SAP AG

Patent Trials and Appeals Board

Mar 19, 2020

13902677 - (D) (P.T.A.B. Mar. 19, 2020)

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.
13/902,677 05/24/2013 Jitender Aswani 000005-036200US 2493
58735 7590 03/19/2020
Fountainhead Law Group P.C.
Chad R. Walsh
900 LAFAYETTE STREET
SUITE 301
SANTA CLARA, CA 95050
EXAMINER
TESSEMA, AIDA Z
ART UNIT PAPER NUMBER
2157
NOTIFICATION DATE DELIVERY MODE
03/19/2020 ELECTRONIC
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.
Notice of the Office communication was sent electronically on above-indicated "Notification Date" to the
following e-mail address(es):
docketing@fountainheadlaw.com
klhussain@fountainheadlaw.com
PTOL-90A (Rev. 04/07)
UNITED STATES PATENT AND TRADEMARK OFFICE
BEFORE THE PATENT TRIAL AND APPEAL BOARD
Ex parte JITENDER ASWANI, RYAN LEASK, and JENS DOERPMUND
Appeal 2019-001390
Application 13/902,677
Technology Center 2100
Before JASON V. MORGAN, ERIC B. CHEN, and
PHILLIP A. BENNETT, Administrative Patent Judges.
BENNETT, Administrative Patent Judge.
DECISION ON APPEAL
STATEMENT OF THE CASE
Pursuant to 35 U.S.C. § 134(a), Appellant1 appeals from the
Examiner’s decision to reject claims 1–9 and 11–20. Claim 10 has been
cancelled. We have jurisdiction under 35 U.S.C. § 6(b).
We reverse.

1 We use the word “Appellant” to refer to “applicant” as defined in
37 C.F.R. § 1.42(a). Appellant identifies the real party in interest as SAP
SE. Appeal Br. 2.
Appeal 2019-001390
Application 13/902,677

2
CLAIMED SUBJECT MATTER
The claims are directed to representing enterprise data in a knowledge
graph. Claim 1, reproduced below with the key limitation in italics, is
illustrative of the claimed subject matter:
1. A computer-implemented method for storing data
comprising steps of:
storing a graph that represents relationships among
enterprise data that comprise one or more databases in the
enterprise, the graph comprising:
a plurality of first-level nodes, the first-level nodes
representative of classes of data among the enterprise
data, the first-level nodes having links to each other based
on relationships among the first-level nodes; and
a plurality of second-level nodes that are children
nodes of the first-level nodes, the second-levels nodes
having links to each other in the same relation as their
respective first level nodes;
receiving data that is not search input data, the received
data to be stored in the one or more databases in an enterprise;
identifying a first-level node from the plurality of first-
level nodes by comparing some of the received data to properties
and attributes of the respective classes of data represented by the
first-level nodes;
using the identified first-level node to identify a database
table in the one or more databases in the enterprise;
storing some of the received data in the identified database
table; and
growing the graph, including:
creating one or more second-level nodes;
storing some of the received data in the one or more
created second-level nodes;
linking the one or more created second-level nodes
as children nodes of the identified first-level node; and
linking the one or more created second-level nodes
to other second-level nodes in the graph in the same
Appeal 2019-001390
Application 13/902,677

3
relation as the identified first-level node linked to the
respective first-level nodes of the other second-level
nodes.
Appeal Br. 23 (Claims Appendix).
REFERENCES
The prior art relied upon by the Examiner is:
Name Reference Date
Das
Lee
Larson
Abbassi
US 2004/0019672 A1
US 2007/0130133 A1
US 2008/0281801 A1
US 2014/0278590 A1
Jan. 29, 2004
June 7, 2007
Nov. 13, 2008
Sept. 18, 2014
REJECTION
Claims 1–9 and 11–20 stand rejected under 35 U.S.C. § 103 as being
unpatentable over Abbassi, Larson, Das, and Lee. Final Act. 5–13.
ISSUE
Has the Examiner erred in finding that Larson teaches or suggests “a
plurality of first-level nodes, the first-level nodes representative of classes of
data among the enterprise data, the first-level nodes having links to each
other based on relationships among the first-level nodes,” as recited in
claim 1?
ANALYSIS
Appellant’s Invention
Appellant’s invention relates to the construction and use of a
knowledge base which is built up by connecting to, and collecting data from,
various data sources in an enterprise. Spec. ¶ 20. These data sources may
Appeal 2019-001390
Application 13/902,677

