Classifications

• • H—ELECTRICITY
  • H04—ELECTRIC COMMUNICATION TECHNIQUE
  • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
  • H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
  • H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
  • H04N21/45—Management operations performed by the client for facilitating the reception of or the interaction with the content or administrating data related to the end-user or to the client device itself, e.g. learning user preferences for recommending movies, resolving scheduling conflicts
  • H04N21/466—Learning process for intelligent management, e.g. learning user preferences for recommending movies
• • H—ELECTRICITY
  • H04—ELECTRIC COMMUNICATION TECHNIQUE
  • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
  • H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
  • H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
  • H04N21/43—Processing of content or additional data, e.g. demultiplexing additional data from a digital video stream; Elementary client operations, e.g. monitoring of home network or synchronising decoder's clock; Client middleware
  • H04N21/442—Monitoring of processes or resources, e.g. detecting the failure of a recording device, monitoring the downstream bandwidth, the number of times a movie has been viewed, the storage space available from the internal hard disk
  • H04N21/44213—Monitoring of end-user related data
  • H04N21/44222—Analytics of user selections, e.g. selection of programs or purchase activity
  • H04N21/44224—Monitoring of user activity on external systems, e.g. Internet browsing
• • H—ELECTRICITY
  • H04—ELECTRIC COMMUNICATION TECHNIQUE
  • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
  • H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
  • H04N21/20—Servers specifically adapted for the distribution of content, e.g. VOD servers; Operations thereof
  • H04N21/25—Management operations performed by the server for facilitating the content distribution or administrating data related to end-users or client devices, e.g. end-user or client device authentication, learning user preferences for recommending movies
  • H04N21/251—Learning process for intelligent management, e.g. learning user preferences for recommending movies
  • H04N21/252—Processing of multiple end-users' preferences to derive collaborative data
• • H—ELECTRICITY
  • H04—ELECTRIC COMMUNICATION TECHNIQUE
  • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
  • H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
  • H04N21/20—Servers specifically adapted for the distribution of content, e.g. VOD servers; Operations thereof
  • H04N21/25—Management operations performed by the server for facilitating the content distribution or administrating data related to end-users or client devices, e.g. end-user or client device authentication, learning user preferences for recommending movies
  • H04N21/258—Client or end-user data management, e.g. managing client capabilities, user preferences or demographics, processing of multiple end-users preferences to derive collaborative data
  • H04N21/25866—Management of end-user data
  • H04N21/25883—Management of end-user data being end-user demographical data, e.g. age, family status or address
• • H—ELECTRICITY
  • H04—ELECTRIC COMMUNICATION TECHNIQUE
  • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
  • H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
  • H04N21/20—Servers specifically adapted for the distribution of content, e.g. VOD servers; Operations thereof
  • H04N21/25—Management operations performed by the server for facilitating the content distribution or administrating data related to end-users or client devices, e.g. end-user or client device authentication, learning user preferences for recommending movies
  • H04N21/258—Client or end-user data management, e.g. managing client capabilities, user preferences or demographics, processing of multiple end-users preferences to derive collaborative data
  • H04N21/25866—Management of end-user data
  • H04N21/25891—Management of end-user data being end-user preferences
• • H—ELECTRICITY
  • H04—ELECTRIC COMMUNICATION TECHNIQUE
  • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
  • H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
  • H04N21/20—Servers specifically adapted for the distribution of content, e.g. VOD servers; Operations thereof
  • H04N21/25—Management operations performed by the server for facilitating the content distribution or administrating data related to end-users or client devices, e.g. end-user or client device authentication, learning user preferences for recommending movies
  • H04N21/266—Channel or content management, e.g. generation and management of keys and entitlement messages in a conditional access system, merging a VOD unicast channel into a multicast channel
  • H04N21/2668—Creating a channel for a dedicated end-user group, e.g. insertion of targeted commercials based on end-user profiles
• • H—ELECTRICITY
  • H04—ELECTRIC COMMUNICATION TECHNIQUE
  • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
  • H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
  • H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
  • H04N21/41—Structure of client; Structure of client peripherals
  • H04N21/414—Specialised client platforms, e.g. receiver in car or embedded in a mobile appliance
  • H04N21/4147—PVR [Personal Video Recorder]
• • H—ELECTRICITY
  • H04—ELECTRIC COMMUNICATION TECHNIQUE
  • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
  • H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
  • H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
  • H04N21/45—Management operations performed by the client for facilitating the reception of or the interaction with the content or administrating data related to the end-user or to the client device itself, e.g. learning user preferences for recommending movies, resolving scheduling conflicts
  • H04N21/4508—Management of client data or end-user data
  • H04N21/4532—Management of client data or end-user data involving end-user characteristics, e.g. viewer profile, preferences
• • H—ELECTRICITY
  • H04—ELECTRIC COMMUNICATION TECHNIQUE
  • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
  • H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
  • H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
  • H04N21/45—Management operations performed by the client for facilitating the reception of or the interaction with the content or administrating data related to the end-user or to the client device itself, e.g. learning user preferences for recommending movies, resolving scheduling conflicts
  • H04N21/454—Content or additional data filtering, e.g. blocking advertisements
• • H—ELECTRICITY
  • H04—ELECTRIC COMMUNICATION TECHNIQUE
  • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
  • H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
  • H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
  • H04N21/45—Management operations performed by the client for facilitating the reception of or the interaction with the content or administrating data related to the end-user or to the client device itself, e.g. learning user preferences for recommending movies, resolving scheduling conflicts
  • H04N21/466—Learning process for intelligent management, e.g. learning user preferences for recommending movies
  • H04N21/4668—Learning process for intelligent management, e.g. learning user preferences for recommending movies for recommending content, e.g. movies
• • H—ELECTRICITY
  • H04—ELECTRIC COMMUNICATION TECHNIQUE
  • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
  • H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
  • H04N21/80—Generation or processing of content or additional data by content creator independently of the distribution process; Content per se
  • H04N21/81—Monomedia components thereof
  • H04N21/812—Monomedia components thereof involving advertisement data
• • H—ELECTRICITY
  • H04—ELECTRIC COMMUNICATION TECHNIQUE
  • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
  • H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
  • H04N21/80—Generation or processing of content or additional data by content creator independently of the distribution process; Content per se
  • H04N21/83—Generation or processing of protective or descriptive data associated with content; Content structuring
  • H04N21/84—Generation or processing of descriptive data, e.g. content descriptors
• • H—ELECTRICITY
  • H04—ELECTRIC COMMUNICATION TECHNIQUE
  • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
  • H04N7/00—Television systems
  • H04N7/16—Analogue secrecy systems; Analogue subscription systems
  • H04N7/162—Authorising the user terminal, e.g. by paying; Registering the use of a subscription channel, e.g. billing
  • H04N7/163—Authorising the user terminal, e.g. by paying; Registering the use of a subscription channel, e.g. billing by receiver means only

