Microsoft Technology Licensing, LLC

Patent Trials and Appeals Board

Jul 6, 2020

14297810 - (D) (P.T.A.B. Jul. 6, 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.
14/297,810 06/06/2014 Benjamin E. Rampson 341532.01 4528
69316 7590 07/06/2020
MICROSOFT CORPORATION
ONE MICROSOFT WAY
REDMOND, WA 98052
EXAMINER
NGUYEN, MAIKHANH
ART UNIT PAPER NUMBER
2176
NOTIFICATION DATE DELIVERY MODE
07/06/2020 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):
chriochs@microsoft.com
usdocket@microsoft.com
PTOL-90A (Rev. 04/07)
UNITED STATES PATENT AND TRADEMARK OFFICE
____________

BEFORE THE PATENT TRIAL AND APPEAL BOARD
____________

Ex parte BENJAMIN E. RAMPSON, CHRISTOPHER J. GROSS,
POORNIMA HANUMARA, and ANUPAM GARG
____________

Appeal 2019-002520
Application 14/297,810
Technology Center 2100
____________

Before JOHN A. JEFFERY, ERIC S. FRAHM, and
JOHNNY A. KUMAR Administrative Patent Judges.

KUMAR, Administrative Patent Judge.

DECISION ON APPEAL
Appellant1 seeks our review under 35 U.S.C. § 134(a) from a final
rejection of claims 1, 4–9, 11–16, and 19–24. 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 (2017). Appellant identifies the real party in interest as Microsoft
Technology Licensing, LLC. Appeal Br. 3.
Appeal 2019-002520
Application 14/297,810

2

STATEMENT OF THE CASE
The Invention
According to the Abstract, in the invention a user interaction input is
detected, indicating that a user is interacting with structured data. The user
interaction input is identified as a pattern for which a summary view is to be
generated. The summary view of the structured data is generated, based
upon the detected pattern, and is displayed to the user.
See Spec., Abstract.
Exemplary Claims
Independent claims 1, 12, and 19 exemplify the claims at issue and
read as follows:
1. A computer-implemented method, comprising:
displaying data items, from a document, in a structured data
display that defines relationships between the data items;
receiving an input indicative of a user interaction that directly
interacts with a set of the data items displayed on the structured
data display;
accessing a set of pattern definition rules, each pattern
definition rule defining when user interaction inputs conform to
a pattern for which a summary view is to be generated by an
automated summary view generation system;
in response to determining that the user interaction with the
structured data display conforms to one of the pattern definition
rules, automatically triggering the automated summary view
generation system to:
automatically generate a set of summary data based on
commonality between attributes of the set of data items;
and automatically generate a summary view that visually
represents the set of summary data;
displaying a user selection mechanism for selection of the
summary view; and
modifying the document to insert the selected summary view
into the document in addition to the data items.
Appeal 2019-002520
Application 14/297,810

3

12. A computer system, comprising:
at least one processor; and
memory storing instructions executable by the at least one
processor, wherein the instructions configure the computer
system to:
display data items in a structured data display that defines
relationships between the data items;
detect one or more user interactions with the structured
data display that directly interacts with a first set of the data
items displayed on the structured data display;
automatically determine that a second set of the data
items in the structure, other than the first set of data items,
correspond to the first set of data items, and have common
values in the structure,
access a pattern definition rule defining a pattern for
which a summary view is to be generated by an automated
summary view generation system;
based on applying the pattern definition rule to the first
and second sets of data items, automatically trigger the
automated summary view generation system to:
automatically generate a set of summary data based
on the first and second sets of data items; and
automatically display a summary view that visually
represents the set of summary data.

19. A computer-implemented method comprising:
displaying a table structure on a user interface display, the
table structure comprising a plurality of data items in rows and
columns;
based on a data item sort user input, sorting a first set of
data items, of the plurality of data items, that are displayed in a
first column of the table structure;
based on a data item selection user input, selecting a range
of the sorted data items in the first column;
automatically identifying a second set of data items, of the
plurality of data items, that are displayed in a second column of
Appeal 2019-002520
Application 14/297,810

4

the table structure and correspond to the selected range of the
sorted data items in the first column;
access a pattern definition rule defining when user inputs
conform to a pattern for which a summary view is to be generated
by an automated summary view generation system;
based on applying the pattern definition rule to the data
item sort user input and the data item selection user input,
automatically triggering the automated summary view
generation system to:
automatically calculate summary data based on the
first and second sets of data items, wherein the summary
data is separate from and corresponds to the first and
second sets of data items; and
automatically display, on the user interface display,
a summary data view showing the summary data.
Appeal Br. 22, 24, and 25–26 (Claims App., claims reformatted for clarity).

