Field of Application
The present invention generally relates to a recommender system for
enhancing overall viewing experience in internet display enabled devices,
like IPTVs. In particular it relates to a method for enhancing the overall user
viewing experience by way of sharing user's device settings associated with
the shared content in a collaborative filtering environment. In the present
invention the recommender system can also be a content based
recommender system. The Internet enabled devices include all networked
multimedia devices like IPTV, mobile phone, audio player and photo
viewer.
The invention provides a system that facilitates sharing of multimedia
content.
Background of the Invention
Recommender systems form a specific type of information filtering (IF)
technique that attempts to present information items (multimedia content in
case of the present invention) that are likely to be of interest to the user.
Recommender systems suggest useful and interesting multimedia content(s)
to users in order to increase their viewing satisfaction.
A collaborative recommendation system has been described in US
2006/0282856 Al.
Document WO 2002/100092 Al provides a method and system for adjusting
video and audio settings for media output devices.
Known recommender systems emphasize only on the quality of the
recommended contents, i.e. how well the content matches the user's liking.

User viewing experience is an aspect that has been neglected in currently
available recommender systems. In the known systems intrinsic qualities of
the multimedia content, e.g., the best suited volume, contrast, brightness,
color balance, etc., are overlooked.
There is therefore, a need for augmenting the capabilities of a recommender
system.
Summary of the Invention
The main object of the present invention is to enhance the capabilities of a
recommender system.
Another object of the invention is to provide a mechanism of capturing the
recommender peer's device settings associated with a particular content that
the user views and adding these setting details to the metadata of the
multimedia content.
Yet another object of the present invention is to add to the metadata the
intrinsic parameters such as various picture and sound settings associated
with a particular content as being viewed by that particular recommender
who has recommended this content to the user.
Metadata can generally be defined as the set of all descriptive elements
related to a particular content. In tfye currently available recommender
systems, metadata consists of extrinsic parameters such as genre, actors,
director and any other related keywords.

Multimedia content includes video content, audio content, still images and
text content.
The device settings include all possible intrinsic video content settings
applicable to the shared multimedia content like volume, bass, treble,
equalizer settings, surround sound, colour, contrast, brightness and tint, etc.
In the present invention the recommender based sharing of multimedia
content includes both solicited and unsolicited recommendations. It includes
sharing of multimedia content on social networks.
Thus, in a preferred embodiment the present invention provides a
recommender system for enhancing overall viewer experience in internet
enabled display devices like IPTVs, said system comprising a device settings
extractor for extracting viewers device settings, with all applicable
multimedia content settings, from the display device, and appending the
extracted device settings to the associated content metadata; a migration
engine for facilitating replacement of current user device settings by the
recommended device settings; a computation device for computing average
device settings from device settings received for a particular content; and a
display unit for displaying the computed average device settings.
The computed average device setting includes all standard acceptable data
sets of expected values, such as weighted mean, median, mode, etc.

BRIEF DESCRIPTION OF THE ACCOMPANYING DRAWINGS
The invention can now be described in detail with the help of the
accompanying drawings, where :
Fig 1: shows a collaborative filtering environment in a
recommender system.
Fig 2: shows an aggregation of device settings for a particular
content.
Fig 3: shows a possible set of device settings.
Fig 4: shows calculation of recommended attribute value.
Fig 5: shows transfer of device settings from a recommender to
• a user.
Fig 6: shows building of recommender list.
Fig 7: shows migration of device settings.
Fig 8: shows system block diagram.
Fig 9: shows flow diagram.
Fig 10: shows method flow diagram for sharing device settings.
DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
Referring to Fig 1, a user 101 interacts with a display device 102 and
consumes a variety of multimedia contents MC (recommended/searched).
These multimedia contents and their associated settings are fetched from the
recommenders (Rl to Rn of 103). Thus, recommenders send pairs consisting
of a particular multimedia content and its device settings as seen in 201, 202,
203 and 204 (Fig. 2). Each of such device settings may consist of several

attributes like various picture settings 302, sound settings 303 (Fig. 3).
There can be three modes in which this scheme can be implemented which
are described as follows :
The user gets device settings for a multimedia content from n best
recommenders i.e. the user can view all the n device settings (different
flavors of device settings) received and choose the best according to users
liking. The user may also adjust the device settings according to ones
preference. The user gets device settings for multimedia content from ones
best recommender, i.e. n =1. The user may also adjust the device settings
according to ones preference. The user gets a hybrid device setting for a
multimedia content (MC) as an average of all the device settings from all the
recommenders (or a pre-determined fixed number of recommenders). Each
setting can be assumed to be comprised of several attributes, e.g. picture
setting 302, sound setting 303.
For each attribute the recommended attribute value (RAV) 401 (Fig. 4) is
calculated using one or more of the aggregated attributes. The output setting
or the recommended device setting is a hybrid setting which is the aggregate
of all the RAVs. After the recommended device setting 703 is decided a
migration engine 702 replaces the current device setting 701 with the
recommended device setting 703 (Fig. 7).
In Figure 5, a recommender Rl consumes a particular multimedia content
and its device settings are extracted by a content and settings extractor 507
using a device interface 508. The collaborative filtering engine 506 then
shares the multimedia content(s) and its device setting via Internet 505 with