4
include, for example, backend systems such as customer relationship
management (CRM) systems and enterprise resource planning systems
(ERM), public data sources, and analytics data. Spec. ¶¶ 20–22. The data
collected for the knowledge base is classified according to defined classes of
data and inserted into a knowledge graph. Spec. ¶ 34.
The knowledge graph includes two levels of nodes. Spec. ¶ 38. The
first-level nodes, referred to as entity nodes, define the classes of data that
are represented in the knowledge graph. Spec. ¶ 38–39. The second-level
nodes, called instance nodes, represent data that first-level nodes define, and
are linked to their respective first-level nodes. Spec. ¶ 40. Thus, first-level
nodes can be thought of as representing and defining a category (such as
“automobile”) and its associated properties (such as “make,” “model,” and
“color”), and the second-level nodes can be thought of as specific items
within the defined category having actual data values associated with the
properties (such as “Ford,” “Mustang,” “red”). Spec. ¶ 40–41. The first-
level nodes in the knowledge graph may be related or linked to other first-
level nodes. Spec. ¶ 39. Those relationship/links are inherited by second-
level nodes such that the second-level nodes relate to each other in the same
way as their respective first-level nodes. Spec. ¶ 50. Figure 5 of
Appellant’s Specification is reproduced below with annotations illustrating
the relationships between the two levels of nodes.
Appeal 2019-001390
Application 13/902,677

5

Figure 5 of Appellant’s Specification (annotations added) illustrates that
each first level node may be linked to several second level instance nodes.
Claim 1 recites a particular method for building a data structure such
as that illustrated in Figure 5 by receiving data, storing the data in an
appropriate enterprise database, and then separately growing the knowledge
graph by adding new nodes representing the received data. Under the
Appeal 2019-001390
Application 13/902,677

6
claimed method, data is received for storing in one or more enterprise
databases. The received data is compared to the first-level nodes defined in
the knowledge graph, and in particular to properties and attributes of the data
categories represented by the first-level nodes. Based on that comparison, a
first-level node in the knowledge graph is identified as the correct
classification for the received data, and the first-level node is used, in turn,
to identify a database table in the enterprise in which to store the received
data. The data is stored in the identified database table. Separately, the
graph database is grown by creating a new second-level node which stores
the received data and by adding two types of links: (1) a link from the
newly-created second-level node to the previously identified first-level node,
and (2) a link from the second-level node to other second-level nodes which
mirrors the links between the previously identified first-level node and other
first-level nodes.
Examiner’s Findings and Conclusion
The Examiner rejects claim 1 as obvious over Abbassi, Larson, Das,
and Lee. Final Act. 5. Relevant to the issues presented here, the Examiner
generally relies on Abbassi to show that the use of graph databases and
graph nodes was known in the prior art. Final Act. 5–6. The Examiner finds
that Abbassi fails to teach a graph database with the specific structure recited
in the claim. Specifically, the Examiner finds Abbassi does not teach “a
plurality of first-level nodes, the first-level nodes representative of classes of
data among the enterprise data, the first-level nodes having links to each
other based on relationships among the first level nodes,” and “a plurality of
second-level nodes that are children nodes of the first-level nodes, the
Appeal 2019-001390
Application 13/902,677

7
second-level nodes having links to each other in the same relation as their
respective first-level nodes.” Final Act. 6.
The Examiner turns to Larson for these limitations, finding that
Larson’s description of “a system where parent nodes are linked based on
some sort of common attribute, and child nodes that are linked based on the
relationship of the parent nodes,” teaches the limitation. Final Act. 7 (citing
Larson Figs. 6, 8, 9; ¶¶ 15, 24, 25, 74, 88, and 89) (italics omitted). The
Examiner further explains that the movie titles under the heading “Movies”
depicted in Larson’s Figure 6 correspond to the recited “first-level nodes”
and that these nodes have links to each other because “[t]here does not seem
to be any limiting factor indicating how the nodes are linked to each other
and therefore regardless of the use of a context node or not, the nodes
disclosed by Larson are in fact linked together.” Ans. 5.
The Examiner concludes it would have been obvious “to have
incorporated the database table as taught by Larson into the system of
Abbassi . . . [because] [b]oth are directed to a means of organizing data
using directed graphs and the combination would further provide[] a
database system callable of analyzing source data to extract data in a timely
and efficient manner.” Final Act. 7 (citing Larson ¶ 29).
Appellant’s Argument
Appellant argues the Examiner has erred because the top level node
shown in Larson’s Figure 8, the node labeled “Movies,” is only a single
node, and the disputed limitation requires “a plurality of first-level nodes.”
Appeal Br. 11–12. Appellant further contends that because there is only a
single first-level “Movie” node, “Larson does not teach “first-level nodes
having links to each other.” Appeal Br. 12. According to Appellant,
Appeal 2019-001390
Application 13/902,677