The Prior Art Supporting the Rejection on Appeal
As evidence of unpatentability under 35 U.S.C. § 103, the Examiner
relies on the following prior art:

Name Reference Date
Noy US 2003/0016252 A1 Jan. 23, 2003
Lewis-Bowen US 2007/0050697 A1 Mar. 1, 2007
Fan US 2013/0097177 A1 Apr. 18, 2013

Appeal 2019-002520
Application 14/297,810

5

The Rejection on Appeal
Claims 1, 4–9, 11–16, and 19–24 are rejected under 35 U.S.C. § 103
as being unpatentable over Lewis-Bowen in view of Noy and Fan.
ANALYSIS
We have reviewed the rejections in light of Appellant’s arguments
that the Examiner erred. For the reasons explained below, we agree with the
Examiner’s unpatentability determinations.

Independent Claim 1
The Examiner determines that Lewis-Bowen teaches, inter alia,
automatically generate summary data based on the commonality between
attributes of the identified set of data items, because Lewis-Bowen teaches
collapsing rows of a selected column such that last names with the first five
letters in common are collapsed together. Final Act. 4 (citing Lewis-Bowen
¶ 69, Figs. 9, 10). The Examiner maps Noy’s teaching of identifying
patterns, for example identical attributes, in the set of last selections by the
user of a user targeted object to the claimed accessing a set of pattern
definition rules. Final Act. 5–6 (citing Noy ¶¶ 130–131). The Examiner
determines that Fan teaches, inter alia, automatically generating a set of
summary data based on the commonality between attributes of the set of
data items, and automatically generating a summary view that visually
represents the set of summary data (Final Act. 5–6; citing Fan ¶ 75, Fig. 18);
and modifying the document to insert the selected summary view into the
document in addition to the data items. Final Act. 6 (citing Fan ¶ 42, Fig. 2).
The Examiner concludes, in relevant part, that it would have been obvious to
Appeal 2019-002520
Application 14/297,810

6

a skilled artisan at the time of the invention to combine the teachings of
Lewis-Bowen with those of Fan, to:
provide the user with optimal chart choices and guide[] the user
to make better choices in creating visualizations (Fan, [0005],
lines 10–14) and the pivot chart recommendation process may
provide the best summarization of data that are of higher value
to the user when the user has [a] repetitive dataset (Fan, [0075],
lines 3–5) such as those taught by Lewis-Bowen.
Final Act. 6–7.
Appellant disagrees that the combination of Lewis-Bowen, Noy, and
Fan teach all of the limitations of claim 1. In particular, Appellant argues
that the express purpose of Lewis-Bowen is to reduce data in a spreadsheet
view (Appeal Br. 9; citing Lewis-Bowen ¶ 10), and it therefore teaches away
from an additional summary view. Appeal Br. 10; Reply Br. 1–2. Appellant
additionally argues that Noy does not teach patterns or using pattern
detection to trigger an automated summary view generation (Appeal Br. 10–
11; Reply Br. 2–3); nor would there be a reason to combine the teachings of
Lewis-Bowen and Noy. Appeal Br. 11–12; Reply Br. 3. Further, Appellant
argues, Fan does not teach generating pivot tables based on direct user
interaction with data items themselves, or triggering automated summary
view generation based on determining a pattern from user interactions.
Appeal Br. 12. We disagree with Appellant’s arguments as explained below.
A. Whether the combination of Lewis-Bowen, Noy, and Fan teaches a
summary view

We disagree with Appellant’s argument that Lewis-Bowen teaches
away from generating a summary view. Appeal Br. 9–10; Reply Br. 1–2.
Notably, the Examiner determines, and we agree, that “[t]he primary
purpose of Lewis-Bowen is to manage presentations of large spreadsheets[,]
Appeal 2019-002520
Application 14/297,810

7

and any solution to summarize or otherwise reduce the amount of
information presented would be motivation to modify Lewis-Bowen.” Ans.
29. Paragraph 10 of Lewis-Bowen teaches that, “[t]he present invention
provides a facility for a user controlled collapsing of rows to facilitate
manipulation of large spreadsheets. . . .This facility greatly reduces the
number of rows that are presented at a time.” Lewis-Bowen ¶ 10. Therefore,
Lewis-Bowen makes clear that the purpose is not, as Appellant argues, a
blanket reduction of data in a spreadsheet view (Appeal Br. 9–10; Reply Br.
1–2), but more specifically, to “facilitate manipulation of large spreadsheets”
and “reduce[] the number of rows that are presented at a time.” Lewis-
Bowen ¶ 10. As such, we agree with the Examiner that the modification of
Lewis-Bowen to automatically generate a summary view, as taught by Fan
(see Final Act. 5–6), is not counter to the purpose of Lewis-Bowen. See Ans.
29, 33.
We further agree that it would have been obvious to a skilled artisan
at the time of the invention to modify Lewis-Bowen to have a summary
view, as taught by Fan, for the reasons proffered by the Examiner. See Final
Act. 6–7. As the Examiner notes, “both [Lewis-Bowen and Fan] create a
visual summary representation of data” (Final Act. 6), and we agree that
Fan’s teaching of displaying information in graphical or chart form (see Fan,
Figs. 2, 18; Final Act. 5–7) accomplishes Lewis Bowen’s goal of facilitating
manipulation of large spreadsheets and reducing the number of rows of data.
See Lewis-Bowen ¶ 10. To the extent that Appellant argues that there must
be a teaching or suggestion in Lewis-Bowen itself to generate a summary
view (Appeal Br. 10), the motivation to combine the teachings of the prior
art need not come from within the references themselves. It is well settled
Appeal 2019-002520
Application 14/297,810

