US20220038541A1 - Methods and apparatus for session building from ping-level data - Google Patents

Abstract

Example apparatus disclosed herein determine a first candidate duration of a session based on a first difference between a first time associated with a first start ping and a second time associated with a first end ping, the first start ping and the first end ping identified from a plurality of pings based on a first partitioning scheme. Disclosed example apparatus also determine a second candidate duration of the session based on a second difference between a third time associated with a second start ping and a fourth time associated with a second end ping, the second start ping and the second end ping identified from the plurality of pings based on a second partitioning scheme. Disclosed example apparatus further select at least one of the first candidate duration or the second candidate duration to represent the session based on a demographics model.

Description

RELATED APPLICATIONS
• This patent arises from a continuation of U.S. patent application Ser. No. 16/702,321, titled “Methods and Apparatus for Session Building from Ping-Level Data,” and filed on Dec. 3, 2019, which is hereby incorporated by reference in its entirety. Priority to U.S. patent application Ser. No. 16/702,321 is claimed.
FIELD OF THE DISCLOSURE
• This disclosure relates generally to audience measurement, and, more particularly, to methods and apparatus for session building from ping-level data.
BACKGROUND
• Audience viewership data is collected and used by audience measurement entities (AMEs) to determine exposure statistics (e.g., viewership statistics) for different media. Some viewership data may be collected through the use of pings from user devices. For example, these pings may be transmitted by the user devices to an AME via one or more networks. Information from the ping is processed by the AME to determine useful media exposure data and associated statistics.
BRIEF DESCRIPTION OF THE DRAWINGS
• FIG. 1 illustrates an example environment in which an example session builder builds sessions from ping-level data in accordance with teachings disclosed herein.
• FIG. 2 is block diagram representative of the example session builder ofFIG. 1.
• FIG. 3 is a block diagram representative of an example session partitioner included in the example session builder ofFIG. 2.
• FIG. 4 is a flowchart representative of example machine-readable instructions which may be executed to implement the session builder ofFIG. 2.
• FIG. 5 is a flowchart representative of example machine-readable instructions which may be executed to implement an example threshold detector included in the session partitioner ofFIG. 3.
• FIG. 6 is a flowchart representative of example machine-readable instructions which may be executed to implement the session partitioner ofFIG. 3.
• FIG. 7 is a flowchart representative of example machine-readable instructions which may be executed to implement an example duration calculator included in the session partitioner ofFIG. 3.
• FIG. 8 is a flowchart representative of example machine-readable instructions which may be executed to implement an example session partition selector included in the session builder ofFIG. 2.
• FIG. 9 is a block diagram of an example processing platform structured to execute the instructions ofFIGS. 4, 5, 6, 7 and/or 8 to implement the example session builder ofFIG. 2.
• The figures are not to scale. In general, the same reference numbers will be used throughout the drawing(s) and accompanying written description to refer to the same or like parts.
• Descriptors “first,” “second,” “third,” etc. are used herein when identifying multiple elements or components which may be referred to separately. Unless otherwise specified or understood based on their context of use, such descriptors are not intended to impute any meaning of priority, physical order or arrangement in a list, or ordering in time but are merely used as labels for referring to multiple elements or components separately for ease of understanding the disclosed examples. In some examples, the descriptor “first” may be used to refer to an element in the detailed description, while the same element may be referred to in a claim with a different descriptor such as “second” or “third.” In such instances, it should be understood that such descriptors are used merely for ease of referencing multiple elements or components.
DETAILED DESCRIPTION
• Audience viewership data may be acquired through the use of event-level data, such as ping-level data. In some examples, audience measurement based on viewership patterns relies on having access to the start times, end times and durations of media viewing sessions or, more generally, media exposure sessions. However, some media clients do not report such session-level data and, instead, report ping-level data. Ping-level data indicates when events occur in the viewing session and may include media identification information (e.g., program name, episode name, genre, application, etc.) in the ping payload data. For example, the pings may be triggered by events such as expiration of a heartbeat interval (e.g., a 5 minute heartbeat interval), when content changes, when advertisements are presented, etc.
• In examples disclosed herein, a session is defined to be a period of continuous exposure to (e.g., viewership of) (e.g., partitioned on the same program, episode, genre, application) media on the same device during which one of more characteristics of the media are unchanged. As such, a session has a defined start and end time. In some examples, the media characteristic(s) defining a session include one or more of a program identifier, an episode identifier, a genre identifier, and application identifier, etc. In examples disclosed herein, a set of ping-level data from a user device is used to build exposure (e.g., viewing) sessions, generally referred to as sessions, that represent the user's media exposure (e.g., viewership) during the time period the pings in the ping-level data were acquired. Examples disclosed herein transform the ping-level data to session-level data identifying the start time, end time and duration of the sessions. In examples disclosed herein, respective sessions are built by determining corresponding session durations from the ping-level data as described in further detail below. For convenience, examples disclosed herein are described from the perspective of building viewing sessions, but such examples also apply to building any other type of media exposure sessions (e.g., corresponding to exposure to media other than visual media, such as audio media, tactile media, etc.).
• In some examples, viewing sessions are built from the collected ping-level data based on a threshold defining an upper limit on the duration of a viewing session. For example, the ping-level data may be transformed to session-level data to define a viewing session according to whether the time difference between two reported pings is greater than or equal to the threshold value. If a subsequent ping is received within a time frame that is less than the threshold value, the subsequent ping may be included into the session with the prior ping (e.g., if the media characteristic(s) used to partition the sessions has remained unchanged between the prior ping and the subsequent ping). However, if the subsequent ping is received within a time frame that is greater than or equal to the threshold value, the subsequent ping is determined to be the start of a new session. In some examples, this threshold value can be defined based on input configuration data (e.g., entered by an operator) or empirically determined from previously determined viewing session duration data. In some examples, the viewing duration for a given viewing session is calculated as the difference between the start time associated with the first ping of the session and the end time determined from the last ping of the session, as described above.
• Thus, in some examples, the viewing session durations are determined based on the time between two consecutive pings. However, this does not account for possible changes in the media identification information of the ping-level data, such as a change in a program name or episode name. This can be problematic in examples where the timing between two consecutive pings is less than the threshold, but the program was changed between these two pings. If just the time difference between the pings is considered when building sessions, these two pings might be classified as belonging to the same viewing session since the threshold was not met to signify a change in viewing session. However, the change in program would correspond to the start of a new viewing session.