Definitions

• This invention relates generally to the delivery of targeted video programming to a viewer, and more particularly to the determination of various viewer characteristics for the linear delivery of video programming targeted to the viewer's characteristics to create a targeted or customized, apparently linear television program.
• VCRs Video Cassette Recorders
• a VCR may be either manually placed into a record mode or may be programmed to record a selected program at a later time.
• the user To program the VCR, the user either enters a date, time and channel of the program desired to be recorded, or enters an identification code ofthe desired program.
• Viewers of television programming increasingly have more choices as to which programs to view.
• cable television provides a dramatic increase in the number of channels available to a viewer in comparison to the channels available by way of a conventional television antenna.
• Digital satellite systems provide even more viewing choices.
• Digital broadcast of programs over cable television systems is expected to further increase the number of channels available to viewers.
• One effect ofthe increase in the number of viewing choices is increased difficulty in deciding which programs to watch. People, particularly those with busy schedules, may not have the time to select and view programs to determine which programs they may or may not like. Programs that may otherwise be desirable to a viewer may never be watched if the program is broadcast at a time that is inconvenient for the viewer. Users may select certain programs for viewing to determine if they like the program. However, with several hundred program selections each week, this task can take a considerable amount of time and is likely to cause certain desirable programs to be overlooked.
• a closed captioned encoding technique transmits text data in synchronization with its associated video data by inserting the closed captioned text data into a vertical blanking interval ofthe video signal.
• the closed captioned text data must be inserted into the vertical blanking interval of the video signal by the producer of the video programming.
• the vertical blanking interval of the video signal cannot be used by the head end operator to insert other text data such as sports, weather, stock market, news, advertising and other data.
• EPG Electronic program guides
• the EPG is provided by an EPG data service.
• EPG data is converted into a video signal at the cable head.
• the EPG data is converted into a video signal at the cable head and transmitted to the viewer's television by a dedicated cable television channel. After tuning to the dedicated cable television channel, the viewer then wait for the programming for the desired time period is displayed.
• the cable head end operator must dedicate a separate cable television channel to the EPG data and create video signals from the EPG data that are provided by the EPG service providers.
• One method of solving this problem is modulation of the EPG data onto an FM carrier and transmitting the FM carrier with a video signal on one ofthe cable television channels.
• a dedicated peripheral device is provided at the viewer's television tuner that demodulates the EPG data from the FM carrier.
• the EPG data is then stored until the viewer requests presentation of the EPG data on the viewer's television.
• the EPG data is then displayed on the viewer's television in place ofthe other video programming.
• a data controller is disclosed in U.S. Patent No. 5,579,055 that manages the flow of text data and electronic EPG data.
• preference agents for television programs have used Bayesian methods. See for example U.S. Patent No. 5,704,017 (hereafter the '"017 Patent").
• a collaborative filtering system is used to predict a desired preference of a television viewer based on attributes ofthe viewer.
• a system implementing '017 would need to communicate to a belief network through a two way communication network, disclosing private viewer information to the network.
• '017 does not leverage the rich EPG information available about television programs which can be used to identify various traits which contribute to viewer's choices.
• An object ofthe present invention is to provide a method and apparatus to select and deliver video data targeted to a specific viewer, typically a television viewer.
• the video data may be targeted in accordance with various characteristics ofthe viewer, including viewing characteristics, demographic characteristics, shopping characteristics, and others.
• an object ofthe present invention is to provide a method and apparatus for determining user preferences for television viewers.
• the present invention defines a method to analyze sample viewing habits along with associated demographic and other characteristics of the general population to generate parameters of a mathematical model that explains the relationship between the viewing patterns ofthe population and the associated characteristics, and that can be communicated to viewer devices using a one way communication channel such as a broadcast network.
• the present invention also defines methods for using these parameters by viewer devices to compute the user preferences of individual viewers without divulging private information about viewing habits to the outside world.
• the invention also details an apparatus that relies upon a viewer's computed preferences to store and present broadcast content that may be of interest to one or more individual viewers utilizing the same apparatus to view video data.
• the invention also discloses a method to compute the preferences of multiple viewers accessing the same viewing device, e.g. a multiple viewer household with a single television set.
• apparatus for causing recordation of television programs comprises a preference agent for causing retrieval of attribute information corresponding to each television program viewed by a user of the apparatus.
• the preference agent generates classification information indicative of viewing preferences of the user as a function of the attribute information.
• a recording manager causes recordation and storage to a storage device of television programs having attribute information that matches the classification information.
• the invention determines viewing preferences of a user by processing attribute information associated with programs viewed by the user and creating a profile of characteristics describing the viewer.
• the characteristics may include, but are not limited to, viewing preferences, demographic information (e.g. age, sex, education, occupation, income, political and religious affiliations, marital status, sexual preferences, ethnic background, geographical location of home and work), reading preferences, shopping preferences, music preferences
• Embodiments employing the principles ofthe present invention advantageously cause recordation of programs that match certain viewing habits ofthe viewer. Such embodiments therefore provide the viewer with stored programs that match certain viewing preferences of the user, which can be viewed at the viewer's leisure. The viewer is therefore relieved of the burden of deciding which programs from among several hundred possible programs to watch.
• programs may be recorded for storage in accordance with available capacity of the storage device.
• programs may be deleted in response to selections by the user or based upon a priority, indicated by viewing preferences of the user, in which programs having lowest priority are deleted first to make room for newly recorded programs.
• the priority of programs may also be a function of time, in which more recently recorded programs are given higher priority.
• determining which programs to record may also be a function of priority in which program s specified for recordation are given highest priority, followed by programs having attribute information corresponding to one or more user specified criteria, then followed by programs having attribute information corresponding to the recordation preference information.
• the user specified requests may be in the form of a first type of request comprising information indicative of a specific program and a second type of request comprising specifications indicative of one or more programs having attribute information corresponding to the user's specifications.
• the user may cause recordation of a currently broadcast program being viewed by the user by causing generation of a pause input. This advantageously allows a user to interrupt viewing of a currently broadcast program by recording the remainder ofthe program for subsequent viewing.
• Program viewing options may be presented to the user in the form of a menu that provides an easy to use interface for selection of programs and viewing and other options including play, pause, delete, fast-forward, rewind and so forth.
• the preference agent organizes the recordation preference information in the form of a database organized in accordance with categorization parameters.
• Programs may be received On either analog or digital formats.
• Programs stored in digital format are advantageously presented to the user in the form of additional channels. This allows the user to easily switch between programs (either recorded or broadcast) simply by switching channels.
• Figure 1 is a high-level block diagram of a Program selection device employing the principles ofthe present invention.
• FIG. 2 is a high-level block diagram of a system employing the principles of the present invention.
• FIG. 3 shows two examples for Program information.
• Figure 4 shows examples for traits and Liking values.
• Figures 5a-b are flowcharts illustrating the Data analysis performed on a representative sample.
• Figure 6 illustrate the process of computing the error in prediction of user choices
• Figure 7 illustrate a step in the process of regression analysis
• Figure 8 shows the relationship between two correlated traits.
• Figures 9a-c illustrate the process of determining trait-ness of a trait in a program.
• Figure 10 is a block diagram-showing an example for the Liking distribution record format.
• Figure 11 lists some sample values for different fields described in Figure 10.
• Figure 12 is a block diagram showing an example for the trait-ness record format.
• Figure 13 shows an example for broadcasting trait-ness as a part of EPG data
• Figure 14 shows an example for user selection record format.
• Figure 15 is a block diagram showing the inputs and output for the generation of user selection history-
• Figure 16 is a flowchart illustrating the process of learning the liking values, performed on a program selection device.
• Figure 17(a) is a block diagram showing the inputs and output for the computation of relevancy value for selection history record.
• Figure 17(b) is a graph representative of relevancy over age.
• Figure 17(c) is a graph representative of repetition rate over relevancy.
• Figure 18(a) is a block diagram showing the inputs and output for the process of updating of past selection history
• Figure 18(b) is a flowchart illustrating the process of updating of past selection history
• Figure 19 is a flowchart illustrating the process of computing liking values performed on a program selection device.
• Figure 20 is a block diagram showing the inputs and output for the computation Liking for programs.
• Figure 21a illustrates distribution of income for different programs.
• Figure 21b illustrates distribution of gender for different programs.
• Figure 22 is a system architecture for providing targeted advertising.
• Figure 23a is a graph illustrating the relationship between Belief function and probability for a user belonging to particular demographic trait value.
• Figure 23b is a flow chart of a demographic trait record format.
• Figure 23c is a flow chart of an advertisement targeting record format.
• Figures 24 and 25 are block diagrams illustrating operation of certain functions performed by the television recording system of Figure 1.
• FIGS 26(a)-b illustrate alternative hardware configurations in systems embodying the principles ofthe present invention.
• Figure 27 is a flowchart illustrating additional aspects of operation of the television recording system of Figure 1.
• Figure 28 is a block diagram of one embodiment of a system for providing EPG data and text data to a viewer.
• Figure 29 is a block diagram that illustrates one embodiment of a data controller for receiving EPG data and text from data providers, formatting the data for display and inserting the data into a vertical blanking interval of a cable television channel.
• FIG 3u illustrates one embodiment of an information field of the EPG data read from the EPG database of Figure 29.
• Figure 31 illustrates a data format of data read from the database for insertion into an assigned cable television channel.
• Figure 32 is a flow chart illustrating the operation ofthe EPG transaction formatter of
• Figure 29 is a flow chart illustrating one embodiment of the operation ofthe text transaction formatters of Figure 29.
• Figure 34 illustrates one embodiment of a set top box for use in receiving text data
• Figures 35-41 illustrate various aspects ofthe process of creating and using multiple viewer profiles according to the present invention.
• Figure 42 illustrates a method for distributing targeted electronic content without compromising the privacy of users.
• Figure 43 illustrates an embodiment of a system according to the present invention.
• Figure 44 illustrates custom linear programming according to the present invention.
• a television control system 100 operates in accordance with the principles of the present invention to cause recordation of television programs in response to user inputs 102 and television signals 104.
• Television control system 100 transmits signals to a television monitor 108 for viewing by the user.
• programs that are recorded by system 100 are presented to the user in the form of additional channels.
• the user can rapidly determine, by changing channels, the stored programs that are available for viewing.
• the user can also change channels between stored programs or between stored programs and currently broadcast programs. If the user changes channels from a recorded program to another program, playback ofthe recorded program is preferably paused.
• whether the playback ofthe recorded program is paused or continued is a user selectable option.
• the user may specify programs for recordation by specification of a particular program, or by specification of particular attributes ofthe program such as comedy/drama, actor(s).
• programs for recordation the user may specify that the program is to be recorded once or repeatedly, such as weekly, when broadcast.
• Signals 104 include a first component 105 which contains the information necessary to display video and audio components of a television program on television monitor 108.
• Signals 104 preferably also include a second component 107 termed herein "attribute information.”
• attribute information 107 is the information available by way ofthe DVB-SI and ATSC-SI formats and various proprietary formats such as StarSight EPG Data and TVData available from StarSight Telecast, Inc., Fremont, CA, and TVData, Glen Falls, NY, respectively.
• Attribute information 107 for any particular program varies depending on the program type, but typically includes a plurality of categories such as start time for the program, duration ofthe program, the title ofthe program and other, attributes (categories) ofthe program, together with an associated value corresponding to each ofthe categories.
• Preference agent 110 processes the attribute information 107 to generate if "category-value" pairs 115. For example, if an attribute for a program is duration, then the category may be duration and the value for that category may be 120 minutes.
• category-value pairs for a movie may include a description category with a short description ofthe movie being the value, a primary actor category with the names ofthe primary stars ofthe movie being the values, a director category with the name ofthe director being the value, a theme category with the theme such as adventure, comedy being the value, and a ratings category with ratings by particular critics being the value.
• Category-value pairs for a sports game such as a football game, may include names ofthe teams who are playing, the location ofthe game, and the specific tournament, such as the play-offs, or Superbowl, etc.
• the category- value pairs 115 are indicative of viewing preferences ofthe user.
• the data shown in Figure 1 as being associated with the category - value pairs 115 contains weighting information for the associated category value, in addition to other information shown by way of example further below.
• Preference agent 110 maintains the preference information 115 in the form of a preference database 1 16.
• Television programs 105 recorded by the system 100 are preferably stored separately together with the associated attribute information 107.
• the category value pairs 115 (with or without the associated values) are stored with the television programs 105 and the raw attribute information 107 is not maintained by the system 100.
• Preference agent 110 generates, in response to user viewing habits, data for each category stored in preference database 116 and for each value of each category.
• the data generated by preference agent 1 10 for each category and value is preferably indicative ofthe amount of time that the particular category and/or value is watched by the user relative to the total amount of time that the particular category and/or value is available to be watched.
• the relative amount of time that a program is watched by a user is a convenient indication ofthe user's relative viewing preference. However, other indications of user viewing preferences may also be used.
• Program source switch 114 operates in response to user inputs 102 to select either presently broadcast programs, by way of television signal 104 or stored programs from storage devices 106.
• Recording manager 1 12 operates to cause recordation and storage of television programs 105 and attribute information 107 in accordance with information generated by preference agent 110 and stored in preference database 116. Recording manager 112 also responds to user requests to record particular programs and to user requests to record programs having specified category-value pairs.
• the signals transmitted to the monitor 108 preferably take a conventional analog form. Alternatively the signals transmitted to the monitor 108 maybe digitally encoded. The exact form ofthe signals transmitted to the monitor is not critical and may take a form as required by a particular monitor.
• the television signals 104 received by the television control system 100 may take one of a variety of signal formats including analog encoded signals that are encoded in accordance with the well known NTSC or PAL standards. Alternatively, the signals 104 may be digitally encoded in a manner as transmitted by commercially available digital satellite systems (DSS) or in accordance with the MPEG-2 (Motion Picture Expert Group - 2) standard. In any given embodiment of television control system 100 the signal 104 may take a variety of the aforementioned forms.
• television control system 100 may be coupled to receive inputs from a digital satellite system, the inputs being digitally encoded.
• the television control system 100 may also be coupled to receive inputs from a Community Antenna Television
• the television control system 100 may also be coupled to receive analog or digital signals from a conventional home antenna.
• the attribute information 107 may be transmitted to the television control system 100 contemporaneously with the television program 105 in a variety of ways including industry standards, such as DVB-SI (Digital Video Broadcasting-Service Information) as defined by the
• ETS European Telecommunication Standards Institute
• ATSC Advanced Television System Committee
• DVB-SI protocol programming for the next six hours is transmitted every eight seconds for each channel.
• program information for the next seven days is available from the interactive on-screen TV program guide available from StarSight Telecast, Inc.
• Programming information further into the future, such as for the next seven days, may also be obtained in other ways. For example, by receiving the information in a time-multiplexed manner over a particular channel. Such information can easily be transmitted when the user is performing an action that does not require a moving video image on the screen, such as when the user has a control menu displayed on the screen.
• television control system 100 can download the attribute information 107 separately from the television program 105 by way of a separate communication session via a modem or the Vertical Blanking Intervals (VBI) contained in television signals.
• VBI Vertical Blanking Intervals
• Such separate communication sessions include data download mechanisms supported by the MPEG-2, DVB- SI and DSS protocols.
• the attribute information 107 can take a form under the DVB-SI protocol as shown below: event_id, start_time. duration,
• the event-id field Is a unique alpha-numeric code assigned to a program.
• DESCRIPTORS can be "Short Event Descriptors," “Extended Event Descriptors” or “Content Descriptors" which include the following information: Short Event Descriptor:
• ITEMs include the following information: ⁇ item_description_length, item_description, item_value_length, item_value ⁇
• CONTENT includes the following information:
• DVB-SI defined theme DVB-SI defined sub-theme, programmer defined theme, programmer defined subtheme,
• theme and subtheme are MOVIE and COMEDY, respectively.
• the programmer defined theme and sub-theme are values that may be provided by the EPG Data provider.