8

that “[t]he motivation [to combine references] need not be found in the
references sought to be combined, but may be found in any number of
sources, including common knowledge, the prior art as a whole, or the
nature of the problem itself.” DyStar Textilfarben GmbH & Co.
Deutschland KG v. C.H. Patrick Co., 464 F.3d 1356, 1361 (Fed. Cir. 2006)
(citation omitted). The pertinent question is whether a person having
ordinary skill in the art at the time of the invention would have found the
combination obvious.
B. Whether the combination of Lewis-Bowen, Noy, and Fan teaches a
pattern for which a summary view is to be generated
1. “a pattern”
We are unpersuaded by Appellant’s attempt to distinguish the pattern
of Noy from that recited in claim 1. Appeal Br. 10–11; Reply Br. 2–3.
Appellant’s Specification discloses that, “pattern detector 130 first detects
whether the structured data has some types of repeating or common values”
(Spec. ¶ 54), and “if the structured data does have some types of repeating
values (or commonality), then pattern detector 130 determines whether the
user is somehow interacting with data items that have some (even partial
commonality).” Id. ¶ 55. Figure 5, reproduced below, “illustrat[es] some
different examples of the types of commonality 258 that can exist in the
structured data, and for which a pattern can be identified.” Id. ¶ 56
(emphasis added).

Appeal 2019-002520
Application 14/297,810

9

Figure 5, above, shows an embodiment of a plurality of different types of
commonalities that can be used in identifying patterns. Spec. ¶ 12.
Furthermore, “pattern detector 130 can detect patterns in a wide variety of
different ways, and those described herein are described for the sake of
example only” Spec. ¶ 27.
Notably, Figure 5, above, discloses “edit history (e.g., recently
changed) 272” as a commonality for which a pattern may be identified. The
Specification indicates that the edit history may be, “[f]or instance, if a user
is selecting only data items that have been recently changed, then pattern
detector 130 can detect this as a pattern.” Spec. ¶ 61 (emphasis added).
Appellant’s characterization that, “Noy mentions using historical user
selections of objects to predict a next likely object that a user will select
(according to the identified pattern)” (Appeal Br. 10–11; see also Reply Br.
2–3), is consistent with the patterns defined by a pattern definition rule
Appeal 2019-002520
Application 14/297,810

10

disclosed in the Specification, as discussed above, and corresponds to the
disclosed edit history pattern. See Spec. ¶ 61, Fig. 5.
2. “for which a summary view is to be generated”
Appellant’s argument that Noy does not teach a pattern for which a
summary view is to be generated (Appeal Br. 10–11; Reply Br. 2–3), attacks
Noy singly for allegedly failing to teach what the Examiner relies upon in
the combined teachings of Lewis-Bowen, Noy, and Fan. Notably, the
Examiner relies upon Lewis-Bowen for teaching triggering collapse of data,
i.e., the automated summary view generation system, conforming to a
detected pattern (Final Act. 3–4; Ans. 29–30); Noy for teaching the claimed
accessing a set of pattern definition rules (Final Act. 4–5); and Fan for
teaching automatically generating a summary view that visually represents
the sets of summary data. Id. at 5–6. Therefore, the combination of Lewis-
Bowen, Noy, and Fan teaches each pattern definition rule defining when
user interaction inputs conform to a pattern for which a summary view is to
be generated by an automated summary view generation system, and
automatically generating a summary view that visually represents the sets of
summary data, as recited in claim 1.
Contrary to Appellant’s argument that a person skilled in the art
would have no motivation to combine the teachings of Lewis-Bowen and
Noy (Appeal Br. 11; Reply Br. 3), we agree with the Examiner that the
combination “resolv[es] pointing ambiguities in human-computer interaction
(HCI) by implicitly analyzing user movements of a pointer. . .and predicting
the user targeted object.” Final Act. 5 (citing Noy ¶ 29). Although Appellant
argues that the Examiner’s proffered motivation to combine has nothing to
do with generating a summary view or pattern definition rules, and that one
Appeal 2019-002520
Application 14/297,810

11

skilled in the art would not understand there to be a pointing ambiguity in
the carat icon selection of Lewis-Bowen (Appeal Br. 11), the Examiner
determines that the combination aids in predicting which of the Lewis-
Bowen columns to sort, on which data sets to produce a summary view, and
whether a user intends to interact with a carat icon. Ans. 32. We agree.
B. Whether the combination of Lewis-Bowen, Noy, and Fan teaches
in response to determining that the user interaction conforms to
one of the pattern definition rules, automatically triggering the
automated summary view generation system to automatically
generate a set of summary data

