Amdocs Development Limited

Eran Yosef Paran

Shmuel Ur

Sarit Chehanowitz

Liat Taub Bahar

Nir Fattal

Yosef Asaf Hermush

Tsvi Haim Lev

As described herein, a system, method, and computer program are provided for network experience optimization using a residential network router. In use, an electronic calendar is accessed by a residential network router. Additionally, the residential network router determines a plurality of events saved to the electronic calendar, wherein the plurality of events require, at least in part, simultaneous network resource (e.g. bandwidth) usage. Further, the plurality of events are prioritized, using the residential network router. Still yet, available network resources are assigned among the plurality of events, using the residential network router, based on the prioritization.

System, method, and computer program for network experience optimization using a home network router

As described herein, a system, method, and computer program are provided for dynamically generating assistance information for customer service agents. In use, presence of a customer at a physical retail store is identified. Additionally, information describing the customer is processed, using a machine learning model, to determine an expected outcome of an interaction with the customer occurring within the physical retail store. It is then determined that the customer is to be assisted by a customer service agent. Further, assistance information for the customer service agent is dynamically generated, based at least in part on the expected outcome of the interaction with the customer.

System, method, and computer program for dynamically generating assistance information for customer service agents

As described herein, a machine learning system, method, and computer program are provided for evaluation of customer service agents. In use, presence of a customer at retail store is identified. Additionally, information associated with the customer is processed, using a machine learning model, to determine an expected outcome of an interaction with the customer occurring within the retail store. Further, after the interaction by an agent of the retail store with the customer, an actual outcome of the interaction is determined. Still yet, the agent is evaluated by comparing the actual outcome with the expected outcome. A result of the evaluation is then output, such as for use in assigning the agent to future in-store customers.

Machine learning system, method, and computer program for evaluation of customer service agents

A method of serving Real Time Bidding (RTB) requests received from client devices based on availability of advertisements (ADs) at an edge storage, comprising using an RTB server for receiving RTB request(s) from one or more client devices for impression(s) in AD placement(s) offered by online content presented by the client device(s) to associated user(s), forwarding the RTB request(s) to AD exchange(s) and receiving in response a plurality of bid offers received from advertisers for purchasing the impression(s) for posting a plurality of ADs, identifying among the plurality of ADs cached AD(s) stored at an edge storage located at an edge of a network serving the client device(s), adjusting bid offers relating to non-cached ADs to reflect cost of network resources required for downloading the non-cached AD(s) and initiating service of AD(s) associated with highest bid offer(s) among the adjusted bid offers and bid offers relating to the cached AD(s).

System, method, and computer program for real-time bidding (RTB) for serving advertisements from edge storage

As described herein, a machine learning system, method, and computer program are provided for making payment related customer predictions using remotely sourced data. A system of a communication service provider (CSP) identifies a customer of the CSP. Additionally, the system collects data from a plurality of data sources independent from the CSP, the data including telephone numbers and/or webpage URLs of other services providers that are associated with making payments. Further, the system processes the collected data to form input data indicating which of the telephone numbers were contacted by the customer and/or webpage URLs were accessed by the customer. Still yet, the system processes the input data using at least one machine learning algorithm to make at least one payment related prediction for the customer. Moreover, the system outputs the at least one payment related prediction made for the customer.

Machine learning system, method, and computer program for making payment related customer predictions using remotely sourced data

A system, method, and computer program product are provided for providing device centric payments in roaming conditions. In operation, a device requests service from a visiting communication network. The device provides an identity associated with the device. The device records an amount of service utilized (e.g. utilizing blockchain technology, etc.). The device submits the amount of service utilized to the visiting network or to a blockchain, such that the visiting network is capable of determining whether the amount of service utilized recorded by the device is similar to an amount of service usage measured by the visiting network (e.g. when an established credit limit available has been utilized, etc.).

