Marshall & Swift/Boeckh, LLC

Patent Trials and Appeals Board

Oct 21, 2020

2020003689 (P.T.A.B. Oct. 21, 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.
14/688,254 04/16/2015 Janakiraman Jagannathan CLOGI.054A 3991
101850 7590 10/21/2020
Knobbe, Martens, Olson & Bear, LLP
CORELOGIC (CLOGI)
2040 Main Street, Fourteen Floor
Irvine, CA 92614
EXAMINER
BARTLEY, KENNETH
ART UNIT PAPER NUMBER
3693
NOTIFICATION DATE DELIVERY MODE
10/21/2020 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):
efiling@knobbe.com
eofficeaction@appcoll.com
jayna.cartee@knobbe.com
PTOL-90A (Rev. 04/07)

UNITED STATES PATENT AND TRADEMARK OFFICE
____________

BEFORE THE PATENT TRIAL AND APPEAL BOARD
____________

Ex parte JANAKIRAMAN JAGANNATHAN

Appeal 2020-003689
Application 14/688,254
Technology Center 3600
____________

Before RICHARD M. LEBOVITZ, FRANCISCO C. PRATS, and
JAMIE T. WISZ, Administrative Patent Judges.

LEBOVITZ, Administrative Patent Judge.

DECISION ON APPEAL
The Examiner rejected claims 1–8, 10, 12–22, and 25–27 under 35
U.S.C. § 103 as obvious and under 35 U.S.C. § 101 as directed to patent
ineligible subject matter. Pursuant to 35 U.S.C. § 134(a), Appellant1 appeals
from the Examiner’s decision to reject the claims. We have jurisdiction
under 35 U.S.C. § 6(b).
We AFFIRM.

1 We use the word “Appellant” to refer to “applicant” as defined in 37
C.F.R. § 1.42. Appellant identifies the real party in interest as is Marshall &
Swift/Boeckh, LLC. Appeal Br. 1.
Appeal 2020-003689
Application 14/688,254

2
STATEMENT OF THE CASE
The Examiner rejected the claims in the Final Office Action as
follows:
1. Claims 1–8, 10, 12, 15–22, 25, and 27 under 35 U.S.C. § 103 as
obvious in view of Swanson et al. (US 2014/0019166 A1, published Jan. 16,
2014) (“Swanson”), Emison (US 2015/0235322A1, published Aug. 20,
2015) (“Emison”), Collins et al. (US 2009/0265193 A1, published Oct. 22,
2009) (“Collins”), and Jeong (US 2015/0043803 A1, published Feb. 12,
2015) (“Jeong”). Final Act. 28.
2. Claims 13, 14, and 26 under 35 U.S.C. § 103 as obvious in view of
Swanson, Emison, Collins, Jeong, and Vande Pol (US 2003/0014342 A1,
published Jan. 16, 2003). Final Act. 44.
3. Claims 1–8, 10, 12–22, and 25–27 under 35 U.S.C. § 101 because
the claimed invention is directed to an abstract idea without significantly
more. Final Act. 24.
Independent claim 1 is directed to a system programmed to carry out
steps in a method. Independent claim 16 is directed to a computer-
implemented method which comprises substantially the same steps recited in
claim 1. We select claim 16 as representative. The claim is reproduced
below, annotated with bracketed numbering for reference to the limitations
in the claim:
16. A computer-implemented method comprising:
[1] receiving identification information associated with a
property, the identification information comprising a location of
the property;
[2] receiving a roofing characteristic associated with a
roofing system associated with the property, the roofing system
comprising a roof of a building on the property;
Appeal 2020-003689
Application 14/688,254

3
[3] receiving a weather characteristic associated with the
location of the property;
[4] receiving an image of the roof of the building on the
property;
[5] analyzing the image using machine learning to
determine a roof image characteristic that indicates whether the
roof of the building includes any missing shingles;
[6] applying a roof condition machine learning model to
the roofing characteristic, the weather characteristic, and the
roof image characteristic to determine a risk of roof damage;
[7] determining a discrepancy between the determined
roof image characteristic and the received roofing
characteristic; and
[8] output an indicator of the discrepancy between the
determined roof image characteristic and the received roofing
characteristic.
Appeal Br. 28 (Claims Appendix).

