PATROLL Winning Submission

U.S. Patent No. 8,490,123

The patent-at-issue, U.S. Patent No. 8,490,123 (“Sisvel”), which was filed on May 26,
2004 and claims a June 3, 2003 priority date, is directed to “a method and device for generating a
user profile on the basis of playlists, the user profile is generated on the basis of the user's own
playlists and properties derived from them. In this way, more playlists are considered in the
generation of a user profile, thereby generating a more reliable user profile. Aspects taken into
account includes frequency of occurrence, creation date of the playlist, and relation between
content items (songs, video, etc.) in one or more playlists.”

WO2002017146A1 (“Schwartzbard”), filed on Jan. 8, 2001 and claims a Jan. 7, 2000

priority date relates to a media management system, apparatus and computer program product
that uses a dynamic profile to suggest uniquely characterized pieces of content. Part of the
content are identified and classified with attributes associated with the portions of the content. A
search method is employed to determine if the pieces of content have been cataloged. The
dynamic profile facilitates media management for use by a user of a client allows the client to
select services offered by service providers.

US 7,096,234 (“Plastina”), published on September 25, 2003 with a filing date of March
21, 2002, relates to a method of monitoring a user’s music playing experience and collecting
metadata to dynamically update the user’s music playing content. Plastina further discloses
collecting metadata from a user’s device based on certain parameters such as the user’s rating,
play count, and time played of particular songs. Plastina also discloses that upon collection of
said parameters, a “smart” playlist may be dynamically and automatically created and updated
for use and enjoyment by that specific content user’s account.

A sample claim chart comparing the Sisvel patent to Schwartzbard, in view of Plastina,
is provided below.

PATROLL – Prior Art Crowdsourcing

http://patroll.unifiedpatents.com
 US 8490123 (“Sisvel”) WO2002017146 (“Schwartzbard”) in view of
US 7,096,234 (“Plastina”)

1.pre. A method of generating a A. WO2002017146

user profile for a given user “A user that has a collection of music on a computer may
wish to build a profile. In this case, it is preferable to have
from at least one first accurate information regarding the actual music listened to
playlist by the user so a program may go through that user's
collection of music, create a signature for each song, and
including a first sequence of send that set of signatures to the server system so that the
content and being associated client need only send a very small amount of information to
with the given user and the server in order for the server system to look up each
stored on a media device, signature in its database and identify whether a particular
signature corresponds to a song or has already been
said method comprising: associated with attributes about that song that are saved in
the server system.” p. 32, par. 2, lines 1-8.

“Information and attributes regarding different songs

(content) is generally a manually intensive process that
involves accurately labeling collections according to artist,
song name, album the song is on, genre, and whatever
other information that may be of interest. That information
may then be sent back from the server system to the client so
that the user of the client automatically receives this
information for use as that particular person sees fit. Having
these attributes associated with the songs listened to by
the user, will help the system develop a dynamic profile
regarding that particular user's listening interest and

PATROLL – Prior Art Crowdsourcing

http://patroll.unifiedpatents.com
US 8490123 (“Sisvel”) WO2002017146 (“Schwartzbard”) in view of
US 7,096,234 (“Plastina”)

patterns.” p. 32, last par., lines 8-13 through p. 33, lines 1-2.

“The flowchart of Figure 11 describes a process employed in

the audio player when using a user-actuated dynamic profile
operation. Moreover, in the player, the user, under user
control may opt to select a dynamic profile to be created to
suggest additional songs to be played in a play list.” p. 38,
lines 3-6.

“The dynamic profile is more comprehensive, discussed

below, and is based on an association of subsets of past song
sequences listened to by that user.” p. 39, par. 2, last 3
lines.

B. US7096234
“The header portion of the file format encapsulates global
playlist metadata. In this example, this includes the title of
the playlist, the author of the playlist, category, genre, one or
more user names and one or more associated user ratings.”

“Various systems and methods described below permit a

user's content experience (e.g. music playing experience)
to be monitored and for metadata describing this
experience to be collected.