• To address such technical problems, examples disclosed herein determine viewing session durations by both comparing the time durations between pings to a threshold and detecting changes in the payload data included in the pings. In some examples, multiple session building procedures are operated in parallel such that different session building procedures perform session partitioning based on different information included in the payload data of each ping (e.g., program names vs. genre identifiers) to enable multiple partitioning schemes to be evaluated. The multiple session building procedures may operate on pipelines running in parallel in order to determine different possible durations for the same viewing session. After the set of session building procedures is completed for the different partitioning schemes, the viewing sessions determined by the different partitioning schemes are compared to determine which of the partitioning schemes built viewing sessions from the ping-level data that most align with the panel data. In some examples, for each partitioning scheme, the viewing sessions are compared to a demographics model generated from the panel data. In some such examples, one of the partitioning schemes is chosen to represent the viewing sessions for the set of ping data based on criteria when compared to the demographic model from the panel data.
• Examples disclosed herein provide technical solutions to the technical problems described above by capturing changes in the content of the ping-level data when determining viewing sessions. Examples disclosed herein also allow for a more flexible approach to determining viewing sessions by testing different partitioning schemes on the ping-level data. This approach enables generation of viewing sessions from a particular set of ping data that satisfy one or more criteria (e.g., a change in the genre content, a change in the program name, a change in the episode name, etc.).
• FIG. 1 illustrates anexample environment100 in which anexample session builder105 build sessions from captured ping-level data in accordance with teachings of this disclosure. Theexample environment100 ofFIG. 1 includes anexample media device110, anexample network115, anexample monitoring system120, and anexample panel database135. Theexample monitoring system120 includes theexample session builder105 and anexample session database130.
• In the illustrated example ofFIG. 1, themedia device110 is used to access and view different media. Theexample media device110 also sends ping data to thenetwork115 to log viewing information associated with the access and/or presentation of the media by themedia device110. Theexample media device110 can be implemented with any device or combinations of devices that are able to connect to media such as, for example, a smart television (TV), a personal computer, a smartphone, a tablet device, a set-top box (STB), a game console, a digital video recorder (DVR), an Apple TV, a Roku device, YouTube TV, an Amazon fire device, etc., or any combination thereof.
• Theexample network115 of the illustrated example ofFIG. 1 provides communication between theuser device110 and themonitoring system120. Theexample network115 is implemented as a public network such as, for example, the Internet. However, any other type of networks (e.g., wired/cabled, wireless, mobile cellular, etc.) which may be public or private, and any combination thereof may additionally and/or alternatively be used.
• Themonitoring system120 of the illustrated example ofFIG. 1 receives ping data, also referred to as pings, from themedia device110 via thenetwork115. Themonitoring system120 captures the payload data included in the ping data for use in the building of viewing sessions. The payload of a ping received by themonitoring system120 includes information on the time the ping was transmitted by and/or received from themedia device110, what device the ping was received from, the media content that was viewed on themedia device110 at the time the ping was transmitted (e.g., genre, program name, episode name, etc.), etc. Thesession builder105 extracts the time information from the payload data of the ping data as well as the media content information from the payload data of the ping data to build the viewing sessions.
• Thesession builder105 of the illustrated example ofFIG. 1 determines viewing sessions from the ping data received by themonitoring system120. For example, thesession builder105 determines viewing session durations from the ping data reported by themobile device110 by determining the respective start and end pings for each viewing session. Thesession builder105 uses the payload data within the reported ping data to determine whether the received ping is a start ping for a new viewing session, an end ping for a current viewing session, or a ping associated with the current viewing session (e.g., is neither a start ping nor an end ping). Thesession builder105 of the illustrated example generates viewing sessions from the ping data by determining durations for each of the viewing sessions using the ping data and corresponding payload data. An example implementation of thesession builder105 is illustrated inFIG. 2, which is described in further detail below.
• Thesession database130 of the illustrated example ofFIG. 1 stores information describing the viewing sessions determined by thesession builder105. However, other data may additionally and/or alternatively be stored by thesession database130. Thesession database130 of the illustrated example ofFIG. 1 is implemented by any memory, storage device, and/or storage disc for storing data such as, for example, flash memory, magnetic media, optical media, solid state memory, hard drive(s), thumb drive(s), etc. Furthermore, the data stored in theexample session database130 may be in any data format such as, for example, binary data, comma delimited data, tab delimitated data, structured query language (SQL) structures, etc. While thesession database130 is illustrated as a single device in the illustrated example ofFIG. 1, thesession database130 and/or any other data storage devices described herein may be implemented by any number and/or type(s) or memories.
• Theexample panel database135 of the illustrated example ofFIG. 1 stores that panel data collected by meters and/or meter applications associated with and/or executed by media devices of panelists. In examples disclosed herein, the panel data is collected from panelists to monitor the operation of the media devices and/or the media presented by the media devices. The panel data can include session level data characterizing the viewing sessions, as well as ping data reported by the panelists' media devices. The ping data can be compared with corresponding session level data for respective ones of the monitored media devices to determine relationships between ping data and sessions.
• Theexample panel database135 provides the panel data information to be used by thesession builder105. As described above, the panel data includes session level data and ping data from panelists' media devices, and this information can include, for example, viewing session data, panelists demographics information, etc. However, other data may additionally and/or alternatively be stored by thepanel database135. Theexample panel database135 provides thesession builder105 with viewing session information that can be used as reference viewing sessions when determining viewing sessions from the ping-data received by themonitoring system120. Thepanel database135 of the illustrated example ofFIG. 1 is implemented by any memory, storage device, and/or storage disc for storing data such as, for example, flash memory, magnetic media, optical media, solid state memory, hard drive(s), thumb drive(s), etc. Furthermore, the data stored in theexample panel database135 may be in any data format such as, for example, binary data, comma delimited data, tab delimitated data, structured query language (SQL) structures, etc. While, in the illustrated example ofFIG. 1, thepanel database135 is illustrated as a single device, thepanel database135 and/or any other data storage devices described herein may be implemented by any number and/or type(s) or memories.
• FIG. 2 is a block diagram illustrating an example implementation of theexample session builder105 ofFIG. 1. Theexample session builder105 ofFIG. 2 includes anexample session partitioner200, anexample threshold calculator205, and an examplesession partition selector210.
