Ex Parte Demuth et al

Patent Trials and Appeals Board

Apr 3, 2019

14331114 - (D) (P.T.A.B. Apr. 3, 2019)

UNITED STA TES p A TENT AND TRADEMARK OFFICE
APPLICATION NO. FILING DATE
14/331,114 07/14/2014
48916 7590
Greg Goshorn, P.C.
9600 Escarpment Blvd.
Suite 745-9
AUSTIN, TX 78749
04/03/2019
FIRST NAMED INVENTOR
Steven J. Demuth
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
ATTORNEY DOCKET NO. CONFIRMATION NO.
AUS920120314US2 2321
EXAMINER
KAZEMINEZHAD, FARZAD
ART UNIT PAPER NUMBER
2657
MAIL DATE DELIVERY MODE
04/03/2019 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 STEVEN J. DEMUTH, CLAUS T. JENSEN,
NICOLAS CHANGHAI KE, and LI XU
Appeal2018-005866
Application 14/331, 114
Technology Center 2600
Before LARRY J. HUME, LINZY T. McCARTNEY, and
JASON J. CHUNG, Administrative Patent Judges.
McCARTNEY, Administrative Patent Judge.
DECISION ON APPEAL 1
Appellants seek review under 35 U.S.C. § 134 of the Examiner's final
rejection of claims 1---6. We have jurisdiction under 35 U.S.C. § 6(b ).
We affirm.
1 This appeal is related to Appeal No. 2018-005839.
Appeal2018-005866
Application 14/3 31, 114
BACKGROUND
This patent application concerns "techniques for the manipulation of
semantic objects within a semantic store." Specification ,r 2, filed July 14,
2014 ("Spec."). Claim 1, the only independent claim, illustrates the claimed
invention:
1. A method for self-programming a semantic reasomng
apparatus, comprising:
evaluating, by a processor implemented semantic driver
that employs a semantic model as input for driving behavior, a
rule, the rule comprising;
a condition part; and
an action part;
wherein the rule is based upon the semantic model
and configured to execute the action part in response to a
determination that the condition part is satisfied by one or
more objects of a plurality of semantic objects of a
semantic store wherein a semantic object is a data object
that captures operation business contexts and intentions in
an actionable fashion; and
evaluating the rule;
modifying the rule based upon the evaluating to generate
a modified rule;
determining that a condition part of the modified rule is
satisfied based upon the evaluating; and, in response to the
determining,
modifying a semantic object of the plurality of semantic
objects in conformity to an action part of the modified rule to
modify the behavior of the semantic driver.
Appeal Brief 16, filed February 6, 2018 ("App. Br.").
2
Appeal2018-005866
Application 14/3 31, 114
1---6
1---62
1-5
6
1---6
REJECTIONS
§ 112
§ 101
§ 103
§ 103
Double Patentin
DISCUSSION
Di Fabbrizio3 and
Gadd4
Di Fabbrizio, Gadd,
and Kim5
Demuth '0396
We have reviewed the Examiner's rejections and Appellants'
arguments. We disagree with Appellants that the Examiner erred. As
consistent with the discussion below, we adopt the Examiner's findings,
conclusions, and reasoning in the Final Office Action mailed August 9, 2017
("Final Act.") and the Answer mailed March 22, 2018 ("Ans."). Appellants
have forfeited arguments that Appellants failed to adequately or timely
present. See 37 C.F.R. § 41.37 (c)(l)(iv).
Section 112 Rejection
Claim 1 recites "modifying the rule based upon the evaluating to
generate a modified rule." App. Br. 16. Claims 2-6 depend from claim 1 and
thus recite this limitation as well. See App. Br. 17. The Examiner found that
2 The Final Office Action incorrectly states that claims 1-13 and 20-25
stand rejected on this ground. Final Office Action 8, mailed August 9, 2017.
We interpret this as a typographical error and view claims 1-6 as standing
rejected on this ground.
3 Di Fabbrizio et al. (US 2012/0253825 Al; Oct. 4, 2012).
4 Gadd et al. (US 2005/0033582 Al; Feb. 10, 2005).
5 Kim et al. (US 2008/0208774 Al; Aug. 28, 2008).
6 Demuth et al. (US Appl. No. 13/932039; July 1, 2013).
3
Appeal2018-005866
Application 14/3 31, 114
this limitation lacks adequate written description support because the written
description and drawings disclose modifying semantic objects, not rules. See
Final Act. 6-8; Ans. 2 ( explaining that "[b ]roadly the rejection was about
lack of written [description] support for 'generat[ing] a modified rule"'
(third alteration in original)).
