WO2020152710A1 - A system and method to attribute household level viewership to individuals of the household. - Google Patents
A system and method to attribute household level viewership to individuals of the household. Download PDFInfo
- Publication number
- WO2020152710A1 WO2020152710A1 PCT/IN2020/050071 IN2020050071W WO2020152710A1 WO 2020152710 A1 WO2020152710 A1 WO 2020152710A1 IN 2020050071 W IN2020050071 W IN 2020050071W WO 2020152710 A1 WO2020152710 A1 WO 2020152710A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- data
- polled
- watermarked
- donor
- channel
- Prior art date
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04H—BROADCAST COMMUNICATION
- H04H60/00—Arrangements for broadcast applications with a direct linking to broadcast information or broadcast space-time; Broadcast-related systems
- H04H60/29—Arrangements for monitoring broadcast services or broadcast-related services
- H04H60/33—Arrangements for monitoring the users' behaviour or opinions
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04H—BROADCAST COMMUNICATION
- H04H2201/00—Aspects of broadcast communication
- H04H2201/50—Aspects of broadcast communication characterised by the use of watermarks
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04H—BROADCAST COMMUNICATION
- H04H60/00—Arrangements for broadcast applications with a direct linking to broadcast information or broadcast space-time; Broadcast-related systems
- H04H60/35—Arrangements for identifying or recognising characteristics with a direct linkage to broadcast information or to broadcast space-time, e.g. for identifying broadcast stations or for identifying users
- H04H60/45—Arrangements for identifying or recognising characteristics with a direct linkage to broadcast information or to broadcast space-time, e.g. for identifying broadcast stations or for identifying users for identifying users
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04H—BROADCAST COMMUNICATION
- H04H60/00—Arrangements for broadcast applications with a direct linking to broadcast information or broadcast space-time; Broadcast-related systems
- H04H60/35—Arrangements for identifying or recognising characteristics with a direct linkage to broadcast information or to broadcast space-time, e.g. for identifying broadcast stations or for identifying users
- H04H60/47—Arrangements for identifying or recognising characteristics with a direct linkage to broadcast information or to broadcast space-time, e.g. for identifying broadcast stations or for identifying users for recognising genres
Abstract
A method for determining viewing behaviour, in terms of behaviour data parameters, based on a plurality of polled currency data items, using a panel per television, each panel being associated with a donor household, said method comprising: identifying constituents of said currency data; determining whether said currency data is correlative to a watermarked channel or a non-watermarked channel; defining a hierarchical and sub-hierarchical list of pre-defined parameters; sampling said polled currency data items to obtain sampled polled currency data items; determining number of donors; iteratively, selecting at least a donor; and determining viewing behaviour parameters (in terms of content and demographic data of a viewer of said content) based on data from said iteratively selected plurality of donors meeting said pre-defined threshold criteria and from said updated currency data.
Description
TITLE OF THE INVENTION
A SYSTEM AND METHOD TO ATTRIBUTE HOUSEHOLD LEVEL VIEWERSHIP TO
INDIVIDUALS OF THE HOUSEHOLD
This application claims priority from Indian Patent Application No. 201921002782 filed on January 23, 2019.
FIELD OF THE INVENTION:
This invention relates to the field of communications engineering.
Particularly, this invention relates to a system and method for determining viewing behaviour.
BACKGROUND OF THE INVENTION:
Tracking subscriber data enables an entity to source large amounts of audience viewing data through set top boxes or any other device with a return path. Having access to this big data is an incredibly useful and accurate measure of subscriber viewing behaviour.
The accuracy of television viewership measurement system is a big loophole.
Measurement of return path data, in the television broadcast and viewing industry, has the following benefits:
Better subscriber management;
Influence over the advertising ecosystem;
Higher CPMs through better targeting and higher ad revenues;
Financial (in exchange for data);
Ability to report on smaller shows/ networks;
Attribution to sales/ demonstration of ROI;
Superior Targeting: Geo based, Addressable TV;
Addressable TV advertising - superior targeting, lower switching during commercial breaks;
New Ad Sales models like guaranteed delivery;
Increased viewership reported with in-house channels;
Demonstrating better ROI to advertisers.
In certain geographies, there are multiple players ad multiple platforms (such as cable, DTH,OTT, and the like).
A household with a television is defined into multiple classes based on research variables. A cell is a class of such households. In order to eliminate or minimize instances of‘zero cell’, sample size needs to be increased. This scaling up is a function of calibration and is a challenge that needs to be addressed.
According to prior art, data collected from television device meters are at individual level viewing, where a user select a code through a remote and then his or her viewing is tracked. It is important to obtain this data at household level viewing.
There is a need to add lifestyle and demographic values to prior art data.
OBJECTS OF THE INVENTION;
An object of the invention is to provide a system and method to convert household level viewing into individual viewing.
Another object of the invention is to provide a system and method to assign viewers to a household level channel viewing.
Yet another object of the invention is to provide a system and method to add lifestyle and demographic values to prior art data.
Still another object of the invention is to provide a system and method for providing tamperproof TV viewership data.
