Ex Parte Yu et al

Patent Trial and Appeal Board

Dec 20, 2012

10891989 (P.T.A.B. Dec. 20, 2012)

UNITED STATES PATENT AND TRADEMARK OFFICE
________________

BEFORE THE PATENT TRIAL AND APPEAL BOARD
________________

Ex parte QINGFENG YU, CHOUDUR K. LAKSHMINARAYAN, and
ALAN BENSON
________________

Appeal 2010-007776
Application 10/891,989
Technology Center 2400
________________

Before KALYAN K. DESHPANDE, JASON V. MORGAN, and
MICHAEL J. STRAUSS, Administrative Patent Judges.

MORGAN, Administrative Patent Judge.

DECISION ON APPEAL

Appeal 2010-007776
Application 10/891,989

2
STATEMENT OF THE CASE
Introduction
This is an appeal under 35 U.S.C. § 134(a) from the Examiner’s Final
Rejection of claims 1 – 11, 16 – 20, 22, and 23. Claims 12 – 15, 21, and 24
have been withdrawn or canceled. App. Br. 2. We have jurisdiction under 35
U.S.C. § 6(b)(1).
We affirm.
Invention1
The invention relates to a method and system for site path evaluation
using web session clustering. Data is acquired from a plurality of sessions
corresponding to at least a portion of the plurality of web page traversal
paths. Portions of the web site traversal paths are grouped into a unified web
page category. The plurality of sessions is clustered into a plurality of web
session clusters according to a similarity measure. See Abstract.
Exemplary Claims (Emphases Added)
1. A method for identifying properties of a plurality of web
page traversal paths, comprising:
acquiring data from a plurality of sessions corresponding
to said plurality of web page traversal paths;
grouping said web page traversal paths into web page
categories;

1 Appellants filed an Amendment after Notice of Appeal relating to the
disclosure of a computer-executable medium in paragraph 13 of the
Specification. See Amend. after Notice of App. 2 (May 27, 2009). While the
Examiner has not considered this amendment, our consideration of the
appealed rejection is not affected by the amendment. Therefore, in the
interest of compact prosecution, we will decide the appeal rather than
remand the case to Examiner to consider the amendment.
Appeal 2010-007776
Application 10/891,989

3
using the web page categories to map the plurality of
sessions to new sessions;
clustering said new sessions according to a similarity
measure into a plurality of web session clusters; and
selecting one of said plurality of web session clusters
most closely exhibiting at least one predefined metric from said
plurality of web session clusters for analysis of properties of a
web page traversal path contained therein.
4. The method of claim 3, wherein said tagging [of each web
page of said plurality of web page traversal paths] includes
embedding a Java script fragment into each of said plurality of
web pages.
10. The method of claim 1, wherein selecting one or more of
said plurality of web session clusters, comprises:
extracting an open sequence as a representative
sequence from each of said web session clusters; and
analyzing properties of a web page traversal path of said
open sequence of each of said web session clusters.
Rejection
The Examiner rejects claims 1 – 11, 16 – 20, 22, and 23 under 35
U.S.C. § 102(b) as being anticipated by Jaideep Srivastava et al., Web Usage
Mining: Discovery and Applications of Usage Patterns from Web Data,
SIGKDD Explorations, Vol. 1, Issue 2, pp. 12 – 23 (“Srivastava”). Ans. 3 –
6.
ISSUES
1. Did the Examiner err in finding that Srivastava discloses “grouping
said web page traversal paths into web page categories; using the web page
categories to map the plurality of sessions to new sessions; [and] clustering
Appeal 2010-007776
Application 10/891,989

