Ex Parte Paul et al

Patent Trial and Appeal Board

Aug 24, 2017

12195246 (P.T.A.B. Aug. 24, 2017)

United States Patent and Trademark Office
UNITED STATES DEPARTMENT OF COMMERCE
United States Patent and Trademark Office
Address: COMMISSIONER FOR PATENTS
P.O.Box 1450
Alexandria, Virginia 22313-1450
www.uspto.gov
APPLICATION NO. FILING DATE FIRST NAMED INVENTOR ATTORNEY DOCKET NO. CONFIRMATION NO.
12/195,246 08/20/2008 Deirdre Paul ATT/2007-0941 8127
83811 7590 08/24/2017
AT & T LEGAL DEPARTMENT - TRBK
PATENT DOCKETING
ROOM 2A-212, ONE AT& T WAY
BEDMINSTER, NJ 07921
EXAMINER
YOUNG, ASHLEY YA-SHEH
ART UNIT PAPER NUMBER
3623
MAIL DATE DELIVERY MODE
08/24/2017 PAPER
Please find below and/or attached an Office communication concerning this application or proceeding.
The time period for reply, if any, is set in the attached communication.
PTOL-90A (Rev. 04/07)
UNITED STATES PATENT AND TRADEMARK OFFICE
BEFORE THE PATENT TRIAL AND APPEAL BOARD
Ex parte DEIRDRE PAUL and CHRISTOPHER VOLINSKY
Appeal 2015-006277
Application 12/195,246
Technology Center 3600
Before ANTON W. FETTING, JOSEPH A. FISCHETTI, and
ROBERT J. SILVERMAN, Administrative Patent Judges.
FETTING, Administrative Patent Judge.
DECISION ON APPEAL
STATEMENT OF THE CASE1
Deirdre Paul and Christopher Volinsky (Appellants) seek review under
35 U.S.C. § 134 of a final rejection of claims 1—5, 9-14, and 18—20, the only
claims pending in the application on appeal. We have jurisdiction over the
appeal pursuant to 35 U.S.C. § 6(b).
The Appellants invented a method for identifying one or more
influences within a social network. Specification para. 2.
1 Our decision will make reference to the Appellants’ Appeal Brief (“Br.,”
filed June 9, 2014) and the Examiner’s Answer (“Ans.,” mailed July 1,
2014), and Final Action (“Final Act.,” mailed January 8, 2014).
Appeal 2015-006277
Application 12/195,246
An understanding of the invention can be derived from a reading of
exemplary claim 1, which is reproduced below (bracketed matter and some
paragraphing added).
1. A method for identifying an influencer, comprising:
[1] determining, via a processor,
a social network for each of a plurality of customers
based upon a group of users who have exchanged
communications with each of the plurality of customers,
wherein the determining the social network for each of the
plurality of customers comprises:
[la] defining a neighborhood for each of the plurality of
customers;
and
[lb] determining a strength of relationships of each of the
plurality of customers with members of the neighborhood
of the customer;
[2] selecting, via the processor,
an attribute associated with communications among the
plurality of customers to be used for predicting a measure
of influence for each of the plurality of customers,
wherein the attribute comprises a willingness of a
customer to adopt new services;
[3] building, via the processor,
a model based upon
the attribute associated with the
communications among the plurality of
customers that is selected,
the social network of each of the plurality of
customers
and
a sequence of product adoption events of
each of the plurality of customers
2
Appeal 2015-006277
Application 12/195,246
for predicting the measure of influence for each of
the plurality of customers,
wherein the model comprises a statistical model,
wherein the statistical model comprises a linear
regression model,
wherein the sequence of product adoption events includes
each of the plurality of customers first adopting a
new service
and subsequently
users in the group of users who have exchanged
communications with each of the plurality of
customers also adopting the new service;
and
[6] determining, via the processor,
the influencer in the social network by applying the
model.