SUMMARY OF THE INVENTION;
According to this invention, there is provided a method for determining viewing behaviour, in terms of behaviour data parameters, based on a plurality of polled currency data items, using a panel per television, each panel being associated with a donor household, said method comprising:
identifying constituents of said currency data;
determining whether said currency data is correlative to a watermarked channel or a non- watermarked channel;
defining a hierarchical list of pre-defined parameters, correlative to said currency data, per recipient household;
defining a sub- hierarchical list for each of said pre-defined parameters of said defined hierarchical list;
sampling said polled currency data items, said sampling being defined in terms of a collapsible hierarchical model and a correlative sub-hierarchical model to obtain sampled polled currency data items;
determining number of donors based on pre-defined thresholds of said constituents of said currency data, based on said sampled polled currency data items;
iteratively, selecting at least a donor, as a combination of at least an item from said defined sub- hierarchical list and at least an item from said hierarchical list in order to, firstly, satisfy a minimum number of similar data items and, secondly, in order to, obtain updated currency data from said selected donor upon satisfaction of said minimum number of similar data items correlative to said sampled polled currency data items; and
determining viewing behaviour parameters (in terms of content and demographic data of a viewer of said content) based on data from said iteratively selected plurality of donors meeting said pre-defined threshold criteria and from said updated currency data.
Typically, said currency data comprises viewership details’ data per individual per household.
Typically, in said method, if said polled panel data comprising a non-watermarked channel, said method comprises:
polhng a panel associated per television per household for polled data, said polled data being defined in terms of watermarked channels (“polled watermarked channel panel data”) and non-watermarked channels (“polled non-watermarked channel panel data”);
identifying closeness (ranking donor households in terms of their closeness to their recipient household) of each non-watermarked channel with that of each watermarked channel, for said pre-defined duration based on a closeness factor;
selecting a closely associated watermarked channel per non-watermarked channel, for a pre-defined duration, said selection being based on pre-defined parameters per identified constituents along with said identified closeness factor;
classifying polled watermarked channel panel data in terms of time duration and associating a higher hierarchy to said classified polled watermarked channel panel data;
classifying polled non-watermarked channel panel data in terms of time duration and associating a lower hierarchy to said classified non-polled watermarked channel panel data; associating data from a classified non-watermarked channel with data from said classified watermarked channel, if threshold criteria of currency data are met, in order to obtain currency data per donor of said panel;
iteratively checking if said currency data constitutes pre-defined number of donor households; and
determining viewing behaviour parameters (in terms of content and demographic data of a viewer of said content) based on data from said iteratively selected plurality of donors meeting said pre-defined threshold criteria and from said updated currency data.
Typically, said collapsible sub-hierarchical model comprises data items pertaining to similarity of channel information in terms of type, genre, language and / or similarity in time-band information.
Typically, said sub- hierarchical model comprises data items pertaining to demographic information, town class information, household information in terms of demographics of individuals in said household, and / or household information in terms of economic classification.
Typically, said pre-defined parameters are parameters selected from a group consisting of language parameter, and genre parameter.
Typically, said constituents are selected from a group of constituents consisting of programs, promotion breaks, and advertisement breaks.
Typically, said step of classifying comprises a step of associating hierarchical levels to each of said polled watermarked channel panel data per time duration and said non-polled watermarked channel panel data per time duration.
Typically, said step of classifying comprising a step of associating hierarchical levels to each of said polled watermarked channel panel data per time duration and said non-polled watermarked channel panel data per time duration comprising an additional step of associating each of said hierarchically defined polled watermarked channel panel data per time duration with a geography.
Typically, said step of classifying comprises a step of associating hierarchical levels to each of said polled watermarked channel panel data per time duration and said non-polled watermarked channel panel data per time duration comprising an additional step of associating each of said hierarchically defined polled watermarked channel panel data per time duration with a household size.
Typically, said step of classifying comprises a step of associating hierarchical levels to each of said polled watermarked channel panel data per time duration and said non-polled watermarked channel panel data per time duration comprising an additional step of associating each of said
hierarchically defined non-polled watermarked channel panel data per time duration with a geography.
Typically, said step of classifying comprises a step of associating hierarchical levels to each of said polled watermarked channel panel data per time duration and said non-polled watermarked channel panel data per time duration comprising an additional step of associating each of said hierarchically defined non-polled watermarked channel panel data per time duration with a household size.
BRIEF DESCRIPTION OF THE ACCOMPANYING DRAWINGS:
The invention will now be described in relation to the accompanying drawings, in which:
FIGURE 1 illustrates a mechanism and Video Audience Measurement (VAM) algorithm for non-watermarked channels where play out data is available;
FIGURE 2 illustrates a mechanism and Video Audience Measurement (VAM) algorithm for non-watermarked channels where play out data is not available; and
FIGURE 3 illustrates a mechanism and Video Audience Measurement (VAM) algorithm for watermarked channels where play out data is not available.
DETAILED DESCRIPTION OF THE ACCOMPANYING DRAWINGS:
According to this invention, there is provided a system and method for determining viewing behaviour. Typically, this system and method determines viewing behaviour, in terms of behaviour data parameters, based on a plurality of polled currency data items, using a panel per television, each panel being associated with a donor household.