Appellants argue that "although the [ written description] may not
explicitly describe modifying a rule," the written description "describes
modifying a constraint, which is part of a rule." App. Br. 8. Appellants point
to paragraph 30 of the written description to support this argument. See App.
Br. 7-8 ( quoting Spec. ,r 30).
Appellants have not persuaded us that the Examiner erred. To satisfy
the written description requirement, "the disclosure of the application relied
upon" must "reasonably convey[] to those skilled in the art that the inventor
had possession of the claimed subject matter as of the filing date." Ariad
Pharm., Inc. v. Eli Lilly & Co., 598 F.3d 1336, 1351 (Fed. Cir. 2010) (en
bane). Appellants have not shown that paragraph 30 or any other part of the
written description satisfies this requirement. Paragraph 30 describes using a
constraint to automatically generate a rule and then using the generated rule
to enforce the constraint on a semantic store. See Spec. ,r 30. The paragraph
explains that the rule enforces the constraint by modifying semantic objects
in the store. See Spec. ,r 30; see also id. ,r 29. Appellants have not
persuasively explained how the rule-generation process modifies a
constraint, nor have Appellants persuasively explained how modifying
semantic objects modifies a constraint. Appellants simply quote part of
paragraph 30 and assert that the written description "describes modifying a
constraint, which is part of a rule." App. Br. 7-8. Given the silence of
4
Appeal2018-005866
Application 14/3 31, 114
paragraph 30 on modifying constraints and rules, Appellants have not shown
that paragraph 30 "reasonably conveys to those skilled in the art that the
inventor had possession of the claimed subject matter as of the filing date."
Ariad, 598 F.3d at 1351.
Appellants also assert that "there is an entire portion of the" written
description "and a corresponding figure" that "describe 'generating rules."'
App. Br. 7 (citing Spec. ,r,r 47-51, Fig. 7).
We agree with Appellants that the cited figure and parts of the written
description disclose generating a rule. But we do not agree that they disclose
modifying an existing rule as required by the claim. The written description
describes rules as including both conditions and actions. See, e.g., Spec.
,r,r 6, ("a rule, comprising; a condition part; and an action part"), 29 ("Rules
122 express conditions and actions that are taken if the conditions are met."),
Abstract ("a rule, comprising; a condition part; and an action part"). This
description tracks both the original and appealed claims, as these claims
recite "a rule ... comprising" both "a condition part" and "an action part."
App. Br. 16; Spec. 16 ( claim 1 as filed). The process described in the cited
figure and parts of the written description generates a new "constraint rule"
by associating a constraint and an action with a rule and then storing the
rule. See, e.g., Spec. ,r,r 4 7 ("Process 300 starts in a 'Begin Generate
Constraint Rules' block 302 .... and proceeds immediately to [a]
'Determine List of Semantic Constraints' block 304 .... Such a list may be
generated by a user .... "'), 48 ("[T]he constraint selected during processing
... is associated with a rule. For example, ... the rule may be 'if type of
attribute X is not Y. "'), 49 ("[T]he action selected during processing ... is
5
Appeal2018-005866
Application 14/3 31, 114
paired with the rule .... the rule action pair generated during processing
.... is stored in rules 122.").
Only after this process associates constraints and actions with a rule is
the rule "generated" within the meaning of the written description. The
intermediate steps of associating a constraint and an action with a rule thus
do not modify a rule; these steps are part of generating a new rule. See, e.g.,
Spec. ,r,r 48 ("[T]he constraint selected during processing ... is associated
with a rule. For example, ... the rule may be 'if type of attribute Xis not
Y. "'), 49 ("[T]he action selected during processing ... is paired with the rule
.... the rule action pair generated during processing .... is stored in rules
122."), Fig. 7 (showing that the "End Generate Constraint Rules" step 319
follows the steps of associating a constraint and an action with a rule). This
understanding is consistent with Appellants' admission that the written
description does not explicitly describe modifying a rule. See App. Br. 8
( arguing that "although the [ written description] may not explicitly describe
modifying a rule," the written description "describes modifying a
constraint").