CLAIM 16
The first four steps of claim 16 involve receiving information about a
property (step [1]), a roof characteristic of a roof on the property (step [2]),
weather at the property location (step [3]), and an image of the roof (step
[4]). In step [5] of the claim, machine learning is used “to determine a roof
image characteristic that indicates whether the roof of the building includes
any missing shingles.” Machine learning is employed to analyze the image,
identify a roof image characteristic, and determine if a shingle is missing
using the roof image characteristic. A “risk of roof damage” is determined in
step [6] using “a roof condition machine learning model” that is applied to
the roofing characteristic of step [2], the weather characteristic of step [3],
and the roof image characteristic determined in step [5]. A discrepancy
between the determined roof image characteristic (step [5]) and the received
Appeal 2020-003689
Application 14/688,254

4
roofing characteristic (step [2]) is determined and then outputted in steps [7]
and [8], respectively.
The Specification lists various “roofing system data” about a roof that
can be stored in a database and provided in step [2], such as roof age, roof
dimensions, roof slope, roof aspect, roof pitch, roof direction, roof shape,
roof type, roof covering material type, evidence of damage, etc. Spec. ¶¶ 36,
47. Whatever “roofing characteristic” is received in step [2], it also must be
the same “roof image characteristic” of step [5] which “indicates whether the
roof of the building includes any missing shingles” because the two are
compared for a discrepancy in step [7].

OBVIOUSNESS REJECTIONS
The Examiner found Swanson describes receiving the information
described in steps [1]–[4] of claim 16. Final Act. 30–32. The Examiner
further found that Swanson analyzed patterns in images to determine roof
condition as in step [5] of the claim (Final Act. 33–34), but not using
machine learning algorithms (Final Act. 36) to detect missing shingles (Final
Act. 39). For machine learning, the Examiner cited Emison, which also
describes determining the condition of a roof on a property (Final Act. 36–
37). For determining missing shingles as required by step [5] of the claim,
the Examiner cited Collins (Final Act. 37–39). For step [6] of determining
the “risk of roof damage” using the roofing characteristic, the weather
characteristic, and the roof image characteristic, the Examiner cited Swanson
and Emison. Final Act. 34–35, 36–37. The Examiner found that Swanson
determined a discrepancy between characteristics as recited in step [7] of the
claim, but not an output of the discrepancy. Final Act. 35. The Examiner
found that Jeong describes outputting differences in images, and determined
Appeal 2020-003689
Application 14/688,254

5
it would have been obvious to do so to show differences between a real and
model image. Final Act. 39.
Appellant contends that Swanson does not suggest determining the
risk of roof damage. Appeal Br. 18. Appellant argues that Swanson focuses
instead on assessing roof aging. Id. Appellant also argues that
“[d]etermining property liability or assessing property value in the context of
Swanson is not the same as determining a risk of roof damage, and Swanson
does not teach or suggest such a connection.” Id. at 19.
This argument does not persuade us that the Examiner erred. Swanson
teaches that “[w]hen insurers are writing a [property insurance liability]
policy on an individual property, the real value roof replacement cost is
often not known and can be greatly misrepresented if the condition of the
roof is poor, old, or otherwise degraded.” Swanson ¶ 2. In writing the policy
to determine the cost of roof replacement Swanson further teaches that
“knowledge of rooftop condition is important because natural hazards, such
as wind, hail, and extreme heat, pose a much higher risk of causing
significant damage to old or aged roof material versus younger or
unweathered roof material that posses[s] more structural integrity to
withstand damaging events.” Id. (emphasis added.) Thus, Swanson teaches
that an insurance liability policy takes into account the roof condition and
the risk of the occurrence of damage to it because of natural hazards, such as
weather conditions (“wind, hail, and extreme heat”).
Swanson further discloses: determining “the risk associated with
selected features in the geographic area of interest is assessed by the
insurance practitioner” (Swanson ¶ 26), obtaining data “for different kinds
of liability, such as wind, flood, and fire phenomenology and risk
Appeal 2020-003689
Application 14/688,254