FIGURE 1 illustrates a mechanism and Video Audience Measurement (VAM) algorithm for non-watermarked channels where play out data is available.
Return Path Data (RPD) viewership data will be at Household level. The objective is to assign viewers to the household level channel viewing for non-watermarked channels for which playout data is available.
As the system does not have RPD household viewership data available with it, the system and method, of this invention, converts individual level viewing of polled panel households (HHs) into household viewing to test the VAM algorithm.
According to an embodiment, the system and method selects 60% of the HH’s as a Recipient HHs (RPD) from the attribution date’s statement file by sorting on State, TC, NCCS, and HH Size. The donors will be all households present on each day of 56 days other than those specific 60% recipient HHs of attribution date.
Step 1: In data polled from a panel, look for the exact Gender X Age group match for each of the recipient individual among individuals of selected donor HH.
Step 2: In case of non-availability of same Gender X Age group for any recipient individual in the donor HH, look for the nearest Age group donor individual WITHIN SAME GENDER of the recipient individual.
Step 3: In case of non-availability of same GENDER in the donor HH for any recipient individual, look for the nearest age group among unassigned donor individuals.
Minutes allocation to Recipient Individuals.
Once, the mechanism and algorithm maps the donor individual to recipient individual, attribute the donor individual viewership (Yes / No) and share of donor individual minutes to donor household minutes to the respective recipient individuals and compute the recipient individual minutes by multiplying the attributed share with recipient household duration on that Channel X Time band X Distinct Programme / Promo / Ad-break.
In at least a first step, the system and method of this invention identifies nearest watermarked channel.
Each non-watermarked channel present in RPD data of a day is mapped to one watermarked channel from same language x genre present in currency data (viewership data) of that day. Mapping is done based on the closeness of the average household minute of non-watermarked
channel with that of watennarked channel. For example, for a given day, a non-watermarked channel“NW1” has average household minutes at 30 minutes and there are four watermarked channels namely“WM1”,“WM2”,“WM3” and“WM4” from same language x genre in the currency data with average minutes as 27 minutes, 35 minutes, 40 minutes, and 20 minutes respectively . In this NW1 is mapped with Wl. Average minutes have to be calculated across 56 days. Based on this average, identified nearest watermarked channel will be fixed as nearest watermarked channel to NW1 for next 56 days. This will ensure that different channel on different days will be identified as nearest channels which will lead to fluctuations of data for NW1.
Typically, the system and method identify constituents of currency data.
Typically, the system and method determine whether said currency data is correlative to a watermarked channel or a non-watermarked channel.
In at least a second step, the system and method of this invention computes HH x Channel X 30 Minute Daypart X Distinct Program/Promo-break/Ad-break.
For each HH x non-watermarked channel x 30 Minute Daypart x Distinct Program/Promo- break/ Ad-break, donor household is searched from the pool of 56 days of currency data for the channel Wl. Hierarchy of search is given below:
In at least an embodiment, at the third step, the system and method selectsdonors for each of Recipient HH x Channel X 30 Minute Daypart X Distinct Program.
Now, for the each of the Recipient HH x Channel x 30 Minute Daypart X Distinct Program cell, the system and method firstcollected the donors from Current Day + Previous 55 Days at same channel, same 30 min Time band and same program. If the collected donor sample from 56 days is less than 15, then the system and method has followed below hierarchy of collapsing levels.
There are four sub-levels within each of the above levels.
Typically, the system and method define a hierarchical list of pre-defined parameters, correlative to said currency data, per recipient household. Additionally, the system and method define a sub- hierarchical list for each of said pre-defined parameters of said defined hierarchical list.
In at least an embodiment, at the third step, the system and method collects the donors for each of Recipient HH x Channel X 30 Minute Daypart X Distinct ad-break.
Now, for the each of the Recipient HH x Channel x 30 Minute Daypart X Distinct Ad-break cell, the system and method firstcollected the donors from Current Day + Previous 55 Days at same channel, same 30 min Time band and same Ad-break. If the collected donor sample from 56 days is less than 15, then the system and method has followed below hierarchy of collapsing levels.
Further, the system and method determine number of donors based on pre-defined thresholds of said constituents of said currency data, based on said sampled polled currency data items.
In at least an embodiment, at the third step, the system and method obtains the donors for each of Recipient HH x Channel X 30 Minute Daypart X Distinct Promo-break.
Now, for the each of the Recipient HH x Channel x 30 Minute Daypart X Distinct Promo cell, the system and method firstcollected the donors from Current Day + Previous 55 Days at same channel, same 30 min Time band and same Promo break. If the collected donor sample from 56 days is less than 15, then the system and method has followed below hierarchy of collapsing levels.