Regarding Appellant’s contention that Fan does not teach generating
pivot tables based on direct user interaction with the data items, or by
triggering automated summary view generation based on determining a
pattern from user interactions (Appeal Br. 12), this argument is similarly
deficient to Appellant’s argument above regarding a pattern for which a
summary view is to be generated. Specifically, the Examiner cites Lewis-
Bowen for teaching the limitations related to direct user interaction with the
data items (see Final Act. 3–4); Noy for teaching determining a pattern from
user interactions (id. at 4–5); and Fan for teaching automated summary view
generation based on commonalities between attributes of the data. Id. at 5–6.
As such, Appellant’s argument, attacking Fan singly for failing to teach what
the combination of Lewis-Bowen, Noy, and Fan teaches, is unpersuasive to
show Examiner error.
Independent Claim 12
Claim 12 recites, inter alia:
detect one or more user interactions with the structured data display
that directly interacts with a first set of the data items displayed
on the structure data display;
Appeal 2019-002520
Application 14/297,810

12

automatically determine that a second set of the data items in the
structure, other than the first set of data items, correspond to the
first set of data items, and. . .

based on applying the pattern definition rule to the first and second
sets of data items, automatically trigger the automated summary
view generation system to:

automatically generate a set of summary data based on the first
and second sets of data items.

Appeal Br. 24 (Claims App.).
The Examiner determines that Lewis-Bowen teaches, in relevant part,
detecting one or more user interactions with a first set of the data items
(Final Act. 12–13; citing Lewis-Bowen ¶ 63, Fig. 4); and automatically
determining that a second set of the data items in the structure, other than the
first set of data items, correspond to the first set of data items (hereinafter
“automatically determine” limitation) (Final Act. 13; citing Lewis-Bowen
¶ 77). The Examiner additionally determines that the combination of Lewis-
Bowen, Noy, and Fan teaches based on applying the pattern definition rule
to the first and second sets of data items (hereinafter “based on applying”
limitation), automatically triggering the automated summary view
generation system to automatically generate a set of summary data based on
the first and second sets of data items (hereinafter “automatically generate”
limitation). See Final Act. 13–15.
Appellant argues that because the Lewis-Bowen expand widget of
Figure 4 and paragraph 63 is unrelated to the Boolean input for collapsing
rows of paragraph 77, Lewis-Bowen does not teach detecting one or more
user actions and the automatically determine limitation. Appeal Br. 14–15.
Appeal 2019-002520
Application 14/297,810

13

Appellant contends that Lewis-Bowen does not teach automatically
calculating summary data based on the first and second sets of data items,
wherein the summary data is separate from and corresponds to the first and
second sets of data items. Id. at 15. Appellant also argues that the alleged
patterns of Noy predict a next object that a user intends to select or target,
but that Noy does not teach a pattern from which a summary view is to be
generated or using pattern detection as a trigger for automated summary
view generation. Id. Appellant further contends that Fan does not discuss
generating the pivot tables based on direct user interaction with the data or
based on determining a pattern from user interactions. Id.
A. Whether the combination of Lewis-Bowen, Noy, and Fan
teaches detect one or more user interactions that directly
interacts with a first set of the data items; and automatically
determine that a second set of the data items in the structure,
other than the first set of data items, correspond go the first set
of data items

We disagree with Appellant’s argument that the Lewis-Bowen expand
widget is unrelated to the Boolean input. Appeal Br. 14–15. Lewis-Bowen
describes the Boolean operator being used for “dairy” in column B and
“yellow” in column C. Lewis-Bowen ¶ 77. Lewis-Bowen also discusses the
expand widget 401 in the context of figure 4. Id. at ¶ 63. Figure 4 of Lewis-
Bowen is reproduced below:
Appeal 2019-002520
Application 14/297,810

14

Figure 4, above, illustrates a view of an example spreadsheet. Id. ¶ 33.
Notably, Figure 4 depicts both the expand widget 401, as well as column B
including entries of “Dairy” and column C including entries of “Yellow.”
Therefore, Appellant’s contention that the teachings in paragraphs 63 and 66
and 77 of Lewis-Bowen are unrelated (Appeal Br. 14–15), is not supported
by a preponderance of the evidence.
Nevertheless, the Examiner further determines, and we agree, that
Lewis-Bowen additionally teaches sorting, i.e., one or more user interactions
as claimed, related to the expand widget embodiment. See Ans. 39–40
(citing Lewis-Bowen ¶ 61). Specifically, Lewis-Bowen teaches that a user
may sort a collapsible column, “name,” and sort a column to the right,
“date,” resulting in the rows sorted by name, and wherever a name is
repeated, sorting the name by date. Lewis-Bowen ¶ 61.
B. Whether the combination of Lewis-Bowen, Noy, and Fan
teaches automatically calculating summary data
Appeal 2019-002520
Application 14/297,810