6
assessment” (id. ¶ 32), and using “library of spectral images for vegetation,
water quality, and other materials . . . to extract those aspects and document
additional hazards or variables that impact property state and potential
liability or intrinsic risk” (id. ¶ 39). Each of these passages describes
determining “risk” based on various types of data collected about a
particular location and the use of the data to determine “potential liability or
intrinsic risk” which in the context of Swanson would be understood to
mean risk of damage to a roof, for example, when there are high winds or
neighboring vegetation that could pose risk of fire.2 In other words, the
liability insurance determination is made based on knowledge of the risk that
a roof might be damaged by weather conditions or nearby vegetation. Thus,
the preponderance of the evidence supports the Examiner’s determination
that Swanson teaches determining the risk of roof damage as recited in the
rejected claims.
Appellant addressed Swanson’s disclosure regarding roof damage and
roof aging (e.g., Reply Br. 7–8), but did not consider that such
determinations are a part of Swanson’s determination about the risk that the
roof could be damaged in determining the liability policy. Appellant
provides no alternative explanation as to why Swanson would be
determining the potential liability and risk that a roof will be damaged.
Appellant acknowledges in the Reply Brief that Swanson “provides a
general statement that certain weather events pose a higher risk of causing

2 “While the above example has been given in terms of assessing rooftop
conditions, the system is equally applicable to vegetation and other features
of a landscape. For example, dry trees and brush very close to a home pose a
greater fire risk than moist healthy trees located 150 ft from a home.”
Swanson ¶ 34.
Appeal 2020-003689
Application 14/688,254

7
significant damage to certain types of roofs,” but contends that Swanson
“does not provide any disclosure as to how roof damage risk would
specifically be determined.” Reply Br. 8. This argument is not persuasive
because the Examiner specifically cited Emison for teaching machine
learning to determine risk of damage as recited step [6] of claim 16. Final
Act. 36–37. Appellant did not identify a defect in the Examiner’s findings
regarding Emison’s disclosure.
Appellant also argues that the cited publications do not teach or
suggest step [7] of the claim of “determining a discrepancy between the
determined roof image characteristic and the received roofing
characteristic.” Appeal Br. 19–20.
The claim requires “[5] analyzing the image using machine learning to
determine a roof image characteristic that indicates whether the roof of the
building includes any missing shingles.” The Examiner did not interpret the
“roof image characteristic” to be the missing shingle, itself. As explained by
the Examiner, “[d]etecting a non-shingle roof by spectra analysis would
indicate there are no shingles on the roof and therefore they might be
missing.” Final Act. 39. In other words, a spectral analysis of the roof could
serve as a roof characteristic that is used to determine whether a shingle is
missing.
The Examiner found that Swanson’s disclosure of rectifying data “by
a series of processes that correct for radiometric, atmospheric, and geospatial
attributes tied to the aerial data” (Swanson ¶ 26) and performing corrections
on radiometric data meet the limitation of step [7] of the claim because
Appeal 2020-003689
Application 14/688,254

8
“[i]nherent with correction is determining a discrepancy between the
received data and the final roof generated image.”3 Final Act. 35.
The data described in paragraphs 26 and 27 of Swanson describes
rectifying and correcting data (such as using “laboratory calibration data”),
but, as explained below, there is not sufficient detail in these passages to
convey step [7] of claim 16 of “determining a discrepancy between the
determined roof image characteristic and the received roofing
characteristic.”
The passage in paragraph 26 states that “at step 14, the data is
rectified by a series of processes that correct for radiometric, atmospheric,
and geospatial attributes tied to the aerial data.” This statement does not
explain that the rectification is done based on a discrepancy between the
determined roof image characteristic and the received roofing characteristic.
The Examiner’s conclusion that this step inherently detects a discrepancy
between the two characteristics is insufficient because the Examiner did not
identify where in Swanson’s disclosure the two recited characteristics are
disclosed.
In paragraph 27, a correction is performed “on the collected data using
laboratory calibration data 44.” The Examiner did not explain how
“laboratory calibration data” serves either as a determined roof image