Finally, Appellants contend that the written description uses the terms
"constraint rules" and "modifying constraints" "to describe a process that ...
is synonymous with modifying a rule." Reply Brief 2, filed May 18, 2018
("Reply Br."). Appellants do not explicitly identify the process that is
supposedly synonymous with modifying a rule. See Reply Br. 2. If
Appellants contend that the disclosed process of generating a constraint rule
is synonymous with modifying a rule, we disagree for the reasons discussed
above. And if Appellants contend that it would have been obvious to modify
a rule given the disclosed process of generating a constraint rule, we note
6
Appeal2018-005866
Application 14/3 31, 114
that "a description that merely renders the invention obvious does not satisfy
the [written description] requirement." Ariad, 598 F.3d at 1352.
As for the term "modifying constraints," the term does not appear in
the written description. The written description uses variations of the word
"modify" when discussing semantic objects, not constraints, and Appellants
have not argued, much less shown, that semantic objects are ( or include)
constraints.
For the above reasons, Appellants have not persuaded us that the
Examiner erroneously rejected claims 1-6 under§ 112.
Section 101 Rejection
Section 101 of the Patent Act provides that "any new and useful
process, machine, manufacture, or composition of matter, or any new and
useful improvement thereof' is patent eligible. 35 U.S.C. § 101. But the
Supreme Court has long recognized an implicit exception to this section:
"Laws of nature, natural phenomena, and abstract ideas are not patentable."
Alice Corp. v. CLS Bankint'l, 573 U.S. 208,216 (2014) (quotingAss'nfor
Molecular Pathology v. Myriad Genetics, Inc., 569 U.S. 576, 589 (2013)).
To determine whether a claim falls within one of these excluded categories,
the Court has set out a two-part framework. The framework requires us first
to consider whether the claim is "directed to one of those patent-ineligible
concepts." Alice, 573 U.S. at 217. If so, we then examine "the elements of
[the] claim both individually and 'as an ordered combination' to determine
whether the additional elements 'transform the nature of the claim' into a
patent-eligible application." Alice, 573 U.S. at 217 (quoting Mayo
Collaborative Servs. v. Prometheus Labs., Inc., 566 U.S. 66, 78, 79 (2012)).
That is, we examine the claim for an "inventive concept," "an element or
7
Appeal2018-005866
Application 14/3 31, 114
combination of elements that is 'sufficient to ensure that the patent in
practice amounts to significantly more than a patent upon the [ineligible
concept] itself."' Alice, 573 U.S. at 217-18 ( alteration in original) ( quoting
Mayo, 566 U.S. at 72-73).
The Patent Office recently issued guidance about this framework. See
2019 Revised Patent Subject Matter Eligibility Guidance, 84 Fed. Reg. 50
(Jan. 7, 2019) ("Revised Guidance"). Under the Revised Guidance, to decide
whether a claim is directed to an abstract idea, we evaluate whether the
claim ( 1) recites one of the abstract ideas listed in the Revised Guidance
("Prong One") and (2) fails to integrate the recited abstract idea into a
practical application ("Prong Two"). See Revised Guidance, 84 Fed. Reg. at
51, 54. If the claim is directed to an abstract idea, as noted above, we then
determine whether the claim has an inventive concept. The Revised
Guidance explains that when making this determination, we should consider
whether the additional claim elements add "a specific limitation or
combination of limitations that are not well-understood, routine,
conventional activity in the field" or "simply append[] well-understood,
routine, conventional activities previously known to the industry, specified
at a high level of generality." Revised Guidance, 84 Fed. Reg. at 56.
With these principles in mind, we tum to the§ 101 rejection.
Appellants argue claims 1-6 together for this rejection, so as permitted by 37
C.F.R. § 41.37, we decide the appeal for this rejection based on claim 1. 37
C.F.R. § 41.37 (c )(1 )(iv).
Abstract Idea
The Examiner determined that claim 1 is directed to the abstract ideas
of "comparison of stored and new information to identify abstract
8
Appeal2018-005866
Application 14/3 31, 114
modification options" and "a method of organizing human activities." Final
Act. 9. For the reasons discussed below, we agree with the Examiner that
claim 1 is directed to an abstract idea. We first consider whether claim 1
recites an abstract idea. We then consider whether claim 1 integrates the
abstract idea into a practical application.
Prong One
Claim 1 recites (1) "evaluating ... a rule" that includes "a condition
part" and "an action part" and satisfies the other requirements specified in
the claim, (2) "evaluating the rule," and (3) "modifying the rule based upon
the evaluating to generate a modified rule." App. Br. 16. The claim also
recites ( 4) "determining that a condition part of the modified rule is satisfied
based upon the evaluating" and, "in response to the determining,"
(5) "modifying a semantic object of the plurality of semantic objects in
conformity to an action part of the modified rule to modify the behavior of
the semantic driver." App. Br. 16.
