CLIAMS:1. A television channel bundle cost and price optimization system (102) comprising:
a processor (110);
a segmentation module (120) coupled to the processor (110) to:
determine a plurality of time intervals for each of a plurality of clusters of households, wherein each of the plurality of clusters of households are determined based on an attribute associated with at least one occupant of a household, and wherein the time intervals are indicative of time duration during which a plurality of households within a cluster are viewing television channels;
estimate frequency of viewership, and average of viewership, in each of the plurality of time intervals, for each of a plurality of households associated with the plurality of clusters of households;
determine, for each of the plurality of time intervals, average viewing time proportion for each household, , based on the estimation, wherein the average viewing time proportion is indicative of a time duration for which a television channel is being watched by the household for a pre-determined duration in the time interval; and
a bundling module (122) coupled to the processor (110) to:
determine bundle purchase propensity of the household based on the average viewing time proportion and the current channel bundle, wherein the bundle purchase propensity is indicative of a degree of acceptance of the household to purchase a channel bundle; and
generate the channel bundle for subscription by the household based on the bundle purchase propensity, average viewing time proportion, and frequency of the at least one channel genre, wherein the bundle comprises of a plurality of channels, and wherein each channel is associated with a genre.

2. The television channel bundle cost and price optimization system (102) as claimed in claim 1, wherein the average viewing time proportion is determined based on frequency of at least one channel genre present in a current channel bundle subscribed by the household.

3. The television channel bundle cost and price optimization system (102) as claimed in claim 1, wherein each of the plurality of time intervals comprises a plurality of time granularities, and wherein each of the plurality of time granularities is indicative of a pre-determined time duration.

4. The television channel bundle cost and price optimization system (102) as claimed in claim 1 further comprising a bundle cost and price determination module (124) coupled to the processor (110) to determine price of each of the plurality of channels in the generated bundle, based on bundle purchase propensity, average viewing time proportion, and frequency of the at least one channel genre.

5. The television channel bundle cost and price optimization system (102) as claimed in claim 4, wherein the bundle cost and price determination module (124) further determines cost of each of the plurality of channels based on bundle parameter, and channel parameter.

6. The television channel bundle cost and price optimization system (102) as claimed in claim 4, wherein the bundle cost and price determination module (124) further:
compares, for each of the plurality of channels, the cost of the channel in the generated bundle with cost of the channel in the current channel bundle;
determines whether the cost of each of the plurality of channels is less than the cost of the channel in the current bundle, based on the comparison; and
initiates a cost negotiation for the cost of each of the plurality of channels with a corresponding channel vendor based on the determining.

7. The television channel bundle cost and price optimization system (102) as claimed in claim 5, wherein the bundle parameter comprises the ascertained bundle purchase propensity and bundle price.

8. The television channel bundle cost and price optimization system (102) as claimed in claim 5, wherein the channel parameter comprises at least one of channel price and channel cost.

9. The television channel bundle cost and price optimization system (102) as claimed in claim 1, wherein the segmentation module (120) further:
obtains the attribute associated with the at least one occupant, from a customer relationship management (CRM) database (134), wherein the attribute comprises at least one of age, ethnicity, income, education, language, and profession; and
segments the plurality of households based on the obtained attribute, to create the plurality of clusters.

10. The television channel bundle cost and price optimization system (102) as claimed in claim 1, wherein the bundling module (122) further:
extracts a current channel bundle data associated with the household, wherein the current channel bundle data comprises channels present in a past channel bundle choice; and
determines the frequency of the at least one genre from the extracted channel bundle data to generate the channel bundle.

11. The television channel bundle cost and price optimization system (102) as claimed in claim 1, wherein each of the plurality of time intervals is ascertained from click stream data obtained from at least one of a set top box (106) and click stream database (136).

12. The television channel bundle cost and price optimization system (102) as claimed in claim 1, wherein the frequency and average of viewership are estimated based on pre-determined time granularity in each of the plurality of time intervals.

13. A computer implemented method for television channel bundle cost and price optimization comprising:
determining, by a processor (110), a plurality of time intervals for each of a plurality of clusters of households, wherein each of the plurality of clusters of households are determined based on an attribute associated with at least one occupant of a household, and wherein the time intervals are indicative of time duration during which a plurality of households within a cluster are viewing television channels;
estimating, by the processor (110), frequency of viewership, and average of viewership, in each of the plurality of time intervals, for each of a plurality of households associated with the plurality of clusters of households;
determine, by the processor (110), for each of the plurality of time intervals, average viewing time proportion for each household, based on the estimation, wherein the average viewing time proportion is indicative of a time duration for which a television channel is being watched by the household for a pre-determined duration in the time interval;
determining, by the processor (110), bundle purchase propensity of the household based on the average viewing time proportion and the current channel bundle, wherein the bundle purchase propensity is indicative of a degree of acceptance of the household to purchase a channel bundle;
generating, by the processor (110), the channel bundle for subscription by the household based on the bundle purchase propensity, average viewing time proportion, and frequency of the at least one channel genre, wherein the bundle comprises of a plurality of channels, and wherein each channel is associated with a genre.

14. The method as claimed in claim 13, wherein each of the plurality of time intervals comprises a plurality of time granularities, wherein each of the plurality of time granularities is indicative of a pre-determined time duration.

15. The method as claimed in claim 13 further comprising:
determining, by the processor (110), price of each of the plurality of channels in the generated bundle, based on based bundle purchase propensity, average viewing time proportion, and frequency of the at least one channel genre.

16. The method as claimed in claim 13 further comprising determining, by the processor (110), cost of each of the plurality of channels based on bundle parameter, and channel parameter.

17. The method as claimed in claim 16, wherein the bundle parameter comprises the ascertained bundle purchase propensity and bundle price.

18. The method as claimed in claim 16, wherein the channel parameter comprises at least one of channel price and channel cost.

19. The method as claimed in claim 16 further comprising:
comparing, by the processor (110), for each of the plurality of channels, the cost of the channel in the generated bundle with cost of the channel in the current channel bundle;
determining, by the processor (110), whether the cost of each of the plurality of channels is less than the cost of the channel in the current bundle, based on the comparison; and
initiating, by the processor (110), a cost negotiation for the cost of each of the plurality of channels with a corresponding channel vendor, based on the determining.

20. The method as claimed in claim 13 further comprising:
obtaining, by the processor (110) the attribute associated with the at least one occupant, from a customer relationship management (CRM) database (134), wherein the attribute comprises at least one of age, ethnicity, income, education, language, and profession; and
segmenting, by the processor (110), the plurality of households, based on the obtained attribute, to create the plurality of clusters of households.

21. The method as claimed in claim 13 further comprising:
extracting, by the processor (110), a current channel bundle data associated with the household, wherein the current channel bundle data comprises channels present in a past channel bundle choice; and
determining, by the processor (110), the frequency of the at least one genre from the extracted channel bundle data to generate the channel bundle.

22. The method as claimed in claim 13, wherein each of the plurality of time intervals is determined from click stream data obtained from at least one of a set top box and a channel broadcasting server.

23. The method as claimed in claim 13, wherein the frequency and the average of viewership are estimated based on pre-determined time granularity in each of the plurality of time intervals.

