Janet Louise. Slifka

Patent Trials and Appeals Board

Sep 30, 2019

13793856 - (D) (P.T.A.B. Sep. 30, 2019)

UNITED STATES PATENT AND TRADEMARK OFFICE
UNITED STATES DEPARTMENT OF COMMERCE
United States Patent and Trademark Office
Address: COMMISSIONER FOR PATENTS
P.O. Box 1450
Alexandria, Virginia 22313-1450
www.uspto.gov
APPLICATION NO. FILING DATE FIRST NAMED INVENTOR ATTORNEY DOCKET NO. CONFIRMATION NO.
13/793,856 03/11/2013 Janet Louise Slifka P7865 4891
136714 7590 09/30/2019
Pierce Atwood LLP (Attn: Amazon Docketing)
Attn: Patent Docketing (Amazon)
100 Summer Street
Suite 2250
Boston, MA 02110
EXAMINER
ZHU, RICHARD Z
ART UNIT PAPER NUMBER
2675
NOTIFICATION DATE DELIVERY MODE
09/30/2019 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):
patent@pierceatwood.com
PTOL-90A (Rev. 04/07)
UNITED STATES PATENT AND TRADEMARK OFFICE
BEFORE THE PATENT TRIAL AND APPEAL BOARD
Ex parte JANET LOUISE SLIFKA
Appeal 2018-005584
Application 13/793,856
Technology Center 2600
Before JOSEPH L. DIXON, JAMES R. HUGHES, and LARRY J. HUME,
Administrative Patent Judges.
DIXON, Administrative Patent Judge.
DECISION ON APPEAL
STATEMENT OF THE CASE
Pursuant to 35 U.S.C. § 134(a), Appellant1 appeals from the
Examiner’s decision to reject claims 1–7, 10–14, 16, and 18–23. Final
Act. 1. Claims 8, 9, 15, and 17 are canceled. We have jurisdiction under
35 U.S.C. § 6(b).
We reverse.

1 We use the word “Appellant” to refer to “Applicant” as defined in 37
C.F.R. § 1.42(a). Appellant identifies the real party in interest as Amazon
Technologies, Inc. (App. Br. 4).
Appeal 2018-005584
Application 13/793,856
2
CLAIMED SUBJECT MATTER
The claims are directed to a domain and intent name feature
identification and processing. Appellant further described the disclosed
invention as:
A natural language processor takes a textual input (one provided
either as the output from an automatic speech recognition (ASR)
or from some other source) and performs named entity
recognition (NER) on the textual input to identify and tag the
meaningful portions of the text so that a later component may
properly form a command to send to another application. To
improve NER processing, additional information beyond the
textual input may be provided to the NER processor ahead of
time. Such information may be referred to as pre-features. Pre-
features may describe any data that may assist natural language
processing such as user identification, user preferences, typical
user queries, etc. As described below, pre-features include
features that are not derived from the text to be processed. Pre-
features may be fed as inputs to an NER processor. Other inputs
to the NER processor may include a domain (a category
describing the context of the textual input) or an intent (an
indicator of the intended command of a user to be executed by a
device). While a domain and/or intent may typically be
determined later in a natural language process, determining them
before NER processing, and offering them as inputs to a NER
processor, may improve overall natural language output. The
domain and intent may be pre-features (that is, not derived from
the text input itself) or may be derived from the text but still input
into the NER processor.

(Spec. ¶ 11).
Claims 1 and 7, reproduced below, are illustrative of the claimed
subject matter:
1. A method of performing natural language processing, the
method comprising:

receiving an audio signal comprising an utterance;
Appeal 2018-005584
Application 13/793,856
3
obtaining first text of the utterance using automatic
speech recognition;

determining at least one of a category of commands or
a potential intent corresponding to the utterance, wherein the
at least one of the category or the potential intent are
determined using data associated with the utterance and are
not determined from the first text or based on a previous
utterance that is associated with the utterance;

performing semantic tagging of the first text using a
named entity recognition model based at least in part on at
least one of the category or the potential intent, wherein the
named entity recognition model was trained using a corpus of
data comprising a first training text, and wherein the first
training text was associated with at least one of a first category
or a first intent;

determining, using the results of the semantic tagging,
a selected intent for the utterance; and

performing an action using the selected intent.

7. A computing device, comprising:

at least one processor;
a memory device including instructions operable to be
executed by the at least one processor to perform a set of
actions, configuring the at least one processor:

to receive audio data corresponding to an input
utterance of a user;
to perform automatic speech recognition (ASR)
on the audio data to obtain text; to determine one or
more pre-features, wherein the one or more pre-features
are data associated with the input utterance and are not
Appeal 2018-005584
Application 13/793,856
4
determined from the text or based on a previous input
utterance of the user;

to associate the one or more pre-features with the
text;
to determine, based at least in part on the one or
more pre-features, at least one of a category of
commands corresponding to the input utterance or a
potential intent corresponding to the input utterance, the
potential intent corresponding to an intended command
to be executed;

to generate a pre-feature vector including the one
or more pre-features;
to associate an entity with at least one word of the
text based at least in part on the pre-feature vector and
the category or the intent; and

to determine, based at least in part on the entity,
a selected intent for the input utterance.
REFERENCES
The prior art relied upon by the Examiner as evidence is:
Suleman et al. US 2015/0039292 A1 Feb. 05, 2015
Thomson et al. US 2008/0037720 A1 Feb. 14, 2008
Arnold et al. US 2005/0234723 A1 Oct. 20, 2005