Claim 1 does not recite how these steps are performed; the claim
essentially recites a series of functions or results and places no limits on the
process used to accomplish them. See App. Br. 16. Although certain steps
recite what the results are based on or intended to do, these steps do not
recite or limit how the results are achieved. See App. Br. 16. And other than
the first "evaluating" step, claim 1 does not require that a processor perform
the recited functions. See App. Br. 16. To the extent that the written
description discusses some of these steps, the written description generally
describes the steps in functional, result-oriented terms. See, e.g., Spec.
,r,r 40("semantic driver ... may change semantic objects"), 42 ("the
conditional portion of the current rule ... is evaluated"), 44 ("a
9
Appeal2018-005866
Application 14/3 31, 114
determination is made as to whether ... the condition evaluated during
processing ... is TRUE. ... the action portion of the rule is executed and
new data is derived .... a rule may necessitate modification of objects").
These steps thus encompass any method of achieving the recited
functions and results, including methods that people can perform in their
heads or using pen and paper. Cf CyberSource Corp. v. Retail Decisions,
Inc., 654 F.3d 1366, 1372 (Fed. Cir. 2011) (determining that a claim
encompassed "essentially any method" of detecting credit card fraud based
on certain information, "even methods that can be performed in the human
mind," when the specification did not disclose a fraud detection algorithm
and the claim did not limit its scope to a particular fraud detection
algorithm); Synopsys, Inc. v. Mentor Graphics Corp., 839 F.3d 1138, 1148
(Fed. Cir. 2016) ("[T]he limited, straightforward nature of the steps involved
in the claimed method make evident that a skilled artisan could perform the
steps mentally."). This is particularly true of the steps that make no mention
of using a processor or a computer to perform the recited functions. Cf
Synopsys, 839 F.3d at 1149 ("Because the Asserted Claims make no mention
of employing a computer or any other physical device, they are so broad as
to read on an individual performing the claimed steps mentally or with
pencil and paper."). For example, people can perform the "evaluating" steps
and "modifying the rule" step by mentally considering whether the condition
part of the rule has been satisfied and, if not, mentally revising the rule so
that the rule meets the recited conditions. People can perform the
"determining that the condition part of the modified rule is satisfied based
upon the evaluating" step by mentally determining whether the condition
part of the modified rule has been met based upon the evaluating.
10
Appeal2018-005866
Application 14/3 31, 114
Finally, people can also mentally perform the "modifying a semantic
object ... in conformity to the action part of the modified rule" step. Claim
1 recites that a "semantic object" is simply a "a data object that captures
operation business contexts and intentions in an actionable fashion." App.
Br. 16. See also Spec. ,r 28 ( explaining that exemplary semantic objects
"capture[] aspects of a business operation model, such as, but not limited to,
business entities, properties of business entities and relationships between
business entities"). As noted above, neither claim 1 nor the written
description limits how the recited "modifying" happens. This limitation thus
encompasses people modifying information "that captures operation
business contexts and intentions in an actionable fashion" in their heads such
that the modified information satisfies the recited conditions.
Considered as a whole, people can perform the steps discussed above
in their heads or using pen and paper. As a result, the claimed concept is no
different from other concepts that courts have determined are mental
processes. See, e.g., CyberSource, 654 F.3d at 1372-73 (determining that a
claim that recites steps that "can be performed in the human mind, or by a
human using a pen and paper" is directed to an unpatentable mental
process). We therefore conclude that claim 1 recites mental processes, one of
the abstract idea groupings in the Revised Guidance. See Revised Guidance,
84 Fed. Reg. at 52. Claim 1 thus recites an abstract idea. 7
7 This outcome is consistent with Federal Circuit decisions that have
determined that some claims reciting broad, functional language may be
drawn to abstract ideas. See, e.g., Two-Way Media Ltd. v. Comcast Cable
Commc 'ns, LLC, 874 F.3d 1329, 1337 (Fed. Cir. 2017) (determining that a
claim was drawn to an abstract idea when the claim required "the functional
results of 'converting,' 'routing,' 'controlling,' 'monitoring,' and
'accumulating records,' but [did] not sufficiently describe how to achieve
11
Appeal2018-005866
Application 14/3 31, 114
We find Appellants' argument to the contrary unpersuasive.
