Abstract: The present disclosure describes systems and methods that provide a hybrid framework for augmenting statistical anomaly detection with contextual features, machine learning and human Subject Matter Expert (SME) input to learn significant characteristics of true anomalies for which alerts should be generated. The framework presented herein is domain agnostic and independent of the underlying statistical anomaly detection technique or the machine learning algorithm. The framework described herein is therefore applicable and adaptable to a number of real world service provider systems and applications, such as, for example, detecting network performance degradation in a service provider network or detecting anomalous conditions from data received from a sensor while filtering out false positives.

Abstract: A method, apparatus and system for improved community discovery that enables the accurate real-time determination of linkages between groups of individuals in a data-driven manner that is insensitive to noise and/or missing/dropped signals for automatically capturing the dynamic real-time interaction between individuals. No pre-defined parameters are necessary and the operations provide a fully data-driven solution that can be applied and adapted to diverse and highly dynamic environments in real-time in an automated fashion.

Abstract: The present disclosure describes systems and methods that provide a hybrid framework for augmenting statistical anomaly detection with contextual features, machine learning and human Subject Matter Expert (SME) input to learn significant characteristics of true anomalies for which alerts should be generated. The framework presented herein is domain agnostic and independent of the underlying statistical anomaly detection technique or the machine learning algorithm. The framework described herein is therefore applicable and adaptable to a number of real world service provider systems and applications, such as, for example, detecting network performance degradation in a service provider network or detecting anomalous conditions from data received from a sensor while filtering out false positives.

Abstract: A method, apparatus and system for improved community discovery that enables the accurate real-time determination of linkages between groups of individuals in a data-driven manner that is insensitive to noise and/or missing/dropped signals for automatically capturing the dynamic real-time interaction between individuals. No pre-defined parameters are necessary and the operations provide a fully data-driven solution that can be applied and adapted to diverse and highly dynamic environments in real-time in an automated fashion.

Abstract: A method for deriving a composite tie metric for an edge between nodes of a telecommunication call graph includes receiving descriptive data with original values for descriptive attributes associated with a telecommunication call graph formed by edges between nodes. Each edge relates to two nodes. Each original value relates to an edge and a descriptive attribute forming an edge-attribute pair for the corresponding original value. The descriptive data is stored in a local storage device. Scaling factors for each descriptive attribute are determined taking into account a distribution of the original values for the corresponding descriptive attribute and a common base for the descriptive attributes. Weighting factors are determined for each descriptive attribute. The composite tie metric is computed for an edge based on the original value, scaling factor, and weighting factor for the descriptive attributes.

Abstract: A capability for classifying a device problem of a customer premises device of a customer is presented. The capability for classifying a device problem of a customer premises device of a customer may include receiving customer trouble ticket information of a trouble ticket associated with a customer, receiving customer device measurement information for a customer premises device of the customer, and classifying a device problem of the customer premises device based on text mining of the customer trouble ticket information and based on the customer device measurement information.

Abstract: A system and method for generating a subscriber churn prediction includes receiving call detail records from a network operator detailing communication between subscribers of the network operator and determining tie-strengths between subscribers based on the call detail records. The system and method further includes generating a net churn influence accumulated at each subscriber from the tie-strengths by propagating churner influence between subscribers due to churn events.

Abstract: A system and method for generating a subscriber churn prediction includes receiving call detail records from a network operator detailing communication between subscribers of the network operator and determining tie-strengths between subscribers based on the call detail records. The system and method further includes generating a net churn influence accumulated at each subscriber from the tie-strengths by propagating churner influence between subscribers due to churn events.

Abstract: Techniques for single-failure protection in load-balanced network architectures are disclosed. For example, in one aspect of the invention, a technique for processing a traffic flow in a communication network comprising a plurality of nodes, the traffic flow being deliverable from a source node to at least one destination node via one or more intermediate nodes, comprises the following steps/operations. The traffic flow is split at the source node into a plurality of parts. The parts are distributed to respective ones of the intermediate nodes such that the parts are routed from the source node to the at least one destination node in a disjoint manner.

Abstract: A method for deriving a composite tie metric for an edge between nodes of a telecommunication call graph includes receiving descriptive data with original values for descriptive attributes associated with a telecommunication call graph formed by edges between nodes. Each edge relates to two nodes. Each original value relates to an edge and a descriptive attribute forming an edge-attribute pair for the corresponding original value. The descriptive data is stored in a local storage device. Scaling factors for each descriptive attribute are determined taking into account a distribution of the original values for the corresponding descriptive attribute and a common base for the descriptive attributes. Weighting factors are determined for each descriptive attribute. The composite tie metric is computed for an edge based on the original value, scaling factor, and weighting factor for the descriptive attributes.

Abstract: Network design techniques and techniques for routing virtually-concatenated data traffic in a network in a manner which distributes delay to intermediate nodes of the network are disclosed. For example, in one aspect of the invention, a technique for routing virtually-concatenated data traffic in a network comprising a plurality of nodes comprises, for a given traffic demand to be routed from a source node to a destination node in the network, the following steps/operations. Two or more paths are determined to route the given traffic demand. Each of the two or more paths correspond to a member of a virtually-concatenated group. At least one path of the two or more paths comprises the source node, the destination node and at least one other node coupled between the source node and the destination node. Further, at least a subset of the source node, the destination node and the one other node buffer at least a portion of the given traffic demand such that a delay is distributed over the at least one path.