15

Appellant’s argument that Lewis-Bowen fails to teach or suggest
“automatically calculating summary data based on the first and second sets
of data items, wherein the summary data is separate from and corresponds to
the first and second sets of data items” (Appeal Br. 15), is not commensurate
with the scope of claim 12. The above limitation appears in claim 19, not
claim 12, and is addressed in the discussion of claim 19 below. To the extent
that Appellant argues Lewis-Bowen does not teach the automatically
generate limitation, such argument attacks Lewis-Bowen singly for what the
Examiner finds in the combined teachings of the prior art. The Examiner has
identified the relevant portions of Lewis-Bowen, Noy, and Fan and has
provided sufficient explanation with corresponding citations to various parts
of the references for teaching the automatically generate limitation. Final
Act. 13–15. In particular, the Examiner finds that Lewis-Bowen teaches that
based on applying a pattern to the first and second sets of data items,
automatically triggering the collapse data based on the first and second sets
of data items (Final Act. 13–14 (citing Lewis-Bowen ¶ 77)); Noy teaches
access a pattern definition rule defining a pattern for which a summary is to
be automatically generated (Final Act. 14 (citing Noy ¶¶ 130–131)); and Fan
teaches automatically generate a set of summary data based on the first and
second sets of data items. Final Act. 15 (citing Fan ¶ 75, Fig. 18)).
Appellant’s remaining arguments for claim 12, namely that Noy does
not teach a pattern from which a summary view is to be generated or using
pattern detection as a trigger for automated summary view generation
(Appeal Br. 13–14), and that Fan does not discuss generating the pivot tables
based on direct user interaction with the data or based on determining a
pattern from user interactions (id.), are similar to arguments presented with
Appeal 2019-002520
Application 14/297,810

16

respect to claim 1, and are unpersuasive for the same reasons discussed
above.
Independent Claim 19
Claim 19 recites, inter alia:
based on a data item sort user input, sorting a first set of data
items. . .that are displayed in a first column of the table
structure;

based on a data item selection user input, selecting a range of the
sorted data items in the first column;

automatically identifying a second set of data items, of the plurality of
data items, that are displayed in a second column of the table
structure and correspond to the selected range of the sorted
data items in the first column. . .

based on applying the pattern definition rule to the data item sort user
input and the data item selection user input, automatically
triggering the automated summary view generation system to:

automatically calculate summary data based on the first and
second sets of data items, wherein the summary data is
separate from and corresponds to the first and second
sets of data items.

Appeal Br. 25–26 (Claims App.) (emphases added to indicate limitations in
dispute).
The Examiner determines that Lewis-Bowen teaches automatically
identifying a second set of data items, of the plurality of data items, that are
displayed in a second column of the table structure and correspond to the
selected range of the sorted data items in the first column (hereinafter
“automatically identifying” limitation), because Lewis-Bowen teaches a
Boolean input that automatically collapses all rows with input “dairy” in
Appeal 2019-002520
Application 14/297,810

17

column B and input “yellow” in column C. Final Act. 20 (citing Lewis-
Bowen ¶ 77). The Examiner additionally determines that the combination of
Lewis-Bowen, Noy, and Fan teaches automatically calculating summary
data based on the first and second sets of data items, wherein the summary
data is separate from and corresponds to the first and second sets of data
items (hereinafter “automatically calculate” limitation). Final Act. 20–23.
Appellant argues, regarding the automatically identifying limitation,
that Lewis-Bowen does not teach that the collapsing of rows is based on a
data item sort user input. Appeal Br. 13. Appellant additionally argues that
Lewis-Bowen does not teach the automatically calculate limitation. Id.;
Reply Br. 3–4. Appellant contends that the alleged patterns of Noy predict a
next object that a user intends to select or target, and that Noy does not teach
a pattern from which a summary view is to be generated or using pattern
detection as a trigger for automated summary view generation. Appeal Br.
13–14. Appellant also argues that Fan does not discuss generating the pivot
tables based on direct user interaction with the data or based on determining
a pattern from user interactions. Id.
B. The automatically identifying limitation
We are unpersuaded by Appellant’s first argument, namely that
Lewis-Bowen does not teach the automatically identifying limitation based
on a data item sort user input. Id. at 13. Here, the Examiner determines, and
we agree, that paragraphs 61 and 92 of Lewis-Bowen teach automatically
collapsing data, i.e., the automatically identifying limitation, based on a data
item sort user input. Ans. 36. Appellant does not rebut the Examiner’s
findings in the Reply Brief, instead reiterating that Lewis-Bowen does not
teach the automatically calculate limitation. Reply Br. 3–4.
Appeal 2019-002520
Application 14/297,810