Appellants contend that the Examiner failed to identify a similar concept that
courts have found abstract. See Reply Br. 2; see also App. Br. 10-11. But
the Examiner determined that the abstract idea recited in claim 1 is similar to
the abstract idea in SmartGene, Inc. v. Advanced Biological Laboratories.,
SA, 555 F. App'x 950 (Fed. Cir. 2014). See Ans. 7. In SmartGene, the
Federal Circuit determined that the claims before it were directed to an
abstract idea when the claims recited a process "in which every step is a
familiar part of the conscious process that doctors can and do perform in
their heads." SmartGene, 555 F. App'x at 955. Similarly, people can
perform each of the steps discussed above in their heads. We thus see no
merit in this argument.
Prong Two
Because we determine that claim 1 recites an abstract idea, we next
consider whether claim 1 integrates the abstract idea into a practical
application. See Revised Guidance, 84 Fed. Reg. at 51. In doing so, we
consider whether there are any additional elements beyond the abstract idea
that, individually or in combination, "integrate the [ abstract idea] into a
practical application, using one or more of the considerations laid out by the
Supreme Court and the Federal Circuit." Revised Guidance, 84 Fed. Reg.
at 54--55.
Along with the steps recited above, claim 1 recites a "processor
implemented semantic driver that employs a semantic model for driving
these results in a non-abstract way"); Affinity Labs of Tex., LLC v.
Amazon.com Inc., 838 F.3d 1266, 1269-70 (Fed. Cir. 2016) ("The purely
functional nature of the claim confirms that it is directed to an abstract idea,
not to a concrete embodiment of that idea.").
12
Appeal2018-005866
Application 14/3 31, 114
behavior" and a "semantic store." App. Br. 18. The written description
makes clear that the recited processor is a generic computer component. See,
e.g., Spec. ,r 23 ( disclosing that a computing system architecture "on which
the claimed subject matter may be implemented" includes a computing
system with a "central processing unit"), Fig. 1 (illustrating an exemplary
computing system architecture and computing system). As for the recited
"semantic store," "semantic model," and "semantic driver," both claim 1 and
the written description describe these elements in largely functional terms
and provide few technical details about these elements. Claim 1 merely
recites, for example, "a plurality of semantic objects of a semantic store," a
"semantic driver that employs a semantic model as input for driving
behavior," the "rule is based upon the semantic model," and "to modify the
behavior of the semantic driver." App. Br. 16. Similarly, the written
description discloses, for example, that semantic store 118 "supports
semantic reasoning and stores semantic objects" and that semantic model
121 "is a representation of domain knowledge and the 'controller' for how
rule engine 134 interprets and interacts with semantic store 118." Spec. ,r 28.
For semantic driver 132, the written description discloses, for instance, that
it "enables rule engine 134 to dynamically interact with and directly
manipulate objects ... and allows rule conditions to be evaluated directly
against objects ... and rule actions to change objects." Spec. ,r 29; see also
id. ,r,r 31-34 (describing a block diagram of semantic driver 132).
Neither claim 1 nor the written description describes how these
elements perform the recited functions. These elements are so lacking in
technical details that the elements are effectively generic components for
performing the recited functions. Cf Intellectual Ventures I LLC v. Erie
13
Appeal2018-005866
Application 14/3 31, 114
Indem. Co., 850 F.3d 1315, 1331 (Fed. Cir. 2017) ("The claimed mobile
interface is so lacking in implementation details that it amounts to merely a
generic component ( software, hardware, or firmware) that permits the
performance of the abstract idea, i.e., to retrieve the user-specific
resources."); In re TL! Commc 'ns LLC Patent Litig., 823 F.3d 607, 615
(Fed. Cir. 2016) ("Such vague, functional descriptions of server components
are insufficient to transform the abstract idea into a patent-eligible
invention."). Indeed, the written description indicates that no particular
components are necessary to implement the claimed invention. See, e.g.,
Spec. ,r,r 26 ("[I]t should be noted there are many possible computing system
configurations."), 32 ("It should be understood that the claimed subject
matter can be implemented in many types of computing systems and data
storage structures.").
Despite the generic nature of these components, Appellants argue
"that the claimed subject matter is more than merely 'implementation of the
abstract idea on a conventional computer' because the disclosed technology
enables an otherwise 'conventional computer' to reprogram itself by means
of modifying the rules that control the computer." App. Br. 9--10 ( citing
Spec. ,r 30). That is, Appellants contend the claimed invention is "self-
leaming," despite the Examiner finding to the contrary. App. Br. 9.
We find these arguments unpersuasive. Although the preamble of
claim 1 recites "[a] method for self-programming a semantic reasoning
apparatus," with the exception of the first "evaluating" step, the claimed
method does not require that a processor or a computer perform the recited
steps. See App. Br. 16. And none of the steps recites that a computer
"reprogram[ s] itself by means of modifying the rules that control the
14
Appeal2018-005866
Application 14/3 31, 114
computer" or that the method results in a "self-learning computer." See App.
