A network administration device may include one or more processors to receive operational information regarding a plurality of network devices; receive flow information relating to at least one traffic flow; input the flow information to a model, where the model is generated based on a machine learning technique, and where the model is configured to identify predicted performance information of one or more network devices with regard to the at least one traffic flow based on the operational information; determine path information for the at least one traffic flow with regard to the one or more network devices based on the predicted performance information; and/or configure the one or more network devices to implement the path information for the traffic flow.

A failure effects estimating device includes an input unit that inputs a log and a traffic amount obtained from a communication system when an abnormality occurs, an estimating unit that estimates a failure effects amount in the communication system, on the basis of the log and the traffic amount, and an output unit that outputs the failure effects amount estimated by the estimating unit.

Various embodiments include network computing devices and methods for computing resource management. A processor of a network computing device may determine a latency metric and a transaction volume metric for a network application, determine an autoscaling cost based on the determined latency metric and transaction volume metric, allocate to the network application computing resources based on the determined autoscaling cost, and provide the allocated computing resources to the network application.

Techniques are described for dynamic bid-driven reconfiguration in a wireless communication network. An analytics front-end system (AFES) can be implemented in a provider network. The AFES can obtain network service experience (NSE) predictions derived by a network data analytics function (NWDAF) based on network status data received from back-end network functions of the provider network. The AFES can generate bid options based on the NSE predictions, each having a respective NSE definition and bid value. The bid options can be output to a user bidding interface, via which a user can review and select the bid options. In response to the user selecting one of the bid options, the AFES can direct a network policy server to update one or more subscriber policy definitions for one or more subscribers based on the respective NSE definition of the selected bid option, thereby dynamically updating at least one subscriber's NSE.

A method for automated web resource deployment is provided. The method comprises creating web resource publication requests, wherein each web resource publication request comprises a number of configuration changes necessary to publish a web resource, on a network, at a particular uniform resource location. A standard format, validation workflow, and an approval workflow are provided for automation of the web resource publication requests. Once validated and approved, web resource publication requests are automatically converted to API calls which are executed on backend servers to implement the configuration changes required in the environment without further human intervention.

Disclosed are systems and methods for a robust Self-Organizing Network (SON) framework that quantifies SON applications' control and management of a network into key performance indicators (KPI) that are leveraged to determine the impact of a SON's application effectiveness in regulating network parameters, which then dictates how the SON application operates. The disclosed framework is configured to receive multiple data streams from existing data sources, determine the performance of a node on a network, and then automatically perform SON operations based therefrom. The disclosed framework can utilize this information to predict additional and/or future opportunities for SON automation on the network.

This disclosure describes systems, methods, and devices related to anomaly detection of CPE firmware revisions. A method may include collecting metrics data for a plurality of customer-provided equipment (CPE) models over a window of time; training a first autoencoder for a first CPE model of the plurality of CPE models using at least a portion of the metrics data to detect anomalies within a plurality of firmware versions of the first CPE model; identifying, using the first autoencoder, that a first firmware version of the plurality of firmware versions is anomalous across a first time series; and storing data indicating that the first firmware version of the plurality of firmware versions is anomalous across the first time series. Metrics data may include one or more of interactive voice response (IVR) session data; calls handled data; and truck schedule data.

The disclosure relates to an electronic apparatus and a method of controlling the same. The electronic apparatus includes: a communication interface; and a processor configured to receive log data of a plurality of devices connected to a network through the communication interface, acquire operation time information of each of the devices from the received log data, calculate similarity of the operation time between the plurality of devices based on the acquired operation time information, and determine a device group including two or more devices with relatively high calculated similarity among the plurality of devices.

A data stream classification model updating method is disclosed. Determining, based on packet information of a current data stream and a behavior classification model, a first data stream class corresponding to the current data stream; determining, based on a target correspondence and a common feature of the current data stream, a second data stream class corresponding to the current data stream, where the target correspondence is a correspondence between a plurality of common features and a plurality of data stream classes; and if the first data stream class is different from the second data stream class, obtaining correction data corresponding to the current data stream, where the correction data includes the packet information of the current data stream and the second data stream class, and the correction data is used as a training sample to update the behavior classification model.

A data stream classification model updating method is disclosed. Determining, based on packet information of a current data stream and a behavior classification model, a first data stream class corresponding to the current data stream; determining, based on a target correspondence and a common feature of the current data stream, a second data stream class corresponding to the current data stream, where the target correspondence is a correspondence between a plurality of common features and a plurality of data stream classes; and if the first data stream class is different from the second data stream class, obtaining correction data corresponding to the current data stream, where the correction data includes the packet information of the current data stream and the second data stream class, and the correction data is used as a training sample to update the behavior classification model.