PATROLL – Prior Art Crowdsourcing

http://patroll.unifiedpatents.com
US 8490123 (“Sisvel”) WO2002017146 (“Schwartzbard”) in view of
US 7,096,234 (“Plastina”)

The metadata is developed by monitoring a number of

different parameters that are associated with the user's
content experience. This metadata can be dynamically
updated as a user experiences media content and then used to
impart to the user a personalized experience that is tailored to
that specific user. The metadata can be actually associated
with the specific media content that a user can play.”

1. One or more computer-readable media embodying a file

structure comprising: a first portion of the file structure
containing data that is associated with a user of a playlist;
and a second portion of the file structure containing data
associated with the playlist; the second portion comprising:
a static portion that contains a list of media content for the
playlist; and a dynamic portion that defines one or more
queries for dynamically updating the playlist; the queries
being executable to produce a result set that is provided into
the static portion to define the list of media content.

PATROLL – Prior Art Crowdsourcing

http://patroll.unifiedpatents.com
 US 8490123 (“Sisvel”) WO2002017146 (“Schwartzbard”) in view of
US 7,096,234 (“Plastina”)

1.a. automatically searching for the A. WO2002017146

at least one first playlist among a “Labels associated with the content may also be used for
plurality of playlists, searching purposes so that work that has been grouped-
together by others as being of a similar genre, for example,
wherein the plurality of may be identified by the user for future play-list inclusion.”
playlists includes at least one p. 4, par. 2, lines 1-3.
of a second playlist and a
third playlist, “A dynamic profile is developed for a particular user so that a
temporal pattern of activity by the user is developed to
form a baseline set of discrete events associated
developedwith a particular user. This baseline set of discrete
events may be used to support "agent" software processes
that are employed to automatically provide services to a
user without the user having to make a specific request for
the service.” p. 10, 2nd full par., lines 5-9.

“Information and attributes regarding different songs

(content) is generally a manually intensive process that
involves accurately labeling collections according to artist,
song name, album the song is on, genre, and whatever other
information that may be of interest. That information may
then be sent back from the server system to the client so
that the user of the client automatically receives this
information for use as that particular person sees fit.” p. 32,
last par., lines 8-13.

PATROLL – Prior Art Crowdsourcing

http://patroll.unifiedpatents.com
US 8490123 (“Sisvel”) WO2002017146 (“Schwartzbard”) in view of
US 7,096,234 (“Plastina”)

“The database server 1113 is connected to the database

1115 and processes requests to store information in and
retrieve information from the database 1115. The
database 1115 stores information that maps usernames
with various groups and maps various groups to
advertisement identifiers (ad IDs). Additionally, the database
1115 stores event logging information for each user. The
event logging information identifies the total length of
time each user is logged into the system as well as the total
length of time that the user listens to different broadcast
shows and advertisements.” p. 63, last par. lines 1-3 through
p. 64 par. 1, lines 1-4.

“In step 1921, the database server 1113 uses the

advertisement list table 1607 to identify the play list of ad
IDs corresponding to the group ID associated with the
client 1125 and sends the client 1125 the identified play
list of ad IDs.” p. 71, 2nd full par., lines 1-3.

“In a preferred embodiment of the invention, users are

assigned to user groups based on their user profiles. The
computer broadcast commercials are associated with the user
groups, and each computer is associated with the user group
of the user at the computer. … The present invention also

PATROLL – Prior Art Crowdsourcing

http://patroll.unifiedpatents.com
US 8490123 (“Sisvel”) WO2002017146 (“Schwartzbard”) in view of
US 7,096,234 (“Plastina”)

enables tracking of users' event logging information,

including the length of time that users listen to various
broadcasts and commercials.” p. 11, 2nd full par.

“Alternatively, the user of the client 3 may request a profile

from other persons, such as a celebrity, so that the user of
client 3 may have listening selections offered to that user,
and a play list that reflects that particular celebrity's
preferences, for example.” p. 23, 1st full par., last four lines.

--------------------------------

B. US7096234
In addition to using one or more filters to execute queries
on local resources, system 1200 can utilize one or more filters
for executing queries on remote databases. For example,
filters 1210 can be used to enable the playlist engine to
query remote database 1212.

The result of the queries by the playlist engine is a playlist

such as the one shown at 1214. This playlist can be rendered
via a suitable user interface (such as user interface 1100 (FIG.
11)).