REJECTIONS
Claims 1–7, 10, 12–14, 16, 18, 19, and 21–23 stand rejected under 35
USC § 102(e)(l) as being anticipated by Suleman.
Claim 11 stands rejected under USC § 103(a) as being unpatentable
over Suleman in view of Arnold.
Appeal 2018-005584
Application 13/793,856
5
Claim 20 stands rejected under USC § 103(a) as being unpatentable
over Suleman in view of Thomson.
OPINION
35 U.S.C. § 102
With respect to independent claims 1, 7, and 14, Appellant sets forth
arguments with regards to independent claim 7 regarding “pre-features,” but
we note that independent claim 1 does not expressly recite “pre-features,”
but includes similar limitations regarding the determining the category of
commands or potential intent corresponding to the utterance. We address
independent claim 7 as the illustrative claim for the group.
With respect to independent claim 7, Appellant contends that the prior
art does not disclose the claimed “pre-features,” and that the pre-features are
associated with text of an utterance but not derived from or determined
based on the text itself. (App. Br. 13). Appellant contends that the claimed
invention is directed to a database of contextual information that is available
to the device, including information associated with multiple users (e.g., a
complete user history for each of the multiple users). (App. Br. 14).
Appellant further contends “determining the pre-features corresponds to
selecting a subset of data from the database (e.g., specific contextual
information at the time the utterance was received), wherein the subset of
data is associated with the utterance but not derived from the text itself.”
(App. Br.16).
The Examiner maintains that the Suleman reference discloses the
claimed “determine one or more pre-features, wherein the one or more pre-
features are data associated with the input utterance and are not determined
from the text or based on a previous input utterance of the user” (Final Act.
Appeal 2018-005584
Application 13/793,856
6
14). See Final Act. 14–15 (citing Suleman ¶ 42); Ans. 5–17 (citing Suleman
¶¶ 32, 41–45, 49–52, 57, 58).
Although we agree with the Examiner that the Suleman reference
discloses the use of pre-stored or pre-calculated data in database(s), the
Examiner has not shown that the Suleman reference discloses the claimed
step “determine one or more pre-features, wherein the one or more pre-
features are data associated with the input utterance and are not determined
from the text or based on a previous input utterance of the user.” From our
review of Figures 3 and 4 of the Suleman reference, each of these disclosed
processes starts with the input of the query, and the Examiner has not
identified a clear disclosure of the claimed step/process to “determine one or
more pre-features, wherein the one or more pre-features are data associated
with the input utterance and are not determined from the text or based on a
previous input utterance of the user.” For example, paragraph 33 of the
Suleman reference discloses “NLP [Natural Language Processing] service
114 analyzes the user query to determine meaning and specific commands
with which to provide the services.” Additionally, in Figure 3, the user
query is input to each of the units 306, 308, and 310 to be used in the
processing. (Suleman ¶ 39). In Figure 4, “step 404, user query 302 is
subjected to binary classification such as via a support vector machine
(SVM)” which again uses the user query. (Suleman ¶ 50).
The Examiner’s anticipation rejection is not well supported by the
express disclosure of the Suleman reference. Accordingly, we agree with
Appellant the Examiner’s determination that the claimed invention is
anticipated is not supported by a preponderance of the evidence. See In re
Caveney, 761 F.2d 671, 674 (Fed. Cir. 1985) (Examiner’s burden of proving
non-patentability is by a preponderance of the evidence); see also In re
Appeal 2018-005584
Application 13/793,856
7
Warner, 379 F.2d 1011, 1017 (CCPA 1967) (“The Patent Office has the
initial duty of supplying the factual basis for its rejection. It may not,
because it may doubt that the invention is patentable, resort to speculation,
unfounded assumptions or hindsight reconstruction to supply deficiencies in
its factual basis.”). We will not resort to such speculation or assumptions to
cure the deficiencies in the factual basis in order to support the Examiner’s
anticipation rejection. Consequently, we cannot sustain the rejection of
independent claim 7 and its dependent claims 10, 12, 13, and 22 based on
anticipation.
The Examiner relies up on the same basis in the anticipation rejection
of independent claims 1 and 14. Independent claim 14 contains the same
claim language found lacking in the anticipation rejection of independent
claim 7 and therefore we cannot sustain the anticipation rejection thereof and
dependent claims 16, 18, 19, and 23. Independent claim 1 does not recite
“pre-features,” but contains the same step of “determining.” As a result, we
cannot sustain the Examiner’s anticipation rejection of independent claim 1
and dependent claims 2–6.

35 U.S.C. § 103
With respect to dependent claims 11 and 20, the Examiner does not
identify how the additional prior art references remedy the noted deficiency
above. (Ans. 18–20). As a result we cannot sustain the Examiner’s
obviousness rejections of dependent claims 11 and 20.
CONCLUSION
We reverse the Examiner’s anticipation and obviousness rejections.
Appeal 2018-005584
Application 13/793,856
8
DECISION SUMMARY
Claims
Rejected
Basis Affirmed Reversed
1–7, 10,
12–14, 16,
18, 19, 21–
23
§ 102
Suleman
1–7, 10,
12–14, 16,
18, 19, 21–
23
11 § 103
Suleman, Arnold
11
20 § 103
Suleman, Thomson
20
Overall
Outcome
1–7, 10–
14, 16, 18–
23
REVERSED