In at least a fourth step, the system and method of this invention computes dataat one single donor for all the recipients - a) Calculated exact match count between each of the donor and the respective recipient with State, TC, NCCS and HH Size. (Critical score -1)
b) Calculated exact match count between donor and the respective recipient with Presence of M 2-14, Presence of M 15-21, Presence of M 22-50, Presence of M 51+, Presence of F 2-14, Presence of F 15-21, Presence of F 22-50, Presence of F 51+ (Critical score -2)
c) Calculated Squared Euclidian Distance between recipient and its respective donor with Highest Education, Presence of each LSAH+LMOS level, Presence of each Working Status level, Presence of Type of Tenement level and HH Duration.
d) Arranged the donors of a respective recipient in the descending order of critical score- 1 and within that, critical score - 2 in the descending order and within that, Donor date from the latest to oldest, and within that calculated distance in the ascending order. Now, Selected the donor which shows highest on critical score- 1, critical score-2, nearest date of the donor from recipient date (attribution day) and least on distance.
In at least a fifth step, the system and method of this inventioncomputesindividual matching for Viewer/Minutes attribution.
Once the selection of a single donor for a recipient out of n number of donors is completed, a) The system and methodlooks for the exact Gender x Age group match for each of the recipient individual among individuals of selected donor HH.
b) In case of non-availability of same Gender x Age group for any recipient individual in the donor HH, the system and methodhas looked for the nearest Age group donor individual WITHIN SAME GENDER of the recipient individual.
c) In case of non-availability of same GENDER in the donor HH for any recipient individual, the system and methodhas looked for the nearest Age group among unassigned donor individuals.
Once the system and methodmatches the recipient individuals with individuals of selected Donor HH, attribute the viewership (Viewer/Non-Viewer) of individuals of selected Donor HH to the respective recipient individuals.
Once the system and methodmaps the donor individual to recipient individual, attribute the donor individual viewership (Yes/No) and share of donor individual minutes to donor HH minutes to
the respective recipient individuals and compute the recipient individual minutes by multiplying the attributed share with recipient household duration on that Channel x Time band x distinct Program/Promo/Ad-break.
Furthermore, the system and method - iteratively select at least a donor, as a combination of at least an item from said defined sub- hierarchical list and at least an item from said hierarchical list in order to, firstly, satisfy a minimum number of similar data items and, secondly, in order to, obtain updated currency data from said selected donor upon satisfaction of said minimum number of similar data items correlative to said sampled polled currency data items.
In at least a sixth step, the system and method of this inventioncomputes Start Time allocation for recipient individuals:
1. For the recipient individuals with attributed minutes same as Recipient HH Duration, assign Start Time as per Recipient HH Start Time.
2. For the recipient individuals with attributed minutes less than Recipient HH Duration,
Step-A: Calculate the difference in minutes between Donor Household Start Time (First minute on that Channel x TB x Program) and respective donor individual Start Time (“A”).
- If A=0 then Recipient Individual Start Time will be same as Recipient HH Start Time. If A is NOT 0 then,
Step-B: Divide“A” with Donor Household duration on that Channel x TB x Program (“B”). Step-C: Now, do ("B" * HH Duration of Recipient HH) +1 to arrive at Starting minute viewing of Recipient individual and add attributed recipient individual minutes to arrive at Ending Time.
Finally, the system and method determine viewing behaviour parameters (in terms of content and demographic data of a viewer of said content) based on data from said iteratively selected plurality of donors meeting said pre-defined threshold criteria and from said updated currency data.
According to a non-limiting exemplary embodiment, in the given example above, the system and methodhas 14 distinct minutes for recipient HH on that Channel x TB x Program. Now, the system and methodhas to figure out which minute of these 14 minutes is the start time for the recipient individual-2.
- The difference in minutes between Donor Household Start Time and respective donor individual Start Time (A) =12
- A / Donor Household duration (B) = 12 / 24 = 0.5
- Recipient HH Duration * B = 14 * 0.5 = 7
- So, do not assign first seven minutes of HH viewing to this individual. 8th minute out of 14 minutes of HH Duration will be recipient individual Start Time. i.e. 8th, 9th, 10th, 11th, 12th, 13th and 14th minutes will be assigned to recipient individual. This logic works for all the collapsing levels.
Particularly, if the polled panel data comprises a non-watermarked channel, the system and method, of this invention, comprises the following steps:
polling a panel associated per television per household for polled data, said polled data being defined in terms of watermarked channels (“polled watermarked channel panel data”) and non-watermarked channels (“polled non-watermarked channel panel data”);
identifying closeness (ranking donor households in terms of their closeness to their recipient household) of each non-watermarked channel with that of each watermarked channel, for said pre-defined duration based on a closeness factor;
selecting a closely associated watermarked channel per non-watermarked channel, for a pre-defined duration, said selection being based on pre-defined parameters per identified constituents along with said identified closeness factor;
classifying polled watermarked channel panel data in terms of time duration and associating a higher hierarchy to said classified polled watermarked channel panel data;
classifying polled non-watermarked channel panel data in terms of time duration and associating a lower hierarchy to said classified non-polled watermarked channel panel data;
associating data from a classified non- watermarked channel with data from said classified watermarked channel, if threshold criteria of currency data are met, in order to obtain currency data per donor of said panel;
iteratively checking if said currency data constitutes pre-defined number of donor households; and
determining viewing behaviour parameters (in terms of content and demographic data of a viewer of said content) based on data from said iteratively selected plurality of donors meeting said pre-defined threshold criteria and from said updated currency data.
FIGURE 2 illustrates a mechanism and Video Audience Measurement (VAM) algorithm for non- watermarked channels where play out data is not available.