3 “At step 42, radiometric correction is performed on the collected data using
laboratory calibration data 44. This correction gives a scale to the pixel
values, e.g. the monochromatic scale of 0 to 255 will be converted to actual
radiance values. In step 46, the HSI data undergoes a process of geo-
rectification using the tagged IMU geo coordinates and attitudinal data 48.
This process aligns image data to a set of geographical coordinates so that
each pixel is assigned a geographical coordinate.” Swanson ¶ 27.
Appeal 2020-003689
Application 14/688,254

9
characteristic obtained from image analysis as in step [5] and the received
roofing characteristic of step [4].
An examiner bears the initial burden of presenting a prima facie case
of obviousness. In re Huai-Hung Kao, 639 F.3d 1057, 1066 (Fed. Cir.
2011). The Examiner did not meet the burden of establishing that step [7] of
claim 16 is made obvious based on the disclosures of Swanson, Emison,
Collins, and Jeong. All the claims require step [7]. The Examiner did not
find that Vande Pol makes up for this deficiency. Consequently, claims 1–8,
10, 12, 15, 17–22, 25, and 27 are reversed for the same reasons.

REJECTION BASED ON § 101
Principles of Law
Under 35 U.S.C. § 101, an invention is patent-eligible if it claims a
“new and useful process, machine, manufacture, or composition of matter.”
However, not every discovery is eligible for patent protection. Diamond v.
Diehr, 450 U.S. 175, 185 (1981). “Excluded from such patent protection are
laws of nature, natural phenomena, and abstract ideas.” Id. The Supreme
Court articulated a two-step analysis to determine whether a claim falls
within an excluded category of invention. Alice Corp. v. CLS Bank Int’l, 573
U.S. 208 (2014); Mayo Collaborative Servs. v. Prometheus Labs., Inc., 566
U.S. 66, 75–77 (2012).
In the first step, it is determined “whether the claims at issue are
directed to one of those patent-ineligible concepts.” Alice, 573 U.S. at 217. If
it is determined that the claims are directed to an ineligible concept, then the
second step of the two-part analysis is applied in which it is asked “[w]hat
Appeal 2020-003689
Application 14/688,254

10
else is there in the claims before us?” Id. (alteration in original). The Court
explained that this step involves
a search for an “‘inventive concept’”—i.e., an element or
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) (citing from Mayo, 566
U.S. at 75–77).
Alice, relying on the analysis in Mayo of a claim directed to a law of
nature, stated that in the second part of the analysis, “the elements of each
claim both individually and ‘as an ordered combination’” must be
considered “to determine whether the additional elements ‘transform the
nature of the claim’ into a patent-eligible application.” Alice, 573 U.S. at
217.
The PTO published revised guidance on the application of 35 U.S.C.
§ 101. USPTO’s January 7, 2019 Memorandum, 2019 Revised Patent
Subject Matter Eligibility Guidance, 84 Fed. Reg. 50 (Jan. 7, 2019)
(“Eligibility Guidance”). This guidance provides additional direction on how
to implement the two-part analysis of Mayo and Alice.
Step 2A, Prong One, of the Eligibility Guidance, looks at the specific
limitations in the claim to determine whether the claim recites a judicial
exception to patent eligibility. In Step 2A, Prong Two, the claims are
examined to identify whether there are additional elements in the claims that
integrate the exception in a practical application, namely, is there a
“meaningful limit on the judicial exception, such that the claim is more than
a drafting effort designed to monopolize the judicial exception.” Eligibility
Guidance, 84 Fed. Reg. 54 (Prong Two).
Appeal 2020-003689
Application 14/688,254

