Patent for “Recommendation engine apparatus, methods”

Software – David Kimble, Fee Ling Chin, Time Warner Cable Enterprises LLC

Search Patent for “Recommendation engine apparatus, methods”

Abstract for “Recommendation engine apparatus, methods”

“Recommendation engine apparatus, and the associated methods, provide content compiled and selected from different sources. One embodiment of the recommendation apparatus includes a headend entity. In another, it is located on a user?s CPE. One embodiment creates content records using content metadata. This is for comparison with a user’s profile. Although the profile is preprogrammed, it can dynamically adapt to the user’s preferences when the user makes content decisions. Client applications are used to create and present content. Feedback mechanisms are used to allow?learning? User activities are used to generate more specific recommendations and to?unlearn? stale preferences. The recommendation content can be displayed as a playlist, a continuous stream on a virtual channels, or in an electronic program guide. An engine for business rules? It is also useful for implementing business or operational goals.

Background for “Recommendation engine apparatus, methods”

“1. “1.

“The invention is related to content delivery over a network. In one aspect, the invention is about methods and apparatus that select content from multiple sources within a satellite or cable network for delivery to a client device.

“2. “2.

“Recent technological advances have allowed for the proliferation of many content sources that offer a variety of content. The sheer number of channels and channels available for purchase (e.g. VOD, pay per view, etc.) can overwhelm viewers. The like offer programming 24 hours a day. The user may find it tedious and frustrating to search channel-by-channel for the desired content. The user might not be able to quickly and easily find the content they are looking for with such a large amount of content available.

“Similarly, technological advances have made it possible to use electronic devices that allow users record content from a bearer network (such a cable TV or satellite network), at their home or any other location within the network. These devices include personal video recorders and digital video recorders (DVR). The user has access to more content than ever before stored on recorders.

There are several ways to provide content that a user might be interested in from the vast amount of content. These include using demographic data and/or explicit viewer identification. A user might have their content pre-selected, or at least narrowed, based on their demographics and/or explicit preferences. These methods only generate targeted content based on information that a user gives or inputs to the system (or can be gleaned via their sub-scriber account, etc ).

“Various other options have been suggested to help a user find content of interest, including the use of computer programs that can generate?playlists? of content that is recommended. These programs use various filtering algorithms that are known from the prior art.

Filtering algorithms can generally be divided into two types: those that use collaborative filtering and those that use content-based filtering. Collaborative filtering gathers user data, such as user-submitted ratings for individual pieces of content within a domain. To make a recommendation for a piece or content, it is possible to compare the similarities and differences between several user profiles. A community-based database is required for collaborative filtering. Content-based filtering, on the other hand, identifies items based upon some correlation between characteristics of a piece or content and user preferences (or profile). These systems heavily rely on user-supplied criteria (??seed). Items), and/or are static in their nature (e.g. they do not change unless the user alters the input criteria). A variety of hybrid content-based and collaborative filtering methods have been created.”

Alternative methods that are able to generate recommended content playlists automatically update themselves based on user feedback and/or implicit actions. These methods can cause playlists to become too narrow and specific and they do not account changes in user preferences over short periods of times, such as within different hours of the day. Some content-based systems of the past recommend content based upon a user profile. This information is provided substantially by the user.

“Several other solutions have been proposed to help a user find content of interest, including the use of a searchable guide like that in U.S. Pat. No. 7,228,556 to Beach et al., published Jun. 5, 2007, and entitled “Distributed Interactive Television Program Guide; Method?”

In the prior art, customized program guides can also be used to deliver targeted content to users. These fall into one of two categories: (i), those where a user must input preference data; and (ii), those that can gather data about a person without the user’s permission. As stated above, the first category of customizable program guide requires that a user manually input preference data or other data. U.S. Pat. 102/399 describes one example of the first type of customizable program guide. No. No. 7,185,355 to Ellis et al., published February 27, 2007, and entitled “Program Guide System With Preference Profiles?”. U.S. Pat. is an example of prior art for the second category of customizable programs guides. No. 7,020 652 to Matz, and others, published Mar. 28.06.2006 and entitled “System and Method of Customizing Content-Access Listings?”

