# US 12,114,029 B1 · Systems and methods of personifying viewership data > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 12,114,029 B1 - Document ID: US12114029B1 - Status: Granted - Kind: B1 - Application number: 18/196,404 - Filed: 2023-05-11 - Granted: 2024-10-08 - Earliest claimed: 2023-05-11 - Assignees: comScore, Inc. - Applicants: Comscore, Inc. - Inventors: Michael J. Vinson - Web page: https://patentgrep.com/patent/US12114029B1 ## CPC - H04N 21/252: {Processing of multiple end-users' preferences to derive collaborative data} - G06N 3/0895: Weakly supervised learning, e.g. semi-supervised or self-supervised learning - G06N 20/00: Machine learning - H04N 21/2407: {Monitoring of transmitted content, e.g. distribution time, number of downloads (arrangements for monitoring programmes for broadcast or for distribution combined with broadcast H04H20/14)} - H04N 21/25883: {being end-user demographical data, e.g. age, family status or address (arrangements for identifying locations of users in broadcast systems H04H60/52)} - H04N: PICTORIAL COMMUNICATION, e.g. TELEVISION - G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS ## Abstract A method may include receiving training data including tuning data, household member data, and initial person level distributions. The method may further include aggregating the tuning data from one or more user devices associated with a household to generate an observed household distribution, and calculating, via a prediction model, an implied household viewership distribution based on the person level distribution associated with one or more members of the household. The method may further include comparing the implied household distribution to the observed household distribution of the household, adjusting the prediction model and/or the person level distributions such that the implied household distribution more closely aligns with the observed household distribution, and generating a report with the person level distributions. A system and a non-transitory computer-readable medium may perform the method. ## Claims ### Claim 1 A method comprising: receiving training data including tuning data, household member data, and initial person level viewership distribution data; aggregating the tuning data from one or more user devices associated with a household to generate an observed household viewership distribution; calculating, via a prediction model, an implied household viewership distribution based on the initial person level viewership distribution data associated with one or more members of the household; comparing the implied household viewership distribution to the observed household viewership distribution; adjusting, based on the comparison of the implied household viewership distribution to the observed household viewership distribution, the prediction model and/or the initial person level viewership distribution data such that the implied household viewership distribution more closely aligns with the observed household viewership distribution; calculating an updated implied household viewership distribution based on the adjusted prediction model and/or the adjusted person level viewership distribution data; comparing the updated implied household viewership distribution to the observed household viewership distribution; adjusting, based on the comparison of the updated implied household viewership distribution to the observed household viewership distribution, the adjusted person level viewership distribution data such that the updated implied household viewership distribution more closely aligns with the observed household viewership distribution; and generating a report with the readjusted person level viewership distribution data. ### Claim 2 (depends on 1) The method of claim 1, wherein the initial person level viewership distribution data is based on demographics including at least one of age or gender. ### Claim 3 (depends on 1) The method of claim 1, further comprising normalizing the tuning data over a predetermined period of time to determine a percentage of time that the household views a content based on the total number of household viewing hours over the predetermined period of time. ### Claim 4 (depends on 1) The method of claim 1, wherein the observed household viewership distribution is independent of the temporal order of the tuning data. ### Claim 5 (depends on 1) The method of claim 1, further comprising: monitoring, via panelist software executing on a user device and/or a receiver, viewing events displayed by the user device and/or the receiver for a panelist household; and generating the tuning data based on the viewing events. ### Claim 6 (depends on 1) The method of claim 1, wherein the prediction model is based on n-way co-viewership, where n is the number of members of the household. ### Claim 7 (depends on 1) The method of claim 1, wherein the initial person level viewership distribution data is based on one or more of household data with a single member, survey data, and/or panel data. ### Claim 8 (depends on 1) The method of claim 1, further comprising selecting the prediction model from among a plurality of prediction models based on a number of members in the household and/or a demographic of at least one of the members. ### Claim 9 (depends on 1) The method of claim 1, wherein the prediction model is used to calculate the implied household viewership distribution according to a machine-learning technique. ### Claim 10 A system comprising: at least one processor; at least one memory comprising instructions, which when executed by the at least one processor, cause the at least one processor to: receive training data including tuning data, household member data, and initial person level viewership distribution data; aggregate the tuning data