• Category- value pairs 115 are generated from the above type of information.
• the category-value pairs 115 take the following format: Category Name - Category Value, where category name cart be “Title”, “Director”, “Theme”, “Program Type” etc. and category values can be "Seinfield” "Martin Scorcese”, “Comedy”, “Sitcom” etc.
• Generation of category-value pairs 1 15 from attribute information 107 allows generation by preference agent 110 of categories that are not explicitly present in the attribute information 107.
• category-value pads 1 15 can be: Title-49ers, Description-football, and Description Search Rule-football(AND) San Francisco.
• preference agent 1 10 is capable of generating category-value pairs 1 15 from attribute information 107 even where there is no field in the attribute information that corresponds to the created category-value pair.
• Preference database 116 is preferably generated initially by downloading category-value pairs from a third-party source such as StarSight Telecast, Inc.
• a third-party source such as StarSight Telecast, Inc.
• such sources may provide information customized for particular geographical areas and dates.
• the database may contain data that gives sporting events involving local teams higher ratings than other sporting events.
• seasonal or holiday programs may be indicated as being preferred during particular seasons or holidays. For example, programs involving summertime activities would be indicated as having higher weighting during the summer than at other times ofthe year.
• the preference database is modified as described herein in accordance with the user's viewing habits.
• the preference database can be periodically updated from third-party sources to reflect the aforementioned seasonal or holiday updates.
• Categories in the preference database 116 are either predefined, such as those received from third-party sources, or are dynamically created from attribute information 107 received for programs 105. Categories, and associated values, that are dynamically created are preferably given a default rating by preference database 116.
• An example of the preference information created by preference agent 110 or downloaded to preference agent 1 10 is shown below.
• the three columns of numbers in the category statistics and value statistics portions indicate weighting (in a range of 0 to 1000) duration watched (in seconds) and amount of time that programs matching that particular category or value was available (in seconds).
• the information is preferably stored in the form of database records.
• fourteen categories are provided (1000 - 1013) followed by thirty- six values.
• the correspondence between the categories and values (category - value pairs) is next shown.
• Data for the categories and then the values is shown next. This data is organized in three columns as described above.
• the ratings for categories and values are dynamically generated by the preference agent 110 instead of being stored in preference database 116.
• the ratings may be stored in preference database together with the category-value pairs.
• preference determination is used to predict a user's preferences in the choice of TV programs. This prediction is based on, (i) analysis ofthe individual users viewing habits, (ii) analysis ofthe viewing habits of a representative sample of users and (iii) EPG data for the television programs available during the period ofthe collected sample.
• the viewing habits of a user are described below in the context of actual programs watched by the user.
• other information may be just as useful for the purposes ofthe invention, and includes information regarding programs that are never watched by the user or watched only very briefly before changing to another channel, programs that the user schedules to record, programs that the user records but does not watch or watches only briefly, and programs for which the user requested information from a program guide (e.g. an on-screen EPG menu) or for which the user never requests information or for which the user requests information but does not choose to watch.
• a program guide e.g. an on-screen EPG menu
• Bayesian network is a data based approach.
• the knowledge engineer and the expert first determine the variables ofthe domain. Data is then accumulated for those variables, and an algorithm applied that creates the belief network to predict user behaviors from this data. The accumulated data comes from real world instances of the domain.
• Bayesian methods can be performed by utilizing an algorithm known as the "EM” algorithm, which, as recognized by those skilled in the art, calculates the maximum a posteriori values ("MAP values") for the parameters ofthe model to predict user behavior.
• the EM algorithm is described in Dempster. Maximum Likelihood From Incomplete Data Via The EM Algorithm, Journal ofthe Royal Statistical Society B, Volume 39 (1977), inco ⁇ orated herein by reference.
• each variable within the belief network is then scored for goodness in predicting the preferences ofthe user by rendering a subscore.
• the subscores are generated by using the marginal likelihood ofthe expected data at the maximum a posteriori values.
• any standard model to predict user behavior inference algorithm can be used to perform this step, such as the one described in Jensen, Lauritzen and Olesen, Bayesian Updating in Recursive Graphical Models by Local Computations, Technical Report R-89-15, Institute of Electronic Systems, Aalborg University, Denmark, inco ⁇ orated herein by reference.
• Television viewing habits ofthe population are sampled to generate representative samples of viewer behaviors, 1 17.
• This task is typically performed by companies such as Nielsen Media Research, who collects user behavioral samples to conduct market research.
• the results ofthe sampling is stored in a viewer behavior database.
• the format ofthe viewer behavior database is proprietary to the company conducting the sampling.
• Viewer behavior database contains demographic information about the people in households participating in the sampling. This information includes but is not limited to race, age, annual income and gender.
• Viewer behavior database also contains information about all the television programs each viewer watched- during the period of sampling.
• Program Information and the schedule of television programs is available in the EPG database, 104.
• Program Information for a television program contains information about various attributes of the program which includes but not limited to the title, program type and program category of the television program, arid also the actors acting in the television program.
• Figure 3 lists two examples of program information, example 1 , 124, describes the program information for a audio visual television program, and example 2, 125 describes program information for a graphical text based television program.
• Viewer behavioral database is analyzed 1 18 to identify traits of TV programs and determine how such identified traits influence viewing habits in the representative sample of users. These traits and their influence amongst the viewing population are then used in aiding the process of predicting an individual user's choice in TV programs.
• Figure 4, 126 lists some > examples of such traits. Some of these traits are derived from program information, 104. Some of these traits are derived by looking for users liking for association of multiple traits or association of traits with other viewing parameters which includes but is not limited to time, day of the week, holiday, working day and weather season. Some of these traits are derived by data analysis of Viewer behavioral database. Any given program would exhibit varying degrees of the above identified traits. This is expressed as the trait-ness of the trait in the program, e.g. trait- ness of comedy in Seinfeld may be 1.2 and trait-ness of comedy in Mad About You may be .79. The trait-ness of a trait in a program is computed as a part of the data analysis, 118.
• the present invention also provides for a methodology for determining viewer preferences.
• an individual viewer picks one show to watch out of the collection of available program by evaluating a stochastic liking function for each program and choosing the program with the highest score.
• the liking function is modeled as an aggregate of liking for a specific trait and the degree to which that trait is exhibited by that program.
• l(p) represents the liking for program p
• ⁇ (tn) represents the liking for a specific trait tn
• tn represents the degree to which the Trait tn is exhibited by p
• Hidden traits are those which influence a user's viewing habits but which can not be derived from the EPG information. For example, sitcoms featuring a specific ethnic background would resonate more with viewers of that ethnic background. Another example of the effect of a hidden trait could be the strong liking of the sitcom "Frasier" amongst the set of users who have a strong liking for the sitcom "Cheers.” Assuming that the names of actors performing in a sitcom are not be available in the EPG data, this affinity towards both the sitcoms by the same set of people may be explained by the presence of some trait commonly exhibited by both the sitcoms, namely the presence of one of the central characters in both the sitcoms. Such hidden preference criteria would have to be captured by adding the hidden traits in the computation of the Liking Function.
• Associated traits - Traits which have a different influence on a user's viewing habits when combined with other traits. For example a user would have a certain liking for any given Seinfeld episode, and a certain liking for any premiere sitcom being aired for the first time. However, its liking for a premiere episode of Seinfeld may be sufficiently large enough to require an additional trait, "new Seinfeld", to fully explain its liking for a premiere episode of Seinfeld.
• the liking of each trait for a given user has to be similarly refined from initial approximations by regression analysis.
• Examples of traits and the associated liking for a sample user include but are not limited to those listed in Figure 4.
• New Channel may be 2 whereas its preference for NEWS on NBC may be 13, i.e., this user does not always watch news programs or programs in general on the NBC channel, however he has a sirong preference for NEWS programs on the NBC channel. If only its liking for ihe "Channel NBC" & "Program Type News" are considered, its preference for "News on NBC" would be unexplainable. This procedure to identify traits is first carried out using the viewing habits ofthe representative sample along with the dete ⁇ riination ofthe distribution of likings of each trait within the representative sample.
• Some ofthe outputs ofthe analysis 102, ofthe viewing habits of a representative sample axe a set of preference determination parameters viz. (i) the traits which are exhibited by recurring program and the degree to which such a trait is exhibited, and (ii) a distribution list of the likings amongst the viewing population of each ofthe above trait.
• This initial liking value may be arrived at in various ways. One of the ways could be to base it on the amount of time for which a given trajj was watched as opposed to the amount of time it was available during the user's viewing period.
• the Liking Function l(p) is computed for all the programs that were available during the times the user watched a given program. These programs are then arranged in descending order, 140, of liking value for a time period during which the viewer watched a program, 141.
• the actual program the user watched at a given time is highlighted, 143. If all the highlighted programs in Figure 6 had the highest score, the Liking Function correctly identified the user's preference. If the program watched by the user ranks below some other competing program available at that time the liking values used do not correctly reflect the user preference.
• the error in the Liking Function for this program may be called N, 142.
• Such values of N are determined for each program watched by the user by comparing it with the Liking Function ofthe other competing programs. In the example illustrated in Figure 6 values of N are computed for the times when the user watched TV between the hours of 10:00-10:30, 10:30-11 :00, 11:00-11:30. Using regression analysis, the liking values of each ofthe traits are adjusted incrementally to reduce the average value of N.
• the set of liking values which yield the lowest value ofthe average of N are considered the best set of liking values for that user.
• FIG. 7 One methodology of he regression process is illustrated in Figure 7.
• liking values for traits a, b, c are ⁇ al, ⁇ bl, ⁇ cl ....(1 4,145).
• ⁇ al represents the liking of trait "a”
• ⁇ bl represents the liking of trait "b” and so on.
• the Liking Function of all the relevant programs are computed using this initial set of liking values.
• the liking value ⁇ al of t he first trait "a” is changed incrementally to determine the average value of N.
• the value ⁇ a2, 147. which yields the lowest average value of N is taken as the new liking value for the trait "a.”
• This new value liking value ⁇ a2, 148 now replaces the old value and is now used to determine the remaining ⁇ values.
• the above set of liking values computed would predict the users preferences with a high degree of accuracy if set ofthe traits considered in the Liking Function includes all the traits that are relevant to a user.
• the average value of N not converging (to 0 or some other acceptable value) would indicate that not all traits that are relevant to the user have been considered in the computation of the Liking Function.
• Introduction of additional traits, either associated or hidden, is used to improve the determination ofthe user's preference.
• a determination of associated traits is achieved by a variety of different methods. One method is through the application of heuristic rules of thumb where the associative value of a number of traits is not reflected in the program information obtained from an EPG but is relevant to human viewing habits.
• a user who has a liking for Seinfeld would , most probably have a much higher preference for a premiere or new episode of Seinfeld.
• Such heuristic rules of thumb regarding the associative value of a number of traits may be passed to each set top box via the Head End.
• Another method of determining associated traits is with an algorithmic search which looks for common traits in programs and introduces new associated traits to try to improve the Liking Function for a user.
• Determination of hidden traits in different programs can be achieved in a number of ways.
• One of these methods is illustrated in Figure 8,
• the y-axis represents an increasing liking value for liigher values of y. Every point on the x-axis in Figure 8 represents a user arranged such that the liking value ofthe trait "tn" increases with higher values of x. If for the same values on the x-axis the liking values of a strongly co-related trait "tm" are also plotted, it will exhibit a relation to the liking graph ofthe trait "tn".
• the co-relation between any two traits may be explained by he presence of a common hidden trait.
• Hidden traits can also identified by applying rules of thumb or some other appropriate manner. While ail programs of a specific genre exhibit some common traits, the degree to which these traits are exhibited vary from program to program This degree of traitness for recurring programs can be quantified such that it best explains the viewing choice of watching or no t watching that program in the representative sample. For example consider a user who has a certain liking G for comedy. The users' decision to watch a comedy program would be influenced by whether the amount of comedmess exhibited by that program crosses some threshold of comediness that is deterrnined by its liking for comedy.
• FIG. 9(a)-(c) One method to determine the traitness of a trait T in a given program N is illustrated in Figures 9(a)-(c).
• the highlighted program represents the programs actually watched 156 by a user All available programs are arranged in decreasmg order of rank 154 as computed by the Liking Function.
• the margin of error 157 in the traitness of trait T is 3.
• the program N watched by the user was ranked the highest.
• the margin of error in the traitness of " trait T is 0.
• user 2 where he did not watch Program N, it was ranked the highest.
• the margin of error in this case can be considered to be a constant K.
• K may be the number of programs available to the user at that time.
• the margin of error may be considered K.-1.
• Such margins of error are computed for all users who watched the Program N and the average margin of error is computed. Using regression analysis, the traitness value ofthe trait T is adjusted incrementally to reduce the average value ofthe margin of error. The value of traitness which yields the lowest value of the average ofthe margin of errors is considered the best value ofthe traitness ofthe trait T exhibited by the program N.
• Traitness values may also be assigned by rules of thumb or some other appropriate manner.
• One of the outputs of analyzing the viewing habits of a representative sample is a distribution list ofthe likings amongst the viewing population of each trait. This information is presented schematically in the Figure 10 and is made available to every individual user Set Top Box along with the broadcast TV programs and program information.
• "Number of traits" 164 represents the total number of traits that have been identified after analyzing the viewing habits of the representative sample.
• the distribution of liking of each trait is provided in a "Trait Record", 165.
• Information included in a trait record include the name of the trait, 167, the type ofthe trait, 168,( whether hidden, associate, etc), the liking distribution ofthe trait 170, and the distribution.pararneters, 172, ofthe trait. Examples of various possible values in a trait record are provided in Figure 11.
• Figure 12 illustrates one schematic representation of traits present in recurring programs This information is passed along with the above liking distribution of traits.
• the "Number of programs” 173 represents the number of programs for which trait information is sent. Relevant traits and the degree to which they are exhibited axe included in the "Program Record” 174.
• the "Program Identification” 177 information is used to identify the TV program to which this record pertains.
• An example of a Program record for a Seinfeld Episode is given in Figures 13.
• certain traits such as "Program Name - Seinfeld", “Channel - NBC", "Program Type - Sitcom” can be derived directly from the EPG and are called static traits. Additional traits such as “comedy” which store the degree of cornediness associated with a Seinfeld Program may also be passed as a part ofthe EPG data.
• the program content and the program information are transmitted from a Broadcast Head End along with preference determination parameters, 119.
• preference determination parameters include but, ere not limited to (i) the traits which are exhibited by each program and the degree to which such a trait is exhibited and (ii) a distribution list ofthe liking values amongst the viewing population of each trait.
• This information is received in each household by a program selection device 100, that include a preference Determination Module, a personal preference database, 116, a storage device and a display device 108.
• the personal preference database 116 is used to store the results ofthe analysis ofthe individual user's viewing habits.
• the storage device stores programs as per the user's specific requests or programs recommended by the preference determination module.
• Program Selection Device 100 monitors each user's viewing actions and selection of TV programs being watched. This is stored in a storage device or in memory in the form of selection history data 189 (see Figure 15).
• the schematic representation of selection history data is given in Figure 14.
• the "Number of selection records", 180 represents the number of user selection choices stored in the selection history data.
• Each selection record 181 contains the information on the actual programs watched 185 along with information on the competing programs available at that time 186. Storing program information is required as the EPG may not be able to provide information on past programs. Information on these program may be obtained directly from the EPG data when the information is obtained while the program is still current.
• the time 183 and duration 182 for which a program was watched also form a part ofthe selection record.
• a uses selection history 189 is derived from each choice 187 made by the user along with program information from the EPG 104.
• Onq process of learning the user preferences of a specific individual is illustrated in Figure 16.
• the liking distribution in the representative sample for traits identified by data analysis, 118, are used by the Liking Function to rm ' riirriize the effect of error introduced because of lack of sufficient sampling in the computation of the liking of the identified traits by an individual.
• An individual user's response to setup question 190 may also be factored in determining initial values of liking for that individual 192.
• a user selection history 189 is maintained for a fixed number of hours.
• the number of hours for which the user selection history is maintained can be preferably changed to have an increased rate of learning during initial days of a new user using the device.