24. A non-transitory computer readable medium having a set of computer readable instructions that, when executed, cause a computing system to:
determine a plurality of time intervals for each of a plurality of clusters of households, wherein each of the plurality of clusters of households are determined based on an attribute associated with at least one occupant of a household, and wherein the time intervals are indicative of time duration during which a plurality of households within a cluster are viewing television channels;
estimate frequency of viewership, and average of viewership, in each of the plurality of time intervals, for each of a plurality of households associated with the plurality of clusters of households;
determine, for each of the plurality of time intervals, average viewing time proportion for each household, based on the estimation, wherein the average viewing time proportion is indicative of a time duration for which a television channel is being watched by the household for a pre-determined duration in the time interval;
determine bundle purchase propensity of the household based on the estimated average viewing time proportion and the current channel bundle, wherein the bundle purchase propensity is indicative of a degree of acceptance of the household to purchase a bundle, and wherein the bundle comprises of a plurality of channels, and wherein each channel is associated with a genre; and
generate a channel bundle for subscription by the household and a price for the channel bundle based on the bundle purchase propensity, average viewing time proportion, and frequency of the at least one channel genre, wherein the bundle comprises of a plurality of channels, and wherein each channel is associated with a genre.
,TagSPECI:

TECHNICAL FIELD
[0001] The present subject matter relates, in general, to television channel bundling
and in particular, to television channel bundle cost and price optimization.
BACKGROUND
[0002] Watching television programs is one of the most popularly known recreational
activities in the world. Today, television programs belonging to a wide variety of genres,
such as entertainment, sports, news, advertisements, and education are available for viewers.
Service vendors, typically use the services of various broadcasters for broadcasting the
television channels to the viewers. The broadcasters typically cater the television channels to
the viewers by way of television channel packages or bundles. Bundles may be understood as
a group of media programs like television programs, and radio broadcasts, watched and/or
listened to by a target group of individuals. The television channel bundles offered to a
household, by the broadcaster, includes varieties of channels which the viewers of the
household may watch.
BRIEF DESCRIPTION OF THE DRAWINGS
[0003] The detailed description is described with reference to the accompanying
figures. In the figures, the left-most digit(s) of a reference number identifies the figure in
which the reference number first appears. The same numbers are used throughout the figures
to reference like features and components. Some embodiments of system and/or methods, in
accordance with embodiments of the present subject matter, are now described by way of
example only, and with reference to the accompanying figures, in which:
[0004] Fig. 1 illustrates a network environment implementing a television channel
bundle cost and price optimization system, in accordance with an embodiment of the present
subject matter.
[0005] Fig. 2 illustrates a method for television channel bundle cost and price
optimization, in accordance with an embodiment of the present subject matter.
3
DETAILED DESCRIPTION
[0006] The present subject matter relates to the systems and methods for television
channel bundle cost and price optimization. In order to cater to needs of the customers,
television (TV) channel broadcasters, hereinafter referred to as broadcasters, form channel
bundles from the channels they subscribe from channel vendors or service providers, such as
Viacom 18, SonyTM, and StarTM TV. The broadcasters provide a wide variety of TV channels,
hereinafter referred to as channels, to their customers through the channel bundles.
[0007] Typically, the channel bundles are formed based on one or more bundling
factors, such as channel genres, geographical location, and price of channels. However,
providing channel bundles to the customers, which are formed based on the bundling factors,
may not always be effective and beneficial to the broadcaster as all customers may not buy
the channel bundles having costly channels. Further, the choice of channel bundle subsciption
primarily depends on customer’s preference for one or more channels in the channel bundle.
In such a case, any channel bundle which includes the channel of customer’s preference, will
eventually be subscribed by the customer. Since, each customer may have liking towards
different channels, and may wish to have such channels in the channel bundle, determining
target customers for subscription of channel bundle becomes difficult.
[0008] Inability in determining target customers may thus affect a broadcaster’s
revenue as the viewers may subscribe to channel bundles from other broadcasters providing
the channels as per viewers’ interest and at lesser price. Thus, more often than not, based on a
competitor’s services and in an attempt to reach to more number of customers, the
broadcasters tend to provide more and more number of channels at lesser prices. However,
providing more number of channels at lesser prices may result in a monetary loss to the
broadcasters, especially in cases where the broadcasters reduce the price of the channels
bought at high costs from the channel vendor. Further, since the broadcasters may not know
the popularity of a channel, the broadcasters may agree to buy the channel from the channel
vendor at a cost decided by the channel vendor, and thus might end up paying more for the
channel.
[0009] Thus, in order to optimize the formation of the channel bundles, and pricing of
such channels and channel bundles, the broadcasters conventionally utilize various methods
to evaluate popularity and utilization of those channels which are catered by them, such as
conducting surveys to inquire random or selected customers about their viewing habits and
viewing patterns. However, such surveys for addressing individual customer-viewing habits
may be time consuming and incur high costs. Further, since the broadcaster is unable to
4
determine the utilization of the channels which are bought at high costs from the channel
vendor, estimation of channel price remains difficult to the broadcaster. Due to the inability
in determination of channel price, opportunity of channel cost negotiation with the channel
vendor remains silent. Furthermore, addition of costly channels may deter customers from
subscribing to channel bundles, which may result in non-purchase of the channel bundles by
the households. Such methods of channel bundling, without consideration of utilization and
household viewing patterns may thus lead to inefficient bundling.
[0010] According to an implementation of the present subject matter, a system and a
method for channel bundle cost and price optimization are described herein. The described
system and method facilitate in ascertaining an optimized price for a channel bundle
distributed by a broadcaster to households of a particular geographical location to which the
broadcaster caters TV channel service. The system and the method further facilitate the
broadcaster in determining optimized cost of the channels based on the optimized price of
channel bundles and negotiating the same with the channel vendors.
[0011] In an implementation, attributes of a household are extracted from a Customer
Relationship Management (CRM) database to determine market segments. The attributes
may be extracted for each occupant of the household and include, but are not limited to, age,
income, education, language, profession, and ethnicity. Based on the extracted attributes, one
or more market segments or clusters of households, such as urban people, business people,
young and mobile, elderly, achievers and singles, may be formed. The cluster formed
corresponds to an attribute selected from amongst the extracted attributes.
[0012] Further, the households are mapped to one or more clusters of households
based on the attributes corresponding to the occupants of the household. Each household may
be mapped to one or more clusters based on different types of occupants in the households. In
an example, household which includes elderly occupants and business people may be mapped
to ‘elderly’ cluster and ‘business people’ cluster, respectively.
[0013] Upon completing the market segmentation, a time segmentation of the
households may be performed based on a click stream data, to determine viewership in one or
more time intervals. Determination of the viewership in the time intervals helps a broadcaster
to analyze time durations in which a particular channel is being watched by multiple
households, and accordingly determine customer preferences. The click stream data may be
defined as a log of all channels and corresponding viewing times, as viewed by the household
in a television. The log of the channels is then aggregated to form 24 time granularities of 60
minutes each. Further, in each time granularity, frequency and average of viewership are
5
determined. Frequency of viewership may be defined as number of households viewing the
TV, while the average of viewership indicates a ratio of sum of viewing times of each
household to the number of households.
[0014] Further, the time granularities are clustered using sequential clustering
technique to form an optimal number of time intervals in a day. Time intervals indicate the
duration during which the households are viewing television channels. In an example, the
time granularities may be clustered to form four time intervals of six hours each, based on