11
If the claim recites a judicial exception that is not integrated into a
practical application, then as in the Mayo/Alice framework, Step 2B of the
Eligibility Guidance instructs us to determine whether there is a claimed
“inventive concept” to ensure that the claims define an invention that is
significantly more than the ineligible concept itself. Eligibility Guidance, 84
Fed. Reg. 56.
With these guiding principles in mind, we proceed to determine
whether the claimed subject matter in this appeal is eligible for patent
protection under 35 U.S.C. § 101.
Discussion
Claim 16 is directed to a computer-implemented method. Methods are
“processes,” which is a statutory category of patent-eligible subject matter
listed in 35 U.S.C. § 101. Thus, because the claim falls into a statutory
category of patent-eligible subject matter, following the first step of the
Mayo analysis, we proceed to Step 2A, Prong One, of the Eligibility
Guidance.
Step 2A, Prong One
In Step 2A, Prong One, of the Eligibility Guidance, the specific
limitations in the claim are examined to determine whether the claim recites
a judicial exception to patent eligibility, namely, whether the claim recites an
abstract idea, law of nature, or natural phenomenon. The Eligibility
Guidance directs examiners to search each limitation in the claim to
determine whether it recites abstract an abstract idea. Eligibility Guidance,
84 Fed. Reg. 54.
The Examiner found the steps in the claim “cover performance of the
[claim] limitation[s] as mental processes” (Final Act. 25), which is one of
Appeal 2020-003689
Application 14/688,254

12
the categories of patent-ineligible abstract ideas identified by the courts and
set forth in the Eligibility Guidance. Eligibility Guidance, 84 Fed. Reg. 52.
Steps [1]–[4] of the claim involve receiving information about a
property (step [1]), a roof on the property (step [2]), weather at the property
location (step [3]), and an image of the roof (step [4]). These steps are
information gathering and not abstract. The last step [8] of the claim in
which the discrepancy of [7] is output, is also not an abstract step, but rather
is extra-solution activity. Eligibility Guidance, 84 Fed. Reg. 55.
Steps [5] and [6] of the claim use machine learning to determine a
roof image characteristic and to determine a risk of roof damage. The
Examiner stated that machine learning is a mathematical algorithm and
“computer processing an algorithm to learn (machine learning) is the same
or similar to a person in their mind processing an algorithm to learn.” Final
Act. 25. We do not agree that steps [5] and [6] recite abstract ideas.
With respect to the Examiner’s finding that machine learning is a
mathematical algorithm and therefore an abstract idea (Eligibility Guidance,
84 Fed. Reg. 52), the claim does not specifically recite the algorithm, but
instead only states it generically. Thus, in accordance with the Eligibility
Guidance, it is not an abstract idea. As explained in the PEG Update:4
A claim does not recite a mathematical concept (i.e., the claim
limitations do not fall within the mathematical concept
grouping), if it is only based on or involves a mathematical
concept. For example, a limitation that is merely based on or
involves a mathematical concept described in the specification

4 Available at
https://www.uspto.gov/sites/default/files/documents/peg_oct_2019_update.p
df (last accessed Jul. 3, 2020) (“PEG Update”).

Appeal 2020-003689
Application 14/688,254

13
may not be sufficient to fall into this grouping, provided the
mathematical concept itself is not recited in the claim.
PEG Update 3 (footnotes omitted.)
Example 39 of the Eligibility Examples discloses a claim for “training
a neural network” and recites steps in which a neural network is trained.
Subject Matter Eligibility Examples (Jan. 7, 2019) 8 (“SMEE”). A neural
network is a type of machine learning algorithm. Id. The neural network is
not identified in the example as either a mental process or a mathematical
concept (another category of abstract ideas identified by the courts and the
Eligibility Guidance as an abstract idea. Eligibility Guidance, 84 Fed. Reg.
52). Therefore, in accordance with the Eligibility Guidance, we conclude
that steps [5] and [6] which recite “machine learning,” without reciting the
specific mathematical algorithm or implementation of it, do not recite an
abstract idea.
The Examiner’s statement that machine learning is a mental process
because it is the “same or similar” to what a person does in their mind to
learn (Final Act. 25) is unreasonable because machine learning uses, for
example, “a logistic regression model, a Generalized Linear Model, a
Support Vector Machine, a Naive Bayes model, a Random Forest, or other
statistical algorithm” (Spec. ¶ 7) and there is no evidence that a person uses
these mathematical models when learning. See also Appeal Br. 8
(“Furthermore, humans do not analyze information in the same manner as a
computing system running supervised or unsupervised pattern recognition
because, as stated above, humans and computing systems do not intake,
analyze, or learn from information in the same manner.”).
Step [7] of the claim recites determining a discrepancy between the
determined roof image characteristic (step [5]) and the received roofing
Appeal 2020-003689
Application 14/688,254