(Note: Fig. 11 below shows “+ My Playlists” under the “All

PATROLL – Prior Art Crowdsourcing

http://patroll.unifiedpatents.com
 US 8490123 (“Sisvel”) WO2002017146 (“Schwartzbard”) in view of
US 7,096,234 (“Plastina”)

Audio” tab.)

"1.c. wherein the second playlist has A. WO2002017146

PATROLL – Prior Art Crowdsourcing

http://patroll.unifiedpatents.com
 US 8490123 (“Sisvel”) WO2002017146 (“Schwartzbard”) in view of
US 7,096,234 (“Plastina”)

a second sequence of content and is “Figure 19 is directed to an embodiment which the profiles
associated with a different user of different users, that were developed from different users
may be exchanged or "swapped" between different users.”
p. 53, 1st full par., lines 1-2.

“In a preferred embodiment of the invention, users are

assigned to user groups based on their user profiles. The
computer broadcast commercials are associated with the user
groups, and each computer is associated with the user group
of the user at the computer. … The present invention also
enables tracking of users' event logging information,
including the length of time that users listen to various
broadcasts and commercials.” p. 11, 2nd full par.

“Alternatively, the user of the client 3 may request a profile

from other persons, such as a celebrity, so that the user of
client 3 may have listening selections offered to that user,
and a play list that reflects that particular celebrity's
preferences, for example.” p. 23, 1st full par., last four lines.

PATROLL – Prior Art Crowdsourcing

http://patroll.unifiedpatents.com
 US 8490123 (“Sisvel”) WO2002017146 (“Schwartzbard”) in view of
US 7,096,234 (“Plastina”)

1.d. and the third playlist has a A. WO2002017146

third sequence of content and is “The services, are not limited to music distribution but may
associated with the given user, also relate to video distribution, ordering of product from
different service providers (both online as well as offline).
Note: The first playlist is also Since the transaction profile includes not only events
associated with the given user. regarding financial transactions, and music listening
patterns, but also sensory information.” p. 51, lines 3-6.

“The following appendices are attached hereto and should be

construed as part of the
^ present disclosure:
…
3. Agrawal, R., "Mining Sequential Patterns".
….” p. 2, lines 1-11.

“A next article is Ramakrishnan, S., et al, "Mining

Generalized Association Rules". In this paper, association
rules are discussed. Association rules have do not have any
mechanism for dealing with the temporal component of
data.” p. 40, lines 9-12.

------------------------

Excerpts from Agrawal, R., "Mining Sequential Patterns"

PATROLL – Prior Art Crowdsourcing

http://patroll.unifiedpatents.com
US 8490123 (“Sisvel”) WO2002017146 (“Schwartzbard”) in view of
US 7,096,234 (“Plastina”)

“To evaluate the performance of the algorithms over a large

range of data characteristics, we generated synthetic customer
transactions. These transactions mimic the transactions in the
retailing environment. In our model of the “real" world,
people buy sequences of sets of items. Each such sequence
of item sets is potentially a maximal large sequence. An
example of such a sequence might be sheets and pillow
cases, followed by a comforter, followed by shams and
ruffles.” p. 15, Sect. 4.1, lines 1-5.

“For instance, some people might buy only sheets and pillow
cases followed by a comforter, and some only comforters. A
customer-sequence may contain more than one such
sequence. For example, a customer might place an order
for a dress and jacket when ordering sheets and pillow
cases, where the dress and jacket together form part of
another sequence.” p. 15, Sect. 4.1, lines 6-8.

“We then assign items to the transactions of the customer.

Each customer is assigned a series of potentially large
sequences. If the large sequence on hand does not fit in
the customer-sequence, the itemset is put in the customer-
sequence anyway in half the cases, and the itemset is moved
to the next customer-sequence the rest of the cases.” p. 16,
1st full par.

PATROLL – Prior Art Crowdsourcing

http://patroll.unifiedpatents.com
US 8490123 (“Sisvel”) WO2002017146 (“Schwartzbard”) in view of
US 7,096,234 (“Plastina”)

“Extension of the algorithms to discover sequential

patterns across item categories. An example of
such a category is that a dish washer is a kitchen
appliance is a heavy electric appliance, etc.