• Average Error is computed N 194, in a similar manner as for users; in the representative sample, as previously described in Figure 6. If the average error is greater than a tolerable limit 195, new liking values are computed 198. Entries in the user selection history are moved to the Past selection history f the Average error remains under a tolerable limit the liking values are computed only after a predefined number of hours 196.
• Another embodiment ofthe selection record stores a program ID instead ofthe entire program information.
• the actual program information is broadcast at predetem ⁇ ned times from the Head End where each program is identified by the same program ID stored in the selection record.
• the computing 198 of liking values for traits are performed only at this predetermined time.
• Selection history is partitioned as user selection history 189 and Past selection history 216 to optimally store the section history records.
• Selection history contains information about all selections made by the user between two computations of Liking values.
• Past selection history maintains the most relevant information about the most relevant selections made by the user in past.
• the number of records in the Past selection history can be advantageously configured to suite the memory available for storage of such records. Relevancy in this context is the importance ofthe record in computing the liking values and is dependant on many parameter including but not limited to the age ofthe record 200, the repetition rate 203 of traits contained in programs 204 in the selection record.
• the relationship between relevancy and age 205 ofthe record is shown in Figure 17(b).
• the most relevant records are the most recent records.
• the relationship between relevancy and the repetition rate of traits contained in programs in the section record is shown in Figure 17(c).
• the relevancy increase 206 As the repetition rate decreases, the relevancy increase 206, till cer ⁇ ain limit after which the relevancy decreases 207.
• a good example to illustrate the reasoning behind this relationship would be relevancy of keeping records on user selection when Seinfeld a weekly program is available and the relevancy of keeping records on user selection when Olympics a 4 yearly program is available. While it is important to keep records about Seinfeld to compute likings of various traits contained in Seinfeld, it is pointless to keep records about viewer's liking for the last Olympics.
• FIG. 18(a) and 18(b) The process of updating of past history is shown in Figures 18(a) and 18(b). The processing is done for every record in the user selection history. If there is enough memory to store the new record in the past selection history 218, the record is removed from the user selection history and added to the past selection history 222. If there is not enough space in the past selection history 218, the records in the past selection history are sorted based on relevancy 220, and :he least relevant record is deleted 221 to make space to add the new entry 222.
• the number of available programs 186 stored in the past selection history can be limited to an optimal number to make best use ofthe memory available (see Figure 18).
• the programs with higher values of liking, representing the traits which have significant liking values, are stored in past selection history record.
• the liking for trait is entered 230 into the preference' database 116.
• Other parameters like the repetition rate for the trait also is entered into the database 230.
• the repetition rate is computed by looking at the repetition rate of the trait in user history and pas t history. If there are no previous selection record with this trait, then the repetition rate is assumed as a predefined value. As new selection records are created, the repetition rate is updated.
• a weighted average ofthe current liking values and the old liking values is computed. This forms the new set of liking values. Learning rate can be increased or decreased by changing the weight ofthe current liking values.
• ⁇ (tn) is computed as a weighted average of liking for the trait tn for the user 234 and the liking for the trait tn for the general population 235.
• the viewer is asked to input the number of viewers in a household during the initial setup process.
• Each viewer can answer a set of setup questions which would capture their liking for representative traits.
• An initial set of possible liking values are created for each viewer.
• Learning process updates each viewer's preference database separately after identifying each viewer session correctly.
• Viewer identification is done either by explicit input from the viewer identifying the viewer or automatically by identifying the viewer by monitoring the television viewing behavior ofthe viewer.
• the error value between predicted and actual is computed for all viewers.
• the viewer with liking value which yields the least error value is chosen as the possible viewer.
• the certainty of this decision is expressed as a probability which is computed as a function ofthe differences between error values for different viewers in this household.
• the objective ofthe present invention is to determine demographic characteristics of a user by analyzing the users viewing habits in juxtaposition with viewing habits of a representative sample of users whose demographic characteristics are known. These demographic characteristics of a user collectively constitute a demographic profile.
• the present invention can receive a collection of possible ads and show individual users only those which are targeted to a matching profile.
• Some example of demographic characteristics include, but are no imited to, a users gender, race, age and income.
• the output ofthe analysis of viewing habits of the representative sample provides a basis for determining demographic characteristics ofthe individual user.
• the TV program viewing preference information chosen in the analysis of viewing habits, ofthe representative sample plays an important role in determination of a demographic characteristic. Different TV programs might be required to detej-m ⁇ ne demographic characteristics. Typically TV programs where a majority ofthe viewing audience shares a common demographic characteristic, in a proportion which is different from what is observed in the general population, are best suited in the discovery of demographic characteristics. For example to determine the "income" demographic characteristic consider graph (i) in Figure 21a. The x-axis represents the annual salary in increasing order of x.
• the y-axis represents the probability of finding an annual salary of x in the representative sample (which is the same in the general population).
• the dotted curve ⁇ plots the same income distribution represented in the graph (i)
• the first solid curve shows the income probability distribution ofthe viewing audience for program PI in the representative sample
• the second solid curve represents the income probability distribution of the viewing audience for program P2 in the representative sample.
• Analysis ofthe viewing habits of an individual user for the above programs PI & P2 could be used to determine his most probable annual income. Using Bayesian inference this probability may be expressed by the following mathematical formula
• the dete ⁇ r ⁇ nation of which programs yield high probability values in determining a demographic characteristic of a user may be done by applying rules of thumb through an algorithm which chooses the program on the basis of how much a demographic trait in it viewing audience differs from what is observed in the general population by some other means.
• the same television program maybe used to determine one or more traits. It is also possible that a completely different set of programs are required to study different demographic characteristics.
• the present invention contemplates analyzing other preferences of a user in addition to or in place o the users television viewing preferences.
• Such information may include, but is not limited to, musical preference, reading preferences, shopping preferences (a very large category that comprises, among many others, preferences in clothing, furniture, jewelry, decorations, appliances, electronic equipment), political and religious tendencies, etc.
• Such information may be obtained from any source, but in one preferred embodiment it is obtained from on-line sources (typically accessible as world wide web sites) that the user is a member of or visits frequently, such as music clubs, book clubs, other special-interest sites (e.g. a site devoted to a particular sport or hobby), news sites, etc.
• Other information descriptive of the users preferences includes user subscriptions, such as to newspapers, magazines, newsletters shopping catalogs, and other special interest publications.
• Bayesian inference While it is possible from Bayesian inference to ascribe a certain preference for a specif i c show based on a users demographic characteristics, it may not lje the only fac t or which explains the users choice of a particular show. For example, consider the graph (ii) in Figure 21 a. It m i ght be possible for a user, Ul, having income less than xl having a strong preference for program P2 for unknown reasons. So just using Bayesian inference to arrive at Ul's most probable annual income based on his viewing habits of only a single program P2 would not yield the most meaningful result.
• a Belief Function for the Bayesian probability for a demographic characteristic dc 1 may be computed in many ways which includes but is not limited to BF(dcl) - MAX( bp(wl), bp(w2) .., bp(wm) ) + crl + MAX(bp(wk), bp(wk-H) .., bp(wk+m) )*cr2 + ... + MAX (bp(wn), bp(wn+l) .., bp(wn+m) )*cm where
• - bp(wn) is the Bayesian probability of a user having that demographic characteristic given the fact that he watched program wn - (wl r w2, ... wm), (wk, wk+1, ..., wk+m), (wn, wn+1, ..., w ⁇ +m) are sets of programs where any member of a set has a high degree of co-relation with another member ofthe same set, but a very low degree of co-relation with a member of another set.
• - crl represents a co-relation co-efficient ofthe set (wl, w2, ... wm)
• c2 represents co- relation co-efficient o the set (wk, wk+1, ..., wk+m) and so on. Elements of the same set have the co-relation co-efficient.
• the Belief Function for any demographic characteristic is illustrated by the S curve in Figure 23 a.
• the ⁇ -a ⁇ is represents increasing values ofthe Belief Function for a demographic characteristic for increasing values of x.
• the y-axis represents increasing values ofthe probability of a user displaying that demographic characteristic for increasing values of y. As shown in Figure, the probability does not increasing any further for higher value ofthe Belief . Function.
• Display of advertisements by a program selection device in accordance with the invention is done when a users computed demographic profile matches the target profile ofthe advertisements.
• Some ofthe elements contained in an advertisement are the advertisement contents (may consists of video clips, audio clips, and/or graphical or textual information) and the meta data which includes information on the demographic profile to which this advertisement is targeted.
• the meta Data contained in an advertisement is represented schematically in the Figure 23c.
• Numberer of Target Records refers to the number of “Target Records” included in this "Target Ad Meta Data”.
• Each "Target Record” refers to a demographic characteristic that this advertisement is targeted to.
• the Target relationship rule is used to determine the relationship rule.
• the advertisement may be targeted towards users who satisfy any one ofthe following criteria: income level is between Ix & Iy - ethnic background is ebl
• the "Target relationship rule” would be used to specify the above relationship.
• the "Demographic Trait Data” determined by analysis of viewing habits of a representative sample with known demographic characteristics and the "Target Ad Meta Data” are broadcast to a Program Selection Device from a Broadcast Head-end along with program content and.EPG data. This is illustrated in Figure 22. This is received by each user at each Program Selection Device to determine the user demographic profile. Broadcast of the "Demographic Trait Data" and “Target Ad Meta Data" along with the advertisement content may be done on a periodic basis, or may be made always available on the broadcast networks or through some other communication mechanisms on an as-needed basis.
• the "Demographic Trait Data” and “Target Ad Meta Data” are collected and stored in a storage device such as, but not limited to, a hard disk, flash ROM, or main memory. Collection of this data may be done at fixed time periods or whenever suitable depending on the embodiment chosen.
• the Program Selection Device goes through past Selection History.
• Each program on the "Trait Value Record” that is available in the past Selection History is used in the computation of the Belief Function.
• the Belief Function Distribution graph in the "Trait Value Record” is then used to determine the probability ofthe user having that demographic characteristic. If an earlier probability for this demographic characteristic is available a weighted average ofthe old probability and the newly determined probability is taken and stored.
• Each targeted advertisement where the "Probability satisfaction criteria" ofthe "Target Record” is met by user's demographic profile is chosen for display at the appropriate time.
• television control system 100 is preferably implemented by way of a general purpose digital computer and associated hardware that executes stored programs to implement the functions shown within block 100. Ihe exact hardware and software platforms on which the television control system 100 is implemented is not important and may take a variety of forms. For example, television control system 100 may be implemented in a set-top box such as may typically used by individuals in the home to receive CAIN signals. Another implementation of television control system 100 is in the form of a personal computer configured with- the requisite hardware and software to receive and display television signals.
• OpenTVTM Operating Environment "OpenTVTM Operating Environment” and "Application Development for OpenTVTM.”
• An advantage of such an operating system is the support provided in the form of function calls to obtain attribute information 107 from the signals 104.
• a general purpose operating system such as the Windows NT operating system from Microsoft Co ⁇ oratio ⁇ may be used in conjunction with additional software that provides the functions required to extract the necessary information from attribute information 107 and to perform other manipulation ofthe received signals 104 and the stored information 105.
• Storage devices 106 may include a, variety of different types of storage devices.
• For _ example preference database 116 may be stored in a non- volatile, random-access semiconductor memory.
• Television programs 105 and attribute information 107 may be stored on storage devices having greater capacity such as a conventional magnetic, hard disk drive.
• storage devices 106 are understood to encompass a variety of storage devices. The exact form of the storage devices 106 is not critical so long as the storage devices have the capacity and speed to store the necessary information.
• Storage devices 106 may also comprise a conventional video cassette recorder (VCR) which operates under control of system 100 to store television programs 105 and attribute information 107 on conventional magnetic tape.
• VCR video cassette recorder
• the television control system 100 is presumed to be integrated into, or coupled to, a system including a tuner and other functions necessary to receive television signals and to extract the attribute information 107 from the television signal and to perform other functions typically associated with the receipt and viewing of television signals.
• television control system 100 may operate in conjunction with a database agent that facilitates interaction with preference database 116 by causing storage and retrieval of information to or from the database in an optimal manner.
• the preference database 116 may be implemented by a commercially available database product • such as the Oracle Light database product available from Oracle Co ⁇ oration which also inco ⁇ orates the functionality to implement the data base agent described above.
• Recording manager 112 causes recording of programs 105 by periodically initiating a sequence of steps shown in Figure 24.
• recording manager 112 sends a request to preference agent 110 for ratings of all programs at a particular time (X), or alternatively, for ratings of all programs within a particular time period (X).
• X time
• t he steps shown in figure 2 may be performed every six hours.
• the frequency with which the steps in Figure 24 are performed may be changeable by the user.
• Preference agent 110 responds at step 202 by providing ratings, from preference database 116, for each program received from recording manager 112. Recording manager 1 12 then causes recordation ofthe programs at time X, or within time period X in accordance with the ratings received from preference agent 110.
• programs having the highest rating are given highest preference for recordation and programs having the lowest rating are given lowest preference to recordation.
• the recordation is subject to storage capacity constraints. For example, if the highest rated program is one-hour long, and only thirty minutes of recording space is available on storage devices 106, then the one-hour program is skipped and the highest rated thirty-rninute program is recorded.
• Highest priority for recording of programs is given to programs specifically requested by the user. For example, if the user identifies a particular program for recording, such as by specifying the date, time and channel, or by specifying an identification code for the program,- - recordation of that program is given priority over programs rated by the preference agent. Next highest priority is given to programs matching particular category-value pairs specified by the user. For example, if the user does not identify a particular program, but specifies that one-hour long documentaries pertaining to travel should be recorded, then recordation of programs matching such category-value pairs is given priority over programs rated by the preference agent 110. In alternative embodiments, relative priority between user-specified programs, user- specified category- value pairs and programs rated by the preference agent 110 is changeable by the user.
• Recording manager 112 manages storage capacity on storage devices 106 by causing deletion of television programs 105 in accordance with ratings of such programs generated by preference agent 110. This is performed in a manner similar to that explained above for dete ⁇ ining which programs to record.
• Figure 25, which shows the steps taken by recording manager 112 to determine which programs to delete, is similar to Figure 24.
• recording manager 301 requests ratings from preference agent 110 of all programs stored on storage devices 106.
• preference agent 110 responds by providing the requested deletion ratings.
• recording manager 112 responds by causing deletion, when needed, of programs in accordance with the deletion ratings received from the preference agent 110.
• recording manager 112 causes deletion, or overwriting of programs having the lowest rating first.
• stored television programs which are determined by preference agent 1 10 to be least preferable, in relation to other stored television programs, are deleted or replaced first, and those determined to be most preferable are deleted or replaced last. Deletion of programs occurs only when required.
• storage device 106 typically being filled to maximum capacity, thus providing the user with as wide a variety of programs as possible.
• the user can specify programs that are to remain on the storage device 106. Such programs are not deleted by the recording system 100 in the steps shown in Figure 25. In addition, the user can specify programs that are to be deleted, and therefore override the steps shown in Figure 25.
• the preference database is used by system 100 to alter the manner in which information about currently broadcast programs is presented to the user. For example, in such embodiments, the preference database is used to rearrange the order in which currently broadcast programs are presented to cause programs having attribute information 107 rated highest by preference database 116 to be presented first.
• the preference database 1 16 can be used to organize information regarding the currently broadcast programs according to the liking of various traits stored in the preference database.
• Figures 26a and 26b illustrate alternative hardware configurations for systems employing the principles ofthe present invention.
• Figure 26a illustrates a hardware configuration that supports storage and retrieval of digitally encoded audio and video.
• Interface 902 is a standard digital cable or digital satellite input interface.
• Interface 906, which is the hardware interface to storage devices 106 preferably takes the form of an IDE or SCSI interface, or the proposed IEEE- 1394 interface.
• Interface 906 is an NTSC or PAL encoded video interface. If the television signal 104 takes the form of an analog signal, as in the case of most current television broadcast signals, and CATV signals, then the signal 104 must be digitized and generally compressed (for example, by the MPEG-II standard) before storage on a digital storage medium such as shown in Figure 26a.
• Figure 26b illustrates an embodiment using an analog storage device 106 such as a conventional VCR. If the television signal 104 is analog then the interface 910 takes the form of a conventional NTSC or PAL interface. If the television signal 104 is digital then the interface 910 takes a form as interface 902 shown in Figure 26a and a digital-to-analog converter is required to convert the received signal to analog form before storage on storage device 106.