14
characteristic (step [2]). The claim does not recite any limitation as to how
the discrepancy is determined. For example, it could be done by simply
observing that there is a difference between the received characteristic and
the determined characteristic. An observation that could be performed in the
human mind is a “mental process” and an ineligible abstract ideas (“Mental
processes – concepts performed in the human mind (including an
observation, evaluation, judgment, opinion).” Eligibility Guidance, 84 Fed.
Reg. 52 (footnotes omitted)). Step [7] therefore recites an abstract idea.
Accordingly, we proceed to Step 2A, Prong Two, of the Eligibility
Guidance.
Step 2A, Prong Two
Prong Two of Step 2A under the 2019 Eligibility Guidance asks
whether there are additional elements that integrate the exception into a
practical application. Integration into a practical application is evaluated by
identifying whether there are additional elements individually, and in
combination, which go beyond the judicial exception. Eligibility Guidance,
84 Fed. Reg. at 54–55. As explained in the Eligibility Guidance, integration
may be found when an additional element “reflects an improvement in the
functioning of a computer, or an improvement to other technology or
technical field” or “applies or uses the judicial exception in some other
meaningful way beyond generally linking the use of the judicial exception to
a particular technological environment, such that the claim as a whole is
more than a drafting effort designed to monopolize the exception.” Id. at 55.
Appellant argues that the claim is not directed to an abstract idea for
the reasons described in McRO, Inc. v. Bandai Namco Games Am. Inc., 837
F.3d 1299, 1307–08 (Fed. Cir. 2016), namely that the rejected claim is
Appeal 2020-003689
Application 14/688,254

15
“directed to a specific set of rules for improving technology related to the
processing of images for roof condition evaluation.” Appeal Br. 10.
Appellant identifies step [1]–[6] of the claim as reciting a “rule-based
process can be used to produce a more accurate determination of the
condition of a roof than can be produced by existing systems that merely
rely on aerial imagery.” Id. at 11.
Steps [1]–[4] of claim 16 involve receiving various type of
information. The claim does not recite a specific way in which the
information is received by the system. We consider these steps to be data
gathering steps. As explained in the Eligibility Guidance, “a mere data
gathering such as a step of obtaining information about credit card
transactions so that the information can be analyzed in order to detect
whether the transactions were fraudulent” is insignificant extra-solution
activity which is insufficient to integrate the judicial exception into a
practical application. Eligibility Guidance, 8 Fed. Reg. 55 (fn. 31).
Step [5] recites “analyzing the image using machine learning to
determine a roof image characteristic that indicates whether the roof of the
building includes any missing shingles.” In step [7] a discrepancy between
the determined roof image characteristic of step [5] and the received roofing
characteristic of step [2] is determined. Step [5] invokes “machine learning,”
but it does not describe a specific way in which machine learning is applied
to determine the “roof image characteristic.” The Specification identifies
various machine learning algorithms by name, but with no specific
information on how they are implemented to determine a roof image
characteristic. Spec. ¶¶ 7, 26, 50. Thus, unlike McRO, there is no specific
Appeal 2020-003689
Application 14/688,254

16
rule in the claim that integrates the judicial exception recited in step [7] into
a practical application.
In McRO, 837 F.3d at 1307–08, the claim recited a series of steps that
“produce[d] lip synchronization and facial expression control of said
animated characters.” The court found that the “claimed process uses a
combined order of specific rules that renders information into a specific
format that is then used and applied to create desired results: a sequence of
synchronized, animated characters.” Id. at 1315. McRO found that the
recited rules “are limiting in that they define morph weight sets as a function
of the timing of phoneme sub-sequences.” Id. at 1313. The claims were
found to be directed to a “technological improvement over the existing,
manual 3-D animation techniques.” Id. at 1316.
Appellant has not directed us to specific rules which are applied to
implement the machine learning process recited in the step [5]. The claim
recites that the image is analyzed by machine learning to determine a “roof
image characteristic” which is compared to the received roof characteristic
in step [7], but there is no disclosure in the claim, or Specification, of how
the image analysis is implemented and what aspects of the image are
subjected to the machine learning algorithm. Thus, steps [5] and [7] do not
“embody a concrete solution to a problem” because they lack “the specificity
required to transform a claim from one claiming only a result to one
claiming a way of achieving it.” Interval Licensing LLC v. AOL, Inc., 896
F.3d 1335, 1343 (Fed. Cir. 2018) (citing SAP Am., Inc. v. InvestPic, LLC,
890 F.3d 1016, 1021–22 (Fed. Cir. 2018).
Step [8], in which an indicator of the discrepancy is output, does not
require a specific graphic interface or functionality to achieve the output.
Appeal 2020-003689
Application 14/688,254

