Human interaction with a webpage may be determined by processing an event stream generated by the client device during the webpage interaction. A classification server receives the event stream and compares components of the event stream, including components of an event header message, with prerecorded datasets. The datasets include prerecorded event streams having a known interaction type. Training clients may be provided for generating the prerecorded datasets.

A method of routing messages includes receiving a request message from an originating device to be forwarded to one of a plurality of target devices, the request message having a first network address as a source address identifying the originating device. The first network address of the request message is dynamically mapped to a second network address of a selected target device, and the first and second network addresses are stored in association with each other as address mapping information. The method also includes forwarding the selected target device using the second network address. The routing device receives from the target device an error message in relation to the request message, and identifies the originating device which originated the request message using the address mapping information and the second network address of the target device which issued the error message.

Systems, methods, and computer-readable media are presented herein for providing lower level physical-layer gateway functionalities and upper-level application functionalities; a system designed with flexible configurations in order to support a wide range of connected applications. The system can include a processor that executes machine instructions to perform operations. The operations can comprise: receiving, from a first device, a first packet representing first data formatted in a first protocol language; transforming the first data to second data formatted in a second protocol language; and transmitting a second packet representing the second data to a second device.

In some embodiments, a method receives a set of packets for a flow and determines a set of features for the flow from the set of packets. A classification of an elephant flow or a mice flow is selected based on the set of features. The classification is selected before assigning the flow to a network resource in a plurality of network resources. The method assigns the flow to a network resource in the plurality of network resources based on the classification for the flow and a set of classifications for flows currently assigned to the plurality of network resources. Then, the method sends the set of packets for the flow using the assigned network resource.

In one embodiment, a device uses a classification model to determine whether implementation of a routing change suggested by a predictive routing engine for a network will result in a violation of one or more network policies. The device computes a trust score, based on performance metrics for the classification model. The device causes, based in part on the trust score, implementation of the routing change in the network, when the classification model determines that application of the routing change will not result in a violation of the one or more network policies.

Systems, apparatuses and methods may provide for technology that adjusts, via a short-term subsystem, a communications parameter for one or more of wireless communication devices based on data from one or more of a plurality of sensors. The technology may also determine, via a neural network, a prediction of future performance of the wireless network based on a state of the network environment, wherein the state of the network environment includes information from the short-term subsystem and location information about the wireless communication devices and other objects in the environment, and determine a change in network configuration to improve a quality of communications in the wireless network based on the prediction of future performance of the wireless network. The technology may further generate generic path loss models based on time-stamped RSSI maps and record a sequence of events that cause a significant drop in RSSI to determine a change in network configuration.

A method for energy efficient routing in wireless sensor network based on multi-agent deep reinforcement learning, predefines a to-be-deployed wireless sensor network and creates a cooperative routing decision system including A decision networks and one sink module, A decision networks deployed on the agents a.sup.i, i=1, 2, . . . , A, of the sensor nodes, the sink module deployed on the sink node n.sup.0. The decision network obtains a probability vector according to its local observation and position vectors. The sink module calculates a routing for each sensor node according the probability vectors of A decision networks and sends the routings to corresponding sensor nodes. A multi-agent deep reinforcement learning algorithm is adopted to train the decision networks of A agents a.sup.i, i=1, 2, . . . , A of the cooperative routing decision system, deploys the to-be-deployed wireless sensor network into an environment and updates the routing policy of the deployed wireless sensor network at each update cycle of routing.

An electronic device and a method of improving a performance of a software-defined networking (SDN) of the electronic device are provided. The method includes: identifying at least one policy from a flow table of the software-defined networking; identifying performance index parameters based on information about the at least one policy; determining whether or not an improvement in the performance of the software-defined networking is required based on the performance index parameters and a target performance index; when the improvement in the performance of the software-defined networking is determined to be required, executing a predetermined algorithm based on information about the performance index parameters to check execution information about the at least one policy; and updating the at least one policy based on the execution information.

A method of managing a controller of a software defined networking (SDN) network is implemented by a computing device in the SDN network. The method includes receiving status information for the controller, receiving usage information for the operating environment, generating at least one failure prediction for the controller based on the received status information, and outputting prediction information for the at least one failure prediction.

Systems, apparatuses and methods may provide for technology that adjusts, via a short-term subsystem, a communications parameter for one or more of wireless communication devices based on data from one or more of a plurality of sensors. The technology may also determine, via a neural network, a prediction of future performance of the wireless network based on a state of the network environment, wherein the state of the network environment includes information from the short-term subsystem and location information about the wireless communication devices and other objects in the environment, and determine a change in network configuration to improve a quality of communications in the wireless network based on the prediction of future performance of the wireless network. The technology may further generate generic path loss models based on time-stamped RSSI maps and record a sequence of events that cause a significant drop in RSSI to determine a change in network configuration.