frequency and average of viewership. In another example, the time granularities may be
clustered to form five or six time intervals. Although, frequency of viewership and average of
viewership are determined in each time granularity, it will be understood that the frequency
of viewership and average of viewership may be determined in each time interval also.
[0015] Subsequently, information regarding current channel bundle(s) subscribed by
the household may be extracted from the broadcaster’s database. The current channel bundle
information indicates the channels present in the current channel bundles being offered by the
broadcaster. In an implementation, channel genres and frequency of each channel genre may
be determined for each current channel bundle based on the current channel bundle
information. Frequency of each genre is defined as the number of times a channel genre
repeats in the current channel bundle information. For example, if an ‘entertainment’ genre is
appearing thrice in the current channel bundle information, then the frequency of the
‘entertainment’ genre may be treated as 3.
[0016] In an implementation, viewing time proportions of a household may be
determined for a time interval, from among the optimal number of time intervals, considered
in a day. The viewing time proportion indicates duration for which TV is being watched by
the household in the time interval. Further, frequency of channel genres viewed by the
household and corresponding viewing time proportions mapped to different parts of the day
may be obtained. Additionally, average viewing time proportion may be determined for a
week from the viewing time proportions of all days in the week.
[0017] Subsequently, utilization of channels present in the current channel bundle
may be determined, for all the time intervals considered in the day. In one implementation,
the utilization can be estimated based on regression analysis. In one implementation,
regression analysis may include regressing viewing time proportion with the channel genres,
household attributes, and events, such as award ceremonies watched by the household, for a
given time interval and a given cluster to which the household is mapped.
6
[0018] Further, purchase propensity of the household may be estimated based on the
average viewing time proportion, and current channel bundle information. The purchase
propensity may be indicative of willingness of the household to pay for the channels in the
current channel bundle subscribed by the household. Based on the purchase propensity,
average viewing time proportion, and frequency of channel genre, an optimal channel bundle
and a corresponding bundle price offering for the household are determined. In one
implementation, the optimal channel bundle and the corresponding bundle price may be
determined based on purchase propensity and utilization of channels. For example, a channel-
1 may be characterized with less utilization and more propensity, and a channel-2 may be
characterized with more utilization and less propensity. In such cases, the channel with more
utilization may be included in the channel bundle, and may be priced accordingly.
[0019] Subsequently, optimal cost of each channel in the optimal channel bundle may
be estimated for negotiation with corresponding channel vendors, based on the price of
channel. In an implementation, the optimal cost of each channel may be determined based on
a bundle parameter and a channel parameter. Examples of the bundle parameter may include,
but are not limited to, the bundle purchase propensity and the bundle price, while the
examples of the channel parameter may include, but are not limited to, channel price of a
channel in the current channel bundle and channel cost of the channel in the current channel
bundle.
[0020] Since the channel bundles are generated based on attribute data of the
household obtained from the CRM database, channel viewing time data obtained from the
click stream database, and viewership of channels watched bythe household, the channel
bundles include channels with high average viewing time proportion. In other words, based
on viewership of each household, channels which are characterized with high utilization are
included in the channel bundles. Such channels with high utilization may be the channels
which the household prefers to watch. Further, since the average viewing time proportion is
determined for each channel, the determination of target customers becomes easier for the
broadcaster, as the broadcaster can ascertain utilization of channels which are bought at high
cost from the channel vendors. Furthermore, since the channel bundles include the channels
based on the purchase propensity of the household for current set of channel bundles, a
broadcaster may realize a high purchase propensity of the household for the offering. Also,
based on the utilization of the channels by the households, pricing of each channel, and
subsequently pricing of the channel bundle becomes easier, thus providing opportunity for
7
better negotiation of channel cost with the channel vendor. Thus, the present subject matter
thus provides an efficient system for channel bundle cost and price optimization.
[0021] While aspects of described system(s) and method(s) of channel bundle cost
and price optimization can be implemented in any number of different computing systems,
environments, and/or configurations, the implementations are described in the context of the
following exemplary system(s) and method(s).
[0022] Fig. 1 illustrates a network environment 100 implementing a channel bundle
cost and price optimization system 102, hereinafter referred to as system 102, in accordance
with an embodiment of the present subject matter. In one implementation, the system 102
may be implemented in a variety of computing devices including a laptop computer, a
desktop computer, a workstation, a mainframe computer, and the like.
[0023] The network environment 100 further includes one or more user devices 104-
1, 104-2, and 104-N, collectively referred to as user device(s) 104, communicating, via settop-
boxes (STB) 106-1, 106-2, and 106-N, hereinafter referred to as STB 106, with the
system 102 through a network 108. The user devices 104 may be implemented as variety of
display devices connected to the STB 106 for displaying TV content, such as TV channels
and video-on-demand provided by STB 106. Examples of the user devices 104 include, but
are not limited to, an electro luminescent display (ELD), a plasma display panel (PDP), an
organic light emitting diode (OLED), a light emitting diode (LED) display, a liquid crystal
display (LCD), and a thin-film transistor LCD (TFT-LCD). As will be understood, the STB
106 receives TV signals, having the TV content in digital format, from TV broadcasters
broadcasting the TV content and provides the TV content to the user devices 104 in an analog
form.
[0024] The network 108 may be a wireless or a wired network, or a combination
thereof. The network 108 may be a combination of wired and wireless networks. The network
108 may be implemented by service provider systems through satellite communication,
terrestrial communication, or may be implemented through the use of routers and access
points connected to various Digital Subscriber Line Access Multiplexers (DSLAMs) of wired
networks. The network 108 can be implemented as one of the different types of networks,
such as intranet, local area network (LAN), wide area network (WAN), and the Internet. The
network 108 may either be a dedicated network or a shared network, which represents an
association of the different types of networks that use a variety of protocols, for example,
Hypertext Transfer Protocol (HTTP), Transmission Control Protocol/Internet Protocol
(TCP/IP), Wireless Application Protocol (WAP), etc., to communicate with each other.
8
[0025] In one embodiment of the present subject matter, the system 102 may be
implemented by a TV broadcaster for channel bundle cost and price optimization. For
instance, a broadcaster broadcasting TV channels obtained from one or more channel vendors
or service providers may implement the system 102 for obtaining optimized channel bundles
that may be broadcasted to a plurality of households implementing the STB 106 and the user
devices 104. Further, the system may facilitate the broadcaster in determining an optimum
cost of the channels based on the optimized channel bundles.
[0026] In one implementation, the system 102 includes processor(s) 110, interface(s)
112, and a memory 114. The processor(s) 110 can be a single processing unit or a number of
units, all of which could include multiple computing units. The processor(s) 110 may be
implemented as one or more microprocessor, microcomputers, digital signal processors,
central processing units, state machines, logic circuitries, and/or any devices that manipulate
signals based on operational instructions. Among other capabilities, the processor(s) 110 are
adapted to fetch and execute computer-readable instructions stored in the memory.
[0027] The functions of the various elements shown in the fig.1, including any
functional blocks labeled as “processor(s)”, may be provided through the use of dedicated
hardware as well as hardware capable of executing software in association with appropriate
software. When provided by a processor, the functions may be provided by a single dedicated
processor, by a single shared processor, or by a plurality of individual processors, some of
which may be shared. Moreover, explicit use of the term “processor” should not be construed
to refer, exclusively, to hardware capable of executing software, and may implicitly include,
without limitation, Digital Signal Processor (DSP) hardware, network processor, Application
Specific Integrated Circuit (ASIC), Field Programmable Gate Array (FPGA), Read Only
Memory (ROM) for storing software, Random Access Memory (RAM), non-volatile storage.
Other hardware, conventional and/or custom, may also be included.
[0028] The interface(s) 112 may include a variety of software and hardware