from one or more user devices associated with a household to generate an observed household viewership distribution; calculate, via a prediction model, an implied household viewership distribution based on the initial person level viewership distribution data associated with one or more members of the household; compare the implied household viewership distribution to the observed household viewership distribution; adjust, based on the comparison of the implied household viewership distribution to the observed household viewership distribution, the prediction model and/or the initial person level viewership distribution data such that the implied household viewership distribution more closely aligns with the observed household viewership distribution; calculate an updated implied household viewership distribution based on the adjusted prediction model and/or the adjusted person level viewership distribution data; compare the updated implied household viewership distribution to the observed household viewership distribution; adjust, based on the comparison of the updated implied household viewership distribution to the observed household viewership distribution, the adjusted person level viewership distribution data such that the updated implied household viewership distribution more closely aligns with the observed household viewership distribution; and generate a report with the readjusted person level viewership distribution data. ### Claim 11 (depends on 10) The system of claim 10, wherein the initial person level viewership distribution data is based on demographics including age and/or gender. ### Claim 12 (depends on 10) The system of claim 10, wherein the instructions, when executed, further cause the at least one processor to normalize the tuning data over a predetermined period of time to determine a percentage of time that the household views a content based on the total number of household viewing hours over the predetermined period of time. ### Claim 13 (depends on 10) The system of claim 10, further comprising a user device and/or receiver including panelist software configured to monitor viewing events displayed by the user device and/or the receiver of a panelist household, wherein the instructions, when executed, further cause the at least one processor to generate the tuning data based on the viewing events. ### Claim 14 (depends on 10) The system of claim 10, wherein the initial person level viewership distribution data is based on one or more of household data with a single member, survey data, and/or panel data. ### Claim 15 (depends on 10) The system of claim 10, wherein the instructions, when executed, further cause the at least one processor to select the prediction model from among a plurality of prediction models based on a number of members in the household and/or a demographic of at least one of the members. ### Claim 16 A non-transitory computer-readable medium storing instructions which, when executed by at least one processor, cause the at least one processor to: receive training data including tuning data, household member data, and initial person level viewership distribution data; aggregate the tuning data from one or more user devices associated with a household to generate an observed household viewership distribution; calculate, via a prediction model, an implied household viewership distribution based on the initial person level viewership distribution data associated with one or more members of the household; compare the implied household viewership distribution to the observed household viewership distribution; adjust, based on the comparison of the implied household viewership distribution to the observed household viewership distribution, the prediction model and/or the initial person level viewership distribution data such that the implied household viewership distribution more closely aligns with the observed household viewership distribution; calculate an updated implied household viewership distribution based on the adjusted prediction model and/or the adjusted person level viewership distribution data; compare the updated implied household viewership distribution to the observed household viewership distribution; adjust, based on the comparison of the updated implied household viewership distribution to the observed household viewership distribution, the adjusted person level viewership distribution data such that the updated implied household viewership distribution more closely aligns with the observed household viewership distribution; and generate a report with the readjusted person level viewership distribution data. ### Claim 17 (depends on 16) The non-transitory computer-readable medium of claim 16, wherein the initial person level viewership distribution data is based on demographics including age and/or gender. ### Claim 18 (depends on 16) The non-transitory computer-readable medium of claim 16, wherein the instructions, when executed, further cause the at least one processor to normalize the tuning data over a predetermined period of time to determine a percentage of time that the household views a content based on the total number of household viewing hours over the predetermined period of time. ### Claim 19 (depends on 16) The non-transitory computer-readable medium of claim 16, wherein the initial person level viewership distribution data is based on one or more of household data with a single member, survey data, and/or panel data. ### Claim 20 (depends on 16) The non-transitory computer-readable medium of claim 16, wherein the instructions, when executed, further cause the at least one processor to select the prediction model from among a plurality of prediction models based on a number of members in the household and/or a demographic of at least one of the members.