18

C. The automatically calculate limitation
Appellant’s argument that Lewis-Bowen does not teach the
automatically calculate limitation is similarly unavailing. Appeal Br. 13;
Reply Br. 3–4. Where the collective teachings of a combination of
references—here, Lewis-Bowen, Noy, and Fan—are cited for teaching the
automatically calculate limitation (see Final Act. 20–23), attacking one
reference singly is not persuasive to show error. More specifically,
Appellant’s argument that “this alleged summary data [of Lewis-Bowen] is
the collapsing of the data items themselves” (Reply Br. 4), fails to
acknowledge that the Examiner relies upon Fan for teaching the claimed
summary data separate from and corresponding to the first and second sets
of data items. Final Act. 22 (citing Fan ¶ 75, Fig. 18).
Appellant’s remaining arguments for claim 19, namely that Noy does
not teach a pattern from which a summary view is to be generated or using
pattern detection as a trigger for automated summary view generation
(Appeal Br. 13–14), and that Fan does not discuss generating the pivot tables
based on direct user interaction with the data or based on determining a
pattern from user interactions (id.), are similar to arguments presented with
respect to claim 1, and are unpersuasive for the same reasons discussed
above.
Dependent Claim 5
Claim 5 depends from claim 4, which depends from claim 1, and
further recites: “wherein determining that the user interaction with the
structured data display conforms to one of the pattern definition rules
comprises: detecting commonality between repeating values in data items
Appeal 2019-002520
Application 14/297,810

19

displayed in the first column and repeating values displayed in the second
column.” Appeal Br. 22–23 (Claims App.).
The Examiner maps the Lewis-Bowen Boolean operator for
collapsing all rows having “dairy” in column B and “yellow” in column C to
the limitations above. Final Act. 7–8 (citing Lewis-Bowen ¶ 77).
Appellant argues that Lewis-Bowen fails to teach or suggest either
generating a summary view to be added to the display, or detecting
commonality between the columns. Appeal Br. 16. We disagree.
Appellant’s argument regarding generating a summary view to be
added to the display, is not commensurate with the scope of claim 5, which
does not require such feature. To the extent that Appellant argues that
Lewis-Bowen does not teach “automatically generate a summary view that
visually represents the set of summary data,” which appears in claim 1,
attacking Lewis-Bowen for allegedly lacking a feature that the Examiner
relies on Fan to teach (see Final Act. 5–6), is not persuasive to demonstrate
error.
We are unpersuaded by Appellant’s argument that Lewis-Bowen does
not teach detecting commonality between the columns. Figure 5 of the
Specification, supra, showing a dearth of types of commonalities, includes a
commonality “other” 276. The Specification discloses that, “[p]attern
detector 130 can, of course, detect patterns in a wide variety of other ways as
well. This is indicated by block 276.” Spec. ¶ 62 (emphasis added).
Therefore, “commonality” as defined in Appellant’s Specification includes a
broad array of patterns. Lewis-Bowenat least suggests detecting
commonality between repeating values displayed in the first and second
columns, because Lewis-Bowen teaches the Boolean operator detecting
Appeal 2019-002520
Application 14/297,810

20

commonality between “dairy” in one column, and “yellow” in another
column. Lewis-Bowen ¶ 77.
We are not persuaded of error in the rejection of claim 5 for the above
reasons.
Dependent Claim 6
Claim 6 depends from claim 5 and further recites: “wherein detecting
commonality, comprises: detecting a first set of repeating values in the first
column that correspond to a corresponding second set of repeating values in
the second column.” Appeal Br. 23 (Claims App.).
The Examiner maps the Boolean operator example of Lewis-Bowen
to the claim limitations. Final Act. 8 (citing Lewis-Bowen ¶ 77).
Appellant contends that Lewis-Bowen fails to teach or suggest either
detecting commonality based on applying a pattern definition rule to direct
user interactions with the data items, or detecting that a first set of repeating
values in the first column correspond to a second set of repeating values in
the second column. Appeal Br. 17.
Appellant’s argument that Lewis-Bowen fails to teach detecting
commonality based on applying a pattern definition rule to direct user
interactions with the data items, is not commensurate in scope with claim 6,
which does not require such feature. Furthermore, as discussed with respect
to claim 1, supra at “Independent Claim 1” § B, Lewis-Bowen is not relied
upon to teach a pattern definition rule. Therefore, to the extent that
Appellant’s argument is directed to “in response to determining that the user
interaction. . .conforms to one of the pattern definition rules,. .
.automatically generate a set of summary data based on commonality
between attributes of the set of data items,” recited in claim 1 (Appeal Br. 22
Appeal 2019-002520
Application 14/297,810

21

(Claims App.)), Appellant’s argument ignores the combined teachings of the
prior art.
Furthermore, Lewis-Bowen teaches detecting when “dairy” in column
B corresponds to “yellow” in column C, and collapsing those rows. Lewis-
Bowen ¶ 77. Therefore, we agree with the Examiner that Lewis-Bowen
teaches detecting a first set of repeating values in the first column that
correspond to a corresponding second set of repeating values in the second
column, as recited in the claim.
We are not persuaded of error in the rejection of claim 6 for the
reasons above.
Dependent Claim 9
Claim 9 depends from claim 4 and further recites, inter alia:
receiving an item sort user input. . .; and

receiving an item selection user input. . .;

wherein determining that the user interaction with the structured
data display conforms to one of the pattern definition
rules comprises determining that the item sort user input
and the item selection user input conform to one of the
pattern definition rules.