interfaces, for example, interface for peripheral device(s), such as a keyboard, a mouse, a
microphone, an external memory, a speaker, and a printer. Further, the interface(s) 112 may
include one or more ports for connecting the system 102 with other computing devices, such
as web servers, and external databases. The interface(s) 112 may facilitate multiple
communications within a wide variety of protocols and networks, such as a network,
including wired networks, e.g., LAN, cable, etc., and wireless networks, e.g., WLAN,
cellular, satellite, etc.
9
[0029] The memory 114 may be coupled to the processor 110 and may include any
computer-readable medium known in the art including, for example, volatile memory, such
as Static Random Access Memory (SRAM) and Dynamic Random Access Memory
(DRAM), and/or non-volatile memory, such as Read Only Memory (ROM), Erasable
Programmable ROMs (EPROMs), flash memories, hard disks, optical disks, and magnetic
tapes.
[0030] The system 102 may also include module(s) 116 and data 118. The modules
116 and the data 118 may be coupled to the processor(s) 110. The modules 116, amongst
other things, include routines, programs, objects, components, data structures, etc., which
perform particular tasks or implement particular abstract data types. The modules 116 may
also be implemented as, signal processor(s), state machine(s), logic circuitries, and/or any
other device or component that manipulate signals based on operational instructions.
[0031] In another aspect of the present subject matter, the modules 116 may be
computer-readable instructions which, when executed by a processor/processing unit,
perform any of the described functionalities. The machine-readable instructions may be
stored on an electronic memory device, hard disk, optical disk or other machine-readable
storage medium or non-transitory medium. In one implementation, the computer-readable
instructions can be also be downloaded to a storage medium via a network connection.
[0032] In one implementation, the module(s) 116 include a segmentation module 120,
a bundling module 122, a bundle cost and price determination module 124, and other
module(s) 126. The other module(s) 126 include programs that supplement applications or
functions of the system 102. The data 118 serves, amongst other things, as a repository for
storing data obtained and processed by one or more module(s) 116. The data 118 includes,
for example, subscriber data 128, bundle data 130, and other data 132. The other data 132
includes data generated as a result of the execution of one or more modules in the other
module(s) 126.
[0033] As previously described, the system 102 is implemented to optimize channel
bundling, thus facilitating the broadcaster in creating channel bundles based on likes and
dislikes of the occupants of the households catered to by the broadcaster. For the purpose, the
segmentation module 120 may segment the households into one or more clusters based on
one or more attribute corresponding to the occupants of the households. In one
implementation, the segmentation module 120 may obtain attributes of a household from a
CRM database 134. The CRM database 134 may be an external database storing, amongst
other data, the attributes of the households belonging to a geographical area. The attributes
10
may include, but are not limited to, age, income, education, language, profession, and
ethnicity.
[0034] Based on the obtained attributes, the segmentation module 120 may form
clusters which include the households, such that each cluster is indicative of an attribute of a
household from among the extracted attributes of the household. For example, the attributes
of households i1, i2, …., i10 obtained by the segmentation module 120 may be age, income,
and education. Accordingly, the segmentation module 120 may form three clusters, such as
elderly, business people, and young and mobile, from the households i1, i2, …., i10, and map
each household to at least one cluster. Further, the segmentation module 120 may indicate the
number of households mapped to each cluster. The segmentation module 120 may further
store data related to the clusters being formed in the subscriber data 128.
[0035] The segmentation module 120 may further determine time intervals in a day
with optimal TV viewership, for determining customer preferences. The term ‘viewership’
may be defined to refer to the viewers of a particular television program or a television
channel. In one implementation, the segmentation module 120 may extract click stream data
of a household from the click stream database 136, on per second basis, which provides log
of viewing history of the household, for each second of a day. In another implementation, the
click stream data may be obtained from the STB 106 which is communicatively, connected to
the user device 104. In order to obtain the click stream data from the STB 106, a click stream
application may be installed in the STB 106 to capture channel information and viewing
duration. Further, in an example, the click stream application may also be implemented in the
system 102, thus making the extraction of click stream data easier and faster. Although the
description herein discusses the extraction of the click stream data from the click stream
database 136 and STB 106, it will be appreciated that the click stream data may be obtained
from other sources as well, such as dish antennas (not shown in the figure) connected to the
STB 106 albeit with few variations in devices present at the other sources.
[0036] For the purpose of determining optimal TV viewership, the segmentation
module 120 may aggregate the log of the channels in the viewing history, to form 24 time
granularities of 60 minutes each. The segmentation module 120 may ascertain frequency and
average of household viewing for each household, in each time granularity, based on click
stream data. Frequency of household viewing indicates the number of households viewing the
cable TV in a particular time granularity, and average of household viewing indicates a ratio
of sum of viewing times of the households in a cluster of households, to the number of
households mapped to the cluster of households. For example, in a time granularity of 60
11
minutes, the segmentation module 120 may determine that 300 households of a particular
cluster are viewing the TV from 10:00 a.m. to 11:00 a.m. The number 300 indicates the
frequency of household viewing for the particular time-granularity. Further, in the same time
granularity, households i1, i2, i3, and i4 may watch the cable TV for 15 minutes, 20 minutes,
30 minutes, and 60 minutes, respectively. The sum of the viewing times divided by four
households, i.e., 31.25 minutes indicates the average of household viewing, and may be
represented by equation (1) given below:
Average of household viewing = �� �������������� ��������
������
�� ….. (1)
where, i represents the household, and n represents the number of households.
[0037] The segmentation module 120 may save the ascertained frequency and
average of household viewing, in the subscriber data 128. Upon ascertaining, the
segmentation module 120 may cluster the time granularities to form an optimal number of
time intervals in a day. In one implementation, the segmentation module 120 may form four
time intervals in a day, of six hours each, using sequential clustering technique. The four time
intervals in the day may be referred to as the major time intervals, such as morning,
afternoon, evening, and night. In an example, the segmentation module 120 may form five or
six time intervals, as chosen by the broadcaster. Further, duration of time interval may be four
hours, five hours, or six hours, depending on number of time intervals formed. Although
determination of the frequency and average of household viewing have been described for a
time granularity, it will be appreciated that the frequency and average of household viewing
may also be determined for each time interval.
[0038] The segmentation module 120 may determine viewing time proportions of
each household, in a particular time interval, from among the optimal number of time
intervals, for determining utilization of channels included in a current channel bundle
currently subscribed by the household. The viewing time proportion indicates duration for
which TV is being watched by a household in the particular time interval. Based on the
viewing time proportions, the time intervals having maximum viewership may be mapped to
different parts of the day, such as morning, noon, evening, and night. In one implementation,
a maximum viewing time proportion associated with a channel may be considered as optimal
viewing time interval for that channel. The segmentation module 120 may further determine
average viewing time proportions for each household. The average viewing time proportion
may be determined by considering the viewing time proportions of all days in a week, for
each of the households.
12
[0039] In one implementation, the segmentation module 120 may extract information
related to the current channel bundles, to determine channel genres present in the current
channel bundle. From the extracted current channel bundle information, frequency of each
channel genre in the current channel bundle may be determined. For example, the current
channel bundle subscribed by the households may include an ABC-1 channel, an ABC-2
channel, an ABC-3 channel, a sports-1 channel, and a sports-2 channel. The ABC-1 channel,
the ABC-2 channel, and the ABC-3 channel may belong to ‘entertainment’ genre while the
sports-1 channel and the sports-2 channel may belong to ‘sports’ genre. Thus, the frequency
of ‘entertainment’ genre may be treated as 3 and the frequency of ‘sports’ genre may be
treated as 2.
[0040] Based on the frequency of the channel genre, the segmentation module 120
may estimate the viewing time proportions in the time interval. The viewing time proportion
may indicate time duration of a particular channel genre watched by a household mapped to a