Br. 16.
The only parts of claim 1 that arguably concern these concepts are the
"modifying the rule" and the "modifying a semantic object" limitations.
App. Br. 16. But as noted above, the claim does not require that a processor
perform these steps, so it is unclear why these steps involve a "self-learning"
computer or a computer that "reprograms" itself. In any event, when these
limitations are considered in light of the written description, it is clear that
these limitations do not recite a computer that reprograms itself or a
computer that self-learns.
The written description does not describe modifying an existing rule,
much less doing so to reprogram a computer or to allow the computer to
self-learn. As for modifying a semantic object, the written description
indicates that the claimed invention modifies semantic objects to ensure that
the semantic store and the semantic model remain consistent. See, e.g., Spec.
,r 40 (explaining that semantic driver 132 may change semantic objects "to
ensure consistency between semantic store 118 and semantic model 121 ").
The written description makes no mention of modifying a semantic object
for reprogramming or self-learning purposes. Because the written
description does not describe modifying a rule or a semantic object to allow
a computer to reprogram itself or to self-learn, one of ordinary skill in the art
would not understand the "modifying the rule" and "modifying a semantic
object" limitations to concern reprogramming or self-learning.
Appellants suggest that paragraph 31 of the written description shows
otherwise. See App. Br. 9--10 ( citing Spec. ,r 31 ). This paragraph simply
provides a general description of Figure 3; nothing in this paragraph
15
Appeal2018-005866
Application 14/3 31, 114
concerns a self-learning computer or a computer that reprograms itself. See
Spec. ,r 31. If Appellants meant to cite paragraph 3 0 of the written
description instead, this paragraph explains that "a constraint rule may be
automatically generated based upon information in semantic model 121, i.e.,
not generated by an administrator" and that the invention uses constraint
rules to modify semantic objects in a semantic store. Spec. ,r 30. But claim 1
does not explicitly recite automatically generating a constraint rule, and the
written description indicates that automatically generating a constraint rule
involves no more than employing a user-defined semantic model to associate
user-defined data. See, e.g., Spec. ,r,r 35-37 (explaining that selecting classes
and properties for a semantic model and defining conditions and actions is
typically "executed by [a] system architect, analyst or administrator"), 4 7
( disclosing that "a user such as a system architect or administrator" may
generate a list of constraints that are defined as part of semantic model 121 ),
48--49 ( describing the process of associating a constraint and an action to
generate a rule).
We see no indication that this process involves a computer
reprogramming itself or self-learning. The Examiner found-and Appellants
have not disputed-that the disclosed process of automatically generating a
constraint rule "merely corresponds to a simple conventional procedural
programming model" and "does not in any way teach or suggest that the
'semantic model' ... itself undergoes changes or is altered once [the
process] begins." Ans. 5. And the written description indicates that the
claimed invention concerns manipulation of semantic objects, not modifying
rules so that a computer can reprogram itself or self-learn. See, e.g., Spec.
,r,r 2 ("The claimed subject matter relates ... to techniques for the
16
Appeal2018-005866
Application 14/3 31, 114
manipulation of semantic objects within a semantic store"), 6 ("Provided are
techniques for the manipulation of semantic objects within a semantic store
.... "), Abstract (same). Indeed, terms such as "reprogram" and "self-learn"
do not appear in the written description.
Because the written description does not describe reprogramming a
computer or a computer that self-learns, one of ordinary skill in the art
would not understand the "modifying the rule" and "modifying a semantic
object" limitations to concern those concepts. Claim 1 thus does not include
limitations that involve reprogramming a computer or a computer that self-
learns. Appellants' arguments that the claimed invention concerns a self-
learning computer or a computer that reprograms itself are therefore
unpersuasive. Cf Intellectual Ventures I LLC v. Capital One Fin. Corp., 850
F.3d 1332, 1342 (Fed. Cir. 2017) (rejecting the argument that the claims "set
forth a unique solution to a problem" and therefore are patent eligible when
"the claims do not recite particular features to yield these advantages").