“Based on what has been said, it is necessary to improve apparatus and methods of recommending or providing content that a user (or group) is most likely will be interested in. This can be done without imposing undue burdens on the user in terms inputs or feedback. These apparatus and methods wouldn’t rely on user-supplied criteria and ratings. They would be adaptable to dynamically and quickly update user preferences to reflect their preferences with high proficiency. This includes the ability to update, including explicit and implicit data.

“Such apparatuses and methods would also create profiles that wouldn’t become too narrow over time but would instead respond to user’s changing preferences, including preference changes over short time periods (such as during different times of the day).

“Additionally, the apparatus and methods mentioned above would allow a user to choose from recommended content and present a navigable content list to the user according to a system that immediately takes into consideration the users activities. This allows a user to derive a more refined profile without becoming too narrow and accounts for changes in a person’s preferences over short periods, such as within different parts of the day.

These features would also be made possible by substantially existing network infrastructure and components. They would also be compatible with a variety of client devices and delivery systems, including wired and wireless.

“The invention addresses the above-mentioned needs by providing an improved apparatus and method for the targeted delivery content over a network.”

“A recommendation apparatus is disclosed in a first aspect. One embodiment of the apparatus includes a storage device that can store information about a plurality content and store a plurality user profiles. A digital processor is also connected to the storage device and is adapted for running a computer program. The computer program is able to: compare information regarding a plurality content with individual user profiles; and, based at most in part on this comparison, produce at minimum one list of content, which contains a relationship with at least one of these plurality user profiles. An interface with the network and with the processor

“In one version, the apparatus consists of a headend entity (or hub) of the network.”

“In another variant, the apparatus includes a consumer premises device.”

“A further variant of the process of comparing the plurality content to the plurality user profiles involves: creating a content record for each one of the plurality content, each content record comprising metadata about the content; and computing a dot product between individual first vectors and individual second vectors to create a scalar amount. Producing at least one content list involves creating a list with a scalar quantity that is greater or equal to a predetermined real-value number and then providing the list to the client device associated to the at least one user account.

“In another variant, the list includes information to identify and retrieve each content in the list.”

“In another variation, at least one user profile from the plurality comprises a template user account.”

“In another variant, the computer software is further adapted for updating at least one user profile, by adding a training vector each to individual second vectors. The training vector comprises weighted data about one or more user activities.”

“A second aspect of the invention discloses a computer-readable apparatus. One embodiment of the apparatus includes media that can be adapted to hold a computer program. The plurality instructions are used to examine metadata associated to a plurality content, generate content records for each content record based at most in part on that metadata, compare the content records with individual user profiles associated to individual client devices, compile at minimum one list, which contains information about individual content with a threshold similarity to at the least one user profile, as well as information for identification and retrieval for each content in the list, and then transmit the list to client devices

“In one variant, metadata includes information about at least one genre, content type or advisory rating, language and era, as well as actor.”

“In another variation, the content records are expressed as vectors with identical numbers of columns or rows. The vectors contain one or more aspects that correspond to metadata associated each individual content record. Comparing the content records with the plurality user profiles involves calculating a dot-product of the user profile vectors and the content record vectors to create a scale quantity. The act of compiling at most one list of content that has a threshold similarity with at least one user profile includes creating a list containing content that is greater or equal to a predetermined real value number.

“In an additional variant, the computer software is further adapted for updating at least one of plurality of user profiles based on user actions occurring at a client device to whom the at least 1 user profile is linked by adding a training vector the user profile vector. The training vector comprises content records weighted according user actions associated with it.”

“Another embodiment of the computer-readable apparatus comprises media that are adapted to include a computer program. The plurality consists of instructions that, when executed, maintain at most one user profile; create a plurality content records at minimum in part using metadata relating to a plurality content; grant access to the plurality content to a person; use the record of at the least one act taken to update the user profile; compare the updated profile with individual content records of the plurality content to identify the individual content records having a prescribed level of similar to the user.

“In one version, the user profile includes a pre-set user profile or template.”