4
said new sessions according to a similarity measure into a plurality of web
session clusters,” as recited in claim 1?
2. Did the Examiner err in finding that Srivastava discloses tagging
that “includes embedding a Java script fragment into each of said plurality of
web pages,” as recited in claim 4?
3. Did the Examiner err in finding that Srivastava discloses
“extracting an open sequence as a representative sequence from each of said
web session clusters,” as recited in claim 10?
ANALYSIS
Claim 1
Claim 1 is directed to a method for identifying properties of a plurality
of web page traversal paths comprising the steps of “grouping said web page
traversal paths into web page categories; using the web page categories to
map the plurality of sessions to new sessions; [and] clustering said new
sessions according to a similarity measure into a plurality of web session
clusters.” The Examiner finds that Srivastava, which is directed to web
usage mining, discloses these recitations. See Ans. 3 (citing Srivastava p. 14,
§ 3.1.2 and p. 16, § 3.2.3). Specifically, the Examiner finds (1) that
Srivastava’s disclosure of the classification of page views discloses the
claimed grouping, (2) that Srivastava’s disclosure that sessions can be
filtered before or after pattern discovery discloses using the web page
categories to map the plurality of sessions to new sessions, and (3) that
Srivastava’s disclosure of clustering discloses the claimed clustering of new
sessions. See id.
Appeal 2010-007776
Application 10/891,989

5
Appellants contend the Examiner erred because “[c]lassifying or
clustering page views based on topic or intended use, as explained in § 3.1.2
of Srivastava, does not constitute using web page categories to map a
plurality of sessions to new sessions.” App. Br. 6. Specifically, Appellants
argue that “the ‘intended use’ of page views does not constitute the web
page categories that have been grouped from web page traversal paths.” Id.;
see also Reply Br. 2 (“the intended use is used for filtering sessions”).
Appellants also argue that “there is no teaching in § 3.1.2 of Srivastava that
these page views are mapped to new page views.” Reply Br. 2. Thus,
Appellants argue that Srivastava fails to disclose the claimed grouping and
using web page categories for mapping.
The Examiner correctly finds that Srivastava discloses that “page
views can be classified according to their intended use.” Ans. 6. Srivastava
discloses that the results of a classification algorithm can be used to “limit
the discovered patterns to those containing page views about a certain
subject or class of products,” with “intended use” being another way for
classifying or clustering page views. See Srivastava p. 14, § 3.1.2. Because
this filtering is based on the results of a classification algorithm, as applied
to page views, we agree with the Examiner that Srivastava discloses
“grouping said web page traversal paths into web page categories,” as
recited in claim 1.
The Examiner also correctly finds that Srivastava discloses sessions
can be filtered before pattern discovery, thus classifying the page views
according to their intended use. See Ans. 3 (citing Srivastava p. 14, § 3.1.2).
For example, “page views conveying information can then be filtered from
other sessions that are irrelevant to the page views” having an intended
Appeal 2010-007776
Application 10/891,989

6
purpose such as “conveying information.” Ans. 6 – 7. Srivastava discloses
that the intended use of a page view (i.e., web page categories) can filter
sessions before pattern discovery. See Srivastava p. 14, § 3.1.2. That is, the
intended use (i.e., the result of a classification algorithm applied to the
plurality of sessions) can be used to filter new web page views (i.e., mapping
the plurality of sessions to new sessions). Therefore, we agree with the
Examiner that Srivastava discloses “using the web page categories to map
the plurality of sessions to new sessions,” as recited in claim 1.
Appellants further argue that “Srivastava refers to clustering users into
groups of users, and clustering pages into groups of pages having related
content. Clustering users or clustering pages, as taught by Srivastava, is
completely different from clustering the new sessions according to a
similarity measure into a plurality of web session clusters.” App. Br. 7; see
also Reply Br. 3. However, the Examiner correctly finds that Srivastava
discloses a session as a “click-stream of page views for a single user for a
particular website.” Ans. 7; see also Srivastava p. 14, § 2.2. Appellants do
not provide persuasive arguments or evidence distinguishing between
Srivastava’s clustering of page views in a session and the claimed clustering
of new sessions according to a similarity measure into a plurality of web
session clusters. In particular, in disclosing the clustering of users,
Srivastava discloses that such clustering “tends to establish groups of users
exhibiting similar browsing patterns.” Srivastava p. 16, § 3.2.3 (emphasis
added). That is, Srivastava discloses the use of clustering as a technique to
form “usage clusters.” Id. Thus, Srivastava discloses clustering of new
sessions (click-streams of page views) according to a similarity measure
(based on similar browsing patterns) into a plurality of web session clusters
Appeal 2010-007776
Application 10/891,989