Appeal Br. 23 (Claims App.).
The Examiner maps selecting the Lewis-Bowen expand widget to sort
rows according to column values, to the recited receiving an item sort user
input. Final Act. 10 (citing Lewis-Bowen ¶ 63). The Examiner also maps an
indicator or highlight widget of Lewis-Bowen, which when selected,
collapses rows having the same column values to collapse, to the recited
receiving an item selection user input. Final Act. 10 (citing Lewis-Bowen
¶ 63). The Examiner additionally cites Noy at paragraphs 130 and 131 for
Appeal 2019-002520
Application 14/297,810

22

teaching the determining limitation above, because Noy teaches identifying
patterns or identical attributes in the set of the last selections by the user of a
user targeted object, and assuming that the user will select a next targeted
object according to the pattern. Final Act. 10–11 (citing Noy ¶¶ 130–131).
Appellant contends that Lewis-Bowen does not teach or suggest
generating a summary view based on an expand widget input, or
determining that user inputs conform to pattern definition rules. Appeal Br.
18. Appellant additionally argues that Noy does not teach that determination
that user interaction conforms to a pattern definition rule comprises
determining that both a sort user input and a selection user input conforms to
a rule.
Appellant’s arguments directed to Lewis-Bowen are not
commensurate with the scope of claim 9, which does not require generating
a summary view. To the extent that Appellant argues features recited in
claim 1, those arguments are deficient for the reasons discussed above.
Furthermore, the Examiner finds that the combination of Lewis-Bowen,
Noy, and Fan teaches determining that user inputs conform to pattern
definition rules, as discussed supra at “Independent Claim 1” § B(2), and
therefore attacking Lewis-Bowen singly for allegedly failing to teach such
feature is not persuasive to show error.
Regarding Appellant’s arguments against Noy for the disputed
limitation above, the Examiner explains that:
The selection of data in Lewis-Bowen provides input on how to
sort and collapse the rows of the spreadsheet[,] and Noy would
be able to detect the pattern in the selection of data and would
know that a sort and collapse selection would follow and thus
provide the targeted objects/functionality.

Appeal 2019-002520
Application 14/297,810

23

Ans. 46. In other words, the combination of Lewis-Bowen and Noy teaches
the limitations of claim 9, above, rather than Lewis-Bowen or Noy singly.
See also Final Act. 9–11. Therefore, Appellant’s arguments against Noy
individually are unpersuasive to show error in the Examiner’s findings that
the combination of Lewis-Bowen and Noy teaches the recited limitations.
We are not persuaded of error in the rejection of claim 9 for the
reasons above.
Dependent Claim 13
Claim 13 depends from claim 12 and further recites: “automatically
generate a summary data structure to display the summary data in the
summary view, wherein the summary data structure is simultaneously
displayed in the summary view with at least a portion of the data items in the
structure.” Appeal Br. 24–25 (Claims App.).
The Examiner determines that Fan teaches the limitations above. Final
Act. 16 (citing Fan ¶ 75). The Examiner concludes that it would have been
obvious to a skilled artisan at the time of the invention to modify Lewis-
Bowen with the teachings of Fan, to “provide the user with optimal chart
choices and guide[] the user to make better choices in creating
visualizations[,] and the pivot chart recommendation process may provide
the best summarization of data that are of higher value to the user when the
user has [a] repetitive dataset.” Final Act. 16–17 (citing Fan ¶¶ 5, 75).
Appellant argues that there is no teaching, suggestion, or motivation
to modify Lewis-Bowen to include the additional summary view as claimed.
Appeal Br. 18. We disagree.
Appellant’s contention that Lewis-Bowen teaches away from an
“additional” summary view (see id.), is unpersuasive for the reasons
Appeal 2019-002520
Application 14/297,810

24

discussed above with respect to the rejections of claims 1 and 12.
Furthermore, the Examiner has articulated the motivation that a person
having ordinary skill in the art at the time of the invention would have had to
modify Lewis-Bowen with the teachings of Fan. Final Act. 16–17.
Specifically, the Examiner determines that the chart options of Fan provide a
better summarization of data that is of higher value to a user. Id. Appellant’s
argument that there is no teaching, suggestion, or motivation (Appeal Br.
18), therefore, without more, is unpersuasive to show error.
We are not persuaded of error in the rejection of claim 13 for the
reasons above.
Dependent Claim 20
Claim 20 depends from claim 19 and further recites: “wherein
applying the pattern definition rule comprises detecting common values in a
corresponding range of the second set of data items in the second column.”
Appeal Br. 26 (Claims App.).
The Examiner determines that Lewis-Bowen teaches the limitations
above, citing the above-discussed Boolean operator. Final Act. 23 (citing
Lewis-Bowen ¶ 77).
Appellant argues that there is no teaching or suggestion in Lewis-
Bowen,
of a sort user input that sorts a first set of items, and then a data
item selection input that selects a range of the sorted data items,
where the system identifies a second set of data items in a
second column by detecting common values in a corresponding
range of that second column.
Appeal Br. 19.
Appellant’s argument that Lewis-Bowen does not teach a sort user
input or a data item selection input (see id.), is not commensurate with the
Appeal 2019-002520
Application 14/297,810