“In another variation, the plurality content records and user profile are expressed in vectors with identical numbers of columns or rows.” Comparing the updated user profile with individual content records involves calculating a dotproduct of individual content records to the profile to create a scale quantity. The act of identifying individual content that has a specified level of relatedness to a user profile includes identifying content that has a scalar amount equal or greater than a predetermined number.

“A further variant of the apparatus can be used to: create a weighted record of the content acted on by multiplying at least one content record with a weighting factor determined based on the nature of each act and an estimate relation of the act to user’s preferences; then add the weighted record to the user profile in order to generate the updated user account.

“Another variant of this is to identify individual content from the plurality having a predetermined level of similarity to the user’s profile by comparing attributes of each individual content to the equivalent attributes of their user profile.”

“A third aspect of the invention discloses a method for recommending content. One embodiment targets content to a specific user in a content-based network. The method includes: creating a plurality content records about a plurality content; comparing individual content records to at most one user profile; storing information about individual content records that bear a substantial relationship to the user profile; and finally, displaying the information concerning the individual content records.

“In one variation, the act to generate the plurality content records involves utilizing metadata associated to the plurality content. The content records and user profiles are expressed in vectors with identical numbers of columns or rows. To compare individual content records to user profiles, it is necessary to calculate a dot product between individual content records and user profiles to create a scale quantity. The information is displayed in a list that is organized based at most in part on the scalar number.

“In another variation, the method also includes providing content associated to at least one individual content record of the plurality that bears a substantial relationship to the user profile.”

“In another variant, the determination of content records with substantial relationship to the user profile involves comparing various aspects of content records to the corresponding aspects in the user profile.”

A fourth aspect of the invention discloses a user-action tracking apparatus. One embodiment of the user-action tracking apparatus is designed for data communication with at most a computer programme adapted recommend content to users. It comprises: an interface to receive user action data from a network; a storage device adapted so that it can store a plurality records regarding user activities; an apparatus for generating the plurality records concerning user actions; and an apparatus for updating the first training records for any user action data.

“In one version, the storage apparatus includes a cache memory.”

“Another variant of user actions records includes at least: (i), identifying information about content to which the user’s action is related, (ii), a chronological reference and (iii), a description or descriptor.

“In another variant, the act for generating the first content record includes: Generating a content file for each individual content to which individual user actions relate; creating a vector with an identical number column and row as a user profile as the user action; associating user actions description with individual content in record regarding user actions with weighting factors; multiplying the content vector by the weighting factors.”

“Another variant of the act of updating a first training document is: Associating user action description with individual content in the records regarding user actions with weighting factors; multiplying the first record by the weighting factors to create an updated training record.”

“In another variation, the apparatus can be further adapted so that the updated training record is distributed to the computer program that’s adapted for recommending content to users.”

“Another variant of the apparatus allows you to: receive at most one user profile; the vector representing the user profile is used as a vector; use the updated training record for the user profile update by adding the training records to the profile. The training record and the user profile are vectors with identical numbers of columns and rows.

A fifth aspect of the invention discloses a method for identifying content that is precisely targeted to a user using one or more actions. One embodiment of the method includes: creating a user profile using a vector; generating multiple content records based on metadata about a plurality content. The content records are expressed as vectors and have identical numbers of columns, rows, and columns. Providing access to the plurality content to the users; generating a weighted record of each user action with respect to individual content records; using the weighted record at minimum in part to create an updated user account; adding a dot product from individual content records to the scalar number; and identifying the individual content

“In one variant, the profile includes a pre-set user profile or template.”

“In another variant, the act generating the weighted contents record involves multiplying the content records associated with at least one act by an appropriate weighting factor.”

“In another variation, the act of identifying individual users of plurality content with a prescribed relationship to the user profile based at minimum in part on the scale quantity entails identifying individual users of plurality content having a scale quantity equal or greater than a predetermined amount.”

“In another variation, the method also includes displaying identifying data regarding the identified individual one of the plurality content having a prescribed relationship to the user profile.”