Abstract: Techniques for designing networks. The techniques utilize network management-based routing (NMS routing) in conjunction with the planning step (design-based routing) of the design process so that an optimal network may be designed. An automated technique for designing a network may comprise the following steps. First, one or more traffic demands are obtained. Then, a network is computed by determining one or more routes for the one or more traffic demands using a design-based routing methodology based on feedback from a network management-based routing methodology.

Abstract: Techniques and systems are disclosed for context-aware migration of a communication network session. For example, a system for migrating a communication session established between a first entity and a second entity in accordance with an application module includes the following elements. A context monitor module supports the application module and is operative to obtain context information associated with at least one of the first entity and the second entity. A migration server module supports the application module and is operative to effectuate a transfer of the communication session from one communication device associated with the first entity to another communication device associated with the first entity. The context monitor module and the migration server module operate in cooperation with the application module to transfer the communication session.

Abstract: A method for providing advice to a service provider includes; a) accessing customer, network, and service data stored in disparate data sources, the data relating to services provided to customers by a service provider over a network infrastructure, a virtualization platform presenting the disparate data in a common logical view; b) processing the data in an analytics platform using a service advisor algorithm to form conclusions and predictions regarding management of customer churn and generate advice; and c) formulating results from the processing to enable the service provider to consider the data, conclusions, predictions, and advice in relation to reduction of customer churn. The service advisor algorithm may also be used for personalization of service provided to customers, targeted marketing of services to customers or non-customers, and management of the network infrastructure. An apparatus for providing advice to a service provider includes the virtualization and analytics platforms.

Abstract: Location-based services are provided in a communication system comprising at least a portion of at least one wireless network. One aspect of the invention relates to generation of user movement statistics that may be utilized to facilitate the development of targeted marketing campaigns or other types of message delivery. In this aspect, the user movement statistics are generated based on location and profile information for mobile user devices, and delivery of at least one message to a given one of the mobile user devices is controlled based on the user movement statistics. The user movement statistics may comprise, for example, information sufficient to determine approximately how many users having particular designated characteristics are likely to be in a given location at a given time, and may be utilized to determine prices charged for delivery of messages to the mobile user devices.

Abstract: Techniques are disclosed for designing optical transmission systems that efficiently compute cost-optimal configurations under one or more constraints. For example, in one aspect of the present invention, a technique for designing an optical transmission system comprises the following steps/operations. A set of one or more demands and a set of optical transmission system elements are obtained. Elements may be consecutively coupled via a span. At least one constraint on the design of the optical transmission system is obtained.

Abstract: A fast shared path allocation technique is disclosed. Network nodes are pre-configured such that data from multiple data sources or multiple primary data paths may be sent via a shared secondary data path. Merge nodes merge input from a plurality of input ports onto an output port. The merge nodes implement a blocking function such that upon receipt of a signal from one of the input ports, the signals from the other input ports are blocked from reaching the output port. Upon a triggering event indicating a need to allocate the shared path, the data is first sent to the merge node where it is appropriately merged onto the output link and transmitted towards its destination. Only after the data has been sent does the merge node block the remaining input ports from reaching the output port. This blocking may be performed automatically by the merge node or by conventional network signaling.

Abstract: Multi-path routing techniques using intra-flow splitting are disclosed. For example, a technique for processing traffic flows at a node in a network comprises the following steps/operations. At least one traffic flow is obtained. The at least one traffic flow comprises multiple packets or bytes. The at least one flow is split into at least two sub-flows, wherein each of the at least two sub-flows comprises a portion of the multiple packets or bytes. The packets or bytes of the at least two sub-flows are respectively routed on at least two paths in the network.

Abstract: Techniques and systems for designing a network providing communication and location identification services are described. A solution point comprising parameters for each of a plurality of base stations is generated. A coverage and locatability performance value for the solution point is computed, as well as derivatives of the performance value. The coverage and locatability performance value and its derivatives are used to indicate favorable directions for searching for subsequent solution points, and subsequent solution points are generated and compared against previous solution points until an optimum solution point is found. The coverage and locatability performance value is a weighted sum of coverage and locatability values for each point in the service area of the network, with coverage values representing forward and reverse link quality and locatability values representing the probability that a point will experience an acceptable power level from at least four base stations.

Abstract: Techniques for network routing and design are provided. A technique for determining a route for a demand in a network, wherein the network comprises primary paths and secondary paths, and at least two secondary paths may share a given link, comprises the following steps/operations. First, a graph representing the network is transformed. Edges of the graph represent channels associated with paths and nodes of the graph represent nodes of the network. The transformation is performed such that costs associated with the edges reflect costs of using channels in secondary paths. Then, the shortest path between nodes corresponding to the demand is found in the transformed graph. The shortest path represents the least-cost path in the network over which the demand may be routed. When the above route determination steps/operations result in a path with at least one loop, an alternative routing process may be executed so as to determine a loopless path for the demand.