• Theexample session partitioner200 of the illustrated example ofFIG. 2 receives the ping data from thenetwork115 ofFIG. 1. In examples disclosed herein, thesession partitioner200 determines viewing sessions from ping-level data using different partitioning schemes. The different partitioning schemes may include partitioning based on genre, program name, episode name, etc. However, on any other content categories may additionally or alternatively be used for partitioning. In the illustrated example ofFIG. 2, threesession partitioners200 are shown. However, any number ofsession partitioners200 may additionally or alternatively be used. For example, asession builder105 depicted inFIG. 2 may contain, for example, fivesession partitioner200 circuits. Utilizingadditional session partitioners200 enables additional partitioning schemes to be evaluated when determining viewing sessions. For example, having threesession partitioners200 allows for comparison of viewing sessions built according to three different partitioning schemes. In the illustrated example, thesession partitioners200 are operated in parallel to improve efficiency of building the viewing sessions.
• Theexample threshold calculator205 of the illustrated example ofFIG. 2 receives the panel data from thepanel database135. Theexample threshold calculator205 determines a threshold value based on the reference viewing sessions data stored in the panel data from thepanel database135. Thethreshold calculator205 selects viewing session durations from the reference viewing sessions data based on a criterion. For example, the criterion may be the10% longest viewing session durations stored in thepanel database135. However, other criterions may additionally or alternatively be used. Thethreshold calculator205 averages the viewing session durations selected and sets the threshold value to the average found. The threshold value determined by thethreshold calculator205 is used by thesession partitioner200 to build the viewing sessions, which is discussed below in connection withFIG. 3.
• The examplesession partition selector210 of the illustrated example ofFIG. 2 determines which of the viewing sessions built in thesession partitioner200 to select to represent the set of ping-level data received. In the examples disclosed herein, the viewing sessions from the different partitioning schemes implemented by the session partitioner(s)200 are tested against a demographics model constructed from the panel data in thepanel database135 to determine which partitioning scheme aligns best with the expected viewing sessions from the demographics model. The performance of the viewing sessions compared to the demographics model is determined by a performance criterion. For example, the performance criterion may be a percentage of similarity between the viewing session from thesession partitioner200 and the viewing session from the demographics model. However, the other criteria may additionally or alternatively be used when determining the performance. The examplesession partition selector210 stores the selected the viewing sessions in thesession database130.
• FIG. 3 is a block diagram illustrating an example implementation of thesession partitioner200 ofFIG. 2. Theexample session partitioner200 ofFIG. 3 includes an example ping identifier305, an examplepayload change identifier310, anexample ping timer315, anexample threshold detector320, an examplestart ping identifier325, an exampleend ping identifier330, anexample duration calculator335, and anexample duration adjuster340.
• The example ping identifier305 of the illustrated example ofFIG. 3 receives a ping from thenetwork115 ofFIG. 1. The example ping identifier305 determines if the ping is the initial ping from a client/device. The example ping identifier305 outputs the ping to thestart ping identifier325 if the ping identifier305 determines that the ping is the first ping from the client/device. The example ping identifier outputs the ping data to thepayload change identifier310 if the ping identifier305 determines that the ping is not the first ping from the client/device.
• The examplepayload change identifier310 of the illustrated example ofFIG. 3 accesses the ping from theping identifier300 when the payload data of the ping indicates the ping is not a first ping from the device/client. The examplepayload change identifier310 determines if the payload data from the ping indicates a change in content, such as a change in genre, program name, or episode name. However, other categories of content may additionally or alternatively be used. In examples disclosed herein, genre is defined to be a subcategory of media content (e.g., action, comedy, mystery, etc.), program name is defined to be a descriptor of the name of the media program such as, for example, a television program (e.g., Game of Thrones, Saturday Night Live, etc.), and episode name is defined to be a descriptor of the program name that indicated the name of the episode of the program (e.g., episode1,101, ep.4, etc.). However, the genre, program name, and episode name are not limited to these examples, and other examples of the content categories may additionally or alternatively be used in determining a change in content. The examplepayload change identifier310 determines if there is a change in the payload data depending on the partitioning scheme being used by thesession partitioner200. For example, if thesession partitioner200 is partitioning viewing sessions based on genre, then thepayload change identifier310 determines if there are changes in the genre content of the payload data from the received ping. For example, if the payload data indicates that the genre of the media changed from action to comedy, the examplepayload change identifier310 would indicate that there was a change in the payload data. In examples disclosed herein, thepayload change identifier310 may use any combination of the partitioning criteria (e.g. genre, program name, episode name, etc.) when determining a change in content of the payload data. The examplepayload change identifier310 outputs the ping to thestart ping identifier325 if thepayload change identifier310 determines that there is a change in the content of the payload data based on the partitioning scheme. The examplepayload change identifier310 output the ping data to theping timer315 if thepayload change identifier310 determines there was no change in the content of the payload data based on the partitioning scheme.
• Theexample ping timer315 of the illustrated example ofFIG. 3 accesses the ping from thepayload change identifier310 when the payload data of the ping indicates no change in the media characteristic(s) according to the partitioning scheme implemented by thepayload change identifier310. Theexample ping timer315 is provided to determine the time difference between the current ping and the preceding ping. Theexample ping timer315 calculates the amount of time between the current ping and the preceding ping using the ping data. In some examples, a ping includes one or more time stamps corresponding to a time or times at which the ping was transmitted by and/or received from themedia device110. In the illustrated example, theping timer315 determines the difference between the time stamp(s) of the current ping and the time stamp(s) of the preceding ping, and outputs the difference to thethreshold detector320.
• Theexample threshold detector320 of the illustrated example ofFIG. 3 accesses the ping and the time difference between the ping and the preceding ping from theping timer315. Thethreshold detector320 accesses a threshold value from thethreshold calculator205 of the exampleFIG. 2. Theexample threshold detector320 determines whether the time difference between the accessed ping and the preceding ping satisfies the threshold value. For example,threshold detector320 determines whether the time difference between the accessed ping and the preceding ping is greater than or equal to the threshold value. Theexample threshold detector320 outputs the ping data to thestart ping identifier325 if the time difference determined by theping timer315 is greater than or equal to the threshold value from thethreshold calculator205. Theexample threshold detector320 output the ping data to theend ping identifier330 if the time difference determined by theping timer315 is less than the threshold value from thethreshold calculator205.
• The examplestart ping identifier325 of the illustrated example ofFIG. 3 accesses the ping from theping identifier300, thepayload change identifier310, or thethreshold detector320 when the ping is indicated to be the start ping for a new viewing session. The examplestart ping identifier325 labels the ping as a start ping for the new viewing session. This label may be a flag that is attached to the ping to signify the ping as a start ping for a viewing session. However, other labels may additionally or alternatively be used such as, for example, the addition of a special character to the ping data, highlighting the ping data, etc. Thestart ping identifier325 outputs the ping data to theduration calculator335 when a viewing session duration needs to be calculated. The viewing session duration is calculated once an end ping is found for the viewing session.