8
“because FIG. 8 does not show any other nodes that are at the same level as
the MOVIES node nor any other classes of data represented by nodes at the
same level as the MOVIES node.” Id. Appellant further argues the
Examiner’s reliance on the movie titles depicted in Larson’s Figure 8 as
corresponding to the recited “first-level nodes” is misplaced, as “Larson
clearly explains that FIG. 6 illustrate a tree-like structure 100 with elements
110, 114, 118 . . . and 150 as being nodes in a tree-like structure.” Reply
Br. 4 (citing Larson ¶ 14). Appellant asserts that “[r]eferring to element 110
as a heading . . . does not change the fact that Larson explicitly teaches
element 110 to be a node,” and that as the top node in the structure, the
“Movies” node is the only node in Larson that can be reasonably
characterized as a “first-level node.” Id.
Our Review
We are persuaded the Examiner has erred. As we noted above, the
data structure recited in claim 1 includes two different types of nodes –
“first-level nodes” which essentially represent classes or categories of data
and “second-level nodes” which can be characterized as representing
instances of the classes defined at the first-level. The first-level nodes are
linked to each other based on relationships between the categories of data
they represent. It is these characteristics that are encompassed by the
disputed limitation “a plurality of first-level nodes, the first-level nodes
representative of classes of data among the enterprise data, the first-level
nodes having links to each other based on relationships among the first-level
nodes.”
The Examiner cites Figure 6 of Larson as teaching this limitation.
Ans. 4–5. Larson describes the tree structure of Figure 6 as “sample movie
Appeal 2019-001390
Application 13/902,677

9
data stored in a tree-like structure 100 of a hierarchical database system.”
Larson ¶ 14. Larson’s Figure 6 is reproduced below.

Larson’s Figure 6 is a block diagram illustrating a tree-like structure of a
hierarchical database system.
The Examiner finds that nodes 114 and 118, labeled “Apollo 13” and “Cast
Away,” respectively, to be “first-level nodes,” and that node 110, labeled
“Movies” is merely a table heading. We agree with Appellant, however, that
the “Apollo 13” and “Cast Away” nodes cannot reasonably be considered
“first-level nodes.” Appellant’s Specification unambiguously states that
“first level nodes do not [ ] represent instances of the data they define.”
Spec. ¶ 40. The nodes relied upon by the Examiner do not meet this
Appeal 2019-001390
Application 13/902,677

10
requirement. The nodes “Apollo 13” and “Cast Away” represent instances
of the movies themselves, and are therefore more akin to the “second-level
nodes” recited in claim 1.
Moreover, we also agree with Appellant that even if the “Apollo 13”
and “Cast Away” nodes could be considered first-level nodes within the
meaning of claim 1, they still would not meet the requirement that the first
level nodes “having links to each other based on relationships among the
first-level nodes.” As illustrated in Appellant’s Figure 5, the relationships
between first-level nodes represent relationships among them. See
Spec. ¶ 49 (“For example, link 5a may represent the relationship that a
customer ‘has purchased’ a product, while the link 5b may represent the
relationship that a product ‘has been purchased by’ a customer.”). In
Larson, the nodes “Apollo 13” and “Cast Away” have no similar
relationship. Instead, these nodes are merely depicted as being instances of
“Movies” without any meaningful connection between them.
Accordingly, we are persuaded the Examiner has erred in finding that
Larson teaches or suggests the disputed limitation in claim 1. We, therefore,
do not sustain the rejection of claim 1 under 35 U.S.C. § 103. Independent
claims 9 and 17 also recite the disputed limitation, and we do not sustain
their rejection under the same reasoning. The remaining claims are
dependent and stand for the same reasons.
CONCLUSION
The Examiner’s rejection of claims 1–9 and 11–20 under 35 U.S.C.
§ 103 is reversed.
Appeal 2019-001390
Application 13/902,677

11
DECISION SUMMARY
Claims
Rejected
35 U.S.C. § References Affirmed Reversed
1–9, 11–20 103 Abbassi, Larson,
Das, Lee
1–9, 11–20

REVERSED