17
Thus, this step is distinguishable from cases in which a graphic interface was
found to confer patent eligibility. For example, in Core Wireless Licensing
S.A.R.L. v. LG Electronics, Inc., 880 F.3d 1356 (Fed. Cir. 2018) the court
found the claim “directed to an improved user interface for computing
devices, not to the abstract idea of an index.” The representative claim in
Core Wireless was directed to a computing device comprising a display
screen “configured to display on the screen a menu listing one or more
applications” and “to display on the screen an application summary that can
be reached directly from the menu.” Core Wireless, 880 F.3d at 1359
(emphasis removed). The court found the claims patent-eligible because the
claim “restrains the type of data that can be displayed in the summary
window” and specified “a particular manner by which the summary window
must be accessed.” Id. at 1362–1363. Thus, there was a recited functionality
associated with the menu. Here, there is no detail or functionality recited in
the claim associated with the output of an indicator of the discrepancy.
A “risk of roof damage” is determined in step [6] using “a roof
condition machine learning model” that is applied to the roofing
characteristic of step [2], the weather characteristic of step [3], and the roof
image characteristic determined in step [5]. This step, alone, as discussed
above, does not recite a mental process. However, this step is also
independent of step [7], and is not involved in the output of step [8], and
therefore cannot serve to integrate the mental process recited in step [7] into
a practical application. Moreover, there is no disclosure in the claim of the
specific rules to implement the roof condition machine learning model. We
also have not been guided to a specific description of it in the Specification.
Appeal 2020-003689
Application 14/688,254

18
Citing steps [1]–[6] of the claim, Appellant argues that the claim
recites “a specific manner of analyzing an image to determine a roof image
characteristic that indicates whether the roof of the building includes any
missing shingles and of determining a risk of roof damage.” Appeal Br. 13.
Appellant also argues that the claim “as a whole is directed to a particular
improvement in electronic analysis of images to detect roof characteristics,
such as missing shingles.” Id. at 14.
The improvement relied upon by Appellant, as explained above, is not
accomplished by a specific rule or implementation. Rather, the claim, and
the written descriptive support in the Specification, refer only to generic
machine learning algorithms with no explanation as to how they are
implemented to improve analysis of images and detection of characteristics
as required by step [5] of the claim. Spec. ¶¶ 7, 26, 50. As explained in
Dropbox, Inc. et al. v. Synchronism Technologies, Inc., 2020 WL 3400682 at
*4 (Fed. Cir. 2020) (nonprecedential), it has been “consistently held that an
‘inventive concept’ exists when a claim ‘recite[s] a specific, discrete
implementation of the abstract idea’ where the ‘particular arrangement of
elements is a technical improvement over [the] prior art.’” [BASCOM Glob.
Internet Servs. v. AT&T Mobility LLC, 827 F.3d 1341, 1350 (Fed. Cir.
2016)].
The patent has to describe how to solve the problem in a
manner that encompasses something more than the “principle in
the abstract.” See [ChargePoint, Inc. v. SemaConnect, Inc., 920
F.3d 759, 769 (Fed. Circ. 2019)] (explaining that an invention
may not be patent eligible if the “claims ‘were drafted in such a
result-oriented way that they amounted to encompassing the
‘principle in the abstract’ no matter how implemented’”
(quoting Interval Licensing LLC v. AOL, Inc., 896 F.3d 1335,
1343 (Fed. Cir. 2018))); see also Finjan, Inc. v. Blue Coat Sys.,
Appeal 2020-003689
Application 14/688,254

