DETECTING SCAM CALLERS USING CONVERSATIONAL AGENT AND MACHINE LEARNING SYSTEMS AND METHODS

    公开(公告)号:US20240040038A1

    公开(公告)日:2024-02-01

    申请号:US18482277

    申请日:2023-10-06

    发明人: Ovidiu Serban

    IPC分类号: H04M3/493 G06N20/00 H04M3/436

    摘要: Systems and methods for detecting indications of a scam caller are disclosed. Call data, such as call audio, is received and used to create a training dataset. Using the training dataset, a machine learning model is trained to detect indications of a scam caller in a phone call. An Interactive Voice Response (IVR) model is trained or configured, using voice samples of speech of a subscriber of a telecommunications service provider, to simulate speech and conversation of the subscriber. A conversational agent is generated using the IVR model and the trained machine learning model. The conversational agent receives a phone call, engages a caller in simulated conversation, and detects indications of whether the caller is a likely scam caller. If the caller is determined to be a likely scam caller, an alert can be generated and/or the call can be disconnected.

    Temporary assignment of emergency route back

    公开(公告)号:US11877342B2

    公开(公告)日:2024-01-16

    申请号:US17190233

    申请日:2021-03-02

    摘要: An emergency route back number (ERBN) may be assigned for use in emergency calls. A telecommunication network associated that may be associated with a wireless service provider, may identify an emergency call directed to an emergency number (e.g., 9-1-1) that originated from a device that is not associated with a unique phone number such as a shared line, or a communication that utilizes some other indicator to establish the emergency call (e.g., an email address, or some other identifier). When an emergency call is detected that does not have a dedicated phone number, an ERBN is assigned to the communication. This ERBN may be reserved for some period of time after the emergency call ends such that emergency contact centers are able to reach callers back in case of dropped calls, fails, and/or additional emergency services that may need to reach the callers directly.

    Handovers between IPV4 public data network sessions and 5G radio access networks

    公开(公告)号:US11877202B2

    公开(公告)日:2024-01-16

    申请号:US17680127

    申请日:2022-02-24

    IPC分类号: H04W36/32 H04W36/00

    摘要: To facilitate a handover of mobile devices from 4th Generation (4G) networks to 5th Generation (5G) networks, a mobility management entity (MME) of the 4G network core inactivates a data session in which the mobile device is assigned an Internet protocol version 4 (IPv4) address. The MME generates a modified context of the mobile device that omits the IPv4 address of the device. The inactivation of the IPv4 session causes the mobile device to initiate a new data session with the 5G core network, in which the device will be assigned a new IPv6 address for use on a data network. The modified context is used by the 5G network to configure communications for the mobile device without interrupting network service to the mobile device.

    TELECOMMUNICATIONS NETWORK USAGE ANOMALY DETECTION WITH SIMULATED USER INTERACTIONS SYSTEMS AND METHODS

    公开(公告)号:US20240015076A1

    公开(公告)日:2024-01-11

    申请号:US17861061

    申请日:2022-07-08

    摘要: Systems and methods for using a user simulation model to facilitate detection of usage anomalies is disclosed. Usage data is received for a session between a user device and an application or service. The usage data is monitored to detect a usage anomaly, such as unusual or suspicious transactions, unexpected user or device attributes, or abnormal usage patterns. In response to detecting a request to terminate the session, the session is instead handed off to a user simulation model that simulates interactions of a user in the session. The user simulation model can be a machine learning model that is trained, using a training dataset, to simulate user interactions. When the user logs into the application or service, the session can be handed off from the user simulation model to the user, such that the session is perpetual or substantially perpetual.

    Estimating cellular site locations
    96.
    发明授权

    公开(公告)号:US11864153B1

    公开(公告)日:2024-01-02

    申请号:US17238143

    申请日:2021-04-22

    发明人: Jong Sung Yoon

    摘要: Solutions for estimating a cellular site location, for example in a roaming network, includes receiving data metrics from a plurality of user devices corresponding to communications using a roaming carrier network, wherein the data metrics are stored on the plurality of user devices during the communication using the roaming carrier network, and the data metrics are received using a home carrier network. The received data metrics is filtered to identify the roaming carrier network and the filtered data metrics group corresponding to a cell within the roaming carrier network. A source of a signal corresponding to the cell is determine and the location of the cellular site corresponding the source of the signal estimated.

    Telecommunications network planning system

    公开(公告)号:US11856412B2

    公开(公告)日:2023-12-26

    申请号:US17554952

    申请日:2021-12-17

    IPC分类号: H04W16/18 H04W24/02 H04W28/02

    摘要: Systems and methods to identify a growth classification/categorization for a geographic area that helps a network provider to solve for what types of planning opportunities are available at various area granularities is disclosed. The system computes values for a set of growth criteria for a geographic area. The growth criteria are related to planning, usability, customer experience, sales, population, and so on. Based on the growth-criteria values, the system identifies a classification/categorization for the area. For example, the system classifies an area as an invest area (e.g., requiring engineering action), a grow area (e.g., requiring sales action/being sales ready), a defend area (e.g., requiring engineering and sales actions to continue current trend), or a fix area (e.g., likely requiring both engineering and sales actions to improve current trend). Based on the area classification, the system can then provide actionable insights to drive improvement in network coverage and customer experience.

    Contextual information inference for incoming calls

    公开(公告)号:US11856133B2

    公开(公告)日:2023-12-26

    申请号:US16780692

    申请日:2020-02-03

    发明人: Parag Garg

    摘要: User profile data of a caller and time-stamped event data from one or more data sources are received. An incoming call initiated by a caller device of the caller is detected, and device data associated with the caller device that includes a geolocation is received. In response to an initiation of the incoming call, an inference framework is applied to infer a situational context of the caller based on an analysis of the time-stamped event data, the user profile data of the caller, and the device data. The contextual information for the situational context of the caller is sent to the recipient device for presentation via the recipient device. Feedback of a call recipient is received from the recipient device following the presentation. Accordingly, one or more inference rules of the inference framework are modified via a machine-learning algorithm to generate a modified inference framework based on the feedback.