• The exampleend ping identifier330 of the illustrated example ofFIG. 3 accesses the ping from thethreshold detector320 when the ping is indicated to not be the start ping for a new viewing session. The exampleend ping identifier330 stores the accessed ping as a candidate end ping for the viewing session. If a next ping after the accessed ping is labeled a start ping for a new viewing session by the examplestart ping identifier325, the exampleend ping identifier330 labels the accessed ping as an end ping for the prior viewing session. The label may be a flag that is attached to the ping data to signify the ping as an end ping for a viewing session. However, other labels may additionally or alternatively be used such as, for example, the addition of a special character to the ping data, highlighting the ping data, etc. If the next ping after the accessed ping is not labeled a start ping for a new viewing session by the examplestart ping identifier325, the example end ping identifier discards the stored candidate end ping and stores the next ping as a candidate end ping. Theend ping identifier330 outputs the ping data of the labeled end ping to theduration calculator335.
• Theexample duration calculator335 of the illustrated example ofFIG. 3 accesses the ping data from thestart ping identifier325 and the ping data from theend ping identifier330. Theexample duration calculator330 takes the difference in time between the ping from thestart ping identifier325 and the ping from theend ping identifier330. This time difference is determined to be the overall duration of the viewing session. Theexample duration calculator335 outputs the viewing session duration to theduration adjuster340.
• Theexample duration adjuster340 of the illustrated example ofFIG. 3 is provided to determine whether additional time is to be added to the viewing session duration calculated by theduration calculator335. Theexample duration adjuster340 generates a probability distribution of end durations from the panel data in thepanel database135. The panel data includes information on the general trends in viewing session durations across panelists from different demographic categories (e.g., age, gender, etc.). The panel data also includes the payload data information on the viewing session durations (e.g., the program name, episode name, genre, etc.). Theexample duration adjuster340 generates a probability distribution that reflects the viewing session durations from different panelists. For example, the probability distribution might reflect the viewing session durations for the same episode of a program that different panelists all watched. The example probability distribution may reflect that the viewing session duration that occurred the most for that particular episode of a program across the panelists has the highest probability of occurring. In the illustrated example, the probability distribution from the duration adjuster240 reflects the end durations for different programs, episodes, genres, etc. that occur the most across the panelists. Theexample duration adjuster340 randomly selects an end duration from the range of possible end durations according to the generated probability distribution. Theexample duration adjuster340 adds the selected end duration to the time of the end ping in the viewing session. An end duration in examples disclosed herein is the additional duration of time after the last ping of the viewing session, and is included into the viewing session duration. The end duration selected by theduration adjuster340 is added to the time of the end ping in order to account for the likelihood that the viewing session ended a certain amount of time after the labeled end ping of the viewing session. Theexample duration adjuster340 outputs the new viewing session duration to thesession partition selector210 ofFIG. 2.
• While an example manner of implementing thesession builder105 ofFIG. 1 is illustrated inFIGS. 2 and 3, one or more of the elements, processes and/or devices illustrated inFIGS. 2 and 3 may be combined, divided, re-arranged, omitted, eliminated and/or implemented in any other way. Further, theexample session partitioner200, theexample threshold calculator205, the examplesession partition selector210, theexample ping identifier300, the examplepayload change identifier310, theexample ping timer315, theexample threshold detector320, the examplestart ping identifier325, the exampleend ping identifier330, theexample duration calculator335, theexample duration adjuster340 and/or, more generally, theexample session builder105 ofFIG. 1 may be implemented by hardware, software, firmware and/or any combination of hardware, software and/or firmware. Thus, for example, any of theexample session partitioner200, theexample threshold calculator205, the examplesession partition selector210, theexample ping identifier300, the examplepayload change identifier310, theexample ping timer315, theexample threshold detector320, the examplestart ping identifier325, the exampleend ping identifier330, theexample duration calculator335, theexample duration adjuster340 and/or, more generally, theexample session builder105 could be implemented by one or more analog or digital circuit(s), logic circuits, programmable processor(s), programmable controller(s), graphics processing unit(s) (GPU(s)), digital signal processor(s) (DSP(s)), application specific integrated circuit(s) (ASIC(s)), programmable logic device(s) (PLD(s)) and/or field programmable logic device(s) (FPLD(s)). When reading any of the apparatus or system claims of this patent to cover a purely software and/or firmware implementation, at least one of theexample session partitioner200, theexample threshold calculator205, the examplesession partition selector210, theexample ping identifier300, the examplepayload change identifier310, theexample ping timer315, theexample threshold detector320, the examplestart ping identifier325, the exampleend ping identifier330, theexample duration calculator335, and/or theexample duration adjuster340 is/are hereby expressly defined to include a non-transitory computer readable storage device or storage disk such as a memory, a digital versatile disk (DVD), a compact disk (CD), a Blu-ray disk, etc. including the software and/or firmware. Further still, theexample session builder105 ofFIG. 1 may include one or more elements, processes and/or devices in addition to, or instead of, those illustrated inFIGS. 2 and 3, and/or may include more than one of any or all of the illustrated elements, processes and devices. As used herein, the phrase “in communication,” including variations thereof, encompasses direct communication and/or indirect communication through one or more intermediary components, and does not require direct physical (e.g., wired) communication and/or constant communication, but rather additionally includes selective communication at periodic intervals, scheduled intervals, aperiodic intervals, and/or one-time events.
• A flowchart representative of example hardware logic, machine readable instructions, hardware implemented state machines, and/or any combination thereof for implementing thesession builder105 ofFIG. 1 is shown inFIGS. 4, 5, 6, 7, and 8. The machine readable instructions may be one or more executable programs or portion(s) of an executable program for execution by a computer processor such as theprocessor912 shown in theexample processor platform900 discussed below in connection withFIG. 9. The program may be embodied in software stored on a non-transitory computer readable storage medium such as a CD-ROM, a floppy disk, a hard drive, a DVD, a Blu-ray disk, or a memory associated with theprocessor912, but the entire program and/or parts thereof could alternatively be executed by a device other than theprocessor912 and/or embodied in firmware or dedicated hardware. Further, although the example program is described with reference to the flowchart illustrated inFIGS. 4, 5, 6, 7, and 8, many other methods of implementing theexample session builder105 may alternatively be used. For example, the order of execution of the blocks may be changed, and/or some of the blocks described may be changed, eliminated, or combined. Additionally or alternatively, any or all of the blocks may be implemented by one or more hardware circuits (e.g., discrete and/or integrated analog and/or digital circuitry, an FPGA, an ASIC, a comparator, an operational-amplifier (op-amp), a logic circuit, etc.) structured to perform the corresponding operation without executing software or firmware.