the target viewer. The collaborative filtering engine 506 in the target
viewer's system sends this information to a recommendation manager 504
which is then stored in local database. On request from the target user to
play a recommended content in the device 501, the content and associated
settings fetcher 503 retrieves the information from the database and presents
it to the user. As a user interacts with a display device 601, the user behavior
is captured into a user profile 602. The recommender list builder 603 finds
similar users (like 604) and builds up the recommender list.
In Figure 8, the block diagram of the system is depicted. The device
interface 801 is the interface to the various device related functionalities. All
the multimedia contents and their device settings, the user views are stored
in the accessed contents database 807 and the accessed contents settings 808
database respectively. It is used by the user behavior mapping engine 803 to
capture user behavior and generate the user profile, extract content metadata
and the device settings of the particular content. All user profile related
information is stored in the user profile database 806. It is also used by the
recommendation manager 804 to recommend multimedia contents and their
related device settings. The collaborative filtering engine 805 builds up the
recommender list and fetches recommendations from the recommenders of
the user through the internet 802. The recommender list is a collection of all
the recommenders and their information including their profiles. The
recommender list, stored in recommender list 809 database, is updated after
a fixed time interval which can be decided by the user or kept preconfigured.
The profiles of the recommenders are stored in the user profile database 806.
The user profile database 808 is updated every time the recommender list is
updated. All the contents fetched from the recommenders are stored in the
recommended contents 810 database and their associated device settings in

recommended contents settings 811 database.
In Figure 9, the flow diagram (conceptual) is shown. All the users are
connected via the Internet 901. The collaborative filtering engine 906 builds
the recommender list 909 by finding similar users or recommenders. Due to
the uncertainty of availability of users in the internet, the recommender list is
continuously updated at the end of a definite interval. Once the
recommender list is populated the collaborative filtering engine 906 fetches
the metadata and device settings of the accessed contents from the
recommenders. These multimedia contents along with their device settings
are then stored in the recommended contents database 905. As soon as the
user selects a recommended content for viewing by querying the device
interface 902 and the relevant device settings is selected, the CE Device is
migrated to the newly selected device settings. The selection of device
settings can be done by any of the three mentioned ways. The device
interface 902 is also responsible to provide interface to the user behavior
mapping engine 904 for the extraction of various user behavior
characteristics and populating the accessed contents 907 database. The
extracted user behavior characteristics are then used by the user profiling
engine 908 for populating the user profile database 910. Whenever a remote
user requests for recommendations the metadata and the device settings of
all/some multimedia contents from the accessed contents are sent to the
remote user.
In Figure 10, the steps in the process are shown. User viewing details are
captured by the device interface and the information is passed on to user
behavior mapping engine (USBE). USBE in turn extracts the content and
setting information and stores them in a database (DB). It also uses the

accessed content information to update the user profile DB. User profile
information is used by collaborative filtering engine (CBF) to find
recommenders by way of comparing user profiles and finding similar users.
When the user requests for recommendations, CBF retrieves multimedia
contents (MC) and associated settings from various recommenders in the
recommender list (RL). Once CBF has retrieved this information, it's the
recommendation manager's responsibility to display the multimedia contents
(MC) to the user along with the associated device settings (DS).

WE CLAIM :
1. A recommender system for enhancing overall viewer experience in
internet enabled display devices, like IPTVs, said system comprising :
- a device settings extractor for extracting viewers device settings, with
all applicable multimedia content settings, from the display device,
and appending the extracted device settings to the associated content
metadata;
- a migration engine for facilitating replacement of current user device
settings by the recommended device settings;
- a computation device for computing average device settings from the
recommended device settings received for a particular content; and
- a display unit for displaying the current/recommended/computed
average, device settings.

2. The system as claimed in claim 1, wherein said system is a content
based recommender system.
3. The system as claimed in claim 1, wherein said display device
comprises all networked multimedia devices like IPTV, mobile phone,
audio player and photo viewer.
4. The system as claimed in claim 1, wherein said multimedia content
comprises video content.
5. The system as claimed in claim 1, wherein said multimedia content
comprises audio content.
6. The system as claimed in claim 1, wherein said multimedia content

comprises still images.
7. The system as claimed in claim 1, wherein said multimedia content
comprises text content.
8. The system as claimed in claim 1, wherein said device settings
comprise all possible intrinsic video content settings applicable to the
shared multimedia content like volume, bass, treble, equalizer
settings, surround sound, colour, contrast, brightness and tint.
9. The system as claimed in claim 1, wherein the recommender-based
multimedia content sharing comprises both solicited as well as
unsolicited recommendations.
10. The system as claimed in claim 1, wherein the sharing of multimedia
content comprises sharing on social networks.

11. The system as claimed in claim 1, wherein said system is for
facilitating sharing of multimedia content.
12. The system as claimed in claim 1, wherein said average device
setting comprises all standard acceptable data sets of 'expected'
values like weighted mean, median and mode.
13. A method for enhancing overall viewing experience in internet
enabled devices, said method comprising the steps of:
- capturing user viewing information with the help of a device interface
and passing on information to a user behavior mapping engine
(USBE);
- extracting the content and setting information by the USBE and
storing them in a database (DB), and updating user profile database
and using the information by a collaborative filtering engine (CBE)
for finding out recommenders by comparing user profiles of similar
users;

getting recommending content and associated settings; and
displaying recommended content along with settings to the viewers.
14. A recommender system for enhancing overall viewer experience in
internet enabled display devices like IPTVs substantially as herein
described and illustrated in the accompanying drawings.

A recommender system for enhancing overall viewer experience in internet enabled display devices, like IPTVs, said system comprising :
- a device settings extractor for extracting viewers device settings, with all applicable multimedia content settings, from the display device, and appending the extracted device settings to the associated content
metadata;
- a migration engine for facilitating replacement of current user device
settings by the recommended device settings;
- a computation device for computing average device settings from the recommended device settings received for a particular content; and
- a display unit for displaying the current/recommended/computed average, device settings.