The Examiner relies upon the following prior art:
Zhuang US 2008/0070209 A1 Mar. 20, 2008
Trusov, Your Members Are Also Your Customers: Marketing For
Internet Social Networks, 9/29/06
Claims 19 and 20 stand rejected under 35 U.S.C. § 112, second
paragraph, as failing to particularly point out and distinctly claim the
invention.
Claims 1—5, 9-14, and 18—20 stand rejected under 35 U.S.C. § 103(a) as
unpatentable over Zhuang and Trusov.
3
Appeal 2015-006277
Application 12/195,246
ISSUES
The issues of definiteness turn primarily on whether breadth of scope
indicates indefiniteness. The issues of obviousness turn primarily on
whether the references describe the recited limitations within the scope of
their breadth.
FACTS PERTINENT TO THE ISSUES
The following enumerated Findings of Fact (FF) are believed to be
supported by a preponderance of the evidence.
Facts Related to Claim Construction
01. The disclosure contains no lexicographic definition of
“adoption.”
02. The ordinary meaning of adoption is to take on.2
Facts Related to the Prior Art
Zhuang
03. Zhuang is directed to viral marketing. Zhuang para. 1.
04. Zhuang’s influential persons identification system and method
includes identifying a set of influential persons (or influencers) in
a social network. The influential persons set is a set of such that
an advertiser can use the set to conduct a viral marketing
campaign with maximum return at a minimal cost. Specifically,
2 American Heritage Dictionary
https: //www. ahdictionary. com/ word/ search.html? q=adopt
4
Appeal 2015-006277
Application 12/195,246
by sending the marketing message to those contained in the
influential persons set, the advertiser achieves large coverage of
the social network in a short amount of time. Zhuang para. 7.
05. Zhuang’s influential persons identification system processes a
list of users in a social network and, based upon links and
interaction with other users on the network, identifies the top
influencers on the network. The influential persons identification
system includes a characteristics ranking module that ranking the
network users based on a certain criteria. The criteria include the
number of friends a user has, a certain topic that the user may be
interested in, as determined, for example, by the user's blog entries
and chat sessions, and the interests and demographics of the users.
Zhuang para. 9.
06. Zhuang’s influential persons identification system processes a
list of users in a social network and outputs an influential persons
set. Input to the influential persons identification system includes
the list of users in a social network. Typically, the social network
is an online social network community. Zhuang para. 21.
07. Input to the influential persons identification system also
includes a set number. This set number corresponds to the
number of persons (or the size) of an influential persons set.
Zhuang para. 22.
08. Zhuang’s influential persons identification system includes a
characteristics ranking module for ranking users contained in the
list of users in a social network. The output is a ranking of the
5
Appeal 2015-006277
Application 12/195,246
users in the social network. This ranking list is processed by a
selection module and the first-ranked user on the ranking list is
selected for inclusion in the influential persons set. Zhuang para.
23.
09. Zhuang’s influential persons identification method selects M
number of influences from the users of a large-scale social
network. The influential persons identification method assumes
that if user A is directly connected with user B in any way, then
they are friends of each other. The influential persons
identification method seeks to identify a set of influences that
maximizes coverage of the social network. In each embodiment
and at each step of the method, the user is selected having the
maximum number of friends that are not covered by previously
selected influences. Zhuang para. 26.
Trusov
10. Trusov is directed to marketing for internet social networks.
Trusov Title.
11. Trusov describes a model which allows social networking site
managers to identify community members whose network
influence makes them important for the business. Trusov links
individual user activity to the online behavior of other users in the
network. Using Bayesian methods and a variable selection
approach, Trusov infers which users in the community are
"influential" in affecting the site usage of other members. The
6
Appeal 2015-006277
Application 12/195,246
developed method is specifically tailored to the variable selection
task with a very large number of predictors. Trusov 1.