• The machine readable instructions described herein may be stored in one or more of a compressed format, an encrypted format, a fragmented format, a compiled format, an executable format, a packaged format, etc. Machine readable instructions as described herein may be stored as data (e.g., portions of instructions, code, representations of code, etc.) that may be utilized to create, manufacture, and/or produce machine executable instructions. For example, the machine readable instructions may be fragmented and stored on one or more storage devices and/or computing devices (e.g., servers). The machine readable instructions may require one or more of installation, modification, adaptation, updating, combining, supplementing, configuring, decryption, decompression, unpacking, distribution, reassignment, compilation, etc. in order to make them directly readable, interpretable, and/or executable by a computing device and/or other machine. For example, the machine readable instructions may be stored in multiple parts, which are individually compressed, encrypted, and stored on separate computing devices, wherein the parts when decrypted, decompressed, and combined form a set of executable instructions that implement a program such as that described herein.
• In another example, the machine readable instructions may be stored in a state in which they may be read by a computer, but require addition of a library (e.g., a dynamic link library (DLL)), a software development kit (SDK), an application programming interface (API), etc. in order to execute the instructions on a particular computing device or other device. In another example, the machine readable instructions may need to be configured (e.g., settings stored, data input, network addresses recorded, etc.) before the machine readable instructions and/or the corresponding program(s) can be executed in whole or in part. Thus, the disclosed machine readable instructions and/or corresponding program(s) are intended to encompass such machine readable instructions and/or program(s) regardless of the particular format or state of the machine readable instructions and/or program(s) when stored or otherwise at rest or in transit.
• The machine readable instructions described herein can be represented by any past, present, or future instruction language, scripting language, programming language, etc. For example, the machine readable instructions may be represented using any of the following languages: C, C++, Java, C#, Perl, Python, JavaScript, HyperText Markup Language (HTML), Structured Query Language (SQL), Swift, etc.
• As mentioned above, the example processes ofFIGS. 4, 5, 6, 7, and8 may be implemented using executable instructions (e.g., computer and/or machine readable instructions) stored on a non-transitory computer and/or machine readable medium such as a hard disk drive, a flash memory, a read-only memory, a compact disk, a digital versatile disk, a cache, a random-access memory and/or any other storage device or storage disk in which information is stored for any duration (e.g., for extended time periods, permanently, for brief instances, for temporarily buffering, and/or for caching of the information). As used herein, the term non-transitory computer readable medium is expressly defined to include any type of computer readable storage device and/or storage disk and to exclude propagating signals and to exclude transmission media.
• “Including” and “comprising” (and all forms and tenses thereof) are used herein to be open ended terms. Thus, whenever a claim employs any form of “include” or “comprise” (e.g., comprises, includes, comprising, including, having, etc.) as a preamble or within a claim recitation of any kind, it is to be understood that additional elements, terms, etc. may be present without falling outside the scope of the corresponding claim or recitation. As used herein, when the phrase “at least” is used as the transition term in, for example, a preamble of a claim, it is open-ended in the same manner as the term “comprising” and “including” are open ended. The term “and/or” when used, for example, in a form such as A, B, and/or C refers to any combination or subset of A, B, C such as (1) A alone, (2) B alone, (3) C alone, (4) A with B, (5) A with C, (6) B with C, and (7) A with B and with C. As used herein in the context of describing structures, components, items, objects and/or things, the phrase “at least one of A and B” is intended to refer to implementations including any of (1) at least one A, (2) at least one B, and (3) at least one A and at least one B. Similarly, as used herein in the context of describing structures, components, items, objects and/or things, the phrase “at least one of A or B” is intended to refer to implementations including any of (1) at least one A, (2) at least one B, and (3) at least one A and at least one B. As used herein in the context of describing the performance or execution of processes, instructions, actions, activities and/or steps, the phrase “at least one of A and B” is intended to refer to implementations including any of (1) at least one A, (2) at least one B, and (3) at least one A and at least one B. Similarly, as used herein in the context of describing the performance or execution of processes, instructions, actions, activities and/or steps, the phrase “at least one of A or B” is intended to refer to implementations including any of (1) at least one A, (2) at least one B, and (3) at least one A and at least one B.
• As used herein, singular references (e.g., “a”, “an”, “first”, “second”, etc.) do not exclude a plurality. The term “a” or “an” entity, as used herein, refers to one or more of that entity. The terms “a” (or “an”), “one or more”, and “at least one” can be used interchangeably herein. Furthermore, although individually listed, a plurality of means, elements or method actions may be implemented by, e.g., a single unit or processor. Additionally, although individual features may be included in different examples or claims, these may possibly be combined, and the inclusion in different examples or claims does not imply that a combination of features is not feasible and/or advantageous.
• FIG. 4 is a flowchart illustrating anexample process400 that is representative of example machine-readable instructions which may be executed to implement theexample session builder105 ofFIG. 2. The program ofFIG. 4 begins execution atblock405 at which thesession partitioner200 accesses the pings and corresponding payload data from thenetwork115. Atblock410, thethreshold calculator205 accesses the panel data from thepanel database135. Theexample threshold calculator205 accesses reference viewing sessions data from the panel data in thepanel database135. Atblock420 thethreshold calculator205 calculates threshold value(s) for the different session partitioning scheme(s). As described in further detail below, theexample flowchart420 ofFIG. 5 represents example instructions that may be implemented to calculate the threshold value(s) for the session partitioning scheme(s).
• Atblock430, thesession partitioner200 uses different partitioning criteria to build corresponding different viewing sessions from the ping data. As described in further detail below, theexample flowchart430 ofFIG. 6 represents example instructions that may be implemented to use different partitioning criteria to build corresponding different viewing sessions from the ping data.
• Atblock440, the session builder125 determines if the session building is complete. If the session builder125 determines that the session building is completed, then process400 continues to block440. If the session builder125 determines that the session building is not complete yet, then process400 returns to block430. Atblock450, thesession partition selector210 performs the viewing session selection process. As described in further detail below, theexample flowchart450 ofFIG. 8 represents example instructions that may be implemented to perform the viewing session selection.
• Atblock460, thesession partition selector210 stores the selected viewing sessions in thesession database130. Once the selected viewing sessions are stored,process400 ends.
• FIG. 5 is a flowchart illustrating theprocess420 ofFIG. 4 and is representative of example machine-readable instructions which may be executed to implement theexample threshold calculator205 ofFIG. 2. Theprogram420 ofFIG. 5 begins execution atblock510 at which thethreshold calculator205 selects durations from the panel data based on a first criteria. In the examples disclosed herein, the first criteria may be a certain percentage of the durations in a cutoff ranked from highest to lowest in the panel data. For example, the first criteria may be ten percent of the durations in the cutoff ranked from highest to lowest in the panel data. However, any other criteria may additionally or alternatively be used for the first criteria.
