Systems and methods for suggesting creative types for online content items to an advertiser

1. A computer-implemented method for generating creative type suggestions for an online content provider, said method using a computing device including a processor and a memory, said method comprising:

training a first model using (i) historical serving performance data of online content items and (ii) online content provider information;

determining a preliminary plurality of creative types for suggestion to the online content provider using at least the first model, each creative type of the preliminary plurality of creative types associated with one or more respective content asset types;

training a second model using adoption rates of previously suggested creative types and the online content provider information;

comparing, using the second model, content asset types associated with each of the preliminary plurality of creative types to content assets provided in the online content provider information;

modifying, using the second model, the preliminary plurality of creative types for suggestion to the online content provider based on (i) the comparing of content asset types associated with each of the preliminary plurality of creative types to content assets provided in the online content provider information and (ii) adoption rates of previously suggested creative types to generate a final creative type suggestion; and

presenting the final creative type suggestion to the online content provider.

2. The method of claim 1 , wherein determining a preliminary plurality of creative types for suggestion to the online content provider further includes generating, using the first model, one or more of (i) a frequency of use factor associated with the preliminary plurality of creative types or (ii) a performance factor associated with the preliminary plurality of creative types.

3. The method of claim 1 , wherein determining a preliminary plurality of creative types for suggestion to the online content provider further includes weighting two or more of a frequency of use factor, a performance factor, or a pre-defined creative type to generate the preliminary plurality of creative types for suggestion to the online content provider.

4. The method of claim 1 , wherein the preliminary plurality of creative types for suggestion to the online content provider includes suggestion values associated with the preliminary plurality of creative types, and wherein modifying the preliminary plurality of creative types for suggestion to the online content provider includes modifying the suggestion values.

5. The method of claim 1 further comprising generating an online content provider profile for each historical online content provider from a plurality of historical online content providers, thereby generating a plurality of online content provider profiles, wherein training the second model using online content provider information further includes training the second model using the plurality of online content provider profiles.

6. The method of claim 1 further comprising generating an online content provider profile for each historical online content provider from a plurality of historical online content providers, thereby generating a plurality of online content provider profiles, wherein training the first model using the online content provider information includes training the first model using at least the plurality of online content provider profiles.

7. A computer system comprising:

an output device;

at least one memory; and

one or more processors configured to:

train a first model using (i) historical serving performance data of online content items advertisement and (ii) online content provider information;

determine a preliminary plurality of creative types for suggestion to an online content provider using at least the first model, each creative type of the preliminary plurality of creative types associated with one or more respective content asset types;

train a second model using adoption rates of previously suggested creative types and the online content provider information;

compare, using the second model, content asset types associated with each of the preliminary plurality of creative types to content assets provided in the online content provider information;

modify, using the second model, the preliminary plurality of creative types for suggestion to the online content provider based on (i) the comparing of content asset types associated with each of the preliminary plurality of creative types to content assets provided in the online content provider information and (ii) the adoption rates of previously suggested creative types to generate a final creative type suggestion; and

present the final creative type suggestion to the online content provider.

8. The computer system of claim 7 , wherein said at least one processor is further configured to generate, using the first model, one or more of (i) a frequency of use factor associated with the preliminary plurality of creative types or (ii) a performance factor associated with the preliminary plurality of creative types.

9. The computer system of claim 7 , wherein said at least one processor is further configured to weight two or more of a frequency of use factor, a performance factor, and a pre-defined creative type to generate the preliminary plurality of creative types for suggestion to the online content provider.

10. The computer system of claim 7 , wherein the preliminary plurality of creative types for suggestion to the online content provider includes suggestion values associated with the preliminary plurality of creative types, and wherein modifying the preliminary plurality of creative types for suggestion to the online content provider includes changing the suggestion values.

11. The computer system of claim 7 , wherein said at least one processor is further configured to:

generate an online content provider profile for each historical online content provider from a plurality of historical online content providers, thereby generating a plurality of online content provider profiles; and

train the second model using the plurality of online content provider profiles.

12. The computer system of claim 7 , wherein said at least one processor is further configured to:

generate an online content provider profile for each historical online content provider from a plurality of historical online content providers, thereby generating a plurality of online content provider profiles; and

train the first model using at least the plurality of online content provider profiles.

13. Non-transitory computer-readable storage media having computer-executable instructions embodied thereon, wherein, when executed by at least one processor, the computer-executable instructions cause the at least one processor to:

train a first model using (i) historical serving performance data of online content items and (ii) online content provider information;

determine a preliminary plurality of creative types for suggestion to an online content provider using at least the first model, each creative type of the preliminary plurality of creative types associated with one or more respective content asset types;

train a second model using adoption rates of previously suggested creative types and the online content provider information;

compare, using the second model, content asset types associated with each of the preliminary plurality of creative types to content assets provided in the online content provider information;

modify, using the second model, the preliminary plurality of creative types for suggestion to the online content provider based on (i) the comparing of content asset types associated with each of the preliminary plurality of creative types to content assets provided in the online content provider information and (ii) the adoption rates of previously suggested creative types to generate a final creative type suggestion; and

present the final creative type suggestion to the online content provider.

14. The computer program product of claim 13 , wherein the computer-executable instructions further cause the at least one processor to generate, using the first model, one or more of (i) a frequency of use factor associated with the preliminary plurality of creative types or (ii) a performance factor associated with the preliminary plurality of creative types.

15. The computer program product of claim 13 , wherein the computer-executable instructions further cause the at least one processor to weight two or more of a frequency of use factor, a performance factor, or a pre-defined type suggestion to generate the preliminary plurality of creative types for suggestion to the online content provider.

16. The computer program product of claim 13 , wherein the preliminary plurality of creative types for suggestion to the online content provider includes suggestion values associated with the preliminary plurality of creative types, and wherein modifying the preliminary plurality of creative types for suggestion to online content provider includes changing the suggestion values.

17. The computer program product of claim 13 , wherein the computer-executable instructions further cause the at least one processor to:

generate an online content provider profile for each historical online content provider from a plurality of historical online content providers, thereby generating a plurality of online content provider profiles; and

train the second model with the plurality of online content provider profiles.

18. The computer program product of claim 13 , wherein the computer-executable instructions further cause the at least one processor to generate an online content provider profile for each historical online content provider from a plurality of historical online content providers, thereby generating a plurality of online content provider profiles, and wherein training the first model using the online content provider information includes training the first model using at least the plurality of online content provider profiles.