particular cluster, in the pre-determined time interval. For example, the segmentation module
120 may determine that in a pre-determined time interval of 2 hours between 10am-12pm, a
household i1 watched the ABC-1 channel from 10 am to 11 am, i.e., for 1 hour. Hence, the
viewing time proportion of the household i1 watching the ABC-1 channel may be estimated
to be 0.5 which indicates 1 hour of viewing in the 2 hours time interval. Similarly, the
viewing time proportion of households mapped to each cluster may also be determined based
on channel genre viewed by the households.
[0041] In one implementation, the segmentation module 120 may determine viewing
time proportion for all the days in a week with respect to a particular channel genre, and may
accordingly determine average viewing time proportion. The average viewing time
proportion indicates an average of all the viewing time proportions determined over a week.
For example, the viewing time proportions for the ‘entertainment’ genre on Monday,
Tuesday,…., and Sunday may be determined as 0.5, 0.4, 0.7, 0.8, 0.5, 0.6, and 0.4,
respectively. Thus, the average viewing time proportion for the week, of the household, may
be determined as sum of the viewing time proportions divided by 7, which accounts to about
0.55. The average viewing time proportion of 0.55 may indicate a utilization of 55% of the
‘entertainment’ genre by the household.
[0042] Similarly, the segmentation module 120 may determine the average viewing
time proportions for each genre in the list of genres present in the current channel bundle
subscribed by the household. Further, the bundling module 122 may perform a regression
analysis to determine the viewing time proportion for a given channel bundle, denoted by
13
T(c). The regression analysis is a statistical measure that attempts to determine the strength of
the relationship between one dependent variable, such as the viewing time proportion; and a
series of other changing variables, such as channel genres, attributes of the household, and
events watched. In one implementation, for a given time interval in a day and for a given
household, the viewing time proportion may be regressed with channel genres, attributes of
the household, and the events watched by the households. The regression equation may be
represented by the equation (2) given below:
����������/�������� �� Σ ���� �� ����
��
������ �� Σ ���� �� ���� �� ���� �� ����������
��
������ ….. (2)
where, ����������/�������� represents viewing time proportion in a day d, for a household i viewing a
channel genre c broadcasting a given event E; ���� and ���� are coefficients for channel and
household factors respectively, and ���������� represents an error term. Further, a set of the
channel genre c, the event E and household attributes h may be represented by the equations
(2), (3), and (4) as given below:
�� �� ������, ����, ����, … … … . . , ������ ….. (3)
�� �� ������, ����, ����, … … … . . , ����] ….. (4)
�� �� ������, ����, ����, … … … , ���������� ….. (5)
where, the term ���� represents an event E which may take place for few days d, for example
Indian Premier League (IPL).
[0043] In an implementation, the bundling module 122 may utilize the current
channel bundle information of the household to determine frequency of each genre in the
current channel bundle. The bundling module 122 may further, extract price of each of the
current channel bundles offered to the household, from a bundle offering database (not shown
in the figure) maintained by the broadcaster. In an example, the system 102 may be coupled
to the bundle offering database or a similar database which may help fetch information
regarding the current channel bundles offered to the household and corresponding price of the
current channel bundles.
[0044] Further, the bundling module 122 may perform a regression analysis on
purchase of the current channel bundle to determine bundle purchase propensity of a
household. The term, bundle purchase propensity indicates a willingness of the household to
14
buy the bundles offered to the household. The regression equation for determining bundle
purchase propensity may be represented by the equation (6) given below:
������������ �� 1| ������, ���������� �� ��
���������� ��������Σ Σ ����������������������
��
������
��
������ ���������������� ….. (6)
where, ���� represents a purchase variable; ������ represents a purchase of a bundle b from among
the current channel bundles offered to the household i such that, ������ �� 1 may indicate a
purchase of the bundle and ������ �� 0 may indicate a no purchase of the bundle; ���������� represents
viewing times of a channel genre c, included in the bundle b, in different time intervals s; ������
represents price offered for the bundle b to the household i; θ0, θibsc, and θ2 represent
coefficients of the attributes. Further, the term ���������� may be determined by the equation (7)
given below:
���������� �� Σ Σ ���������������� ������������/��������
������
������ ….. (7)
[0045] The bundling module 122 may subsequently, generate a channel bundle for the
household based on the determined bundle purchase propensity of the household, optimal
time interval, and average viewing time proportion of each cluster. In case the bundling
module 122 determines a low bundle purchase propensity of any bundle from among the
current channel bundles subscribed by the household, the bundling module 122 may prepare a
bundle for offering to the household, which includes similar channels, of the same channel
genre, to those present in the current channel bundle. For example, the household may be
currently subscribed to bundle b1 which may contain channel-1, channel-2, channel-3,……,
channel-7. The segmentation module 120 may determine that the channel-3, channel-4, and
channel-5 are not associated with considerable viewing time proportions. In other words, a
low percentage of average viewing time proportion may be indicated against the channel
genres of the three channels, i.e., channel-3, channel-4, and channel-5. The bundling module
122 may thus, ascertain that the household is not willing to pay for the bundle b1 as 3
channels among the 7 channels are not watched, thus indicating that the purchase propensity
of the household for the bundle b1 may be low. In such a scenario, the bundling module 122
may substitute the channel-3, channel-4, and channel-5 with similar channels.
[0046] In an implementation, the channels that may be used for substituting the
channels with low percentage of average viewing time proportion may also be determined by
the bundling module 122 based on the average viewing time proportion of each cluster. Since
the channel bundle is generated according to the determined viewing time proportions at each
time interval of the household, a high purchase propensity may be associated with the
15
channel bundle. In one implementation, the channel bundle for offering to the household may
be generated by the bundling module 122 using the equation (8) given below:
������Σ Σ ������ �� Σ ������
��
������ ��Σ Σ ��������
��
������
��
������
��
������
��
������ ….. (8)
where, ������ represents price of bundle b for household i; ������ represents advertisement
revenue for different time interval s and �������� represents viewing time of bundle b, by the
household i in time interval s. The bundling module 122 may further, save the generated
channel bundles in the bundle data 130.
[0047] In one implementation, the generation of channel bundle b may be subjected
to one or more pre-determined conditions saved in the bundle data 130.
[0048] For instance, the first condition may indicate that a channel n in the channel
bundle b may be allocated at most to one bundle b for a given household as represented by
equation (9) mentioned below:
Σ ����,��
�� �� �� 1, ����, ����
������ ….. (9)
where, ����,��
�� is 1, if channel n is linked to bundle b for household i, else 0. In other words, if
the segmentation module 120 determines a high viewing time proportion of a channel n based
on the click stream data, then the channel n may be allocated to a bundle b generated by the
bundling module 122. The value of ����,��
�� =1 may indicate that the channel n has been allocated
to the bundle b, based on which the bundling module 122 may allocate other channels to the
bundle b. At a particular instance, if the bundling module 122 determines the value of ����,��
�� to
be 0, then the bundling module 122 may decide a new bundle for allocation of channel n.
[0049] A second condition may indicate that a channel n may be considered as a base
for the bundle b, using equations (10) and (11) mentioned below:
����,��
�� �� ����
�� ����, ���� ….. (10)
����,��
�� �� ����,��
�� ����, ���� ….. (11)
where, ����
�� is 1 if channel n is considered as center for the bundle b for the household i. A
channel n may be considered as center for bundling, when the channel n is associated with a
high average viewing time proportion. For example, if a channel 10 indicates a high average
viewing time proportion, then the channel genre of the channel 10 may be considered as
center for the bundle b.
16
[0050] A third condition may indicate maximum number of channel bundles that may
be formed by the bundling module 122, as represented by equation (12) mentioned below:
Σ ����
�� ��
������ �� ������������������ ���� ….. (12)
where, Maxbundle indicates the maximum number of channel bundles which includes channels
having high average viewing time proportion as center of each bundle. Further, in an
example, the number of channel bundles to be offered to a household i, may be based on the
number of channel genres c watched by the cluster to which the household I is mapped to.
[0051] A fourth condition may indicate maximum number of channels in the channel
bundle, as represented by equation (13) mentioned below:
Σ ����,��
�� �� �� ��������������������, ����, ����
������ ….. (13)
where, �������������������� indicates the maximum number of channels which may be linked to the
channel bundle b.
[0052] A fifth condition may indicate that a bundle price of the channel bundle,
