Managing computing resources by predicting resource usage

A method comprising:

• identifying, by a processor, a first resource having a first attribute with a first value;
• identifying, by the processor, a second resource that is selectable as an alternative to the first resource and having a second attribute with a second value that is lower than the first value of the first attribute;
• determining, by the processor, a configuration parameter based on at least one of (i) historical data that indicates a number of times the second resource was selected instead of the first resource, or (ii) a predetermined probability indicating a rate in which the first resource was used or another rate in which the first resource was not used after being selected, the configuration parameter indicating a likelihood that the second resource will be selected over the first resource;
• predicting, by the processor and using a predetermined model, a first amount of interest in the first resource based on the first value, the second value, and the configuration parameter;
• predicting, by the processor and using the predetermined model, a second amount of interest for the second resource based on the first value, the second value, the configuration parameter, an elasticity parameter representing how sensitive a base-interest parameter is to a change in another parameter on which the base-interest parameter depends, and a reference value for the first attribute and the second attribute;
• and
• transmitting, by the processor, an electronic communication indicating the predicted first amount of interest in the first resource and the predicted second amount of interest in the second resource.

A system comprising:

• a processing device;
• and
• a memory device in which instructions executable by the processing device are stored for causing the processing device to:
• identify a first resource having a first attribute with a first value;
• identify a second resource that is selectable as an alternative to the first resource and having a second attribute with a second value that is lower than the first value of the first attribute;
• determine a configuration parameter based on at least one of (i) historical data that indicates a number of times the second resource was selected instead of the first resource, or (ii) a predetermined probability indicating a rate in which the first resource was used or another rate in which the first resource was not used after being selected, the configuration parameter indicating a likelihood that the second resource will be selected over the first resource;
• predict, using a predetermined model, a first amount of interest in the first resource based on the first value, the second value, and the configuration parameter;
• predict, using the predetermined model, a second amount of interest for the second resource based on the first value, the second value, the configuration parameter, an elasticity parameter representing how sensitive a base-interest parameter is to a change in another parameter on which the base-interest parameter depends, and a reference value for the first attribute and the second attribute;
• and
• transmit an electronic communication indicating the predicted first amount of interest in the first resource and the predicted second amount of interest in the second resource.