A computer-implemented method, a computer program product, and a computer system for detecting, verifying and preventing unauthorized use of a Voice over Internet Protocol (VoIP) service. A computer rates a VoIP call based on a database including information of the caller number, in response to determining that no record of a caller number exists in a database including the information of unauthorized uses. The computer sets a predetermined time period for the VoIP call based on a rating of the VoIP call, adds the predetermined time period to a session initiation protocol (SIP) invite, and connects the VoIP call to a called party. In response to that the predetermined time period is reached, the computer interrupts the VoIP call and prompts the caller to conduct user verification. In response to that the caller is successfully verified, the computer reconnects the VoIP call to the called party.

A sequence of call volume measurements is accessed, where each of the call volume measurements is associated with respective metadata. The respective metadata may provide information regarding a time period during which a call volume measurement was made. A window of the sequence of call volume measurements with the respective metadata is input to a machine learning model to obtain a prediction of a call volume. The machine learning model includes embedding functions that are applied to the respective metadata for the call volume measurements in the window.

A computer-implemented method, a computer program product, and a computer system for detecting, verifying and preventing unauthorized use of a Voice over Internet Protocol (VoIP) service. A computer rates a VoIP call based on a database including information of the caller number, in response to determining that no record of a caller number exists in a database including the information of unauthorized uses. The computer sets a predetermined time period for the VoIP call based on a rating of the VoIP call, adds the predetermined time period to a session initiation protocol (SIP) invite, and connects the VoIP call to a called party. In response to that the predetermined time period is reached, the computer interrupts the VoIP call and prompts the caller to conduct user verification. In response to that the caller is successfully verified, the computer reconnects the VoIP call to the called party.

A method for allocating resources to modules of a contact center includes: receiving a first interaction in a first state; determining a first load of a first module of the contact center to be low; in response to determining that the first load is low, routing the first interaction to the first module of the contact center, the first module transitioning the first interaction from the first state to a second state; receiving a second interaction in the first state; determining a second load on the first module of the contact center to be high; and in response to determining that the second load is high, routing the second interaction to a second module configured to transition the second interaction from the first state to the second state, the second module having different resource requirements than the first module.

A system, method and non-transitory computer readable storage medium comprising instructions that when read by a processor perform receiving a telephony connection request, determining a purported sender telephony number from the telephony connection request, determining additional request information from the telephony connection request, accessing an information source using the purported sender telephony number to determine source information regarding the sender, comparing the source information with the additional request information, based on the comparison, and determining whether the purported sender telephony number is incorrect.

A sequence of call volume measurements is accessed, where each of the call volume measurements is associated with respective metadata. The respective metadata may provide information regarding a time period during which a call volume measurement was made. A window of the sequence of call volume measurements with the respective metadata is input to a machine learning model to obtain a prediction of a call volume. The machine learning model includes embedding functions that are applied to the respective metadata for the call volume measurements in the window.

A system for proactive intrusion protection comprises a memory operable to store data identifying a plurality of compromising entities, comprising at least one of a device identifier or a contact identifier, and a processor communicatively coupled to the memory and operable to receive, from a remote application associated with a remote device and with the system, information regarding a destination of the outgoing communication. The processor is further operable to determine an entity associated with the destination of the outgoing communication and to determine that the entity associated with the destination matches at least one of the plurality of compromising entities based on comparing the data identifying the plurality of compromising entities and the entity associated with the destination of the outgoing communication. Furthermore, the processor is operable to send to the remote application, before the outgoing communication is sent, a signal configured to block the outgoing communication.

A method for allocating resources to modules of a contact center includes: receiving a first interaction in a first state; determining a first load of a first module of the contact center to be low; in response to determining that the first load is low, routing the first interaction to the first module of the contact center, the first module transitioning the first interaction from the first state to a second state; receiving a second interaction in the first state; determining a second load on the first module of the contact center to be high; and in response to determining that the second load is high, routing the second interaction to a second module configured to transition the second interaction from the first state to the second state, the second module having different resource requirements than the first module.

A machine learning model (e.g., including a deep learning neural network) with learned embeddings is applied to time series data with associated metadata to obtain predictions of the time series value. For example, a call volume in a period of time may be predicted based on call volume data for a sequence of time bins in a window of preceding time. Time bins may be associated with respective metadata, such as day of week, hour of day, day of month, holiday, part of business cycle, weather, and/or tide. These pieces of metadata may be mapped to embedding vectors using trained embedding functions. The resulting embedding vectors may be input to a neural network along with the corresponding time series data (e.g., call volumes) to make a prediction for future time bin. For example, the prediction may be used to provision servers in a network infrastructure.

A system, method and non-transitory computer readable storage medium comprising instructions that when read by a processor perform receiving a telephony connection request, determining a purported sender telephony number from the telephony connection request, determining additional request information from the telephony connection request, accessing an information source using the purported sender telephony number to determine source information regarding the sender, comparing the source information with the additional request information, based on the comparison, and determining whether the purported sender telephony number is incorrect.