Return Path Data (RPD) viewership data will be at Household level. Our objective is to assign viewers to the household level channel viewing for non- watermarked channels for which playout data is not available.
As the system and method does not have RPD household viewership data available with us, the system and method has converted the individual level viewing of BARC panel HHs into household viewing to test the YAM algorithm.
Select 60% of the HH’s as a Recipient HHs (RPD) from the attribution date’s statement file by sorting on State, TC, NCOS and HH Size. The donors will be all households present on each day of 56 days other than those specific 60% recipient HHs of attribution date.
Step 1: Look for the exact Gender X Age group match for each of the recipient individual among individuals of selected donor HH.
Step 2: In case of non-availability of same Gender X Age group for any recipient individual in the donor HH, look for the nearest Age group donor individual WITHIN SAME GENDER of the recipient individual.
Step 3: In case of non-availability of same GENDER in the donor HH for any recipient individual, look for the nearest age group among unassigned donor individuals.
Minutes allocation to Recipient Individuals.
Once, the mechanism and algorithm maps the donor individual to recipient individual, attribute the donor individual viewership (Yes / No) and share of donor individual minutes to donor household minutes to the respective recipient individual minutes by multiplying the attributed share with recipient household duration on that Channel X Time band.
In at least a first step, the system and method of this invention identifies nearest watermarked channel.
Each non-watermarked channel present in RPD data of a day is mapped to one watermarked channel from same language x genre present in currency data of that day. Mapping is done based on the closeness of the average household minute of non-watermarked channel with that of watermarked channel. For example, for a given day, a non-watermarked channel“NWl” has average household minutes at 30 minutes and there are four watermarked channels namely “WM1”,“WM2”,“WM3” and“WM4” from same language x genre in the currency data with average minutes as 27 minutes, 35 minutes, 40 minutes and 20 minutes respectively . In this NWl is mapped with Wl. Average minutes have to be calculated across 56 days. Based on this average, identified nearest watermarked channel will be fixed as nearest watermarked channel to NWl for next 56 days. This will ensure that different channel on different days will be identified as nearest channels which will lead to fluctuations of data for NWl.
In at least a second step, the system and method of this invention computes HH x Channel X 30 Minute Dayparti
For each HH x non-watermarked channel x 30 Minute Daypart, donor household is searched from the pool of 56 days of currency data for the channel Wl. Hierarchy of search is given below:
The system and method has converted the statement file which is at individual channel session level, into HH x Channel X Each Minute level. (i.e. HH x Channel x Distinct minutes). The system and method has mapped the information to HH x Channel x Each Minute. Then, the system and method has aggregated the Minute level data at HH x Channel X 30 Minute Daypart for both Donor HHs and Recipient HHs.
In at least an embodiment, at the third step, the system and method selectsdonors for each of Recipient HH x Channel X 30 Minute Daypart.
Now, for the each of the Recipient HH x Channel x 30 Minute Daypart cell, the system and method firstcollected the donors from Current Day + Previous 55 Days at same channel, same 30 min Time band. If the collected donor sample from 56 days is less than 15, then the system and method has followed below hierarchy of collapsing levels.
Taking these above seven sub-levels into consideration, there will be 28 levels. For all the recipient HH x Channel x 30 Minute Time band cell, the system and method will collect the donors from Current Day + Previous 55 days at Level-1.1. If the system and method does not have a 15-donor sample for any cell, then the system and method will move to Level-1.2 and collect the donors from Current Day + Previous 55 days. The system and method will go from Level 1.1 to Level VII.4 until the system and method hits the 15-donor sample.
In at least an embodiment, at the third step, the system and method computes dataat one single donor for all the recipients - a) Calculated exact match count between each of the donor and the respective recipient with State, TC, NCCS and HH Size. (Critical score -1).
b) Calculated exact match count between donor and the respective recipient with Presence of M 2-14, Presence of M 15-21, Presence of M 22-50, Presence of M 51+, Presence of F 2-14, Presence of F 15-21, Presence of F 22-50, Presence of F 51+ (Critical score -2).
c) Calculated Squared Euclidian Distance between recipient and its respective donor with Highest Education, Presence of each LSAH+LMOS level, Presence of each Working Status level, Presence of Type of Tenement level and HH Duration.
d) Arranged the donors of a respective recipient in the descending order of critical score- 1 and within that, critical score - 2 in the descending order and within that, Donor date from the latest to oldest, and within that calculated distance in the ascending order. Now, Selected the donor which shows highest on critical score- 1, critical score-2, nearest date of the donor from recipient date (attribution day) and least on distance.
In at least a fifth step, the system and method of this invention computesindividual matching for Viewer/Minutes attribution.
Once the selection of a single donor for a recipient out of n number of donors is completed, a) The system and method has looked for the exact Gender x Age group match for each of the recipient individual among individuals of selected donor HH.
b) In case of non-availability of same Gender x Age group for any recipient individual in the donor HH, the system and method has looked for the nearest Age group donor individual WITHIN SAME GENDER of the recipient individual.
c) In case of non-availability of same GENDER in the donor HH for any recipient individual, the system and method has looked for the nearest Age group among unassigned donor individuals.