In sum, we see no indication that these additional elements,
considered alone or together, are anything but generic tools used to perform
an abstract idea. We thus determine that these additional elements do not
integrate the recited abstract ideas into a practical application. See Affinity
Labs of Tex., LLC v. DIRECTV, LLC, 838 F.3d 1253, 1262 (Fed. Cir. 2016)
("In this case, the claims are directed not to an improvement in cellular
telephones but simply to the use of cellular telephones as tools in the aid of a
process focused on an abstract idea. That is not enough to constitute
patentable subject matter."); Revised Guidance, 84 Fed. Reg. at 55
( explaining that courts have identified "merely us[ing] a computer as a tool
17
Appeal2018-005866
Application 14/3 31, 114
to perform an abstract idea" as an example of when a judicial exception may
not have been integrated into a practical application).
Inventive Concept
Finally, we consider whether claim 1 has an inventive concept, that is,
whether the additional claim elements "'transform the nature of the claim'
into a patent-eligible application." Alice, 573 U.S. at 217 (quoting Mayo,
566 U.S. at 78, 79). As discussed above, this requires us to evaluate whether
the additional claim elements add "a specific limitation or combination of
limitations that are not well-understood, routine, conventional activity in the
field" or "simply append[] well-understood, routine, conventional activities
previously known to the industry, specified at a high level of generality."
Revised Guidance, 84 Fed. Reg. at 56.
For the reasons discussed above, the additional elements recited in
claim 1 are generic components. There is no indication that these generic
components perform anything but well-understood, routine, and
conventional activities. Yet Appellants quote DDR Holdings, LLC v.
Hotels.com, L.P., 773 F.3d 1245 (Fed. Cir. 2014) and argue that claim 1 is
not directed to an abstract idea because the claim is "sufficiently rooted in
the computer implementation of semantic processing and, therefore
computer processing itself." App. Br. 10.
We find this argument unpersuasive. In DDR Holdings, the court
determined that the claims at issue "specif[ied] how interactions with the
Internet are manipulated to yield ... a result that overrides the routine and
conventional sequence of events." DDR Holdings, 773 F.3d at 1258. That is,
"[i]nstead of the computer network operating in its normal, expected
18
Appeal2018-005866
Application 14/3 31, 114
manner," "the claims recite[d] an invention that is not merely the routine or
conventional use of the Internet." DDR Holdings, 773 F.3d at 1258-59.
That is not the case here. As noted above, the written description does
not suggest that the additional claim elements are anything but generic
components that operate in a well-understood, routine, and conventional
fashion. These additional elements therefore do not provide "a specific
limitation or combination of limitations that are not well-understood,
routine, conventional activity in the field." Revised Guidance, 84 Fed. Reg.
at 56. Rather, these elements "simply append[] well-understood, routine,
conventional activities previously known to the industry, specified at a high
level of generality." Revised Guidance, 84 Fed. Reg. at 56. As a result, we
agree with the Examiner that claim 1 does not recite an inventive concept.
Conclusion
For at least the above reasons, we agree with the Examiner that claim
1 is directed to an abstract idea and does not contain an inventive concept.
We thus sustain the Examiner's rejection of claim 1 under§ 101.
Section 103 Rejections
Semantic Object
Claim 1 recites "a plurality of semantic objects of a semantic store,
wherein a semantic object is a data object that captures operation business
contexts and intentions in an actionable fashion." App. Br. 16. Appellants
contend that the Examiner erroneously determined that Gadd's user question
teaches or suggests this limitation. See App. Br. 12. According to
Appellants, Gadd's user question "cannot be characterized as a 'semantic
object' that might comprise [sic] a 'semantic store' ... because a question
19
Appeal2018-005866
Application 14/3 31, 114
submitted by a user is not part of the disclosed system but rather external to
that system." App. Br. 12.
We find Appellants' arguments unpersuasive. First, the Examiner
found that Di Fabbrizio teaches "a plurality of semantic objects of a
semantic store" and Gadd teaches "wherein a semantic object is a data object
that captures operation business contexts and intentions in an actionable
fashion." See Final Act. 11, 13. The Examiner concluded it would have been
obvious to combine these teachings to arrive at the disputed limitation. See
Final Act. 11-13. Appellants' argument that Gadd alone does not teach or
suggest "a 'semantic object' that might comprise [sic] a 'semantic store"'
has not persuaded us the Examiner erred because "one cannot show non-
obviousness by attacking references individually where, as here, the
rejections are based on combinations of references." In re Keller, 642 F.2d
413,426 (CCPA 1981).
Second, in response to Appellants' arguments, the Examiner found
that Gadd's user questions are part of Gadd's system because Gadd captures
and stores the questions for future processing. See Ans. 8-9 ( citing Gadd
,r 70, Abstract). The Examiner made a similar finding about Di Fabbrizio's
questions. See Ans. 8-9 (citing Di Fabbrizio ,r,r 30, 34). Appellants have not
addressed either finding. Appellants have therefore not persuaded us the
Examiner erred.