• Atblock520, thethreshold calculator205 calculates an average value from the selected durations based on the first criteria. After the average is calculated,process420 continues to block530 where thethreshold calculator205 sets the threshold value to be the average from the selected durations. Once the threshold value is set,process420 completes and returns to process400 ofFIG. 4.
• FIG. 6 is a flowchart illustrating theexample process440 ofFIG. 4 that is representative of machine-readable instructions which may be executed to implement theexample session partitioner200 ofFIG. 3. The program ofFIG. 6 begins execution atblock610 at which thesession partitioner200 chooses a partitioning scheme to use when building viewing sessions. The partitioning scheme may be chosen to be the genre name, the program name, the episode name, etc., or any combination thereof, included in the ping payload data. Atblock615, theping identifier300 receives a ping and corresponding payload data from thenetwork115. Atblock620, theexample ping identifier300 determines if the current ping and corresponding payload data indicate that the current ping is the first ping received from the client/device. If theexample ping identifier300 determines that the current ping is the first ping from the client/device, then process440 continues to block645 where the current ping is labeled as the start ping of a new viewing session. If the examplefirst ping identifier300 determines that the current ping is not the first ping from the client/device, then process440 continues to block625.
• Atblock625, thepayload change identifier310 determines if the current ping and corresponding payload data from thefirst ping identifier300 indicate that there has been a change in the payload data. Thepayload change identifier310 determines if there has been a change in the payload data for the partitioning scheme chosen atblock610. For example, if the genre name partitioning scheme was chosen atblock610, the payload change identifier would determine if there was a change in the genre content of the payload data. If the examplepayload change identifier310 determines that the payload data of the current ping has changed, then process440 continues to block645 at which the current ping is labeled as the start ping of a new viewing session. If the examplepayload change identifier310 determines that the payload data of the current ping has not changed, then process440 continues to block630.
• Atblock630, theping timer315 determines the time difference between the current ping and the previous ping received. Atblock635, thethreshold detector320 accesses a threshold value for a viewing session from thethreshold calculator205. Atblock640, thethreshold detector320 determines if the time difference between the current ping and the previous ping calculated by theping timer315 is greater than or equal to the threshold value received from thethreshold calculator205. If theexample threshold detector320 determines that the time difference between the current ping and the previous ping is greater than or equal to the threshold value, then process440 continues to block655 where the previous ping is labeled as the end ping of the existing viewing session and the current ping is labeled as the start ping of a new session. If theexample threshold detector320 determines that the time difference between the current ping and the previous ping is less than the threshold value, then process440 continues to block650 where the current ping is included in the existing viewing session.
• Atblock645, the current ping is set to be the start ping of a new viewing session.Block645 executes when the current ping and corresponding payload data is determined to be a start ping by theexample ping identifier300 or the examplepayload change identifier310. Atblock650, the current ping is included in the existing session.Block650 executes when the current ping and corresponding payload data is determined to be included in the existing session by theexample threshold detector320. Atblock660, the previous ping is set to be the end ping in the existing session and the current ping is set to be the start ping of a new session.Block660 executes when the current ping and corresponding payload data is determined to be an end ping by theexample threshold detector320. After execution of any one ofblocks645,650, and655,process440 continues to block660 where thesession partitioner200 determines if there is a next ping in the ping data set. If thesession partitioner200 determines that there is another ping left in the ping data set, then theprocess440 returns to block615 where theping identifier300 accesses a ping and its corresponding payload data from the ping data set. If thesession partitioner200 does not determine that there is a next ping in the ping data set, then process440 continues to block665.
• Atblock665, thesession partitioner200 determines if there are any partitioning schemes left that have not been run. If thesession partitioner200 determines that there are partitioning schemes that still need to be run, then process440 returns to block610 where thesession partitioner200 selects another partitioning scheme to run. If thesession partitioner200 determines that there are no partitioning schemes left to run, then process440 continues to block670.
• Atblock670, theduration calculator335 calculates the viewing session durations for the different partitioning scheme(s) run. Afterblock670, theprocess440 returns to process400 ofFIG. 4.
• FIG. 7 is a flowchart illustrating aprocess700 that is representative of machine-readable instructions which may be executed to implement theduration adjuster340 of thesession partitioner200 ofFIG. 2. The program ofFIG. 7 begins atblock705 at which theduration adjuster340 accesses the panel data from thepanel database135. Atblock710, theduration adjuster340 generates a probability distribution of end durations from the panel data. The panel data includes information on the general trends in viewing session durations across panelists from different demographic categories (e.g., age, gender, etc.). The panel data also includes the payload data information on the viewing session durations (e.g., the program name, episode name, genre, etc.). Theduration adjuster340 generates a probability distribution from the viewing session duration information in the panel data. For example, the probability distribution may reflect the viewing session durations for the same episode of a program that different panelists all watched. The example probability distribution may reflect that the viewing session duration that occurred the most for that particular episode of a program across the panelists has the highest probability of occurring. The probability distribution from the duration adjuster240 reflects the end durations for different programs, episodes, genres, etc. that occur the most across the panelists. An end duration in example disclosed herein is the duration of time after the last ping of the viewing session that is included into the viewing session duration. The end duration accounts for the time between the end ping of one viewing session and the start ping of a second viewing session.
• Atblock720, theduration adjuster340 randomly selects an end duration according to the probability distribution generated atblock710. Theexample duration adjuster340 randomly selects an end duration from a range of possible end duration from the probability distribution. Atblock730, theduration adjuster340 adds the selected end duration to the time of the end ping to augment the session duration. As discussed above in connection withblock710, the end duration is used to account for the time between the end ping of one viewing session and the start ping of a second viewing session. The end duration selected by theduration adjuster340 is added to the time of the end ping in order to account for the likelihood that the viewing session ended a certain amount of time after the labeled end ping of the viewing session. Afterblock730 is completed,process700 then ends.