Once the system and method matches the recipient individuals with individuals of selected Donor HH, attribute the viewership (Viewer/Non-Viewer) of individuals of selected Donor HH to the respective recipient individuals.
Once the system and method maps the donor individual to recipient individual, attribute the donor individual viewership (Yes/No) and share of donor individual minutes to donor HH minutes to the respective recipient individuals and compute the recipient individual minutes by multiplying the attributed share with recipient household duration on that Channel x Time band.
In at least a sixth step, the system and method of this invention computes Start Time allocation for recipient individuals:
1. For the recipient individuals with attributed minutes same as Recipient HH Duration, assign Start Time as per Recipient HH Start Time.
2. For the recipient individuals with attributed minutes less than Recipient HH Duration,
Step-A; Calculate the difference in minutes between Donor Household Start Time (First minute on that Channel x TB) and respective donor individual Start Time (“A”).
- If A=0 then Recipient Individual Start Time will be same as Recipient HH Start Time. If A is NOT 0 then,
Step-B: Divide“A” with Donor Household duration on that Channel x TB (“B”).
Step-C: Now, do ("B" * HH Duration of Recipient HH) +1 to arrive at Starting minute viewing of Recipient individual and add attributed recipient individual minutes to arrive at Ending Time.
According to a non-limiting exemplary embodiment, in the given example above, the system and method has 14 distinct minutes for recipient HH on that Channel x TB. Now, the system and method has to figure out which minute of these 14 minutes is the start time for the recipient individual-2.
- The difference in minutes between Donor Household Start Time and respective donor individual Start Time (A) =12
- A / Donor Household duration (B) = 12 / 24 = 0.5
- Recipient HH Duration * B = 14 * 0.5 = 7
- So, do not assign first seven minutes of HH viewing to this individual. 8th minute out of 14 minutes of HH Duration will be recipient individual Start Time. i.e. 8th, 9th, 10th, 11th, 12th, 13th and 14th minutes will be assigned to recipient individual. This logic works for all the collapsing levels.
FIGURE 3 illustrates a mechanism and Video Audience Measurement (VAM) algorithm for watermarked channels where play out data is not available.
Step 1: Look for the exact Gender X Age group match for each of the recipient individual among individuals of selected donor HH.
Step 2: In case of non-availability of same Gender X Age group for any recipient individual in the donor HH, look for the nearest Age group donor individual WITHIN SAME GENDER of the recipient individual.
Step 3: In case of non-availability of same GENDER in the donor HH for any recipient individual, look for the nearest age group among unassigned donor individuals.
Minutes allocation to Recipient Individuals.
Once, the mechanism and algorithm maps the donor individual to recipient individual, attribute the donor individual viewership (Yes / No) and share of donor individual minutes to donor household minutes to the respective recipient individuals by multiplying the attributed share with
recipient household duration on that Channel X Time band X Distinct Program / Promo / Ad- break.
In accordance with an embodiment of this invention, a hierarchy definition is configured to define hierarchy for donor samples for each of Recipient HH x Channel X 30 Minute Daypart X Distinct Program. According to an exemplary embodiment these hierarchies are as follows:
In accordance with an embodiment of this invention, a hierarchy definition is configured to define hierarchy for donor samples for each of Recipient HH x Channel X 30 Minute Daypart X Distinct Ad-break cell. According to an exemplary embodiment these hierarchies are as follows:
In accordance with an embodiment of this invention, a sub-hierarchy definitionis configured to define sub-hierarchies or sub-levels within the above-mentioned hierarchies or above-mentioned levels for each of Recipient HH x Channel X 30 Minute Daypart X Distinct Program. According to an exemplary embodiment these sub-hierarchies are as follows:
In at least an embodiment, a score determination mechanism is configured to determine score for each donor and respective recipient based on pre-determined criteria.
Mechanism for arriving at one single donor for all the recipients is as follows:
e) Calculate exact match count between each of the donor and the respective recipient with State, TC, NCCS and HH Size. (Critical score -1)
f) Calculate exact match count between donor and the respective recipient with Presence of M 2-14, Presence of M 15-21, Presence of M 22-50, Presence of M 51+, Presence of F 2-14, Presence of F 15-21, Presence of F 22-50, Presence of F 51+ (Critical score -2)
g) Calculate Squared Euclidian Distance between recipient and its respective donor with Highest Education, Presence of each LSAH+LMOS level, Presence of each Working Status level, Presence of Type of Tenement level and HH Duration.
h) Arrange the donors of a respective recipient in the descending order of critical score- 1 and within that, critical score - 2 in the descending order and within that, Donor date from the latest to oldest, and within that calculated distance in the ascending order. Now, Selected the donor which shows highest on critical score- 1, critical score-2, nearest date of the donor from recipient date (attribution day) and least on distance.
Mechanism for individual matching for Viewer/Minutes attribution is as follows:
Once the selection of a single donor for a recipient out of n number of donors is completed, d) The system is configured to look for the exact Gender x Age group match for each of the recipient individual among individuals of selected donor HH.
e) In case of non-availability of same Gender x Age group for any recipient individual in the donor HH, the system looks for the nearest Age group donor individual WITHIN SAME GENDER of the recipient individual.
f) In case of non-availability of same GENDER in the donor HH for any recipient individual, the system looks for the nearest Age group among unassigned donor individuals.
