Methods and systems for determining reach information

1. A method, comprising:

using a processor to obtain campaign spot plans, each campaign spot plan comprising multiple campaign spots;

using the processor to generate simulated campaign spot plans utilizing campaign structure attributes of the campaign spot plans;

analyzing, using machine learning with the processor, duplicated watching habits of viewers viewing patterns for a combination of three or more campaign spots in a designated market using the campaign spot plans, the simulated campaign spot plans, and panel data comprising observations derived over time for the viewers;

applying, using the machine learning with the processor, the campaign spot plans to single viewer data to calculate campaign spot plan Television Average Ratings Points (TARP) viewership pattern information, the TARP viewership pattern information calculated using feature vectors and control parameters to determine a reach increase after a new campaign spot is added to a campaign, the feature vectors comprising: market information, station information, daypart information, time information, day information, program information, length information, rate information, or ratings information, or any combination thereof;

translating, using the machine learning with the processor, the TARP viewership pattern information into reach information, the reach information comprising a number of viewers uniquely exposed to each campaign spot;

determining, using the processor, cumulative reach information comprising the reach information for all campaign spots in the campaign spot plan, wherein the cumulative reach information predicts a number of viewers uniquely exposed to all campaign spots in the campaign spot plan;

iteratively updating and optimizing the machine learning used with the processor; and

displaying, using the processor, the cumulative reach information.

2. The method of claim 1 , wherein the reach information is determined using an inclusion-exclusion principal comprising intersection information between campaign spots.

3. The method of claim 1 , wherein the reach information comprises a cumulative reach calculation for every step of the campaign spot plan.

4. The method of claim 1 , wherein a normalization reach is used so that the machine learning can train using similar campaigns to improve learning the TARP viewership pattern information.

5. The method of claim 1 , further comprising: determining frequency information using the TARP viewership pattern information and the reach information.

6. The method of claim 1 , wherein grid-searches are used to find hyper-parameters for the model.

7. The method of claim 1 , wherein the simulated campaigns have machine-learned campaign patterns learned from real life campaign patterns.

8. The method of claim 1 , wherein a campaign spot comprises a program and/or a promotion.

9. The method of claim 1 , wherein the cumulative reach information predicts the number of new and repeating viewers exposed to all campaign spots in the campaign spot plan.

10. The method of claim 1 , wherein the TARP pattern information comprises duplicated watching habit information.

11. The method of claim 10 , wherein the duplicated watching habit information comprises: watching the same time of day; watching the same type of content; or watching the same channels; or any combination thereof.

12. A system, comprising:

a processor configured for:

obtaining campaign spot plans, each campaign spot plan comprising multiple campaign spots;

generating simulated campaign spot plans utilizing campaign structure attributes of the campaign spot plans;

analyzing, using machine learning, duplicated watching habits of viewers viewing patterns for a combination of three or more campaign spots in a designated market using the campaign spot plans, the simulated campaign spot plans, and panel data comprising observations derived over time for the viewers;

applying, using the machine learning, the campaign spot plans to single viewer data to calculate campaign spot plan Television Average Ratings Points (TARP) viewership pattern information, the TARP viewership pattern information calculated using feature vectors and control parameters to determine a reach increase after a new campaign spot is added to a campaign, the feature vectors comprising: market information, station information, daypart information, time information, day information, program information, length information, rate information, or ratings information, or any combination thereof;

translating using the machine learning, the TARP viewership pattern information into reach information, the reach information comprising a number of viewers uniquely exposed to each campaign spot;

determining, using the processor, cumulative reach information comprising the reach information for all campaign spots in the campaign spot plan, wherein the cumulative reach information predicts a number of viewers uniquely exposed to all campaign spots in the campaign spot plan;

iteratively updating and optimizing the machine learning used with the processor; and

displaying the cumulative reach information.

13. The system of claim 12 , wherein the reach information is determined using an inclusion-exclusion principal comprising intersection information between campaign spots.

14. The system of claim 12 , wherein the reach information comprises a cumulative reach calculation for every step of the campaign spot plan.

15. The system of claim 12 , wherein a normalization reach is used so that the machine learning can train using similar campaigns to improve learning the TARP viewership pattern information.

16. The system of claim 12 , wherein the processor is configured for: determining frequency information using the TARP viewership pattern information and the reach information.

17. The system of claim 12 , wherein grid-searches are used to find hyper-parameters for the model.

18. The system of claim 12 , wherein the simulated campaigns have machine-learned campaign patterns learned from real life campaign patterns.

19. The system of claim 12 , wherein a campaign spot comprises a program and/or a promotion.

20. The system of claim 12 , wherein the cumulative reach information predicts the number of new and repeating viewers exposed to all campaign spots in the campaign.

21. The system of claim 12 , wherein the TARP pattern information comprises duplicated watching habit information.

22. The system of claim 21 , wherein the duplicated watching habit information comprises: watching the same time of day; watching the same type of content; or watching the same channels; or any combination thereof.