System, method, and computer program for providing device centric payment authorization in roaming conditions

A method of serving Real Time Bidding (RTB) requests received from client devices at a network edge node, comprising an edge node located at an edge of a network serving one or more client devices for obtaining a dataset associating each of a plurality of cached ADs stored in edge storage device(s) locally accessible to the edge node with one or more user attributes, receiving, from the client device(s), RTB request(s) for impression(s) in AD placement(s) presented by the client device(s) to respective user(s), each RTB request comprising user attribute(s) of the respective user, serving in response to each RTB request, to the respective client device, one of the cached ADs associated with the user attribute(s) of the respective RTB request or forwarding the respective RTB request to an RTB server when none of the cached ADs is associated with the user attribute(s) of the respective RTB request.

System, method, and computer program for edge real-time bidding (RTB)

Presented herein are methods and systems for alerting users identified in a monitored location, by an edge node located at an edge of a network, to be preoccupied by interaction with client devices of potential risk events, comprising analyzing, at the edge node, images captured by one or more imaging sensors deployed to monitor a monitored location in order to estimate a motion pattern of one or more passerby users preoccupied by watching a screen of a client device, analyzing the images with respect to the estimated motion pattern(s) to identify a potential risk event involving one or more of the preoccupied passerby users and transmitting one or more warning messages alert one or more of the preoccupied passerby users of the potential risk event.

Safety service by using edge computing

As described herein, a system, method, and computer program are provided for deriving user attention data. In use, user attention data is collected for a user. The user attention data includes first information describing content being viewed by a user on a first device, and second information describing user activity occurring on the first device and/or one or more second devices while the content is being viewed by the user on the first device. Further, the first information and the second information are processed, using a machine learning model, to predict a degree to which the user likes the content. Still yet, the prediction is output for use in making one or more content recommendations.

System, method, and computer program for using user attention data to make content recommendations

A method and system for analyzing real world information of users to derive physical (real world) user attributes used for selecting online content for presentation by client devices correlated with the users, comprising, detecting a presence of one or more users in a physical registration location by analyzing sensory data captured by one or more sensors deployed to monitor the physical registration location, correlating between the user(s) and an identifier of client device(s) associated with the user(s), identifying one or more user attributes of the user(s) by analyzing the sensory data and transmitting the user attribute(s) coupled with the identifier to one or more remote servers adapted to use the user attribute(s) for selecting one or more online content items served to the client device(s) via a network for presentation to the user(s). The sensor(s) is physically and communicatively disconnected from the client device(s).

System, method, and computer program for serving online content to users based on real world user information

As described herein, a system, method, and computer program are provided for edge management of geographically dispersed sensors. An edge device within a network accesses observations of a plurality of geographically dispersed sensors. Additionally, the edge device processes the observations to determine overlapping portions of the observations. Further, the edge device optimizes the observations to form optimized observations for transmission to a cloud processing system, wherein the optimizing is based on the determined overlapping portions of the observations.

System, method, and computer program for edge management of geographically dispersed sensors

A system, method, and computer program product are provided for automatic management of intent classification. In operation, an intent classification system identifies a conversation including a plurality of utterances for which to identify intent of a user associated with the plurality of utterances. The intent classification system identifies one or more actions performed in association with the plurality of utterances. Further, the intent classification system correlates at least one of the one or more actions to at least one of the plurality of utterances based on which of the one or more actions were performed as a result of the plurality of utterances. The intent classification system automatically determines an intent of the user based on the correlation of the at least one of the one or more actions to the at least one of the plurality of utterances. Additionally, the intent classification system stores an indication of the intent based on the at least one of the plurality of utterances. Moreover, the intent classification system associates one or more specific actions with the indication of intent such that subsequent utterances are capable of being used by the intent classification system to automatically identify an intent of one or more users and to automatically identify specific actions that should be performed based on the automatically identified intent of the one or more users.

Growing community of inventors

Hod Hasharon, Israel

Eran Yosef Paran