Evaluating the Rule and Modifying the Rule
Claim 1 also recites "evaluating the rule" and "modifying the rule
based upon the evaluating to generate a modified rule." App. Br. 16.
Appellants contend that the Examiner erroneously found Di Fabbrizio
teaches or suggests these limitations for three reasons. First, Di Fabbrizio
20
Appeal2018-005866
Application 14/3 31, 114
concerns "modifying a system customized for one domain into a different
domain rather than enabling a system for one particular domain to be
updated based upon changing conditions." App. Br. 12. Second, Di
Fabbrizio implies that it lacks the claimed functionality because Di
Fabbrizio uses the words "which is time and human-effort consuming." See
App. Br. 12. Third, Di Fabbrizio does not "modify" its follow-up question
because the follow-up question "already includes a reference to a previous
question and therefore does not contain new additional information." App.
Br. 12-13 (emphasis omitted).
We find Appellants' arguments unpersuasive. First, claim 1 does not
preclude "modifying a system customized for one domain into a different
domain" or recite "enabling a system for one particular domain to be
updated based upon changing conditions" as suggested by Appellants. The
claim simply requires "evaluating the rule" and "modifying the rule based
upon the evaluating to generate a modified rule." See App. Br. 16. Because
this argument is not commensurate with the scope of claim 1, we find the
argument unpersuasive.
Second, Di Fabbrizio uses the words "time and human-effort
consuming" to describe a drawback of a prior art algorithm, not
Di Fabbrizio's invention. See Di Fabbrizio ,r 25 (explaining that "[f]or a new
corpus from a different domain," a known "rule-based algorithm" would
"very likely ... have to go over the data set and modify the rules, which is
time and human-effort consuming"). Di Fabbrizio discloses that "[t]he
present inventors propose an alternative, data driven approach to
automatically" perform this function. Di Fabbrizio ,r 25 ( emphases added).
That is, Di Fabbrizio teaches using machine learning to modify rules for a
21
Appeal2018-005866
Application 14/3 31, 114
new corpus from a different domain, thereby avoiding the drawbacks of the
prior art approach. Appellants have thus failed to persuade us that Di
Fabbrizio lacks the claimed functionality because Di Fabbrizio uses the
words "time and human-effort consuming."
Third, the Examiner relied on Di Fabbrizio's follow-up question for
"modifying a semantic object," not "modifying the rule." See Final Act. 12-
13. The Examiner found that Di Fabbrizio's "SimilarityMeasure(Qi)" and
"data-driven machine learning" teach or suggest "modifying the rule." See
Final Act. 12; Ans. 10-11. Appellants have not addressed these findings.
Appellants have therefore not persuaded us that the Examiner erred.
If Appellants contend that Di Fabbrizio's follow-up question does not
correspond to a modified semantic object because the question does not
contain new information, the Examiner also found that Di Fabbrizio teaches
or suggests this limitation because Di Fabbrizio discloses modifying the
context associated with the question. See Ans. 12 (Di Fabbrizio ,r,r 23, 70).
Appellants have not addressed this finding and thus have not persuaded us
the Examiner erred.
Conclusion
For the above reasons, Appellants have not persuaded us the
Examiner erroneously rejected claim 1. We therefore sustain the Examiner's
rejection of claim 1. Because Appellants have not presented separate,
persuasive patentability arguments for claims 2-6, we also sustain the
Examiner's rejections of these claims.
22
Appeal2018-005866
Application 14/3 31, 114
Double Patenting
The Examiner rejected claims 1---6 on the ground of nonstatutory
double patenting over claims 1-6 of Demuth '039. See Final Act. 16-18.
Appellants have not contested this rejection. See App. Br. 15 ("Appellants
are not arguing the rejections based upon double patenting."). We therefore
summarily sustain this rejection.
DECISION
1---6 § 112
1---6 § 101
1-5
6
1---6
Summary
§ 103
§ 103
Double
Patenting
Di Fabbrizio
and Gadd
Di
Fabbrizio,
Gadd, and
Kim
Demuth '039
1---6
1---6
1-5
6
1---6
1---6
No period for taking any subsequent action in connection with this
appeal may be extended under 3 7 C.F .R. § 1.13 6( a)( 1 )(iv). See 3 7 C.F .R.
§ 41.50(±).
AFFIRMED
23