7
(to form usage clusters). Therefore, we agree with the Examiner that
Srivastava discloses “clustering said new sessions according to a similarity
measure into a plurality of web session clusters,” as recited in claim 1.
Accordingly, we affirm the Examiner’s 35 U.S.C. § 102(b) rejection
of claim 1, as well as the rejection of claims 2, 3, 5 – 11, 16 – 19, 22, and 23,
which are not argued with sufficient specificity to constitute separate
arguments. See App. Br. 7 – 8.
Claim 4
Claim 4 recites tagging of web pages that “includes embedding a Java
script fragment into each of said plurality of web pages.” The Examiner
finds that Srivastava’s use of remote agents such as Javascripts discloses the
claimed embedding. See Ans. 4 (citing Srivastava p. 13, § 2.1.2).
Appellants contend the Examiner erred because Srivastava “refers to
client-side data collection. However, there is nothing here that hints at
embedding the Javascripts . . . into each of the plurality of web pages of the
plurality of web page traversal paths.” App. Br. 7; see also Reply Br. 4.
However, the Examiner correctly finds that the use of Javascript provides for
the collection of single-site browsing behavior. See Ans. 8; see also
Srivastava p. 13, § 2.1.2. While Srivastava discloses that using Javascript
“cannot capture all user clicks (such as reload or back buttons),” and is not
as versatile as a modified browser, Srivastava notes using Javascript
“consume[s] little interpretation time” and still allows for the collection of
“single-user, single-site browsing behavior.” See Srivastava p. 13, § 2.1.2.
Thus, Srivastava distinguishes between client-side data collection based on
use of a modified browser and based on use of Javascript (i.e., embedded
Appeal 2010-007776
Application 10/891,989

8
script fragments). Therefore, we agree with the Examiner that Srivastava
discloses web page tagging that “includes embedding a Java script fragment
into each of said plurality of web pages,” as recited in claim 4.
Accordingly, we affirm the Examiner’s 35 U.S.C. § 102(b) rejection
of claim 4.
Claim 10
Claim 10 recites that “selecting one or more of said plurality of web
session cluster” comprises “extracting an open sequence as a representative
sequence from each of said web session clusters.” The Examiner finds that
the pattern discovery algorithms of Srivastava disclose this extracting. See
Ans. 5 (citing Srivastava p. 16, §§ 3.2.1 – 3.2.5). In particular, the Examiner
finds that Srivastava’s use of the Apriori algorithm (or variants thereof) to
correlate between users viewing different pages discloses the claimed
extracting. See Ans. 8 (citing Srivastava p. 16, § 3.2.2).
Appellants contend the Examiner erred because none of the
techniques in Srivastava refer “to developing web session clusters, or
selecting one of the web session clusters by extracting an open sequence as
a representative sequence from each of the web session clusters.” App. Br. 8.
Appellants further argue that Srivastava’s use of the Apriori algorithm to
correlate web page visits has nothing to do with the claimed extracting. See
Reply Br. 5. However, Srivastava discloses the generation of association
rules using algorithms such as the Apriori algorithm to “relate pages that are
most often referenced together in a single server session.” Srivastava p. 16,
§ 3.2.2 (emphasis added); see also Ans. 8. Appellants do not provide
arguments or evidence that persuasively distinguish between relating the
Appeal 2010-007776
Application 10/891,989

9
page views in a user session for a particular website and “extracting an open
sequence as a representative sequence from each of said web session
clusters.” Therefore, we agree with the Examiner that Srivastava discloses
“extracting an open sequence as a representative sequence from each of said
web session clusters,” as recited in claim 10.
Accordingly, we affirm the Examiner’s 35 U.S.C. § 102(b) rejection
of claim 10, and claims 11 and 20, which are not argued separately. See
App. Br. 8.
DECISION
The Examiner’s decision to reject claims 1 – 11, 16 – 20, 22, and 23 is
affirmed.
No time period for taking any subsequent action in connection with
this appeal may be extended under 37 C.F.R. § 1.136(a)(1)(iv).

AFFIRMED

msc