Transposition of constraints into the discovery

algorithms. There could be item constraints (e.g.
sequential patterns involving home appliances) or
time constraints (e.g. the elements of the patterns
should come from transactions that are at least d1 and
at most d2 days apart.”

p. 21 ,1st full par.

--------------------------

PATROLL – Prior Art Crowdsourcing

http://patroll.unifiedpatents.com
 US 8490123 (“Sisvel”) WO2002017146 (“Schwartzbard”) in view of
US 7,096,234 (“Plastina”)

Figure 1, Ramakrishnan, S., et al, "Mining Generalized

Association Rules"

1.e. and each playlist of the plurality A. WO2002017146

of playlists “Each content is labeled with a conventional label such as a
including at least one name of the piece of content (e.g., the title of a song, for
identifying characteristic of example). Associated with the content may be other

PATROLL – Prior Art Crowdsourcing

http://patroll.unifiedpatents.com
US 8490123 (“Sisvel”) WO2002017146 (“Schwartzbard”) in view of
US 7,096,234 (“Plastina”)

content stored on the media attributes associated with that content, such as the genre,
device; artist, producer, etc. These labels are intended to be
unique so that each content may be uniquely identified,
provided that the user knows what work the user is interested
in receiving.” p. 3, last par. last 4 lines through p. 4, line 1.

“Another feature of the present invention is that once stored,

attributes associated with a particular content are
permanently and reliably attached to that content.” p. 10,
1st full par., lines 1-2.

“Having these attributes associated with the songs listened

to by the user, will help the system develop a dynamic
profile regarding that particular user's listening interest
and patterns.” p. 32, last line through p. 33, lines 1-2.

“After all the transactions have been identified by making

sequential passes, of varying window sizes, over the
transaction data, the process proceeds to step S97. In step
S97, every one of the discrete events is expanded
according to the attributes associated with that event,
based on a predetermined taxonomy. This taxonomy
describes the attributes of every event and attributes of
those attributes etc. For example, in the context of the
preferred embodiment, the taxonomy may be that for a

PATROLL – Prior Art Crowdsourcing

http://patroll.unifiedpatents.com
 US 8490123 (“Sisvel”) WO2002017146 (“Schwartzbard”) in view of
US 7,096,234 (“Plastina”)

certain song X there may be a certain artist, as well as a

certain album on which the song appears, and the song
also having a certain genre.” p. 41, last par., lines 1-6
through p. 42, line 1.

1.f. analyzing the at least one first A. WO200201714

playlist and automatically deriving “The static profile is based on a probability analysis of the
from the at least one analyzed first user's overall perceived listening habits, in relationship to
playlist attributes of songs included on a play list, or in a larger
at least one playlist feature data base of content available for download, perhaps from the
expressing at least one server system. The dynamic profile is more comprehensive,
property of the at least one discussed below, and is based on an association of subsets
first playlist, of past song sequences listened to by that user.” p. 39, 1st
full par., last 5 lines.

“These transactions may relate to a preferred set of music

listened to by the user or other events (such as purchased
items) that have been recorded in the transaction table.”
p. 49, 1st full par., lines 3-6.

PATROLL – Prior Art Crowdsourcing

http://patroll.unifiedpatents.com
 US 8490123 (“Sisvel”) WO2002017146 (“Schwartzbard”) in view of
US 7,096,234 (“Plastina”)

“The profile 231 , as will be discussed in more detail herein,

is a user-unique association of discrete events that are used
to predict future events, based on a user's present and past
activities. For example, in the audio player embodiment, the
profile 231 includes an organized set of information
regarding the audio files and an association between
audio files that are associated with the user's listening
habits, and other events performed by the user, or
environmental events regarding a state in which the files were
used by the user.” p. 22, 2nd full par.

1.g. the at least one playlist feature A. WO2002017146

comprising an occurrence “The dynamic profile mechanism and process are based
frequency or at least a content on an understanding of association of rules.” p. 39, 2nd full
relationship of the plurality of par., lines 1-2.
playlists; and
“A particular example of an association rate developed
may look like the following:
[0.6153 0.0516] genre = pop genre = rock album = the fury of
the aquabats! — > genre = ska.