12. Any two users can create a connection between their profiles by
exchanging electronic invitations. For the most part, users are
involved in two kinds of activities on the site: they either create
new content by editing their profiles (adding pictures, uploading
music, writing blogs and messages, etc.), or they browse through
profiles consuming content created by others (looking at pictures,
downloading music, reading blogs and messages, etc.).
Essentially, popular social networking sites are simply massive
collections of continuously changing profiles and emerging
connections. Trusov 2—3.
13. A social networking site is typically initiated by a small group
of founders who send out invitations to join the site to the
members of their own personal networks. In turn, new members
send invitations to their personal networks, and so on. Trusov 6.
14. To identify the influential friends within an ego-centered
network, we model a user's login activity as a function of the
user's characteristics, the user's past behavior on the site and, most
importantly, the login activity of the user's friends. Trusov 13.
15. Trusov applies Poisson regression to create its model.
Trusov 13.
7
Appeal 2015-006277
Application 12/195,246
ANALYSIS
Claims 19 and 20 rejected under 35 U.S.C. § 112, second paragraph, as
failing to particularly point out and distinctly claim the invention
We are persuaded by Appellants' argument that the nature of the
computer-readable medium is not an issue of indefiniteness, but rather one
of claim scope. App. Br. 6. See Application of Miller, 441 F.2d 689, 693
(CCPA 1971) (“breadth is not to be equated with indefiniteness”). The
Examiner finds that the claim encompasses both structural and signal
embodiments and that a signal is not a physical structure. Ans. 15. Again,
this is a matter of breadth of scope. One of ordinary skill understands the
scope.
Claims 1—5, 9—14, and 18—20 rejected under 35 U.S.C. § 103(a) as
unpatentable over Zhuang and Trusov
We are not persuaded by Appellants' argument that
none of the cited portions of Trusov describes a sequence of
product adoption events that includes each of the plurality of
customers first adopting a new service and subsequently users
in the group of users who have exchanged communications
with each of the plurality of customers also adopting the new
service, as recited in the present claims.
App. Br. 18. The key to understanding the Examiner’s finding is in seeing
that the Examiner is equating the creation of content in Trusov with the
product adoption events recited in the claims. Ans. 16. Trusov describes
sending invitations to a first group, some of whom adopt the service, and
who in turn send invitations to other users, again, some of whom adopt the
service. While this message as product is self-referential, nothing in the
8
Appeal 2015-006277
Application 12/195,246
claim excludes this as a form of product adoption. Further, even once a user
creates a profile, the user adds individual items of content to that profile.
Each such addition is an adoption of the use of the next item of content as
product to the profile, and is therefore a product adoption event.
We are not persuaded by Appellants' argument that in Trusov, there is no
tracking of a first user or any subsequent users' adoption of a new service.
App. Br. 18—19. Trusov tracks this by way of logins. The claim does not
recite or narrow the manner in which the model is based on such events.
We are not persuaded by Appellants' argument that
the term "sequence of product adoption events" is further
defined in the claims as including "each of the plurality of
customers first adopting a new service and subsequently users
in the group of users who have exchanged communications
with each of the plurality of customers also adopting the new
service". As such, the Examiner's broad interpretation of the
term "sequence of product adoption events" also ignores this
further definition that appears in the claims. The present
rejection also convolutes "adoption of a new service" with
"increased usage of a certain product or service". (See Final
Office Action p. 4, Response to Arguments). For example, the
Examiner suggests that "the product usage of Trusov is
analogous to the product adoption of the instant application.
The two are analogous as the ultimate end goal is the same in
that users deemed as influential have the ability to sway other
users they communicate with to act/react in a similar manner,
whether it be higher usage rates for an existing service, or
adopting a new product altogether." (See Final Office Action p.
4, emphasis added). However, the "product usage" alleged to
be found in Trusov is usage of a social network by a user who
is already a participant in the social network. It is simply not a
new service, nor is it a new service that is adopted. Nor is there
a measure of a willingness to adopt new services to be found in
a user simply increasing usage of a social network.