generated by the bundling module 122, to be less than sum of a-la-Carte prices, as
represented by equation (14) mentioned below:
���� ��
�� Σ ���� �� ����,��
�� �� , ����, ����
������ ….. (14)
where, ���� ��
represents price of bundle b for household i; ���� represents a-la-Carte price of each
channel in the bundle b. The term ‘Ala-Carte’ indicates quoted price of each channel in a
channel menu, which may be available with the channel vendor.
[0053] A sixth condition may indicate utilization of the bundle b in term of channel
genre c for time interval s, as represented by equation (15) mentioned below:
���������� �� Σ �������� �� ����,��
�� �� ����, ����
������ , ����, ���� ….. (15)
where, ���������� represents viewing times of genre c of bundle b in different time intervals s by
the household i; �������� represents viewing time proportion, in time interval s, of channel n
belonging to genre c. When a channel n is included in the bundle b, the value ����,��
�� is
considered as 1, thus the viewing time of the channel n can be determined. In case, the
channel n is not included in the bundle b, the ����,��
�� is considered as 0, and thus the viewing
time of the channel n remains 0. Similarly, the viewing times for all the channels in the
bundle b may be determined by the bundling module 122 and summation of the viewing
times represents the utilization of the channel n.
���������� �� ���� ��
�� ������������������ , ����, ����, ����, ���� ….. (16)
17
where, ������������������ represents an operational constant which may indicate a maximum
utilization. The value of ������������������ may be (24 * 60 * 60) which indicates utilization of a
bundle b in throughout the day.
[0054] A seventh condition may indicate utilization of the bundle b in terms for time
interval s, as represented by equation (17) mentioned below:
�������� �� Σ ���������� �� ����, ����
������ , ���� ….. (17)
From the viewing times obtained from the equation (16), the bundling module 122
determines viewing times of the bundle b in different time intervals s by the household i for
each channel genre c. Similar to equation (16), the equation (17) may consider value of ����������
as 1 and 0, if channel genre c is present and not present in the bundle b, respectively.
�������� �� ���� ��
�� ������������������ , ����, ����, ���� ….. (18)
[0055] An eight condition may indicate bundle Purchase Propensity of the bundle b,
as represented by equation (19) mentioned below:
���� ��
�� ��1 �� ���� ��
�� �� ������������������
��
10 �� ������������������������ �� ���� ��Σ Σ ���������� �� ����������
��
������
��
������ �� ���� �� ���� ��
��, ����, ���� ... (19)
where, ���������������������� indicates a threshold value for purchase of the bundle b. A value below
the ���������������������� may indicate ‘no purchase’ or less purchase propensity and a value above the
���������������������� may indicate a ‘purchase’ or ‘high purchase propensity’.
���� ��
�� ����
�� ����, ���� ….. (20)
In above equation, the value of ���� ��
and ����
�� equal to 1 indicates a purchase of the bundle b when
the bundle b includes a channel n as the center for bundle.
[0056] A ninth condition may indicate constraints to the pricing of the bundle b, as
represented by equation (21) mentioned below:
���� ��
�� ���� ��
�� ������������������ ����, ���� ….. (21)
The bundle cost and price determination module 124 may generate the price of the bundle b
based on the equation (21) which indicates that the price ���� ��
of the bundle b should be
generated as a value equal to the purchase propensity of the bundle b. In other words, in order
to realize a high purchase propensity of the bundle b, the price of the bundle b needs to be
low.
���� ��
�� ���� ��
���� �������������� ���� �������������� ������������, ���� ….. (22)
18
Further, the bundle cost and price determination module 124 may consider the price of the
current channel bundle subscribed by a household i to generate the price for bundle b. In
order to have ���� ��
= 1, the price of bundle b needs to be less than the price of current channel
bundle.
����,��
�� , ����
��, ���� ��
�� ��0,1�� ….. (23)
��������, ����������, ���� ��
�� ��0, ��∞�� ….. (24)
[0057] In an implementation, the bundle cost and price determination module 124
may determine cost of each channel n in the generated channel bundle b, based on bundle
parameter and channel parameter. The bundle parameter may include the ascertained bundle
purchase propensity and bundle price of the generated bundle. The channel parameter may
include channel cost and channel price in the current channel bundle. The term, channel cost
may be understood as a sum for which a service provider is willing to provide his channel to
the broadcaster, and the term channel price may be understood as the price for which the
broadcaster is willing to provide the channel to the household.
[0058] Further, the bundle cost and price determination module 124 may estimate a
margin for negotiation of channel cost with the channel vendor. In an implementation, the
bundle cost and price determination module 124 may generate a list having maximum cost
and minimum cost for each of the channels in the channel bundle to facilitate negotiation of
the channel cost between the broadcasters and the channel vendors. The maximization of
margin may be represented by an equation as mentioned below:
������ Σ ������ �� Σ ������
��
������
��
������ ….. (25)
where, ������ represents optimal bargain cost of channel n; ������ represents optimal price of
channel n. Further, the bundle cost and price determination module 124 may save the
generated list of prices in the bundle data 130. In one implementation, the maximization of
margin may be subjected to one or more pre-determined conditions.
[0059] For instance, the first condition may indicate that a sum of individual channel
price may be less than the price of the channel bundle b, as represented by equation (26)
mentioned below:
Σ���������������������� �� ������, ���� ….. (26)
where, ������ represents the price of the channel bundle b generated by the bundling module
122. In one implementation, the bundling module 122 may determine the price of all the
channel bundles such that the price is less than the sum of price of all the channels included
in the channel bundle.
19
[0060] A second condition may indicate that a total margin of a particular channel to
be greater, in case of bundling than a-la-carte, as represented by equation (27) mentioned
below:
������ �� �������� �� �������� �� ���� �� ������ �� ������ ….. (27)
where, ������ represents number of households for the channel n during bundling; ���� represents
number of households for the channel n; ���� represents price of the channel n; and ����
represents cost of the channel n. The left hand side of the equation (27) may represent a total
margin after the channel bundle has been generated by the bundling module 122, and the
right hand side of the equation (27) represents a total margin obtained for the current channel
bundle. In an implementation, ������ may be considered as the number of households which
show a high average viewing time proportion for the channel n.
[0061] A third condition may indicate that a total revenue of a particular channel to be
greater in case of bundling than Ala-carte, as represented by equation (28) mentioned below:
Purchase of the generated channel bundle may be represented by an equation as mention
below:
������ �� ���������� �� ���� �� �������� ….. (28)
The left hand side of the equation (28) represents a purchase value of the channel bundle
generated by the bundling module 122, and the right hand side of the equation (28) represents
a purchase of the current channel bundle. As the channel bundle is generated based on bundle
purchase propensity and average viewing time proportion of the household for the current
channel bundles, the purchase of the channel bundle remains higher than that of the current
channel bundle.
[0062] A fourth condition may indicate negotiation of the channel cost with the
channel vendor, as represented by equation (21) mentioned below:
���������� �� �������� ….. (29)
More often, the channel vendor offers the channel for a high cost. In order to determine a
higher margin than the margin determined during current channel bundle, the cost of channel
included in the channel bundle should be negotiated at a lower cost as compared to the cost
offered by the channel vendor.
[0063] A fifth condition may indicate determination of optimal price of channel n, as
represented by equation (30) mentioned below:
���������� �� �������� ….. (30)
20
In order to realize a high bundle purchase propensity of the generated bundle, the bundling
module 122 may estimate a lower price for each channel of the channel bundle. Based on the
estimated price of each channel, the channel bundle price may thus be determined by the
bundle cost and price determination module124.
[0064] Fig. 2 illustrates method 200 for channel bundle optimization, in accordance
with an embodiment of the present subject matter. According to an aspect, the concept of
channel bundling optimization are described with reference to the channel bundling cost and
price optimization system 102 described above.
[0065] The method 200 may be described in general context of computer executable
instructions. Generally, computer executable instructions can include routines, programs,
objects, components, data structures, procedures, modules, functions, etc., that perform
particular functions or implement particular abstract data types. The method 200 may also be
practiced in a distributed computing environment where functions are performed by remote
processing devices that are linked through a communications network. In a distributed
computing environment, computer executable instructions may be located in both local and
remote computer storage media, including memory storage devices.
[0066] The order, in which the method 200 is described, is not intended to be
construed as a limitation, and any number of the described method blocks can be combined in