PATROLL – Prior Art Crowdsourcing

http://patroll.unifiedpatents.com
US 8490123 (“Sisvel”) WO2002017146 (“Schwartzbard”) in view of
US 7,096,234 (“Plastina”)

This association rale would indicate that 5.16% of the

time, the user listened to a set of songs that contained the
attributes genre = pop, genre = rock, and the album = the
fury of the aquabats!. Then, as a consequential action, the
user listened to something with the attribute genre = ska.
Furthermore, this association rule indicates that 61.53% of
the time, that the user listened to a set of songs that
contained the attributes genre = pop, genre = rock, and
album = the fury of the aquabats!, the next song the user
listened to had the attribute genre = ska.

Another example of an association rule may look like

[0.5555 0.0322] group = Sara McLachlan group =
Tori Amos = = < = genre = pop.”
p. 42, last 12 lines through p. 43, lines 1-2.

“For example, a supermarket may have built a support set

over many transactions for a particular customer. The
supermarket may find that this customer purchases
whole milk in 20% of the purchaser's transactions, buys
milk in 25% of its transactions, and buys dairy products
in 50% of his transactions. If the supermarket used this data
to make special offers to the customer, the supermarket may
make special offers on dairy products 50% of the time, it may
make special offers on milk an additional 25% of the time

PATROLL – Prior Art Crowdsourcing

http://patroll.unifiedpatents.com
 US 8490123 (“Sisvel”) WO2002017146 (“Schwartzbard”) in view of
US 7,096,234 (“Plastina”)

and it may make special offers on whole milk an additional

20% of the time.” p. 46, 1st full par., lines 2-9.

“Now, it is possible to determine the independent supports of

each item set; the support of an item set is the frequency
with which it appears in the transaction data while the
independent support is the support of an item set independent
of other item sets.” p. 46, last par., last 3 lines through p. 47,
line 1.

1.h. automatically generating a user A. WO2002017146

profile for the given user based on “The static profile is based on a probability analysis of the
the analyzed at least one first user's overall perceived listening habits, in relationship to
playlist and the derived at least one attributes of songs included on a play list, or in a larger
playlist feature; data base of content available for download, perhaps from the
server system. The dynamic profile is more comprehensive,
discussed below, and is based on an association of subsets
of past song sequences listened to by that user.” p. 39, 1st
full par., last 5 lines.

PATROLL – Prior Art Crowdsourcing

http://patroll.unifiedpatents.com
 US 8490123 (“Sisvel”) WO2002017146 (“Schwartzbard”) in view of
US 7,096,234 (“Plastina”)

“For example, in the audio player embodiment, the profile

231 includes an organized set of information regarding
the audio files and an association between audio files that
are associated with the user's listening habits, and other
events performed by the user, or environmental events
regarding a state in which the files were used by the user.” 2nd
full par., lines 3-6.

1.i. wherein at least one of the said A. WO2002017146

searching, analyzing, and 1. A media management system, comprising:
generating comprises use of (a) a server connected to a network wherein said network
computerized hardware including a supports communication between said server and at least one
processing element. client processor;
(b) said server to respond to commands from at least one of
said client processors and supply digital data to said client
processor through said network;
(c) at least one client processor including a decoder to
decode encoded discrete digital data messages, an
evaluator to create a value corresponding to the digital
data, and an assignor that associates a value to digital data;

PATROLL – Prior Art Crowdsourcing

http://patroll.unifiedpatents.com
US 8490123 (“Sisvel”) WO2002017146 (“Schwartzbard”) in view of
US 7,096,234 (“Plastina”)

and,
(d) a transceiver controlled by said client processor that sends
and receives messages from the network for evaluation by
said client processor.
claim 1.

15. A media input/output device, comprising :

(a) an input that is electrically coupled to a channel;
(b) a storage that receives digital data from the input;
(c) the storage having capacity to maintain digital data for
timed retrieval by processor;
(d) a processor that is enabled to retrieve data from
storage and includes an examiner that calculates a value
of said digital data wherein said examiner enables the
confirmation of the integrity of said digital data;
(e) said processor controlling the location of said digital data;
and (f) an output coupled to said processor or storage for
converting said digital data in human perceptible form.
claim 15.

PATROLL – Prior Art Crowdsourcing

http://patroll.unifiedpatents.com