Call volume prediction
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.
1. A method comprising:
accessing a sequence of call volume measurements, wherein each of the call volume measurements is associated with respective metadata, and wherein the respective metadata provides information regarding a time period during which a call volume measurement was made; and
inputting a window of the sequence of call volume measurements with the respective metadata 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, wherein the embedding functions are implemented with a set of parameters that control nonlinear functions applied to a discrete variable of the respective metadata for the call volume measurements in the window to map the discrete variable to a continuous vector space.
2. The method of claim 1 , wherein the embedding functions have been trained using a sequence of network traffic measurements other than call volume measurements.
3. The method of claim 1 , wherein the embedding functions have been trained using a sequence of website traffic measurements.
4. The method of claim 1 , wherein the respective metadata for a call volume measurement in the sequence of call volume measurements includes at least one of a day of the week, an hour of the day, or day of the month for the call volume measurement.
5. The method of claim 1 , comprising:
activating one or more telephony servers based on the prediction of the call volume.
6. The method of claim 1 , wherein the respective metadata for a call volume measurement in the sequence of call volume measurements includes an indication of weather.
7. The method of claim 1 , wherein the respective metadata for a call volume measurement in the sequence of call volume measurements includes an indication of tide level.
8. A system comprising:
a processor, and
a memory, wherein the memory stores instructions executable by the processor to:
access a sequence of call volume measurements for one or more telephony servers, wherein each of the call volume measurements is associated with respective metadata, and wherein the respective metadata provides information regarding a time period during which a call volume measurement was made; and
input a window of the sequence of call volume measurements with the respective metadata 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 respective metadata for the call volume measurements in the window, wherein the embedding functions are implemented with a set of parameters that control nonlinear functions applied to a discrete variable of the respective metadata for the call volume measurements in the window to map the discrete variable to a continuous vector space.
9. The system of claim 8 , wherein the embedding functions have been trained using a sequence of network traffic measurements other than call volume measurements.
10. The system of claim 8 , wherein the embedding functions have been trained using a sequence of website traffic measurements.
11. The system of claim 8 , wherein the respective metadata for a call volume measurement in the sequence of call volume measurements includes at least one of a day of the week, an hour of the day, or day of the month for the call volume measurement.
12. 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.
13. The system of claim 8 , wherein the respective metadata for a call volume measurement in the sequence of call volume measurements includes an indication of weather.
14. The system of claim 8 , wherein the respective metadata for a call volume measurement in the sequence of call volume measurements includes an indication of tide level.
15. A non-transitory computer-readable storage medium, comprising executable instructions that, when executed by a processor, facilitate performance of operations, comprising:
accessing a sequence of call volume measurements for a unified communications as a service platform, wherein each of the call volume measurements is associated with respective metadata, and wherein the respective metadata provides information regarding a time period during which a call volume measurement was made;
inputting a window of the sequence of call volume measurements with the respective metadata 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 respective metadata for the call volume measurements in the window, wherein the embedding functions are implemented with a set of parameters that control nonlinear functions applied to a discrete variable of the respective metadata for the call volume measurements in the window to map the discrete variable to a continuous 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.
16. 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.
17. The non-transitory computer-readable storage medium of claim 15 , wherein the embedding functions have been trained using a sequence of website traffic measurements.
18. The non-transitory computer-readable storage medium of claim 15 , wherein the respective metadata for a call volume measurement in the sequence of call volume measurements includes at least one of a day of the week, an hour of the day, or day of the month for the call volume measurement.
19. The non-transitory computer-readable storage medium of claim 15 , wherein the respective metadata for a call volume measurement in the sequence of call volume measurements includes an indication of weather.
20. The non-transitory computer-readable storage medium of claim 15 , wherein the respective metadata for a call volume measurement in the sequence of call volume measurements includes an indication of tide level.