# US 2024/0267457 A1 · Predicting Call Volume Using Call Volume Data

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2024/0267457 A1
- Document ID: US20240267457A1
- Status: Published application
- Kind: A1
- Application number: 18/636,637
- Filed: 2024-04-16
- Published: 2024-08-08
- Earliest claimed: 2021-07-30
- Assignees: Zoom Video Communications, Inc.
- Applicants: Zoom Video Communications, Inc.
- Inventors: Davide Giovanardi; Andrew Miller-Smith
- Web page: https://patentgrep.com/patent/US20240267457A1

## CPC

- G06Q 30/0202: Market predictions or forecasting for commercial activities
- H04M 3/362: {Traffic simulation}
- G06N 20/00: Machine learning
- H04M 3/365: {Load metering of control unit}
- H04M 3/5238: {with waiting time or load prediction arrangements}
- H04M 2203/55: related to network data storage and management
- G06Q: INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- H04M: TELEPHONIC COMMUNICATION (circuits for controlling other apparatus via a telephone cable and not involving telephone switching apparatus G08)
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS

## Abstract

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.

## Claims

### Claim 1

A method comprising:
accessing call volume data that includes time period data; and
inputting the call volume data to a machine learning model to obtain a prediction of a call volume, wherein the machine learning model includes embedding functions that are applied to the time period data, wherein the embedding functions are implemented with a set of parameters that control functions applied to a discrete variable of the time period data to map the discrete variable to a vector space.

### Claim 2 (depends on 1)

The method of claim 1, wherein the embedding functions have been trained using a sequence of network traffic measurements other than call volume measurements.

### Claim 3 (depends on 1)

The method of claim 1, wherein the embedding functions have been trained using a sequence of website traffic measurements.

### Claim 4 (depends on 1)

The method of claim 1, wherein the time period data includes at least one of a day of a week, an hour of the day, or day of a month for the call volume data.

### Claim 5 (depends on 1)

The method of claim 1, comprising:
activating one or more telephony servers based on the prediction of the call volume.

### Claim 6 (depends on 1)

The method of claim 1, wherein metadata for the call volume data includes an indication of weather.

### Claim 7 (depends on 1)

The method of claim 1, wherein metadata for the call volume data includes an indication of tide level.

### Claim 8

A system comprising:
a processor, and
a memory, wherein the memory stores instructions executable by the processor to:
access call volume data that includes time period data for one or more telephony servers; and
input the call volume data to a machine learning model to obtain a prediction of a call volume for the one or more telephony servers, wherein the machine learning model includes embedding functions that are applied to the time period data, wherein the embedding functions are implemented with a set of parameters that control functions applied to a discrete variable of the time period data to map the discrete variable to a vector space.

### Claim 9 (depends on 8)

The system of claim 8, wherein the embedding functions have been trained using a sequence of network traffic measurements other than call volume measurements.

### Claim 10 (depends on 8)

The system of claim 8, wherein the embedding functions have been trained using a sequence of website traffic measurements.

### Claim 11 (depends on 8)

The system of claim 8, wherein the time period data includes at least one of a day of a week, an hour of the day, or day of a month for the call volume data.

### Claim 12 (depends on 8)

The system of claim 8, wherein the memory stores instructions executable by the processor to:
activate one or more telephony servers based on the prediction of the call volume for the one or more telephony servers.

### Claim 13 (depends on 8)

The system of claim 8, wherein metadata for the call volume data includes an indication of weather.

### Claim 14 (depends on 8)

The system of claim 8, wherein metadata for the call volume data includes an indication of tide level.

### Claim 15

A non-transitory computer-readable storage medium, comprising executable instructions that, when executed by a processor, facilitate performance of operations, comprising:
accessing call volume data that includes time period data fora unified communications as a service platform;
inputting the call volume data to a machine learning model to obtain a prediction of a call volume for the unified communications as a service platform, wherein the machine learning model includes embedding functions that are applied to the time period data, wherein the embedding functions are implemented with a set of parameters that control functions applied to a discrete variable of the time period data to map the discrete variable to a vector space; and
activating one or more telephony servers based on the prediction of the call volume for the unified communications as a service platform.

### Claim 16 (depends on 15)

The non-transitory computer-readable storage medium of claim 15, wherein the embedding functions have been trained using a sequence of network traffic measurements other than call volume measurements.

### Claim 17 (depends on 15)

The non-transitory computer-readable storage medium of claim 15, wherein the embedding functions have been trained using a sequence of website traffic measurements.

### Claim 18 (depends on 15)

The non-transitory computer-readable storage medium of claim 15, wherein the time period data includes at least one of a day of a week, an hour of the day, or day of a month for the call volume data.

### Claim 19 (depends on 15)

The non-transitory computer-readable storage medium of claim 15, wherein metadata for a the call volume data includes an indication of weather.

### Claim 20 (depends on 15)

The non-transitory computer-readable storage medium of claim 15, wherein metadata for the call volume data includes an indication of tide level.
