Methods, systems, and media for identifying automatically refreshed advertisements

1. A method for modifying content spending, the method comprising:

receiving, using a hardware processor, content delivery information associated with a plurality of content items displayed on a web page;

generating, using the hardware processor, a distribution of an amount of time that the plurality of content items were displayed on the web page using the content delivery information, wherein a continuous probability distribution is fitted to the generated distribution;

automatically identifying, without user intervention, using the hardware processor, a statistically significant deviation within the generated distribution;

determining, using the hardware processor, whether the statistically significant deviation correlates to an automatic refresh command performed by one or more of the browser applications by monitoring the content delivery information following the statistically significant deviation;

determining, using the hardware processor, whether a number of views correlated with the automatic refresh command is greater than an auto-refresh threshold value; and

inhibiting, using the hardware processor, one or more of the plurality of content items from being presented on the web page in response to determining that the number of views correlated with the automatic refresh command is greater than the auto-refresh threshold value.

2. The method of claim 1 , wherein the content delivery information is received from monitoring code that tracks advertisement delivery and that is loaded on browser applications executing on a plurality of client devices on which the plurality of content items were presented.

3. The method of claim 1 , wherein the statistically significant deviations indicates a sharp increase in the number of views of the plurality of content items displayed on the web page and wherein a portion of the sharp increase of the statistically significant deviation is determined to correlate to the automatic refresh command.

4. The method of claim 1 , further comprising:

presenting an interface for receiving (i) the web page on which the plurality of content items associated with an content provider device are to be displayed and (ii) a content budget designated for displaying the plurality of content item on the web page; and

receiving information identifying the web page and the content budget via the interface.

5. The method of claim 1 , wherein the content delivery information includes a content load time and a content unload time.

6. The method of claim 1 , wherein generating the distribution further comprises:

calculating content display data based on the content delivery information; and

fitting a Weibull distribution to content display data associated with the web page.

7. The method of claim 1 , further comprising extracting pattern features from the generated distribution.

8. The method of claim 7 , further comprising determining whether at least one of the extracted pattern features is a sharp peak in the number of content items that were displayed on the web page over a short duration.

9. The method of claim 8 , further comprising determining whether the sharp peak is followed by a significantly smaller number of content items that were displayed on the web page over a longer duration.

10. The method of claim 8 , further comprising determining that a first portion of the sharp peak corresponds to the automatic refresh command and a second portion of the sharp peak corresponds to user commands to the one or more browser applications.

11. A system for modifying content spending, the system comprising:

a hardware processor that is configured to:

receive content delivery information associated with a plurality of content items displayed on a web page;

generate a distribution of an amount of time that the plurality of content items were displayed on the web page using the content delivery information, wherein a continuous probability distribution is fitted to the generated distribution;

automatically identify, without user intervention, a statistically significant deviation within the generated distribution;

determine whether the statistically significant deviation correlates to an automatic refresh command performed by one or more of the browser applications by monitoring the content delivery information following the statistically significant deviation;

determine whether a number of views correlated with the automatic refresh command is greater than an auto-refresh threshold value; and

inhibit one or more of the plurality of content items from being presented on the web page in response to determining that the number of views correlated with the automatic refresh command is greater than the auto-refresh threshold value.

12. The system of claim 11 , wherein the content delivery information is received from monitoring code that tracks advertisement delivery and that is loaded on browser applications executing on a plurality of client devices on which the plurality of content items were presented.

13. The system of claim 11 , wherein the statistically significant deviations indicates a sharp increase in the number of views of the plurality of content items displayed on the web page and wherein a portion of the sharp increase of the statistically significant deviation is determined to correlate to the automatic refresh command.

14. The system of claim 11 , wherein the hardware processor is further configured to:

present an interface for receiving (i) the web page on which the plurality of content items associated with an content provider device are to be displayed and (ii) a content budget designated for displaying the plurality of content item on the web page; and

receive information identifying the web page and the content budget via the interface.

15. The system of claim 11 , wherein the content delivery information includes a content load time and a content unload time.

16. The system of claim 11 , wherein generating the distribution further comprises:

calculating content display data based on the content delivery information; and

fitting a Weibull distribution to content display data associated with the web page.

17. The system of claim 11 , wherein the hardware processor is further configured to extract pattern features from the generated distribution.

18. The system of claim 17 , wherein the hardware processor is further configured to determine whether at least one of the extracted pattern features is a sharp peak in the number of content items that were displayed on the web page over a short duration.

19. The system of claim 18 , wherein the hardware processor is further configured to determine whether the sharp peak is followed by a significantly smaller number of content items that were displayed on the web page over a longer duration.

20. The system of claim 18 , wherein the hardware processor is further configured to determine that a first portion of the sharp peak corresponds to the automatic refresh command and a second portion of the sharp peak corresponds to user commands to the one or more browser applications.

21. A non-transitory computer-readable medium containing computer-executable instructions that, when executed by a processor, cause the processor to perform a method for content spending, the method comprising:

receiving content delivery information associated with a plurality of content items displayed on a web page;

generating a distribution of an amount of time that the plurality of content items were displayed on the web page using the content delivery information, wherein a continuous probability distribution is fitted to the generated distribution;

automatically identifying, without user intervention, a statistically significant deviation within the generated distribution;

determining whether the statistically significant deviation correlates to an automatic refresh command performed by one or more of the browser applications by monitoring the content delivery information following the statistically significant deviation;

determining whether a number of views correlated with the automatic refresh command is greater than an auto-refresh threshold value; and

inhibiting one or more of the plurality of content items from being presented on the web page in response to determining that the number of views correlated with the automatic refresh command is greater than the auto-refresh threshold value.