• FIG. 8 is a flowchart illustrating aprocess450 that is representative of machine-readable instructions which may be executed to implement thesession partition selector210 ofFIG. 2. The program ofFIG. 8 begins execution atblock810 at which thesession partition selector210 accesses the different viewing session durations from the selected partitioning scheme(s) from thesession partitioner200. Atblock820, thesession partition selector210 compares one of the different viewing session durations to a demographics model. In examples disclosed herein, the demographics model represents general trends seen in viewing session durations across different demographics categories accounted for in the panel data from thepanel database135. For example, the panel data may include viewing session duration statistics for the demographic of women ages12-18. The demographics model would reflect the trends seen in the viewing session duration data for women ages12-18. The examplesession partition selector210 compares one of the possible viewing session durations built in thesession partitioner200 based on the different partitioning schemes to the demographics model that matches the demographics of the household that the ping data is from.
• Atblock830, thesession partition selector210 determines the performance of one of the different viewing session durations from the demographic model. In example disclosed herein, the performance may be a percentage value for how similar the viewing session duration from thesession partitioner200 is to the viewing session durations found in the demographics model. For example, the viewing session duration from thesession partitioner200 may be80% (or some other value) similar to the expected viewing session duration based on the demographics model for the household. However, other determinations of the performance may additionally or alternatively be used.
• Atblock840, thesession partition selector210 determines if there are partitioning schemes left. If thesession partition selector210 determines that there are partitioning schemes left,process450 returns to block810 where thesession partition selector210 access different viewing session durations from the selected partitioning schemes. If thesession partition selector210 determines that there are not any partitioning schemes left,process450 continues to block850.
• Atblock850, thesession partition selector210 selects the viewing session duration based on a performance criterion. In examples disclosed herein, the performance criterion may be the highest percentage of similarity across the different viewing session durations. For example, if there was a viewing session duration partitioned on genre, another viewing session duration partitioned on program name, and a third viewing session duration partitioned on episode name, the performance criterion would be which of the three had the highest percentage of similarity found when compared to the demographics model atblock840 above. However, any other performance criterion may additionally or alternatively be used. After the selection of the viewing session durations,process450 returns to process400.
• FIG. 9 is a block diagram of anexample processor platform900 structured to execute the instructions ofFIGS. 4, 5, 6, 7, and 8 to implement thesession builder105 ofFIG. 1. Theprocessor platform900 can be, for example, a server, a personal computer, a workstation, a self-learning machine (e.g., a neural network), a mobile device (e.g., a cell phone, a smart phone, a tablet such as an iPad^Tm), or any other type of computing device.
• Theprocessor platform900 of the illustrated example includes aprocessor912. Theprocessor912 of the illustrated example is hardware. For example, theprocessor912 can be implemented by one or more integrated circuits, logic circuits, microprocessors, GPUs, DSPs, or controllers from any desired family or manufacturer. The hardware processor may be a semiconductor based (e.g., silicon based) device. In this example, the processor implements theexample session builder105, theexample session partitioner200, theexample threshold calculator205, the examplesession partition selector210, theexample ping identifier300, the examplepayload change identifier310, theexample ping timer315, theexample threshold detector320, the examplestart ping identifier325, the exampleend ping identifier330, theexample duration calculator335, theexample duration adjuster340.
• Theprocessor912 of the illustrated example includes a local memory913 (e.g., a cache). Theprocessor912 of the illustrated example is in communication with a main memory including avolatile memory914 and anon-volatile memory916 via abus918. Thevolatile memory914 may be implemented by Synchronous Dynamic Random Access Memory (SDRAM), Dynamic Random Access Memory (DRAM), RAMBUS® Dynamic Random Access Memory (RDRAM®) and/or any other type of random access memory device. Thenon-volatile memory916 may be implemented by flash memory and/or any other desired type of memory device. Access to themain memory914,916 is controlled by a memory controller.
• Theprocessor platform900 of the illustrated example also includes aninterface circuit920. Theinterface circuit920 may be implemented by any type of interface standard, such as an Ethernet interface, a universal serial bus (USB), a Bluetooth® interface, a near field communication (NFC) interface, and/or a PCI express interface.
• In the illustrated example, one ormore input devices922 are connected to theinterface circuit920. The input device(s)922 permit(s) a user to enter data and/or commands into theprocessor912. The input device(s) can be implemented by, for example, an audio sensor, a microphone, a camera (still or video), a keyboard, a button, a mouse, a touchscreen, a track-pad, a trackball, isopoint and/or a voice recognition system.
• One ormore output devices924 are also connected to theinterface circuit920 of the illustrated example. Theoutput devices924 can be implemented, for example, by display devices (e.g., a light emitting diode (LED), an organic light emitting diode (OLED), a liquid crystal display (LCD), a cathode ray tube display (CRT), an in-place switching (IPS) display, a touchscreen, etc.), a tactile output device, a printer and/or speaker. Theinterface circuit920 of the illustrated example, thus, typically includes a graphics driver card, a graphics driver chip and/or a graphics driver processor.
• Theinterface circuit920 of the illustrated example also includes a communication device such as a transmitter, a receiver, a transceiver, a modem, a residential gateway, a wireless access point, and/or a network interface to facilitate exchange of data with external machines (e.g., computing devices of any kind) via anetwork926. The communication can be via, for example, an Ethernet connection, a digital subscriber line (DSL) connection, a telephone line connection, a coaxial cable system, a satellite system, a line-of-site wireless system, a cellular telephone system, etc.
• Theprocessor platform900 of the illustrated example also includes one or moremass storage devices928 for storing software and/or data. Examples of suchmass storage devices928 include floppy disk drives, hard drive disks, compact disk drives, Blu-ray disk drives, redundant array of independent disks (RAID) systems, and digital versatile disk (DVD) drives. In some examples, the mass storage device(s)928 can implement thesession database130 and/or thepanel database135. In some examples, thevolatile memory914 can additionally or alternatively be used to implement thesession database130 and/or thepanel database135.
• The machineexecutable instructions932 ofFIGS. 4, 5, 6, 7 and/or 8 may be stored in themass storage device928, in thevolatile memory914, in thenon-volatile memory916, and/or on a removable non-transitory computer readable storage medium such as a CD or DVD.
• From the foregoing, it will be appreciated that example methods, apparatus and articles of manufacture have been disclosed that build viewing session from ping-level data. The disclosed methods, apparatus and articles of manufacture improve the efficiency of using a computing device by providing the ability to build viewing sessions from ping-level data according to different partitioning schemes. The disclosed methods, apparatus, and articles of manufacture build viewing sessions according to changes in the content being viewed by a media device. The disclosed methods, apparatus and articles of manufacture are accordingly directed to one or more improvement(s) in the functioning of a computer.
• Although certain example methods, apparatus and articles of manufacture have been disclosed herein, the scope of coverage of this patent is not limited thereto. On the contrary, this patent covers all methods, apparatus and articles of manufacture fairly falling within the scope of the claims of this patent.
• The following claims are hereby incorporated into this Detailed Description by this reference, with each claim standing on its own as a separate embodiment of the present disclosure.