• an analog storage device 106 such as a conventional VCR.
• Figure 27 illustrates operation of an automatic pause-record feature of preferred embodiments.
• the recording system 100 advantageously allows the program to be recorded so the user can resume viewing the program at a subsequent time.
• the user is viewing a currently broadcast program.
• Generation of a pause input at 1004 by the user such as by pressing of an appropriate button on a remote control coupled to the recording system 100, causes the system 100 to cause at 1006, recordation ofthe program being viewed by the user. The user is then free to watch another program or stop watching the television 108 altogether.
• the recording system 100 causes the program recorded at step 1006 to be retrieved and shown on the television 108 from the point the recordation was initiated at step 1006. If the program is still being broadcast when step 1010 is initiated, then recordation ofthe program continues by the system 100. The user thus can easily interrupt viewing of a currently broadcast program and resume subsequent viewing.
• the recording system 100 supports a variety of functions such as fast-forward, rewind and visual scan of stored programs, and other functions supported by the storage medium 106.
• functions such as fast-forward, rewind and visual scan of stored programs, and other functions supported by the storage medium 106.
• the storage medium 106 takes the form of a VCR then the program viewing and manipulation functions will be limited to the standard VCR functions of fast- forward, rewind, forward or reverse visual scan.
• the storage device 106 takes the form of a digital storage medium then more advanced program search and retrieval functions can be supported.
• the functions performed by the preference agent 110 and the recording manager 112 are illustrative of a particular embodiment. However, the division of tasks between the two modules 110 and 112 may be changed.
• the data formats 115, 116, 105 and 107 may also take a variety of forms.
• FIG 28 illustrates an EPG and text information service in accordance with one embodiment ofthe present invention.
• a local cable television company's billing vendor 10 communicates via a billing link to an RS-232 port of a system manager 12 located at a cable head end.
• Billing vendor 10 includes a subscriber database and generates a monthly bill for the subscribers in the system based on the level of service and any pay-per-view purchases.
• System manager 12 can be a personal computer or other processing device which receives viewer authorization transactions from billing vendor 10 and generates transactions for delivery to the distribution apparatus or the subscribers.
• Such transactions include text channel definition transactions which instruct the subscriber's set top box which group of channels it is entitled to receive, which frequency to tune for a particular text data channel, whether to mute the audio for that text channel, the pagination delay between pages, and the like.
• System manager 12 also communicates via a head end link to an RS-232 port of a head end controller (HEC) 14 which controls the transmission of television programming to the subscribers.
• HEC 14 communicates via a control link to an RS-232 port of an information services processor (or data controller) 16 which manages the flow of EPG and text data in accordance with the invention.
• ISP 16 information services processor 16 is preferably located at the cable head end with system manager 12, HEC 14 and the signal scramblers. However, those skilled in the art will appreciate that all of the head end equipment need not be located at one site.
• EPG data is supplied from one or more local or remote EPG suppliers 18 via a satellite link, modem link or other communication link to an RS-232 port of ISP 16.
• text data from one or more text channel suppliers 20 is provided via a satellite link, modem link, or other communication link to another RS-232 port of ISP 16.
• ISP 16 has a plurality of identical RS-232 ports for accepting data from a plurality of EPG suppliers 18 and text channel suppliers 20.
• ISP 16 manages EPG and text source databases in response to control signals from HEC 14 in order to provide EPG data and/or text channel data to selected viewers.
• HEC 14 also provides control data directly to the viewer's set top box via an RS-485 output port.
• the control data from HEC 14 can include text channel definition transactions as well as EPG definition transactions for instructing the set top box at which frequency to tune for the EPG data and the like.
• the control data may also include software for downloading into the viewer's set top box for reprogramming the viewer's set top box as necessary.
• the control data from HEC 14 can be inserted into the vertical blanking interval of the selected cable television signal by daisy-chained scramblers 22, 24 and 26 using known in- band techniques, although the control data from HEC 14 may also be modulated on an out-of- band carrier or an in-band audio carrier for transmission as.
• scramblers 22-26 are daisy-chained so that the scramblers may be addressed individually or globally.
• EPG data and text channel data from ISP 16 are provided to the viewer's television set top box via an RS-485 output port of ISP 16.
• EPG data and text channel data are similarly inserted into the vertical blanking intervals of selected cable television signals by EPG scrambler 28 and text channel scramblers 30 and 32, respectively.
• Scramblers 22-32 may insert the control data, EPG data, and text channel data into other portions of the video signals such as the horizontal blanking intervals or else replace the video entirely.
• the number of scramblers depends on the number of premium channels for which scrambling is used.
• EPG scrambler 28 and text channel scramblers 30 and 32 are identical to control data scramblers 22-26 and are similarly daisy-chained for individual or global addressing.
• scramblers 28-32 receive a single serial data channel which carries the combined EPG data and text data and display control transactions, as described in more detail with respect to Figure 29) for all data streams in use.
• Each scrambler is also equipped with memory for storing a predetermined amount of this data in an internal memory so as to minimize the number of database accesses.
• scramblers 28-32 have internal memory sufficient to store a significant number of transactions.
• scrambler 30 may have enough internal memory to score a day's sports scores for display on a sports text channel.
• the data received and stored in scramblers 28-32 is preferably in RS-485 format, and the protocol in a preferred embodiment is SDLC. All data transactions to scramblers 28-32 are sent on individual data streams specifying the target scrambler (station addresses in SDLC protocol), and the control data is sent on a global data stream which is filtered in the scramblers 28-32 based on the address ofthe scrambler so that the data streams can be configured by a transaction from ISP 16.
• the individual EPG data and text data streams are preferably generic in the scramblers so that they can be allocated as desired.
• scramblers 28-32 have baud rates of at least 9600.
• the subscriber's set top box is a set top box 34 which comprises an EPG memory 36 for storing the EPG data from ISP 16.
• EPG memory 36 may store one or two weeks of EPG data for selective access by the viewer via a menu ofthe set top box 34. This menu preferably allows the viewer to scroll through the EPG data stored in EPG memory 36 using the key pads ofthe viewer's television remote control device.
• Set top box 34 may also comprise a nonvolatile template memory 38 for storing the template in which the EPG data is to be inserted for display to the viewer on the viewer's television 40. In this manner, a video signal containing the template display data need not be continuously retransmitted to the set top box 34, thereby saving more bandwidth.
• set top box 34 may also comprise a text data memory 42 for storing a page of text data for presentation to the screen. Thus, while one page of text data is displayed to the subscriber, the next page of text data may be loaded into the text data memory 42.
• ISP 16 manages the flow of text data and EPG data from the data service provider to the viewer's set top box 34.
• ISP 16 manages this data by accepting data only from one or more authorized text data and/or EPG data sources, processing the text data and EPG data in its internal database manager, and formatting the processed data into a common data transaction format for output to the scramblers for transmission to the set top box 34. Provision of EPG data and text data to the subscribers is controlled by the head end controller 14 via the control link.
• One example of a suitable ISP 16 is an IBM PS model 7546 personal computer with a plurality of RS-232 serial input ports for EPG data and/or text data inputs and at least one RS-485 HDLC serial link at its output ofthe type used by HEC 14. As shown in Figure 28, the control link can be a single RS-232 serial port.
• the hardware and software components of ISP 16 are then configured as illustrated in Figure 29.
• ISP 16 is illustrated in Figure 29 as a plurality of RS-232 ports which provide a common interface for the EPG data and text channel data asynchronously provided by the EPG supplier(s) 18 and text channel suppliers 20.
• the EPG data and text channel data is transmitted to ISP 16 via a satellite link (when the interface is operated in simplex mode) or by modem (when the interface is operated in half duplex mode).
• the data is transmitted at a baud rate of at least 1200.
• ISP 16 functions as a "gate keeper" which only allows access by authorized data sources. Accordingly, when ISP 16 receives a message from an EPG supplier 18 or a text channel supplier _0, it first checks the data for authorization. If that supplier is not authorized, the data is ignored. When the supplier is authorized to access ISP 16, ISP 16 performs the requested action and returns a command response message. If the communications link is simplex, the response is ignored. In this manner, access to ISP 16 can be limited by authorization codes, but as will be described below, access is also limited by whether the data provider provides EPG data or text data in the transmission protocol expected by ISP 16.
• Messages sent between an EPG supplier 18 or a text channel supplier 20 and ISP 16 can be formatted to include a start of text byte, a data block of ASCII characters, checksum bytes and an ASCII carriage return. This format is used in commands sent to ISP 16 from the data suppliers as well as in responses sent to the data suppliers.
• the checksum can be a two byte CRC of all bytes in the message field beginning with the first character following the start of text character up to but not including the checksum field.
• the checksum is transmitted in the message as the hexadecimal ASCII representation (four bytes) ofthe CRC computation.
• the data blocks can be configured differently depending upon whether the input data is EPG data or text data.
• EPG data from the EPG supplier 18 can be formatted in accordance with an EPG command set including, for example, a Define Program Command which is used to identify all data relating to a single program, a Define Category Command which is used to establish a category for identifying different types of programs, and a Delete Category Command which is used to delete an unused category to make room in the database of ISP 16 for new programming categories.
• the EPG data may be formatted on a per program basis by these commands.
• Delimiter characters can be used for variable length fields such as the title and program description blocks to identify the length of the field. For example, a NUL (0 hexadecimal) means the field is null, SOH (1 hexadecimal) means the field is valid, and ETX (31 hexadecimal) means the end of the current record.
• EPG data is formatted into transactions for transmission to the set top box 34.
• This command may also be used to update a program definition since it will overwrite a corresponding entry in the EPG database of ISP 16.
• ISP 16 may respond to such commands from the EPG supplier 18 by sending an appropriate response such as: no error (normal response), service provider not found (not authorized), type of service not found (not authorized), category ID riot found, unrecognized command, checksum error, insufficient disk space, and the like.
• Other EPG commands and command responses may be provided as desired.
• the text channel data can originate from many different text channel suppliers 20 and arrive at the ISP 16 through different communications links such as satellite, dial up modem, direct connect modem or with a direct connect to the system manager 12.
• ISP 16 can include a plurality of databases and database managers. As shown in Figure 29, there may be two types of databases maintained in ISP 16 — one type for EPG data and one for text channel data.
• the EPG database is designed to collect data from each EPG supplier and to sort each EPG program record by channel and time of day.
• a separate database is created for each text channel for collecting text data from the associated text channel supplier 20 and formatting the received text data for transmission on individual text channels using the techniques to be described below.
• Each database that is created is identified by the service provider and type of service codes listed in the Define Program Command for use in the control link commands provided to ISP 16 from HEC 14.
• a received command is checked for its command code, the service provider, type of service and authorization code, as appropriate, by router and formatter
• router and formatter 43 routes the data to the appropriate database within ISP 16. For example, if EPG data is received, it is routed via EPG database manager 44 to EPG database 46.
• EPG database manager 44 sorts the received EPG data by channel and time of day and stores the received EPG data in the appropriate location in EPG database 46 for later recall. EPG database manager 44 may also perform garbage collection on the EPG database 46 as records are deleted. EPG database manager 44 may also call a data compression software routine such as the Huffman Compression Algorithm which, as known to those skilled in the art, maps more frequently used characters to fewer bits than the usual eight bits used in normal ASCII, while giving the less frequently used characters more bits. The number of bits used for a character is based on its probability of appearing in the data stream. Huffman encoding is described in detail in an article entitled “Lossless Data Compression", Byte, March, 1991, pp. 309-314, inco ⁇ orated herein by reference. Such a routine is desired to maximize storage efficiency at EPG database 46. Similarly, each text database manager can store the text information in the associated text database and performs data compression.
• Huffman Compression Algorithm maps more frequently used characters to
• Router and formatter 43 and database managers 44, 48 and 52 are all controlled by configurator 56, which is, in turn, responsive to control data from HEC 14.
• Configurator 56 responds to control commands from HEC 14 to provide updated authorization information to router and formatter 43 for comparison with the incoming data and for adding/ subtracting database managers and databases and the like as EPG suppliers 18 and text channel suppliers 20 are added and subtracted from the system.
• ISP 16 Access to ISP 16 is can be carefully controlled through the use of authorization codes. In addition, ISP 16 can maintain control over the information services provided to the viewer by storing the EPG data and text data in a particular format in the appropriate database within ISP 16.
• the EPG database key can be a combination of the date and time field and the channel number from the EPG data. Following these fields are the duration field, the repeat field, the program rating field, the program category field, the critique field, the attributes flag field, the program traits flag field, the text data compressed flag and lastly the text data.
• the text data field may further consist of several optional subfields with a delimiter between each field.
• EPG database manager 44 can access the EPG database 46 through shared library routines such as add a record, delete a record, read a record, and the like. As the EPG database
• EPG database manager 44 may further include garbage collection routines for periodically performing the garbage collection function on the EPG database 46.
• the text databases are similarly configured except that garbage collection is not necessary.
• EPG transaction formatter 58 reads the database records of EPG database 46 and formats them to program-based transactions having a predetermined number of bytes which are transmitted to the EPG scrambler 28 for insertion into the vertical blanking interval of a video signal and transmission to the set top box 34. These transactions are then sent via a transaction arbitrator 64 to the EPG scrambler 28 shown in Figure 28 for insertion into the appropriate video channel.
• the transactions from transaction arbitrator 64 can be output to a single RS-485 output port of ISP 16 which is connected to multiple scramblers of the type used to scramble premium cable channels.
• the transactions are segmented into EPG data and text data streams for transmission to the EPG scrambler 28 (if the transaction includes EPG data) or to the text channel scramblers 30 and 32 (if the transaction includes text data).
• the EPG transactions generated by EPG a transaction formatter 58 are formatted into SDLC frames as noted above.
• a sample SDLC format for the EPG transaction data is shown in Figure 31.
• the beginning flag delineates the beginning ofthe SDLC frame
• the station address delineates the scrambler to be addressed
• the control byte is a command code that defines what is to be processed
• the information field contains the EPG data formatted as in Figure 30
• the frame check contains the CRC for all data between the beginning and ending flags
• the ending flag delineates the end of the SDLC frame.
• a transmission from EPG transaction formatter 58 will address a specific data stream and a response from the EPG scrambler 28 will identify its data stream in the station address location. As noted above, such transmissions may or may not require a response from the EPG scrambler 28.
• the EPG transactions typically include an Add EPG Block command including a byte specifying that the following data is from the EPG data stream, a control code byte specifying, for example, whether a reply from the scrambler is expected, two bytes setting forth the EPG data block number, a flag setting forth whether the EPG data is Short Term or Long Term data, the number of transactions which make up the EPG data block, and the actual transactions.
• EPG transaction formatter 58 may also generate a Delete EPG Block command which specifies that the data is to be deleted from the EPG data stream, the control code byte, and the EPG block number to be deleted.
• Figure 32 illustrates a flow chart for one embodiment of software that can be used and embodied in EPG transaction formatter 58.
• the software starts at step 500 and gets the system time and date from the ISP system clock 63 at step 502.
• An expired EPG data block is then deleted from the memory of the EPG scrambler 28 at step 504.
• An expired EPG data block is defined as a data block representing a program which has been completely aired prior to the current system time or a program which was aired before the time window used for the EPG.
• current EPG data blocks having a time and date within the EPG time window are read from the EPG database 46.
• the current EPG data blocks are then formatted into Add EPG Block commands and associated transactions at step 508.
• a block/time map of EPG transaction formatter 58 is then updated at step 510.
• the block/time map preferably stores the time that each EPG data block was sent to EPG scrambler 28.
• the transactions representing the EPG data are then transmitted the EPG scrambler 28 at step 512.
• EPG transaction formatter 58 then waits at step 514 for the next EPG update (which should occur when the system time enters a new half hour) or the next EPG change (which may occur at any time). Upon receipt of such an update or change, control returns to step 502.
• Text transaction formatters 60 and 62 similarly generate text transactions for the text data, which as noted above, is defined on a per screen (rather than per program) basis.
• an Add Text Screen command is similar to an Add EPG Block command except that the text channel number and screen number are provided in place of the EPG block number and Short Term/Long Term data bytes.
• the text transaction formatters 60 and 62 may also request the time from the scrambler so that proper pagination may be maintained.
• Figure 33 illustrates a flow chart for one embodiment of software that can be used and embodied in text channel transaction formatters 60, 62.
• the software starts at step 600 and reads a text screen record from the text database 50 or 54 at step 602.
• the text screen is formatted into Add Text Screen transactions for transmission to the text channel scramblers 30, 32 at step 606.
• such transactions are formatted such that the display characters are sent as display commands rather than as separate characters for every display coordinate ofthe text display screen.