any order to implement the methods 200, or an alternative method. Additionally, individual
blocks may be deleted from the method 200 without departing from the spirit and scope of
the subject matter described herein. Furthermore, the method 200 can be implemented in any
suitable hardware, software, firmware, or combination thereof. The method 200 is explained
with reference to the channel bundling optimization system 102, however, it will be
understood that the method 200 can be implemented in other systems as well.
[0067] At block 202, attributes of at least one occupant of each of a plurality of
households are obtained to form one or more clusters of households. In one implementation
each cluster of household may be indicative of an attribute of the household. For example, a
channel bundling optimization system, such as the system 102 may create one or more
clusters of households based on the attributes, such that each cluster of households
corresponds to a particular attribute; and map each of the household having the particular
attribute to the cluster of households corresponding to the attribute. The household may
contain one or more occupants, and all the attributes associated with each occupant of the
household, when considered together may be understood as the attribute of the household.
The attributes of the household may be obtained from a Customer Relationship Management
21
(CRM) database. In an example, the attributes may include, but are not limited to, age,
income, education, language, profession, and ethnicity. In one implementation, the household
may be mapped to one or more clusters of households depending on the different types of
occupants in the households. For example, a household which includes children and adults,
may be mapped to both ‘young and mobile’ cluster and ‘business people’ cluster.
[0068] At block 204, one or more time intervals are determined for each cluster of
households, where the time interval indicates the time duration for which the households are
watching television channels. In one implementation, the system 102 may extract click
stream data, of all the households belonging to a particular cluster of households, from a click
stream database to determine the one or more time intervals for each cluster. In an example,
the click stream data may be obtained from a Set-Top-Box (STB) connected to a user device,
such as a television (TV). The click stream data provides TV viewing history of the
household in a form of log, which includes information regarding channels viewed and
corresponding viewing time for each second of a day. The system 102 aggregates the
extracted data to form time granularities of 60 minutes each. Further, the time granularities
are clustered to form one or more time intervals in a day. In one implementation, the system
102 may cluster the time granularities to form four time intervals in a day, and these four time
intervals may be referred to as the major time intervals which indicate morning, afternoon,
evening, and night. In an example, the number of intervals formed by the system 102 may be
five or six intervals.
[0069] At block 206, frequency and average of viewership may be estimated in each
time interval from among the one or more time intervals, for each household in the cluster of
households. In each time granularity, the system 102 may ascertain frequency and average of
viewership, to determine user preferences. Frequency of viewership may be defined as
number of households viewing the TV, while the average of household viewing indicates a
ratio of sum of viewing times of each household to the number of households. In one
implementation, the system 102 may estimate the frequency and average of viewership in
each time interval.
[0070] At block 208, average viewing time proportion in a time interval, from among
the plurality of time intervals, is determined for each household mapped to the cluster of
households. In one implementation, viewing time proportions of each household, in a
particular time interval, is determined for ascertaining utilization of channels included in a
current channel bundle currently subscribed by the household. Viewing time proportion may
22
indicate duration of viewing a particular channel by the household, in the particular time
interval.
[0071] In one implementation, information related to the current channel bundle
subscribed by the household may be extracted. The current channel bundle information
includes channels, included in the current channel bundle. From the extracted channel bundle
information, channel genres and frequency of each channel genre in the current channel
bundle may be determined. Based on the channel genres, and their frequency, viewing time
proportion may be determined. Further, average viewing time proportion, in a week, of the
household may be determined from viewing time proportions from all days of the week.
[0072] At block 210, bundle purchase propensity of the household is determined
based on average viewing time proportion and the current channel bundle. In one
implementation, price of each of the current channel bundles offered to the household, may
be extracted from a bundle offering database which may be maintained by the broadcaster. In
an example, the bundle purchase propensity may be determined by the system 102. The
bundle purchase propensity may indicate willingness of the household to purchase or pay for
the channels in the current channel bundle.
[0073] At block 212, a channel bundle for subscription by the household, and a price
for the channel bundle are generated. In an example, the system 102 may generate the
channel bundle and the price for the channel bundle based on the bundle purchase propensity,
average viewing time proportion, and frequency of the channel genre in the current channel
bundle. The generated channel bundle may include channels which are aligned with the
viewing time proportions and the channels genres watched by the household. In another
example, channel bundle may include channels based on analysis of the cluster. In other
words, channels having high average viewing time proportions may also be included in the
channel bundle.
[0074] In an implementation, price for the generated channel bundle may be
estimated based on the bundle purchase propensity of the household for current channel
bundle and price of the current channel bundle subscribed by the household. In an example,
the price of the generated channel bundle may be determined based on bundle purchase
propensity, average viewing time proportion, and frequency of channel genre. Further, the
system 102 may determine bundle purchase propensity of the household for the generated
channel bundle. In case, the bundle purchase propensity is determined as a high value, the
price may be tagged to the bundle accordingly.
23
[0075] At block 214, cost of each of the plurality of channels in the channel bundle is
determined. In an example, the system 102 may determine cost of each channel based on
bundle parameter and channel parameter. The bundle parameter may include the ascertained
bundle purchase propensity and bundle price of the generated bundle. The channel parameter
may include channel cost and channel price of channels in the current channel bundle. In
another example, cost of each channel may be determined based on viewing times associated
with the channel. The determined cost may be used for negotiating with a channel vendor
providing the channel.
[0076] In an implementation, one or more of the method blocks, of the method 200,
described herein, may be implemented at least in part, as instructions, embodied in a nontransitory
computer-readable medium and executable by one or more computing devices. In
general, a processor (for example a microprocessor) receives instructions, from a nontransitory
computer-readable medium (for example, a memory), and executes those
instructions, thereby performing one or more method(s), including the method 200 described
herein. Such instructions may be stored and/or transmitted using any of a variety of known
computer-readable media.
[0077] Although embodiments for channel bundling optimization have been
described in language specific to structural features and/or methods, it is to be understood
that the subject matter is not necessarily limited to the specific features or methods described.
Rather, the specific features and methods are disclosed as exemplary embodiments for
channel bundling optimization.
24
I/We claim:
1. A television channel bundle cost and price optimization system (102) comprising:
a processor (110);
a segmentation module (120) coupled to the processor (110) to:
determine a plurality of time intervals for each of a plurality of clusters
of households, wherein each of the plurality of clusters of households are
determined based on an attribute associated with at least one occupant of a
household, and wherein the time intervals are indicative of time duration
during which a plurality of households within a cluster are viewing television
channels;
estimate frequency of viewership, and average of viewership, in each
of the plurality of time intervals, for each of a plurality of households
associated with the plurality of clusters of households;
determine, for each of the plurality of time intervals, average viewing
time proportion for each household, , based on the estimation, wherein the
average viewing time proportion is indicative of a time duration for which a
television channel is being watched by the household for a pre-determined
duration in the time interval; and
a bundling module (122) coupled to the processor (110) to:
determine bundle purchase propensity of the household based on the
average viewing time proportion and the current channel bundle, wherein the
bundle purchase propensity is indicative of a degree of acceptance of the
household to purchase a channel bundle; and
generate the channel bundle for subscription by the household based on
the bundle purchase propensity, average viewing time proportion, and
