or more user accounts can also optionally be identified. A list can optionally be presented of the network resources that may be affected by the computer intrusion. The affected network resources can be, for example, servers, services and/or client machines.

Abstract: The present invention provides methods and apparatus for classifying and demultiplexing packets in a network protocol stack. It provides extendibility for packet processing in the network protocol stack by defining a standard method for adding new functionality. It provides a method to obtain external information, from an application scheduled outside of the forwarding or interrupt context of the kernel, in order to augment packet classification and/or augment packet disposition. In some embodiments, external information augments a criteria of a node in a classification tree with additional information. It presents a way of augmenting which suspends the classification process until an application, scheduled outside of the forwarding or interrupt context of the kernel, completes. The resulting external information is used to augment the packet classification.