• text transaction formatter 60, 62 waits a period of time specified by system manager 12 (if auto-pagination is used) before the next text page is formatted and transmitted to the text channel scramblers 30, 32.
• control returns to step 604 and the next text screen record in the text database 50, 54 is formatted for transmission to text scramblers 30, 32 for insertion in the vertical blanking interval of a particular video signal.
• Text data can be passed to the screen by sending a separate character for each display location of a page.
• a text screen is represented by sending 384 (24X16) characters, one for each display location for that display screen.
• a blank space character is sent to indicate that no character is present in a particular text screen location.
• the text data is transmitted to the screen along with appropriate commands for controlling the display ofthe text data. For example, a first display command in a sequence identifies the following data as text data and instructs the set top box 34 to fill the television screen with a blue background or some other background or template over which the text will be displayed.
• the text data is then converted into a series of commands which together identify the separate screens of text.
• the text data is grouped on a per screen basis, which allows the appropriate delay mechanism to be inco ⁇ orated into the display commands to provide the necessary delay between the presentation of respective text screens.
• transaction formatters 60 and 62 can include software for scanning the text data for actual characters, skipping extra spaces in the text data, and grouping the actual text for transmission in transactions of a designated size which will fit in the vertical blanking interval of a field of a video signal. Since spaces are eliminated, the display commands include a coordinate specifying the row and column address ofthe first display character on the screen and a number of contiguous characters follow that character in the same transaction until the transaction is filled or a number of successive spaces are encountered.
• Attribute information such as underline, blinking, or luminance inversion associated with the characters may also be transmitted using these display commands.
• These display commands are used to read the text data for a text screen from the appropriate database, and at the end of the text data for a text screen, a display command is transmitted to indicate that all data for that text screen has been transmitted.
• the transaction formatter 60, 62 also includes a wait loop or "timeout" command at the end ofthe transmission which builds in a delay (on the order of 7 seconds) which gives the viewer sufficient time to view a text screen before the text data for the next text screen is displayed, thereby providing auto-pagination ofthe text screen.
• Auto-pagination permits the viewer to automatically advance from one text screen to the next without any intervention by the viewer.
• the cable operator can specify the time duration between screens and forward this information to the transaction formatters; 60, 62. Then, during operation, when the viewer selects a text channel, the current page of text data is displayed by extracting the selected text channel data from the vertical blanking interval of the video signal in which it is inserted and mapping that text data to a channel ofthe viewer's television which is designated for display of that text channel. The next screen of text data will be displayed after a predetermined delay which gives the viewer sufficient time to read the displayed text data for the current screen (approximately 7 seconds).
• Configurator 56 can respond to control commands from HEC 14 to provide updated authorization information to router and formatter 43 for comparison with the incoming data and to add/subtract database managers and databases and the like as EPG suppliers 18 and text channel suppliers 20 are added and subtracted 57 from the system.
• the control link between HEC 14 and configurator 56 can be used to report the status of the ISP 16 to system manager 12. Additionally, the control link may accept text data from system manager 12 for displaying system messages and the like.
• the interface between configurator 56 and HEC 14 is an RS-232 port with a data format fixed at, for example, 9600 baud. All control data is preferably transmitted as ASCII characters.
• configurator 56 Upon receipt of a message from HEC 14, configurator 56 checks the data, performs the requested action, and returns a command response message in a message format of the type described above for communications between router and formatter 43 and the EPG and text channel suppliers.
• Sample commands sent from HEC 14 to configurator 56 over the control link include a Set Date and Time command (for synchronization purposes), Request Configuration commands, Request Status commands, Get Category Record commands, Scrambler Control commands, and Database Control commands.
• ISP 16 monitors all input ports data from the EPG and text data service providers and builds a list of all available EPG and text data services. This list is sent to the system manager 12 using a Request Configuration command.
• This command specifies the available service providers, the type of service (EPG or text data) from each provider, the communications port associated with each service, the scrambler address or data stream (EPG or text data) for each service, the authorization code from the supplier for each service, the time and date of the last update from the service provider, the time and date of the last update to the scramblers, the time and date ofthe latest EPG data in the EPG database, and the like.
• EPG or text data the type of service
• the communications port associated with each service the communications port associated with each service
• the scrambler address or data stream EPG or text data
• the Request Status command can contain flags indicating whether there are errors present in the error log and if the category list has changed since the last Request Status command was received. Get Error Record and Get Category Record commands may then be used to extract the error and category information.
• configuration commands are separated into EPG and text service configuration commands.
• a Configure EPG Service command specifies the service provider, the type of service, whether the service is to be enabled or disabled, the authorization code from the EPG supplier 18, the scrambler data stream for Short Term data, the scrambler data stream for Long Term data, the length of the Short Term data in hours (1 -255), and the length of the Long Term data in hours (1-4096).
• the Configure Text Service command specifies the service provider, the type of service (weather, sports, and the like), whether this service is to be enabled or disabled, the authorization code from the text channel supplier 20, the scrambler address or data stream for the text data, the channel number, and the pagination delay time in seconds) before the next page of text data is to replace the current page of text data on the screen for auto- pagination.
• the scrambler control commands can include, for example, a Rebuild Scrambler Memory command which is used when a scrambler replaced and needs data to be reloaded in its memory and a Scrambler Configuration command for specifying the amount of scrambler memory in kilobytes.
• the database control commands include, for example, a Clear Database command which is used to clear the database associated with a particular service and a Delete Database command which is used to delete the database associated with a particular service.
• Other database control commands such as a Download Category Map command may also be provided for establishing a list of the specified categories of program data in the EPG data.
• Figure 34 illustrates one embodiment of a set top box 34 that can be used with the present invention.
• set top box 34 includes EPG memory 36, template memory 38, text page memory 42, a set top box 700, and a set top processor 702 which reads commands from the vertical blanking interval ofthe incoming video signal and performs the appropriate action. For example, if the incoming command is a text channel definition or EPG definition command from HEC 14, the appropriate update of bit map 704 is performed. Similarly, if the incoming command is a display command including EPG data, that data is stored in EPG memory 36 and is displayed with the template stored in template memory 38 when the user makes a menu selection via television remote control unit 706 and remote receiver 708 requesting display of the EPG data.
• the template data may be sent as part of EPG display commands if no template memory is provided.
• the incoming command is a display command including text data
• a page of that data is stored in text page memory 42 for presentation to the display a page at a time.
• the text page memory can be automatically updated every few seconds by virtue ofthe delay built into the display commands from the text formatters 60, 62 if auto-pagination is enabled. Alternatively, the user can be allowed to manually access the text data in the memory.
• set top processor 702 preferably has the ability to request the next text page while the current page is being displayed so that the next text page is already loaded for display at the end of the screen delay time.
• the selected text, EPG or video signal is then modulated at modulator 710 for display on television screen 40 at the channel specified in bit map 704.
• Bit map 704 of set top processor 702 of the set top box 34 maps the received text information to the designated cable channel for display by designating the frequency that must be tuned by box 700 to receive the desired text data. This information is received in the aforementioned text channel definition transactions from HEC 14.
• the viewer may specify via television remote 706 that she wishes to view a sports text data channel which her program guide indicates to be available by tuning the set top box 34 to channel 181.
• Set top processor 702 checks bit map 704 for channel 181 to determine that it must tune the frequency for channel 29 in order to extract the sports text data for the viewer's channel 181 from the vertical blanking interval of channel 29, set top processor 702 ten sets set top box TOO to tune channel 29 but the video signal for channel 29 is not displayed. Instead, the video screen is blanked by set top processor 702 and the text data extracted from the vertical blanking interval by set top processor 702 is displayed. Any necessary descrambling of the received video is performed by set top processor 702.
• electronic content is tagged and aggregated at a server and becomes electronic content with targeted information.
• Suitable electronic content includes but is not limited to, advertising, news segments, video programs, audio programs, WEB pages, and the like.
• the electronic content with targeted information is send to a broadcast server and broadcast widely.
• Set top box 34 has created profiles representing different people in the household. Set top box 34 matches the tags in the broadcast stream with the profiles created and determines which of the tagged content should be downloaded. A subset of all of the content that is broadcast that are relevant to the personalities in the household are downloaded and displayed to the user.
• set top box 34 can be used to match the tags in the broadcast stream with a demographic profile of a viewer previously created and determine which of the tagged content should be downloaded. An individual viewer receives from set top box 34 only.* Set top box 34 matches the tags in the broadcast stream with the demographic profile of the viewer created and determines which of the tagged content should be downloaded.
• set top box 34 reads commands of the incoming tagged and aggregated electronic content and performs the appropriate action. For example, if the incoming command is a text channel definition or EPG definition command from HEC 14, the appropriate update of bit map 704 is performed. Similarly, if the incoming command is a display command including EPG data, that data is stored in EPG memory 36 and is displayed with the template stored in template memory 38 when the user makes a menu selection via television remote control unit 706 and remote receiver 708 requesting display of the EPG data.
• the template data may be sent as part of EPG display commands if no template memory is provided. If the incoming command is a display command including text data, a page of that data is stored in text page memory 42 for presentation to the display a page at a time.
• the text page memory can be automatically updated every few seconds by virtue of the delay built into the display commands from the text formatters 60, 62 if auto-pagination is enabled. Alternatively, the user can be allowed to manually access the text data in the memory. If manual access is provided, it is preferred that the text page memory hold at least the currently displayed text page, the previous text page and the subsequent text page in order to give the user the ability to scroll through the text data.
• set top processor 702 preferably has the ability to request the next text page while the current page is being displayed so that the next text page is already loaded for display at the end of the screen delay time.
• the tagged and aggregated electronic content is then modulated at modulator 710 for display on television screen 40 at the channel specified in bit map 704.
• Bit map 704 of set top box 34 maps the received tagged and aggregated electronic content to a designated cable channel for display.
• Set top processor 702 then checks bit map 704 for channel 18 1 to determine that it must tune the frequency for channel 29 in order to extract the. Any necessary descrambling of the received tagged and aggregated electronic content is performed by set top processor 702.
• the present invention also provides a mechanism to automatically create multiple profiles corresponding to multiple users with or without any information being explicitly providing by the users about themselves.
• Fig 35 shows an embodiment of the invention that causes the automatic creation of multiple profiles and automatic identification of profiles when they are active on the device.
• number of profiles to be created is initially determined. Several methods may be used to determine the number of profiles. In one embodiment ofthe current invention the number of profiles is optimally determined by applying the Minimum Message Length (MML) criterion. The process of applying MML criterion for determining the optimal number of clusters in described in reference (ref..). In one embodiment of the current invention, a user ofthe device explicitly specifies; the number of profiles to be created.
• MML Minimum Message Length
• the system monitors user interactions with the device.
• the user interactions may include but are not limited to channel change requests, requests to view more information, configuration of device parameters, requests to performs recording or deletion of programs from a storage device.
• All actions due to user interactions are recorded in a history database that stores history of viewer actions.
• Data records in the history database describing viewer action can be of different formats.
• Figure 37a shows one ofthe possible formats for the data record.
• Data records may contain information about the action, time at which the action occurred, and some parameters that further describes the action.
• only actions matching certain criteria are recorded in the history database. For example, all user action about watching channels can be ignored if the duration of watching is less than a configurable threshold, or all user action about watching a particular channel like preview channel can be ignored.
• a usage session in this context can be the period during which there are user activities between two periods of inactivity, or the period during which the device is used between two periods during which the device is not in use. it will be left to the device to determine who the actual user is, in order to provide a very personalized environment to the user. In such a scenario the user action records will not contain any information about who the user is.
• the user can identify himself or herself explicitly for all or some ofthe usage sessions by using some mechanism provided by the system. These mechanisms may include pressing a sequence of keys, or choosing a user name from a menu and logging-in as that user.
• the user can identify himself or herself during certain sessions and may choose not to identify himself or herself during some other sessions.
• certain user action records can have the user name or some other identification data as a parameter which specifies that a user was togged-in for the session from which the user action record was generated.
• the current invention provides a mechanism to create multiple profiles automatically.
• Each of these profiles may correspond to the entire preference of a single user, a group of related preferences of a single user, a group of related preferences of a group of users or the entire set preferences of a group of users.
• the mechanism for creating multiple profiles is described in figure 36.
• Sets of consecutive user action records are grouped together to form usage pattern records.
• Usage pattern records can be in the form of arrays of user action records. Only user actions that occur contiguously are grouped together in a single usage pattern record.
• the usage pattern records can be formed using many methods, some of which are:
• Each usage pattern record is mapped to a point in an N-dimensional space, each axis of this N-dimensional space representing a parameter that is of significance in identifying multiple profiles.
• the N-dimensions for this space are called cluster-dimensions.
• the parameters for this N-dimensional space are chosen either manually by people skilled at identifying the most significant parameters in identifying multiple profiles, or automatically by identifying the aspects of profiles that differs significantly between profiles.
• a set of these parameters are identified and configured as the apparatus is configured during initialization, and a new set is added periodically by looking at aspects which differ in the multiple profiles that are stored in a device.
• initial parameters are the channel names in a channel line up and viewing times.
• these initial parameters are possible values of program description fields of television programs.
• the rate of channel change is also one of the parameters.
• the mapping of the usage pattern record to a point in the space defined by cluster-dimensions can be done by many methods. One method determines the quantity of a particular characteristic which is defined by a clustering parameter, exhibited in a usage pattern record, e.g. number of channel changes in a usage pattern, and uses this as the value of the axis- for the corresponding dimension. One method determines the rate of consumption of a particular characteristics in a usage pattern record, e.g. rate of consumption for "NBC" in a usage pattern record is .5 which indicates that programs on "NBC" were watched 50% of times during the period of this usage pattern record.
• All the points in the space defined by cluster dimension are clustered into a number of clusters using standard clustering algorithms. Any clustering algorithm can be used to perform the clustering.
• EM clustering is applied to group the points in the cluster- dimension space into a predetermined number of clusters. The number of profiles being created decides the number of clusters formed.
• Each cluster formed as a resulting ofthe clustering represents a single profile.
• the clustering process also provides the mixture weights for each of the clusters as one of the outputs. The mixture weight for a cluster can be used to compute the percentage of time the profile was active, ofthe total amount of time for which the device was used.
• the clustering is performed periodically using the user actions records accumulated in the history database.
• user action records are periodically removed from the history database based on certain criteria such as the age ofthe record, size ofthe record, relevancy ofthe record etc.
• the current invention also provides a mechanism to predict profile that is active at any given time.
• Figure 40 illustrates the method for identifying the profile currently active on the device. Current user actions are monitored and user action records are created. The most recent user action records are used to create a usage pattern record. The usage pattern record is mapped to a point in the cluster dimension. Using the Information about clusters representing multiple profiles, the, probability of each ofthe clusters being active given the usage pattern is computed. This is computed by using Bayesian Probability theory that can be used to compute the posterior probability using the prior probability and mixture weights. The probability of a cluster being active is computed using the probability ofthe usage-pattern record given the profile
• P(profile/usage_pattern) P(usage_pattern, profile) * P(profile)) P(usage_pattern)
• probability of usage pattern record occurring given profile P(usage- pattern, profile)
• the cluster probability distribution can be assumed to be the Gaussian probability distribution, so that the P(usage_pattern profile) can be computed using the cluster center and cluster variation.
• Cluster center and the cluster variation are the output generated by the clustering process performed for generating multiple profiles.
• switching the device on can be one of the user actions recorded and "time of usage" one of the clustering dimension.
• the probability for any cluster being active can be determined, even with out any further user actions.