9
Appeal 2015-006277
Application 12/195,246
App. Br. 19-20. The Specification, at 15 pages of 62 paragraphs, does not
bulge with disclosure. There is no guidance provided as to the metes and
bounds of what constitutes newness or adoption. A newly added item of
content is a new product. Adding an item is taking it on, which is the plain
meaning of adoption. Using an existing service in a new period of time is a
service new to that time period. Using the service in the new time period is
taking it on in that time period.
To say that there is no measure of a willingness to adopt new services to
be found in a user simply increasing usage of a social network is entirely
conclusory. Willingness is in the mind of the originator, not even the
beholder, and so any evidence of willingness is an attribute indicating
willingness. Trusov uses logins as an indication of willingness to continue
to add new content and otherwise use the services.
We are not persuaded by Appellants' argument that
Specifically, the linear regression model of the present claims 1,
10 and 19 for predicting the measure of influence for each of
the plurality of customer accounts for three parameters of (1)
the attribute associated with the communications among the
plurality of customers that is selected, (2) the social network of
each of the plurality of customers and (3) a sequence of product
adoption events. While, the Examiner cites Trusov with respect
to a "Poisson regression model" and "Bayesian" methods, the
rejection fails to describe a linear regression model that
accounts for the above three variables/parameters.
App. Br. 21.
First, Trusov applies Poisson regression because the variables it tracks
are integers and not continuous. One of ordinary skill knew that the choice
between linear regression and Poisson regression was a matter of matching
10
Appeal 2015-006277
Application 12/195,246
the method to the input characteristics,3 and so it was predictable to use
linear regression with a different input tracked. The claim does not recite
three parameters as being linear regression parameters, but only being
somehow used as a basis for the recited model. The claim does not recite
how the parameters are used as such a basis. Trusov bases its model on the
logins (attribute) to the social network of the users as the users login to adopt
by registering or by adding new content or by using in the new time period.
While these may be broad interpretations, the lack of circumscribing
disclosure allows for such breadth.
As to separately argued claims 2 and 9 reciting providing an action to be
taken for the influencer and that action being a directing marketing action,
the Examiner finds that all three references describe the need for acting so as
to keep influencers. Final Act. 12. Appellants rely on their arguments in
claim 1 that the art never finds an influencer. App. Br. 24. We found in
claim 1 that Trusov’s model finds influencers based on its model.
As to separately argued claim 3, reciting truncating the neighborhood for
each of the plurality of customers by selecting k neighbors, we agree with
Appellants that the Examiner presents no findings to support this. Claims 4
and 5 depend from claim 3.
Claims 10 and 19 are apparatus and software variants of claim 1. Claims
11, 18, and 20 are similarly variants of claims 2 and 9. Claims 13—14 are
variants of claims 3—5.
3 See, for example, a tutorial on the comparison between linear and Poisson
regression http://www.lisa.stat.vt.edu/sites/default/files/Poisson.and
_.Logistic.Regression.pdf
11
Appeal 2015-006277
Application 12/195,246
CONCLUSIONS OF LAW
The rejection of claims 19 and 20 under 35 U.S.C. § 112, second
paragraph, as failing to particularly point out and distinctly claim the
invention is improper.
The rejection of claims 1, 2, 9—11, and 18—20 under 35 U.S.C. § 103(a)
as unpatentable over Zhuang and Trusov is proper.
The rejection of claims 3—5 and 12—14 under 35 U.S.C. § 103(a) as
unpatentable over Zhuang and Trusov is improper.
DECISION
The rejection of claims 1, 2, 9-11, and 18—20 is affirmed.
The rejection of claims 3—5 and 12—14 is reversed.
No time period for taking any subsequent action in connection with this
appeal may be extended under 37 C.F.R. § 1.136(a). See 37 C.F.R.
§ 1.136(a)(l)(iv) (2011).
AFFIRMED-IN-PART
12