“Another aspect of the disclosure is a computerized system for recommending content to a specific user in a content delivery platform. One embodiment of the computerized method involves: receiving user data about the particular user; creating a plurality data records concerning user actions associated with that particular user; using the plurality data records to create a first data record; updating this first data record for subsequent user action data related to individual digitally rendered content elements; and, based at minimum on the update of the first data record, algorithmically creating data representative of a particular user’s user profile.

“A computerized network apparatus is also disclosed in another aspect of this disclosure. One embodiment of the invention is a computerized network apparatus that tracks interactions between a user and one or more content elements to generate recommendations for additional content elements to be delivered to the user via streaming content delivery transport. One variant of the computerized networking apparatus comprises: storage apparatus in data communications with a digital processor device and at most one computer program. The computerized network apparatus generates a first structure that includes first training data and the generation of the first structure. It then generates a second structure that is based on at least the second user interaction with the content elements. Finally, it generates an updated version the first structure based on at least the second data received.

“In another aspect, the present disclosure discloses computerized apparatus. One embodiment of the computerized apparatus is designed to recommend content to a user in a content distribution system. It includes: a digital processing apparatus in data communication with storage apparatus, adapted to run at most one computer program, the program being able to: use data representative of at minimum a portion of the plurality to generate a training record; update that training record for subsequent data relating user action to create an updated training track; and generate the recommended content to be sent to the user based at the very least on the updated

“A non-transitory computer readable apparatus is also disclosed in another aspect of this disclosure. One embodiment of the non-transitory computer readable apparatus comprises a storage medium. The storage medium contains at most one computer program with a plurality instructions. These instructions are configured to cause a computerized device to: receive user action information from a content delivery system; generate a plurality records about user actions; use the plurality to generate at minimum one training record; update the content data records associated with individual content elements to create an updated content data records; and use the at-least one updated training record to generate a user profile.

“Other features or advantages of the invention will be immediately recognised by persons of ordinary skill with reference to attached drawings and detailed description below of exemplary embodiments.”

“Refer to the drawings, where like numerals refer throughout to like parts.”

“Advertisement” is the term used herein. Similar forms and audiovisual messages, as well as any other communication that can be perceptible by humans, including for-profit or non-profit, are all examples of advertisements. Advertisements include the so-called “bumper” type. Advertisements (advertisements that are inserted before or following a client-requested program), ‘pause? Advertisements (presented by a client when they send a pause command to a server or similar), as well as additional and replacement advertisements.

“As used in this document, the term ‘application? A unit of executable code that implements a particular functionality or theme is generally referred to as an application. Applications can have many themes. They may be used in a variety of functions and disciplines (e.g., on-demand content management, brokerage transactions or home entertainment), etc. One application might have multiple themes. Executable software runs in a predetermined environment. For example, it could include a downloadable Java Xlet. JavaTV? environment.”

“Capacity” is the term used herein. “Capacity” refers to the ability of a network or portion of a system to perform a requested service, act or other performance. Bandwidth is a common measure of capacity. It is roughly equivalent to the size of a channel or ‘pipe. Capable of carrying content and other information. However, capacity limitations may be imposed by any number of factors, such as the unavailability of the content from a provider (e.g., studio or television network), delays imposed by transmission, filtering, transcoding, encryption/decryption, conditional access establishment and/or download (e.g., according to a ?DCAS? You can also download conditional access system paradigms, and so forth.

“As used herein the terms ‘client device? and ?end user device? These include but aren’t limited to set-top boxes (e.g. DSTBs), personal computer (PCs), and minicomputers. They can also be used on mobile devices like smartphones, PDAs and personal media devices (PMDs), as well as desktop computers, laptops, and other types of computers.

“Codec” is the term used herein. “Codec” refers to any video, audio, or data coding and/or encoding algorithm, process, or apparatus, including those of the MPEG (e.g. MPEG-1, MPEG-2 or MPEG-4). ), Real (RealVideo, etc. ), AC-3, DiVX/ViD, Windows Media Video (e.g. WMV 7, 8 or 9, ATI Video codec or VC-1 families (SMPTE standard 4221M).