25

scope of claim 20, which does not require those features. To the extent that
Appellant’s argument is directed to limitations in claim 19, we agree with
the Examiner, as discussed supra at “Independent Claim 19” § A, that
Lewis-Bowen teaches the recited data item sort user input. See Final Act.
19–20; Ans. 36. We additionally agree with the Examiner that Lewis-Bowen
teaches the data item selection input recited in claim 19. Final Act. 20 (citing
Lewis-Bowen ¶ 77). In particular, the Examiner determines, and we agree,
that the Boolean operation, whereby all rows having “dairy” in column B
and “yellow” in column C are collapsed, is appropriately mapped to the
limitation of based on a data item selection user input, selecting a range of
the sorted data items in the first column, and automatically identifying a
second set of data items displayed in a second column which correspond to
the selected range of the sorted data items in the first column. See Final Act.
20; Lewis-Bowen ¶ 77.
Furthermore, we agree with the Examiner that the Lewis-Bowen
Boolean operator detects common values in a corresponding range of the
second set of data items in the second column. Final Act. 23 (citing Lewis-
Bowen ¶ 77). Specifically, Lewis-Bowen teaches detecting a common value,
“yellow,” in a corresponding range of the second set of data in the second
column. Lewis-Bowen ¶ 77.
We not persuaded of error in the rejection of claim 20 for the reasons
above.
Dependent Claim 23
Claim 23 depends from claim 1 and further recites: “wherein the
automated summary view generation system is configured to generate a
plurality of different sets of summary data based on the commonality
Appeal 2019-002520
Application 14/297,810

26

between attributes of the data items, wherein the plurality of different sets of
summary data are separate from, and correspond to, the data items.” Appeal
Br. 26–27 (Claims App.).
The Examiner determines that Fan teaches the limitations of claim 23,
because Fan teaches pivot table suggestions based on a dataset. Final Act. 25
(citing Fan ¶ 75, Fig. 18).
Appellant contends that a skilled artisan would find no motivation to
modify Lewis-Bowen to include the generation of multiple summary data
sets, i.e., more than one way to collapse the rows into a single collapsed row,
and that the sets of summary data would not be separate from the data items
from which they are generated. Appeal Br. 19–20. We disagree.
Figure 18 of Fan is reproduced below:
Appeal 2019-002520
Application 14/297,810

27

Figure 18, above, shows an embodiment of PivotTable suggestions for
a dataset that contains aggregates. Fan ¶ 30. Figure 18 illustrates generation
of multiple summary data sets #1820, 1830, 1840, 1850 based on a
commonality, e.g., identical data input, between attributes of the data items,
wherein the plurality of different sets of summary data are separate from,
and correspond to, the data items. Therefore, we agree with the Examiner
that Fan teaches the limitations of claim 23. Final Act. 25.
We are unpersuaded by Appellant’s argument that a skilled artisan
would not have been motivated to modify Lewis-Bowen to include more
Appeal 2019-002520
Application 14/297,810

28

than one way to collapse the items into a single collapsed row. Appeal Br.
19–20. Here, Appellant’s argument is deficient because it ignores the
combined teachings of Lewis-Bowen, Noy, and Fan, specifically Fan’s
teaching of a summary view, which is discussed in the Examiner’s rejection
of claim 1. See Final Act. 3–7. In other words, Appellant’s argument focuses
solely on Lewis-Bowen, where the Examiner instead determines that the
combined teachings of the prior art render obvious the limitations of claim
23.
Appellant’s additional argument that the summary data of Lewis-
Bowen as modified by Noy and Fan would not be separate from the data
items from which it is generated, is similar in scope to arguments previously
addressed for independent claim 19. Because the Examiner cites the same
portions of the prior art for teaching the summary views of claim 1 and
claim 19 (compare Final Act. 22 (citing Fan ¶ 75, Fig. 18), with Final Act.
5–6 (citing Fan ¶ 75, Fig. 18)), we do not repeat our determinations, and find
Appellant’s argument unpersuasive for the same reasons as discussed above
with respect to claim 19.
We are not persuaded of error in the rejection of claim 23 for the
reasons above.
For at least the foregoing reasons, we sustain the Examiner’s
obviousness rejections of claims 1, 4–9, 11–16, and 19–24.
CONCLUSION
We affirm the Examiner’s decision to reject claims 1, 4–9, 11–16, and
19–24.
Appeal 2019-002520
Application 14/297,810

29

In summary:
Claims
Rejected 35 U.S.C. § Reference(s)/Basis Affirmed Reversed
1, 4–9, 11–
16, 19–24 103
Lewis-Bowen, Noy,
Fan
1, 4–9, 11–
16, 19–24
TIME PERIOD FOR RESPONSE
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). See
37 C.F.R. § 41.50(f).
AFFIRMED