frequency of the at least one channel genre, wherein the bundle comprises of a
plurality of channels, and wherein each channel is associated with a genre.
2. The television channel bundle cost and price optimization system (102) as claimed in
claim 1, wherein the average viewing time proportion is determined based on
frequency of at least one channel genre present in a current channel bundle subscribed
by the household.
25
3. The television channel bundle cost and price optimization system (102) as claimed in
claim 1, wherein each of the plurality of time intervals comprises a plurality of time
granularities, and wherein each of the plurality of time granularities is indicative of a
pre-determined time duration.
4. The television channel bundle cost and price optimization system (102) as claimed in
claim 1 further comprising a bundle cost and price determination module (124)
coupled to the processor (110) to determine price of each of the plurality of channels
in the generated bundle, based on bundle purchase propensity, average viewing time
proportion, and frequency of the at least one channel genre.
5. The television channel bundle cost and price optimization system (102) as claimed in
claim 4, wherein the bundle cost and price determination module (124) further
determines cost of each of the plurality of channels based on bundle parameter, and
channel parameter.
6. The television channel bundle cost and price optimization system (102) as claimed in
claim 4, wherein the bundle cost and price determination module (124) further:
compares, for each of the plurality of channels, the cost of the channel in the
generated bundle with cost of the channel in the current channel bundle;
determines whether the cost of each of the plurality of channels is less than the
cost of the channel in the current bundle, based on the comparison; and
initiates a cost negotiation for the cost of each of the plurality of channels with
a corresponding channel vendor based on the determining.
7. The television channel bundle cost and price optimization system (102) as claimed in
claim 5, wherein the bundle parameter comprises the ascertained bundle purchase
propensity and bundle price.
8. The television channel bundle cost and price optimization system (102) as claimed in
claim 5, wherein the channel parameter comprises at least one of channel price and
channel cost.
26
9. The television channel bundle cost and price optimization system (102) as claimed in
claim 1, wherein the segmentation module (120) further:
obtains the attribute associated with the at least one occupant, from a customer
relationship management (CRM) database (134), wherein the attribute comprises at
least one of age, ethnicity, income, education, language, and profession; and
segments the plurality of households based on the obtained attribute, to create
the plurality of clusters.
10. The television channel bundle cost and price optimization system (102) as claimed in
claim 1, wherein the bundling module (122) further:
extracts a current channel bundle data associated with the household, wherein
the current channel bundle data comprises channels present in a past channel bundle
choice; and
determines the frequency of the at least one genre from the extracted channel
bundle data to generate the channel bundle.
11. The television channel bundle cost and price optimization system (102) as claimed in
claim 1, wherein each of the plurality of time intervals is ascertained from click
stream data obtained from at least one of a set top box (106) and click stream database
(136).
12. The television channel bundle cost and price optimization system (102) as claimed in
claim 1, wherein the frequency and average of viewership are estimated based on predetermined
time granularity in each of the plurality of time intervals.
13. A computer implemented method for television channel bundle cost and price
optimization comprising:
determining, by a processor (110), a plurality of time intervals for each of a
plurality of clusters of households, wherein each of the plurality of clusters of
households are determined based on an attribute associated with at least one occupant
of a household, and wherein the time intervals are indicative of time duration during
which a plurality of households within a cluster are viewing television channels;
27
estimating, by the processor (110), frequency of viewership, and average of
viewership, in each of the plurality of time intervals, for each of a plurality of
households associated with the plurality of clusters of households;
determine, by the processor (110), for each of the plurality of time intervals,
average viewing time proportion for each household, based on the estimation, wherein
the average viewing time proportion is indicative of a time duration for which a
television channel is being watched by the household for a pre-determined duration in
the time interval;
determining, by the processor (110), bundle purchase propensity of the
household based on the average viewing time proportion and the current channel
bundle, wherein the bundle purchase propensity is indicative of a degree of
acceptance of the household to purchase a channel bundle;
generating, by the processor (110), the channel bundle for subscription by the
household based on the bundle purchase propensity, average viewing time proportion,
and frequency of the at least one channel genre, wherein the bundle comprises of a
plurality of channels, and wherein each channel is associated with a genre.
14. The method as claimed in claim 13, wherein each of the plurality of time intervals
comprises a plurality of time granularities, wherein each of the plurality of time
granularities is indicative of a pre-determined time duration.
15. The method as claimed in claim 13 further comprising:
determining, by the processor (110), price of each of the plurality of channels
in the generated bundle, based on based bundle purchase propensity, average viewing
time proportion, and frequency of the at least one channel genre.
16. The method as claimed in claim 13 further comprising determining, by the processor
(110), cost of each of the plurality of channels based on bundle parameter, and
channel parameter.
17. The method as claimed in claim 16, wherein the bundle parameter comprises the
ascertained bundle purchase propensity and bundle price.
28
18. The method as claimed in claim 16, wherein the channel parameter comprises at least
one of channel price and channel cost.
19. The method as claimed in claim 16 further comprising:
comparing, by the processor (110), for each of the plurality of channels, the
cost of the channel in the generated bundle with cost of the channel in the current
channel bundle;
determining, by the processor (110), whether the cost of each of the plurality
of channels is less than the cost of the channel in the current bundle, based on the
comparison; and
initiating, by the processor (110), a cost negotiation for the cost of each of the
plurality of channels with a corresponding channel vendor, based on the determining.
20. The method as claimed in claim 13 further comprising:
obtaining, by the processor (110) the attribute associated with the at least one
occupant, from a customer relationship management (CRM) database (134), wherein
the attribute comprises at least one of age, ethnicity, income, education, language, and
profession; and
segmenting, by the processor (110), the plurality of households, based on the
obtained attribute, to create the plurality of clusters of households.
21. The method as claimed in claim 13 further comprising:
extracting, by the processor (110), a current channel bundle data associated
with the household, wherein the current channel bundle data comprises channels
present in a past channel bundle choice; and
determining, by the processor (110), the frequency of the at least one genre
from the extracted channel bundle data to generate the channel bundle.
22. The method as claimed in claim 13, wherein each of the plurality of time intervals is
determined from click stream data obtained from at least one of a set top box and a
channel broadcasting server.
29
23. The method as claimed in claim 13, wherein the frequency and the average of
viewership are estimated based on pre-determined time granularity in each of the
plurality of time intervals.
24. A non-transitory computer readable medium having a set of computer readable
instructions that, when executed, cause a computing system to:
determine a plurality of time intervals for each of a plurality of clusters of
households, wherein each of the plurality of clusters of households are determined
based on an attribute associated with at least one occupant of a household, and
wherein the time intervals are indicative of time duration during which a plurality of
households within a cluster are viewing television channels;
estimate frequency of viewership, and average of viewership, in each of the
plurality of time intervals, for each of a plurality of households associated with the
plurality of clusters of households;
determine, for each of the plurality of time intervals, average viewing time
proportion for each household, based on the estimation, wherein the average viewing
time proportion is indicative of a time duration for which a television channel is being
watched by the household for a pre-determined duration in the time interval;
determine bundle purchase propensity of the household based on the estimated
average viewing time proportion and the current channel bundle, wherein the bundle
purchase propensity is indicative of a degree of acceptance of the household to
purchase a bundle, and wherein the bundle comprises of a plurality of channels, and
wherein each channel is associated with a genre; and
generate a channel bundle for subscription by the household and a price for the
channel bundle based on the bundle purchase propensity, average viewing time
proportion, and frequency of the at least one channel genre, wherein the bundle
comprises of a plurality of channels, and wherein each channel is associated with a
genre.