Claims (20)

What is claimed is:

1. An apparatus comprising:

at least one memory;

instructions in the apparatus; and

processor circuitry to execute the instructions to:

determine a first candidate duration of a session based on a first difference between a first time associated with a first start ping and a second time associated with a first end ping, the first start ping and the first end ping identified from a plurality of pings based on a first partitioning scheme;

determine a second candidate duration of the session based on a second difference between a third time associated with a second start ping e and a fourth time associated with a second end ping, the second start ping and the second end ping identified from the plurality of pings based on a second partitioning scheme; and

select at least one of the first candidate duration or the second candidate duration to represent the session based on a demographics model.

2. The apparatus ofclaim 1, wherein the first partitioning scheme is based on genre information included in payload data of the plurality of pings, and the second partitioning scheme is based on media name information included in the payload data of the plurality of pings.

3. The apparatus ofclaim 1, wherein the processor circuitry is to determine a threshold based on an average of durations of reference sessions represented in panel data obtained from panel meters.

4. The apparatus ofclaim 3, wherein the processor circuitry is to determine a first one of the plurality of pings is the first start ping when at least one of (i) the first one of the plurality of pings is an initial ping from a device associated with the plurality of pings, (ii) payload data from the first one of the plurality of pings indicates a change in session partition based on the first partitioning scheme, or (iii) a time difference between the first one of the plurality of pings and a preceding one or the plurality of pings satisfies the threshold.

5. The apparatus ofclaim 3, wherein the processor circuitry is to determine a first one of the plurality of pings is the second start ping when at least one of (i) the first one of the plurality of pings is an initial ping from a device associated with the plurality of pings, (ii) payload data from the first one of the plurality of pings indicates a change in session partition based on the second partitioning scheme, or (iii) a time difference between the first one of the plurality of pings and a preceding one of the plurality of pings satisfies the threshold.

6. The apparatus ofclaim 1, wherein the processor circuitry is to determine the first candidate duration of the session and the second candidate duration of the session in parallel.

7. The apparatus ofclaim 1, wherein the demographics model includes session duration statistics for demographic categories represented in panel data obtained from panel meters.

8. The apparatus ofclaim 7, wherein the processor circuitry is to compare the first candidate duration and the second candidate duration to the session duration statistics associated with demographics information of a household associated with a device that provided the first start ping, the first end ping, the second start ping, and the second end ping.

9. The apparatus ofclaim 1, wherein the processor circuitry is to:

determine a first performance criterion for the first candidate duration based on a first comparison of the first candidate duration and the demographics model;

determine a second performance criterion for the second candidate duration based on a second comparison of the second candidate duration and the demographics model; and

select at least one of the first candidate duration or the second candidate duration to represent the session based on the first performance criterion and the second performance criterion.

10. At least one non-transitory computer readable medium comprising instructions that, when executed, cause at least one processor to at least:

determine a first candidate duration of a session based on a first difference between a first time associated with a first start ping identified from a plurality of pings based on a first partitioning scheme and a second time associated with a first end ping identified from the plurality of pings based on the first partitioning scheme;

determine a second candidate duration of the session based on a second difference between a third time associated with a second start ping identified from the plurality of pings based on a second partitioning scheme and a fourth time associated with a second end ping identified from the plurality of pings based on the second partitioning scheme; and

select at least one of the first candidate duration or the second candidate duration to represent the session based on a demographics model.

11. The at least one non-transitory computer readable medium ofclaim 10, wherein the first partitioning scheme is based on genre information included in payload data of the plurality of pings, and the second partitioning scheme is based on media name information included in the payload data of the plurality of pings.

12. The at least one non-transitory computer readable medium ofclaim 10, wherein the instructions, when executed, cause the at least one processor to determine a threshold based on an average of durations of reference sessions represented in panel data obtained from panel meters.

13. The at least one non-transitory computer readable medium ofclaim 12, wherein the instructions, when executed, cause the at least one processor to determine a first one of the plurality of pings is the first start ping when at least one of (i) the first one of the plurality of pings is an initial ping from a device associated with the plurality of pings, (ii) payload data from the first one of the plurality of pings indicates a change in session partition based on the first partitioning scheme, or (iii) a time difference between the first one of the plurality of pings and a preceding one or the plurality of pings satisfies the threshold.

14. The at least one non-transitory computer readable medium ofclaim 12, wherein the instructions, when executed, cause the at least one processor to determine a first one of the plurality of pings is the second start ping when at least one of (i) the first one of the plurality of pings is an initial ping from a device associated with the plurality of pings, (ii) payload data from the first one of the plurality of pings indicates a change in session partition based on the second partitioning scheme, or (iii) a time difference between the first one of the plurality of pings and a preceding one of the plurality of pings satisfies the threshold.

15. The at least one non-transitory computer readable medium ofclaim 10, wherein the demographics model includes session duration statistics for demographic categories represented in panel data obtained from panel meters, and wherein the instructions, when executed, cause the at least one processor to compare the first candidate duration and the second candidate duration to the session duration statistics associated with demographics information of a household associated with a device that provided the first start ping, the first end ping, the second start ping, and the second end ping.

16. The at least one non-transitory computer readable medium ofclaim 10, wherein the instructions, when executed, cause the at least one processor to:

determine a first performance criterion for the first candidate duration based on a first comparison of the first candidate duration and the demographics model;

determine a second performance criterion for the second candidate duration based on a second comparison of the second candidate duration and the demographics model; and

select at least one of the first candidate duration or the second candidate duration to represent the session based on the first performance criterion and the second performance criterion.

17. A method comprising:

determining, by executing an instruction with at least one processor, a first candidate duration of a session based on a first difference between a first time associated with a first start ping and a second time associated with a first end ping, the first start ping and the first end ping identified from a plurality of pings based on a first partitioning scheme;

determining, by executing an instruction with the at least one processor, a second candidate duration of the session based on a second difference between a third time associated with a second start ping and a fourth time associated with a second end ping, the second start ping and the second end ping identified from the plurality of pings based on a second partitioning scheme; and

selecting, by executing an instruction with the at least one processor, at least one of the first candidate duration or the second candidate duration to represent the session based on a demographics model.

18. The method ofclaim 17, further including determining a threshold based on an average of durations of reference sessions represented in panel data obtained from panel meters.

19. The method ofclaim 18, further including determining a first one of the plurality of pings is the first start ping when at least one of (i) the first one of the plurality of pings is an initial ping from a device associated with the plurality of pings, (ii) payload data from the first one of the plurality of pings indicates a change in session partition based on the first partitioning scheme, or (iii) a time difference between the first one of the plurality of pings and a preceding one or the plurality of pings satisfies the threshold.

20. The method ofclaim 18, further including determining a first one of the plurality of pings is the second start ping when at least one of (i) the first one of the plurality of pings is an initial ping from a device associated with the plurality of pings, (ii) payload data from the first one of the plurality of pings indicates a change in session partition based on the second partitioning scheme, or (iii) a time difference between the first one of the plurality of pings and a preceding one of the plurality of pings satisfies the threshold.

Images