• the current usage pattern record is refined to include more user action records and the probability of profiles which may be active is refined with the addition of each new user action record.
• a set of the oldest user action records may be removed from the current usage pattern record. This process ensures that the effect of individual user actions in the identification of active profiles decreases as time passes.
• the clusters generated by the clustering procedures are used to create multiple profiles using a method described in Figure 41.
• the clusters are used to compute the probability of each clusters being active during the period of each usage pattern record.
• User preferences for each profile are created using processes described above using the user actions weighted by the probability of the corresponding cluster being active.
• multiple profiles may be created for each user.
• Each of such multiple profiles may be used to describe the user's viewing preferences at various time periods, such as weekday or weekend, or morning or evening.
• the profiles may also vary according to other variables, such as showing more sports during football season, more gardening shows during spring, more political commentaries every four years during the primaries, etc.
• the broadcast network amasses content from the many different content providers and combines it at a network broadcast head end 29 into a broadcast signal that is transmitted to, and received by, the television control system 2 of the present invention.
• the control system 2 sorts through the content received from the broadcast network based upon the profiles of the different viewers in the particular household, compiles lists of preferred programs available to each user, records and stores programs of interest, and makes the programs available for viewing when required. Equally important, the control system 2 receives and sorts through various advertising content and stores the content most relevant to the viewers in the house hold based upon their demographic profiles and the target audience of the advertising content.
• content providers include EPG suppliers 18, text channel suppliers 20, conventional programming suppliers 23, and alternative content suppliers 25.
• alternative content refers to highly targeted programming that may be delivered through the use of a control system 2 as described herein.
• Conventional programming encompasses all types of programming currently available and including, but not limited to, movies, news, sport events, educational programs, pay-per-view shows, and shopping channels.
• Advertising content suppliers 19 provide promotional material such as short video clips or graphical and text information to programming suppliers 23 for inclusion in their conventional programming content, in a manner familiar to TV viewers worldwide.
• advertising content may be supplied to alternative content suppliers (see discussion below) and may also be profiled through a profiler 21 to develop meta data descriptive of each individual advertising message.
• the meta data refers to demographic characteristics that describe the viewers to whom each advertisement is specifically targeted.
• the profiler 21 may be a software based system or may rely upon human intervention to watch and analyze each advertising message and develop meta data descriptive of the advertisement. Alternatively, the advertising content suppliers 19 may provide this information along with the advertisement.
• the meta data is developed based upon the parameters of interest to the clients for whom the ads are developed, and will typically include the age, gender, ethnic background, education, profession, and other demographic information related to the viewer. The meta data will thus describe the viewers to which each advertising message is targeted in terms of the viewers' demographic information as defined by the parameters mentioned previously.
• the meta data thus generated is then appended to each advertisement and the thus- tagged advertisements are provided to the broadcast network for inclusion in the broadcast signal.
• the tagged advertisements will preferably be provided in a plurality of channels, wherein each channel may carry a specific type of advertisement in terms of the target audience.
• each channel may be defined according to one demographic parameter or an often recurring combination, e.g. a channel with ads for men, a channel with ads for women, a channel with ads for male sports fans, etc.
• each channel may carry a stream of all types of ads, wherein each type of ad is present in a ratio dependent upon the demographic parameters to which it is targeted.
• ads targeted to male sports fans on weekend afternoons will be more numerous than ads targeted to young children, because of the large number of sports events broadcast at such time and the high probability that young children will be playing outside rather than watching TV indoors at such times.
• the signal broadcast by the network may further include information provided by the administrator of the service provided by the invention, e.g. Metabyte Inc., the assignee of this application. As described in more detail elsewhere in the application, this information may include updates of the software ofthe control system 2.
• the network broadcast signal is transmitted to, and received by, the control system 2.
• the broadcast signal may be analog or digitally encoded, and may be transmitted via cable, satellite, telephone line, or any other practicable manner.
• practical considerations will most likely dictate that the network signal be broadcast in digital form, in which case both the network head end 29 and the control system 2 will need to inco ⁇ orate ADC and DAC circuitry, in a manner very well known to those skilled in the art.
• control system 2 will need to inco ⁇ orate channel tuning circuitry that, although not shown in the figure, will be necessary to separate the numerous signals multiplexed in the broadcast signal and select any one of these signals (i.e. channels) as necessary.
• the broadcast signal is supplied to various components ofthe system.
• the program switch 1 14 responds to the viewers input as provided via a remote control or similar unit to select the desired channel and direct the channel signal to the TV monitor 108.
• the preference agent 1 10 monitors the viewing selection ofthe various viewers using the control system 2 and creates viewing profiles of each viewer that are stored in the preference database 1 16. Based upon these profiles, the preference agent 1 10 sorts through the incoming programming content as described in the EPG information to compile lists such as "Top 10" lists of viewing choices available at any given time to each viewer, and directs the recording manager 1 12 to record the top-ranked program being broadcast at any given time (including any programs selected by the viewers for recording) and store it in the stored programs memory device 35.
• the preference agent further contains software that allows it to create a demographic profile for each viewer, based upon the viewing profile of the viewer and certain algorithms or associative rules. These algorithms may be adjusted over time as the model employed by the system administrator 27 is enhanced and its accuracy improves.
• the system update information channel included in the broadcast signal may include periodic software updates, including new preference database parameters that may need to be included at the request of the advertising suppliers 19.
• the control system 2 of the invention may be remotely upgraded to meet any new demands that may arise as advertising content providers become familiar with the system and the process of custom tailoring narrowly focused, targeted advertisements.
• the demographic profile created for each viewer is stored in the demographic database 31 , which resides in the control system 2 and thus ensures the viewers privacy.
• the preference agent 1 10 also sorts through the advertising content streaming in through multiple advertising channels contained within the broadcast signal and, based upon the demographic profiles ofthe viewers and the meta data contained in each advertisement to describe the target audience for the particular advertisement, stores and/or causes the display of particular advertisements.
• the control system 2 may utilize any of a variety of methods to manipulate the advertising content, as described below.
• the advertising channels each carry the same type of advertising.
• the preference agent determines which viewer is watching TV at any given time and stores in stored ads memory device 33 those tagged advertisements that are targeted to the particular viewer. At appropriate times during the program that is being watched by the viewer, such as during the commercial breaks typically inserted in most TV programs, the preference agent directs the program source switch 114 to access the stored ads and play a selected advertisement on the TV monitor 108. If the program being watched by the viewer contains information regarding the length of the commercial break, the preference agent may select stored ads of appropriate length to insert in the allotted time slot. The preference agent may further keep track of the ads that have been previously played to ensure that all stored ads are displayed equally.
• tagged ads may contain the desired number of times that the advertisement provider wants the ad to be aired during any given day, or perhaps the specific times at which the ad should be shown.
• the preference agent may review all ofthe stored ads meta data and develop a strategy for showing these ads to the viewer including when and how often.
• the preference agent may cause a certain number of ads to be displayed and direct the recording manager 112 to record the selected program if the stored ads being displayed run longer than the allotted time slot for the commercial break.
• the control system 2 ofthe present invention can therefore manipulate the broadcast schedule of a program to a certain extent by modifying the amount of advertising content that the viewer is subjected to.
• the advertising channels may be operational for only a limited time during the day, typically at off-peak hours (e.g. 2 a.m.), during which all advertisements for the coming day may be downloaded and stored.
• an advertising channel may be turned on at a certain time and stream all advertisements to the control system during the space of an hour or two.
• a regular programming channel may suspend programming for an hour or two at a convenient time and supply the advertisements for the following day.
• a dedicated connection such as a phone line or an internet connection may be used for periodic downloads of advertisements.
• multiple advertising channels carry a mixture of advertisements such that at any given time the preference agent has the option of selecting at least one advertisement targeted to the viewer from one of these channels.
• the control system need only to store one advertisement at any given time to ensure continuity between the program being watched and the advertisements.
• the commercial break in the program may occur at a time that does not correspond to the beginning of an advertisement on any ofthe advertising channels.
• the advertisement stored by the control system is directed through the program source switch to the TV monitor while another targeted advertisement is concurrently stored for subsequent display. If the commercial break happens to coincide with the start of a targeted advertisement on any ofthe advertising channels, the preference agent can simply cause the program source switch to direct the particular channel to the TV monitor while another advertisement from another advertising channel is being stored.
• each regular programming channel can carry multiple, multiplexed versions of an advertisement.
• the preference agent selects the most appropriate version ofthe advertisement and directs it to the program source switch. This embodiment would require additional circuitry to de-multiplex the various versions of each add and apply the particular version selected by the preference agent to the program source switch.
• the novel method of delivering targeted advertising that is provided by the invention is extremely advantageous to the advertising community.
• the invention allows greater freedom in producing advertisements because they no longer need to be appealing to as wide and varied an audience.
• the alternative advertisements are delivered in real time, all viewers are reached by individualized advertisements at the same time, thus providing significant time savings in comparison with the prior art, wherein five different advertisements would require five different time slots during which to be broadcast.
• the cost savings thereby realized, especially during very popular events such as the Super Bowl, can be significant.
• preference agent 110 may sort through available channels to select the ten (or any other number) programs currently playing that most closely match the viewer's viewing profile.
• preference agent may also build customized listings of future programs based upon the viewer's profile as well as any additional criteria specified by the user.
• the user will have the ability to fully customize his viewing profile, including the values assigned to the different parameters that make up his or her viewing profile.
• the user may even be allowed to specify what kind of advertising he prefers. Such a configuration would likely generate an alternative billing arrangement, whereby the user agrees to watch advertisements in exchange for the ability to specify the types of advertising he wishes to be subjected to and perhaps some sort of financial incentive.
• Another novel possibility engendered by the present invention is the creation of a 'personal channel' that is always showing the most interesting program being broadcast at any given time and/or previously recorded and stored programs that were very close matches to the viewer's profile.
• the personal channel may thus show a collage of movies, news segments, sports events, and any other programming content that was broadcast during the preceding 24 hours and that matched the viewer's profile to within a specified degree.
• control system 2 An additional element of control system 2 that may be inco ⁇ orated only in selected control systems is a privacy filter 37 that deletes any personal information from the demographic database and transmits this anonymous information to the system administrator 27 for pu ⁇ oses of maintaining and updating the model used to generate the demographic database.
• Such models of the control system will require a feedback line for transmitting the demographic information to the system administrator, and in a preferred embodiment is a telephone line or a dedicated internet line such as DSL, cable, etc.
• the system administrator may offer financial or other incentives to users to convince them to supply this type of information. By carefully selecting users across a wide demographic cross section, the system administrator can use the information thus gathered to enhance the model used to develop a viewer's demographic profile based upon his or her viewer profile by comparing these users' actual demographic data with the demographic profile developed by the preference agent.
• the targeted advertising developed for delivery by the control system of the invention may include novel elements such as coupons or highly targeted descriptions.
• targeted advertising developed for distribution via the present invention may include specific information, couched in specific language, that is intended to be especially appealing to the target audience.
• the invention further allows the creation of highly targeted content other than advertisements that can be delivered only to a very specific audience.
• movies, shows, religious programs, video magazines, infomercials, etc. may be developed to reach a very specific audience without the restrictions typically imposed on the content developers when the program will reach, or at least be accessible to, a very wide audience.
• This embodiment will require that such specific content be supplied via dedicated channels that cannot be tuned to directly by a conventional TV tuner, and thus may only be accessed through the control system 2.
• Such highly targeted content may be provided by alternative content suppliers 25 or even be developed for alternative distribution by conventional programming suppliers 23.
• use of the present invention may create a new distribution medium that will allow the content providers to not only reach a very specific audience but also to remotely, automatically exclude certain segments of the audience from accessing this material.
• Such alternative content could be broadcast exclusively on a viewer's personal channel, as described previously.
• a time line 800 is shown tracking the progress of a typical television program 850.
• program 850 is composed of scenes such as Scene 1, Scene 2, etc., and advertisements such as Ad 1, Ad 2, etc.
• scenes are shown sequentially as predetermined by the producer of the program, and are interspersed with showing of the advertisements as determined by the head-end operator.
• customized linear programming is enabled due to the development of demographic and/or viewer preference profiles for each user.
• a customized linear program 860 may still be comprised of scenes interspersed with advertisements, but some or all of the scenes as well as the advertisements may now be targeted to various demographic or viewing traits of the user.
• Scene 2 may actually have been shot in two different versions by the producer of the television program, Scene 2 and Scene 2a, wherein each version of the scene is more appropriate, or of more interest, or less offensive, to a particular target audience.
• Scene 2a may be a less violent or less graphic version of Scene 2.
• the system of the present invention receives TV signals carrying both scenes, and selects among the available alternatives based upon the profile of the currently watching user and the target data of the alternative scenes.
• Program 860 as shown in Fig.
• Ad 1 may be provided in four variations, each targeting a different demographic characteristics, be it the users sex, income, geographic location, or any other desired characteristic. As previously mentioned, providing such targeted ads provides a more interesting and enjoyable watching experience for the user, and thus enhances the likelihood that the user will watch the advertisement rather than switch channels or walk away.
• the present invention enables showing advertisements in an apparently conventional, linear manner whereby the broadcast of a television program is halted periodically by the head-end to broadcast advertisements.
• the advertisements thus shown are targeted to the individual user watching a particular TV set, but the user is not prompted in any manner to select between alternative ads or in any other way informed of the fact that alternative ads are being received by the set-top box.
• This is a different manner of providing targeted advertisements from providing targeted advertisements by a programming guide because, in a conventional programming guide context, the TV screen is typically split in at least two different areas, wherein one area shows program listings and the other area displays advertisements.
• the user does not receive any programming at all, but rather only program listings and advertisements.
• alternate scenes and advertisements are preferably of the same length and thus occupy the same time slots within the program time line.
• alternative advertisements and scenes of different lengths such as those comprising television program 870.
• the set top box would additionally fulfill the role of scheduler by calculating the length of each scene and advertisement and scheduling the scenes and advertisements for showing to the user in a seamless, apparently linear program with no interruptions or overlaps.
• both the scenes and the advertisements may be comprised of any combination of broadcast and previously recorded, stored scenes and advertisements.
• alternative Scene 2a may be shorter than alternative Scene 2, which would necessitate showing Scene 3b and Ad la at an earlier time.
• Ad A may be shown between Ad la and Ad 2c, where Ad A is selected by the set top box because its target profile fits the demographic profile of the user and because it is of the appropriate length of time.
• providing such customized content may also allow greater artistic and expressive freedom for producers of programming, as they will be able to explore different variations ofthe same story line and edit various scenes for various audiences.
• Another possible use for the customized linear programming enabled by the present invention is providing highly individualized services such as news, weather, stock market quotes, shopping events, instructional videos, etc.
• the popularity and acceptance of the system ofthe present invention will depend largely upon the cost to the end users, i.e. the viewers.
• the user pays a set price for the control system, i.e. the hardware, connects the control system to his TV and incoming cable or satellite dish line, and enjoys the personalized services provided by the control system at no further cost.
• the system administrator will provide tagged advertisements for broadcast by the broadcast network and charge a fee to the advertisement client.
• the actual advertisements may be provided by an advertising content supplier or the advertising client itself (e.g. a truck manufacturer), to which the system administrator tags the meta data.
• the fee may be based upon the total number of control systems installed and/or the target audience that the advertising client wishes to reach.