“As used herein the term ‘computer program? “As used herein, the term?computer program? It can include any sequence of human- or machine-recognizable steps that performs a function. This program can be written in any programming environment, including C/C++ and Fortran. (including J2ME, Java Beans, etc. ), Binary Runtime Environment, (e.g. BREW), or the like.

“Consideration” is the term used herein. It refers to any payment, incentive, option or forbearance of any debt, credit or other thing or act that conveys monetary value between two or more people, such as cash or credit/debit payments or credits to account.

“Similarly, the terms ‘Consumer Premises equipment (CPE),? “Consumer Premises Equipment (CPE)” and “host device?” Any type of electronic equipment that is located in a user’s or consumer’s home and connected to a network. The term “host device” is used. The term “host device” generally refers to a terminal device with access to digital TV content via satellite, cable, and terrestrial networks. A digital television (DTV), set may include host device functionality. “Consumer premises equipment” is the term used. CPE includes electronic equipment like set-top boxes and televisions. It also includes digital video recorders (DVR), gateway storage device (Furnace) and ITV Personal Computers.

“Display” as used in this document means: Any type of device that can display information, such as CRTs and LCDs, TFTs or plasma displays, LEDs, fluorescent, incandescent, and other devices, is considered a display device. Display devices can also include non-dynamic devices like printers and e-ink, or the like.

“DOCSIS” is the term used herein. Any of the variants or plans of the Data Over Cable Services Interface Specification (DOCSIS) is referred to. DOCSIS (version 1.0), is a standard and protocol that allows internet access via a?digital’ network. DOCSIS 1.1 is a standard and protocol for internet access using a?digital? cable network. DOCSIS 1.1 can be interoperable with DOCSIS 1.0 and offers data rate and latency guarantees, as well as better security than DOCSIS 1.0. DOCSIS 2.0 can be interoperable with 1.0 or 1.1 but offers a wider upstream band (6.4 MHz) and new modulation formats, including TDMA/CDMA. It also offers symmetric services (upstream at 30 Mbps).

“DVR” is the abbreviation of digital video recorder. Digital video recorder is a general term that refers to any type or combination of recording mechanisms and/or software environments, located at the headend, user premises, or anywhere else where content can be selectively recalled and recorded over a network. These DVRs can be either dedicated or part of a multi-function or non-dedicated system.

“As used in this document, the term ‘headend? A networked system that is controlled by an operator (e.g. an MSO or multiple-systems operator) and distributes programming to MSO clients using client devices. This programming can include virtually any information source/receiver, including free-to-air TV channels and pay TV channels, interactive television, and the Internet. DSTBs can come in any configuration and be sold as retail devices. This means that consumers might not get their DSTBs exclusively from the MSO. It is possible that MSO networks will have clients devices from multiple vendors. These client devices will also have a wide range of hardware capabilities. Multiple regional headends could be located in different cities or may be in the same place.

“Integrated circuit (IC)” is the term used herein. Any type of device with any level of integration (including VLSI and LSI), and regardless of process or base material (including Si, SiGe and CMOS) is considered an integrated circuit (IC). ICs can include memory devices (e.g. DRAM/SRAM, EEPROM/Flash and ROM), digital processors as well as FPGAs. They also include ASICs. ADCs. DACs. Transceivers, memory controllers and any combination thereof.

“Internet” and “internet” are used herein. “Internet?” and “internet?” are interchangeable terms that refer to inter-networks, including the Internet. They are interchangeable to refer to internetworks, including the Internet.

“Memory” is used in this document. Any type of integrated circuit, or other storage device, that is adapted to storing digital data, including ROM, is included. PROM, EEPROM, DRAM, SDRAM, DDR/2 SDRAM, EDO/FPMS, RLDRAM, SRAM, ?flash? memory (e.g. NAND/NOR) and PSRAM.”

“As used herein the terms’microprocessor? “Microprocessor” and “digital processor?” These terms generally refer to all digital processing devices, including digital signal processors (DSPs), reduced instructions set computers (RISC), general purpose (CISC), processors, microprocessors and gate arrays (e.g. FPGAs), reconfigurable compute fabrics, array processors, secure processors, and application specific integrated circuits. These digital processors can be contained in a single IC die or distributed over multiple components.