19
Inc., 879 F.3d 1299, 1305 (Fed. Cir. 2018) (“[A] result, even an
innovative result, is not itself patentable.”).
Dropbox at *4.
While Appellant states that the claim is for determining missing
shingles and determining risk of roof damage, the claim also recites steps [7]
and [8] of determining a discrepancy between characteristics and then
displaying the output. Thus, the claim can also be characterized as method of
determining discrepancies between a characteristic of a received image and
an image analyzed by conventional machine learning algorithms (step [5] of
the claim). As discussed above, the additional element of machine learning
and the lack of any rule by which the discrepancy is determined is not
sufficient to integrate the mental process recited in claim 16 into a practical
application.
In sum, Appellant has not identified an additional element in the
claim, beyond the abstract idea, that integrates the judicial exception into
practical application.
Step 2B
Because we determined that the judicial exception is not integrated
into a practical application, we proceed to Step 2B of the Eligibility
Guidelines, which asks whether there is an inventive concept beyond the
judicial exception. In making this Step 2B determination, we must consider
whether there are specific limitations or elements recited in the claim “that
are not well-understood, routine, conventional activity in the field, which is
indicative that an inventive concept may be present” or whether the claim
“simply appends well-understood, routine, conventional activities previously
known to the industry, specified at a high level of generality, to the judicial
exception, indicative that an inventive concept may not be present.”
Appeal 2020-003689
Application 14/688,254

20
Eligibility Guidance, 84 Fed. Reg. 56. We must also consider whether the
combination of steps performs “in an unconventional way and therefore
include[s] an ‘inventive concept,’ rendering the claim eligible at Step 2B.”
Id. In this part of the analysis, we consider “the elements of each 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.
Appellant states that the limitations recited in the claim “are directed
to improving computing systems that evaluate roof conditions by using
weather data and image processing based on machine learning.” Appeal Br.
15. Appellant states that the claimed features recited in the claim “can
improve the accuracy of computing systems that evaluate roof conditions by
applying weather data and the results of machine learning-based image
analysis to a machine learning model.” Id. at 16. Appellant also argues that
existing systems “do not evaluate roof conditions using weather data in
combination with machine learning-based image analysis that identifies
missing shingles based on roof images.” Id. at 17.
The focus of our analysis is on step [7] because it recites a mental
process, and therefore it must be determined whether the additional elements
in the claim provide an inventive concept when considered in combination
with step [7]. The additional element cited by Appellant is step [5] which
applies machine learning in a conventional way to perform the image
analysis. Spec. ¶¶ 7, 26, 50. Thus, this step alone does not provide an
inventive concept. The step of receiving the roofing characteristic is recited
in step [2]. The roofing characteristic is used in determining the discrepancy
in step [7]. Step [2] does not limit how the roofing characteristic is received
Appeal 2020-003689
Application 14/688,254

21
and thus could be accomplished by conventional processes. Appellant did
not explain how the steps [2] and [5], using known and conventional
techniques, when combined, provide an inventive concept.
The claim also determines the risk of roof damage in step [6]. This
step of the claim is not integrated with step [7], but it is separate step which
relies on the roof image characteristic of step [5]. Step [6] does not add
“significantly more” to the judicial exception of step [7] because it is
independent and unrelated to how the discrepancy is determined.
The rejection based on Section 101 of all the claims is affirmed.
Claims not separately argued fall with claim 16. 37 C.F.R. § 41.37(c)(1)(iv).

CONCLUSION
In summary:
Claims
Rejected
35 U.S.C.
§
Reference(s)/Basis Affirmed Reversed
1–8, 10, 12,
15–22, 25,
27
103 Swanson, Emison,
Collins, Jeong
1–8, 10, 12,
15–22, 25,
27
13, 14, 26 103 Swanson, Emison,
Collins, Jeong,
Vande Pol
13, 14, 26
1–8, 10, 12–
22, 25–27
101 Subject matter
eligibility
1–8, 10, 12–
22, 25–27

Overall
Outcome
1–8, 10, 12–
22, 25–27

Appeal 2020-003689
Application 14/688,254

22

TIME PERIOD
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