Once the system matches the recipient individuals with individuals of selected Donor HH, attribute the viewership (Viewer/Non-Viewer) of individuals of selected Donor HH to the respective recipient individuals.
Once the system maps the donor individual to recipient individual, attribute the donor individual viewership (Yes/No) and share of donor individual minutes to donor HH minutes to the respective recipient individuals and compute the recipient individual minutes by multiplying the attributed share with recipient household duration on that Channel x Time band x distinct Program/Promo/Ad-break.
Mechanism for start time allocation, for recipient individuals, is as follows:
1. For the recipient individuals with attributed minutes same as Recipient HH Duration, assign Start Time as per Recipient HH Start Time.
2. For the recipient individuals with attributed minutes less than Recipient HH Duration, perform the following steps:
Step-A: Calculate the difference in minutes between Donor Household Start Time (First minute on that Channel x TB x Program) and respective donor individual Start Time (“A”).
- If A=0 then Recipient Individual Start Time will be same as Recipient HH Start Time. If A is NOT 0 then,
Step-B: Divide“A” with Donor Household duration on that Channel x TB x Program (“B”). Step-C: Now, do ("B" * HH Duration of Recipient HH) +1 to arrive at Starting minute viewing of Recipient individual and add attributed recipient individual minutes to arrive at Ending Time.
The TECHNICAL ADVANCEMENTof this invention lies in providing a mechanism and method in order to scale / translate household viewing data to individual-level viewing data without using dedicated meters.
While this detailed description has disclosed certain specific embodiments for illustrative purposes, various modifications will be apparent to those skilled in the art which do not constitute departures from the spirit and scope of the invention as defined in the following claims, and it is to be distinctly understood that the foregoing descriptive matter is to be interpreted merely as illustrative of the invention and not as a limitation.
Claims
1. A method for determining viewing behaviour, in terms of behaviour data parameters, based on a plurality of polled currency data items, using a panel per television, each panel being associated with a donor household, said method comprising:
- identifying constituents of said currency data;
- determining whether said currency data is correlative to a watermarked channel or a non-watermarked channel;
- defining a hierarchical list of pre-defined parameters, correlative to said currency data, per recipient household;
- defining a sub- hierarchical list for each of said pre-defined parameters of said defined hierarchical list;
- sampling said polled currency data items, said sampling being defined in terms of a collapsible hierarchical model and a correlative sub-hierarchical model to obtain sampled polled currency data items;
- determining number of donors based on pre-defined thresholds of said constituents of said currency data, based on said sampled polled currency data items;
- iteratively, selecting at least a donor, as a combination of at least an item from said defined sub- hierarchical list and at least an item from said hierarchical list in order to, firstly, satisfy a minimum number of similar data items and, secondly, in order to, obtain updated currency data from said selected donor upon satisfaction of said minimum number of similar data items correlative to said sampled polled currency data items; and
- determining viewing behaviour parameters (in terms of content and demographic data of a viewer of said content) based on data from said iteratively selected plurality of donors meeting said pre-defined threshold criteria and from said updated currency data.
2. The method as claimed in claim 1 wherein, said currency data comprising viewership details’ data per individual per household.
3. The method as claimed in claim 1 wherein, if said polled panel data comprising a non- watermarked channel, said method comprising:
- polling a panel associated per television per household for polled data, said polled data being defined in terms of watermarked channels (“polled watermarked channel panel data”) and non-watermarked channels (“polled non-watermarked channel panel data”);
- identifyingcloseness (ranking donor households in terms of their closeness to their recipient household) of each non-watermarked channel with that of each watermarked channel, for said pie-defined duration based on a closeness factor;
- selecting a closely associated watermarked channel per non-watermarked channel, for a pie-defined duration, said selection being based on pie-defined parameters per identified constituents along with said identified closeness factor;
- classifying polled watermarked channel panel data in terms of time duration and associating a higher hierarchy to said classified polled watermarked channel panel data;
- classifying polled non-watermarked channel panel data in terms of time duration and associating a lower hierarchy to said classified non-polled watermarked channel panel data;
- associating data from a classified non-watermarked channel with data from said classified watermarked channel, if threshold criteria of currency data are met, in order to obtain currency data per donor of said panel;
- iteratively checking if said currency data constitutes pre-defmed number of donor households; and
- determiningviewing behaviour parameters (in terms of content and demographic data of a viewer of said content) based on data from said iteratively selected plurality of donors meeting said pre-defmed threshold criteria and from said updated currency data.
4. The method as claimed in claim 1 wherein, said collapsible sub-hierarchical model comprising data items pertaining to similarity of channel information in terms of type, genre, language and / or similarity in time-band information.
5. The method as claimed in claim 1 wherein, said sub- hierarchical model comprising data items pertaining to demographic information, town class information, household information in terms of demographics of individuals in said household, and / or household information in terms of economic classification.