“MSO” and “Multiple Systems Operator” are used herein. or?multiple system operator? Or?multi-systems operators? Refers to a cable, fiber-to-the home (FTTH), fiber at the curb (FTTC), or satellite provider with the infrastructure necessary to provide services over these mediums.

“Network” and “bearer network” are used herein. “Network” and “bearer network” are interchangeable terms. All types of telecommunications and data networks are included, including hybrid fiber coax networks (HFC), satellite networks, telco networks and data networks (including intranets, MANs and WANs), LANs and WLANs. These networks and portions of them may use any number of topologies, including ring, bus or star, loop, or other variations. ), transmission media (e.g., wired/RF cable, RF wireless, millimeter wave, optical, etc.) ), and/or communications protocols (e.g. SONET, DOCSIS or IEEE Std. 802.3, ATM, X.25, Frame Relay, 3GPP, 3GPP2, WAP, SIP, UDP, FTP, RTP/RTCP, H.323, etc.).”

“Network agent” and?network entity are used herein. “Network agent” and “network entity” are interchangeable. Any network entity, whether software, firmware, or hardware-based, that is adapted to serve a specific purpose. A network agent, or entity, may be a computer program that runs on a server of a network operator and is communicating with one or more processes on a CPE, or another device.

“Network interface” is the term used herein. Any signal, data, and/or software interface that is used with a component or network, including those of FireWire (e.g. FW400, FW800), etc. ), USB (e.g., USB2), Ethernet (e.g., 10/100, 10/100/1000 (Gigabit Ethernet), 10-Gig-E, etc. ), MoCA (e.g. USB2, Serial ATA (e.g. SATA, eSATA, SATAII), Ultra ATA/DMA (Gigabit Ethernet), 10-Gig-E, etc. ), radio frequency tuner, e.g. in-band, OOB, cable modem. ), WiFi (802.11a.b.g.n), WiMAX (802.26), PAN (802.15), and IrDA families.

“The term “node” as used herein refers to any location, functional entity or component within a network. “Node” can refer to any location, functional entity or component of a network without limitation.

“As used herein the term ‘on demand? “On demand” or ‘OD? Any service that allows real-time, quasi-real time (e.g. ?trick? ?trick? This content can be stored on a server or temporarily cached. It may also be streamed from a source.

“As used in this document, the term ‘QAM? Modulation schemes that transmit signals over cable networks are referred to as QAM. This modulation scheme can use any constellation level (e.g. QPSK, QAM-16, QAM-64, QAM-256 etc.) Dependent on the details of a cable system, QPSK, QAM-16 and QAM-64 are all possible. The QAM could also refer to a physical channel that is modulated in accordance with the schemes.

“Server” is the term used herein. Any computerized component, system, or entity that is capable of providing data, files or content to other devices or entities via a computer network, regardless of its form, is called a “server”.

“As used herein the term’service?, content?, or?program?” “Service”, “content”, and “program” are synonyms for “stream”. Sometimes, the terms?stream? and?service are used interchangeably to describe a sequence or packetized data that is delivered in what a subscriber might perceive as a service. A?service? A?service? (or?content? or?stream?). In the former, the specialized sense could refer to different types services in the latter, which is non-technical. A?service? could be an example. A?service? in its specialized sense could refer to, among other things, video broadcast, audio only broadcast, pay per view, or video on-demand. What is the perceivable content of such a’service? It could be live, prerecorded or delimited in duration, or may contain other descriptions. Sometimes, a “service” may be used in some cases. In some cases, a “service” may be used to refer to what a subscriber might call a “channel”. Traditional broadcast television.

“Service group” is the term used herein. A group of service users (e.g. Subscribers) or the resources they share in the form of whole cable RF signals, which only includes the RF channels that are used to receive the service.

“As used herein the terms’storage device? “storage device” and “storage media?” These terms refer to computer hard drives, DVR devices, memory, RAID arrays, memory, DVR devices, and optical media (e.g. CD-ROMs. Laserdiscs. Blu-Ray). Any other media or devices capable of storing data or content.