Landscapes

• Engineering & Computer Science (AREA)
• Databases & Information Systems (AREA)
• Multimedia (AREA)
• Signal Processing (AREA)
• Social Psychology (AREA)
• General Health & Medical Sciences (AREA)
• Health & Medical Sciences (AREA)
• Computer Networks & Wireless Communication (AREA)
• Computer Graphics (AREA)
• Business, Economics & Management (AREA)
• Marketing (AREA)
• Computing Systems (AREA)
• Computer Security & Cryptography (AREA)
• Two-Way Televisions, Distribution Of Moving Picture Or The Like (AREA)
• Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Priority Applications (1)

Applications Claiming Priority (9)

Related Child Applications (1)

Publications (1)

Family

ID=27496004

Family Applications (1)

Country Status (4)

Cited By (1)

Families Citing this family (49)

Citations (1)

Family Cites Families (11)

• 2000
  • 2000-08-31 AU AU69478/00A patent/AU6947800A/en not_active Abandoned
  • 2000-08-31 EP EP00957929A patent/EP1230798A1/en not_active Ceased
  • 2000-08-31 JP JP2001521064A patent/JP5068910B2/ja not_active Expired - Fee Related
  • 2000-08-31 WO PCT/US2000/023868 patent/WO2001017250A1/en not_active Application Discontinuation
• 2010
  • 2010-09-16 JP JP2010208382A patent/JP5591638B2/ja not_active Expired - Fee Related
• 2012
  • 2012-02-03 JP JP2012021577A patent/JP5475030B2/ja not_active Expired - Fee Related
• 2013
  • 2013-12-16 JP JP2013259144A patent/JP2014078995A/ja active Pending
• 2014
  • 2014-03-17 JP JP2014054015A patent/JP6067609B2/ja not_active Expired - Fee Related

Patent Citations (1)

Also Published As

Similar Documents

Legal Events