6. The method as claimed in claim 1 wherein, said pre-defined parameters being parameters selected from a group consisting of language parameter, and genre parameter.
7. The method as claimed in claim 1 wherein, said constituents being selected from a group of constituents consisting of programs, promotion breaks, and advertisement breaks.
8. The method as claimed in claim 1 wherein, said step of classifying comprising a step of associating hierarchical levels to each of said polled watermarked channel panel data per time duration and said non-polled watermarked channel panel data per time duration.
9. The method as claimed in claim 1 wherein, said step of classifying comprising a step of associating hierarchical levels to each of said polled watermarked channel panel data per time duration and said non-polled watermarked channel panel data per time duration comprising an additional step of associating each of said hierarchically defined polled watermarked channel panel data per time duration with a geography.
10. The method as claimed in claim 1 wherein, said step of classifying comprising a step of associating hierarchical levels to each of said polled watermarked channel panel data per time duration and said non-polled watermarked channel panel data per time duration comprising an additional step of associating each of said hierarchically defined polled watermarked channel panel data per time duration with a household size.
11. The method as claimed in claim 1 wherein, said step of classifying comprising a step of associating hierarchical levels to each of said polled watermarked channel panel data per time duration and said non-polled watermarked channel panel data per time duration comprising an additional step of associating each of said hierarchically defined non-polled watermarked channel panel data per time duration with a geography.
12. The method as claimed in claim 1 wherein, said step of classifying comprising a step of associating hierarchical levels to each of said polled watermarked channel panel data per time duration and said non-polled watermarked channel panel data per time duration comprising an additional step of associating each of said hierarchically defined non-polled watermarked channel panel data per time duration with a household size.
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| IN201921002782 | 2019-01-23 | ||
| IN201921002782 | 2019-01-23 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO2020152710A1 true WO2020152710A1 (en) | 2020-07-30 |
Family
ID=71736877
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/IN2020/050071 WO2020152710A1 (en) | 2019-01-23 | 2020-01-22 | A system and method to attribute household level viewership to individuals of the household. |
Country Status (1)
| Country | Link |
|---|---|
| WO (1) | WO2020152710A1 (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20160269766A1 (en) * | 2015-03-09 | 2016-09-15 | Rentrak Corporation | Attribution of household viewership information to individuals |
| US20180131996A1 (en) * | 2012-11-06 | 2018-05-10 | Comscore, Inc. | Demographic attribution of household viewing events |
-
2020
- 2020-01-22 WO PCT/IN2020/050071 patent/WO2020152710A1/en active Application Filing
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20180131996A1 (en) * | 2012-11-06 | 2018-05-10 | Comscore, Inc. | Demographic attribution of household viewing events |
| US20160269766A1 (en) * | 2015-03-09 | 2016-09-15 | Rentrak Corporation | Attribution of household viewership information to individuals |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN100551031C (en) | In the project recommendation device, a plurality of items are divided into the method and the device of similar group | |
| US11507588B2 (en) | Media names matching and normalization | |
| US20210084386A1 (en) | Flagging Advertisement Frames for Automatic Content Recognition | |
| US11507580B2 (en) | Methods and apparatus to partition data | |
| US20070083887A1 (en) | Commercial augmentation | |
| US20210281925A1 (en) | Dynamic viewer prediction system for advertisement scheduling | |
| WO2011005155A1 (en) | A method and arrangements for analysing multimedia content | |
| CN102708497A (en) | VideoBag feature-based accurate advertisement release system and method | |
| US20230230107A1 (en) | Methods and apparatus to apply household-level weights to household-member level audience measurement data | |
| US20210144222A1 (en) | Methods and apparatus to facilitate meter to meter matching for media identification | |
| US8055547B2 (en) | Method for site selection for retail and restaurant chains | |
| US11184682B2 (en) | Obtaining viewer demographics through advertisement selections | |
| US20150193822A1 (en) | Apparatus and method for inferring user profile | |
| CN113222653A (en) | Programmed advertisement user audience expansion method, system, equipment and storage medium | |
| WO2020152710A1 (en) | A system and method to attribute household level viewership to individuals of the household. | |
| JP2011015355A (en) | Attribute estimation apparatus and method, and program | |
| US11934455B2 (en) | Systems, methods, and apparatuses for audience metric determination | |
| Clark et al. | Who’s Watching TV? | |
| KR20230051531A (en) | Methods, Systems, and Mediums for Identifying Relevant Content | |
| CN105229684B (en) | System for controlling and optimizing information distribution between users in information exchange | |
| WO2018128254A1 (en) | Method and device for recommending usr group for new user | |
| KR20190044592A (en) | Dynamic multi-dimensional analysis based advertisements recommendation apparatus and method | |
| CN104333777A (en) | Radio and television program making system and method | |
| CN109949078B (en) | Promotion information processing method and device | |
| WO2021260933A1 (en) | Estimation device, estimation method, and recording medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| 121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 20745688 Country of ref document: EP Kind code of ref document: A1 |
|
| NENP | Non-entry into the national phase |
Ref country code: DE |
|
| 122 | Ep: pct application non-entry in european phase |
Ref document number: 20745688 Country of ref document: EP Kind code of ref document: A1 |