“As used herein the terms ‘user channel? “User channel” and “Program channel?” All of these terms are generally used interchangeably with the idea of a perceived stream information. For example, a program/user channel might comprise ?Channel 3? Which carries the content from a particular network (e.g. NBC). This must be distinguished from a physical cable channel, which is used to physically transport and distribute the content.

“User interface” is the term used herein. It can be visual, tactile, audible or sensory information that is provided to or received by a user or another entity.

“WiFi” is the term used herein. Any of the IEEE-Std. variants can be used without limitation. 802.11 and related standards, including 802.11a/b/g/n.

“Wireless” is used in this document. “Wireless” can be used to refer to any wireless signal, data or communication, including but not limited WiFi, Bluetooth, 3G and HSDPA/HSUPA TDMA, CDMA (e.g. IS-95A, WCDMA) ), FHSS. DSSS. GSM. PAN/802.15. WiMAX (802.16), 80.20.

“Overview”

“In one aspect, the invention discloses methods, apparatus, and processes for identifying and recommending content that is targeted at a specific user (or group) within a content-based network such as a cable TV or satellite network. The present invention allows for the selection of content that is most closely aligned with a viewer’s preferences. This mechanism does not need to be entered manually. The content that is provided to the user comes from many sources, such as DVR, broadcasts and VOD systems. This invention allows the user to learn and unlearn their preferences, as well as which content they will enjoy, based on the actions taken in relation to the content. In one embodiment, the recommended content is displayed as a table or list of titles with related information. Another alternative is to feed it to the user via a continuous stream of content on a virtual channel. Another embodiment presents the user with the compiled content in conjunction an electronic program guide (EPG), which can be customized to suit the user’s needs.

“Another aspect is that client applications are used to compile the playlist using user-imputed and pre-programmed user profiles. Another embodiment uses feedback mechanisms to allow the client application to “learn” The application can learn from user activities to generate finer recommendations.

“Another embodiment allows a user to establish a connection via the Internet to the client applications. The user can modify or create a playlist and remotely set up programs to record/erase from their recording device (DVR), etc. Optionally, the MSO or the user can modify the dynamics of the recommendation engine that generated the user’s playlist.

Methods and apparatus for dynamic content insertion (e.g. recommendation and/or insert of secondary content such advertisements, promotions and FVOD content) are described. These methods are described based on the user profile.

“Another aspect is that the client applications described above are implemented by a network entity such as a hub, headend server, adapted to perform these functions for various user profiles (e.g. individual user profiles or user accounts).

“The advantages of the invention include the ease with which the methods and apparatus can be implemented using existing infrastructure, i.e., software upgrades, which obviate the need for significant modifications or additional expense when implementing these capabilities.”

“An operational and/or business rules engine?” These rules can be used to implement various business or operational goals. They also provide methods for doing business.

“Exemplary embodiments and methods of this invention are now described in depth. These exemplary embodiments were described in relation to the hybrid fiber coax cable (HFC), which has a multi-system operator, digital networking capability and plurality of clients devices/CPE. However, the general principles of the invention can be extended to other networks and architectures such as wired, wireless, broadband, or content. The following description is only an example. The invention could be used over a fiber to-the home (FTTH), fiber-to the-curb system (FTTC), or satellite- or millimeter wave-based network.

“It will also be appreciated that while described generally in the context of a network providing service to a consumer (i.e., home) end user domain, the present invention may be readily adapted to other types of environments including, e.g., commercial/enterprise, and government/military applications. There are many other possible applications.

“Moreover, while the primary embodiments described herein describe predominantly the distribution of programs or similar content, other types of content including without limitation advertisements/promotions, instructional videos, or even data applications or files may likewise be distributed using the techniques of the present invention.”

“It should be noted that although aspects of the invention were described in relation to 6 MHz radio channels within the HFC network’s HFC network, the present invention can be applied to any frequency/bandwidth such as 8 MHz channels. Furthermore, although the invention is generally described as content delivery over discrete QAMs and RF channels, certain portions of it can be used with multiplexing algorithms and wideband tuner devices such as those described in U.S. patent application ser. No. No.

Summary for “Recommendation engine apparatus, methods”