NICE-SYSTEMS LTD.

Patent Trials and Appeals Board

Jun 24, 2020

14980056 - (D) (P.T.A.B. Jun. 24, 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.
14/980,056 12/28/2015 Hila WEISMAN P-79063-US 5901
49443 7590 06/24/2020
Pearl Cohen Zedek Latzer Baratz LLP
1500 Broadway
12th Floor
New York, NY 10036
EXAMINER
SPOONER, LAMONT M
ART UNIT PAPER NUMBER
2657
NOTIFICATION DATE DELIVERY MODE
06/24/2020 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):
Arch-USPTO@PearlCohen.com
USPTO@PearlCohen.com
PTOL-90A (Rev. 04/07)
UNITED STATES PATENT AND TRADEMARK OFFICE
BEFORE THE PATENT TRIAL AND APPEAL BOARD
Ex parte HILA WEISMAN, PETER IZSAK, INNA ACHLOW, and
VICTOR SHAFRAN
Appeal 2019-002382
Application 14/980,056
Technology Center 2600

Before JOSEPH L. DIXON, ST. JOHN COURTENAY III, 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, 3–5, 7, 9, 10, 12–14, 16, 18, and 19.
Claims 2, 6, 8, 11, 15, 17, and 20 have been canceled. We have jurisdiction
under 35 U.S.C. § 6(b). An oral hearing was held on May 21, 2020. A
transcript of the oral hearing was entered into the administrative record on
June 2, 2020.
We REVERSE.

1 We use the word Appellant to refer to “applicant” as defined in 37 C.F.R.
§ 1.42(a) (2018). Appellant identifies the real party in interest as NICE
LTD. Appeal Br. 1.
Appeal 2019-002382
Application 14/980,056

2
CLAIMED SUBJECT MATTER
The claims are directed to a method and apparatus for text
normalization in noisy channels. Claim 1, reproduced below, is illustrative
of the claimed subject matter:
1. A method for text normalization in a plurality of noisy
channels, performed on a computing device having a processor,
memory, and one or more code sets stored in the memory and
executing in the processor, the method comprising:
generating a set of models for a respective set of
communication channels;
generating, based on the set of models, a respective
set of channel specific error-type adapter frameworks,
wherein each of the channel-specific error-type adapter
frameworks is optimized for a respective communication
channel;
receiving, by the processor, a text entry and channel
origin data of the text entry;
determining, by the processor, whether the text
entry matches an in-vocabulary (IV) entry or whether the
text entry is an out-of-vocabulary (OOV) entry;
if the text entry is determined to have a matching IV
entry;
outputting, by the processor, the matching IV entry;
and
if the text entry is determined to be an OOV entry;
implementing, by the processor, the relevant
channel-specific error-type adapter framework based on
the channel origin data;
wherein the channel-specific error-type adapter
framework is optimized for the specific channel from
which the text entry originated; and
Appeal 2019-002382
Application 14/980,056

3
wherein the channel-specific error-type adapter
framework further comprises a probabilistic model of one
or more error-type priors in the specific channel;
normalizing, by the processor, the text entry using
the channel-specific error-type adapter framework;
wherein the channel-specific error-type adapter
framework comprises one or more error-type adapters;
wherein the one or more error-type adapters
comprise at least one of a spelling error-type adapter, an
abbreviation error-type adapter, an acronym error-type
adapter, a phonetic shorthand error-type adapter, and a word
concatenation error-type adapter; and
wherein each error-type adaptor [sic adapter] is
configured to model a different type of error to be normalized;
and
outputting one or more candidate normalized forms of
the text entry.
REFERENCES
The prior art relied upon by the Examiner as evidence is:
Name Reference Date
Nasri et al. US 2012/0072204 A1 Mar. 22, 2012
Beaufort et al. US 2013/0096911 A1 Apr. 18, 2013
Stoytchev et al. US 2015/0193431 A1 July 9, 2015

REJECTIONS
Claims 1, 3–5, 9, 10, 12–14, 18, and 19 stand rejected under 35 U.S.C.
§ 103 as being unpatentable over Nasri in view of Beaufort.
Claims 7 and 16 stand rejected under 35 U.S.C. § 103 as being
unpatentable over Nasri in view of Beaufort as applied to claim 1 above, and
further in view of Stoytchev.
Appeal 2019-002382
Application 14/980,056

4
OPINION
35 U.S.C. § 103
With respect to claims 1, 3–5, 9, 10, 12–14, 18, and 19, Appellant
contends the application-type specific modules of the Nasri reference are not
interchangeable between communication channels or channel types. Appeal
Br. 5. Appellant contends the relied upon paragraphs 85–107 of the Nasri
reference uses “context,” as opposed to “channel” to determine how to
normalize. Appeal Br. 7. Appellant argues the Nasri reference does not
discuss the concept of in-vocabulary (IV) versus out of vocabulary (OOV).
Appeal Br. 8. Appellant also argues the “normalization portions of Nasri do
not differentiate between different channels from which the text is received
as is asserted in the Office Action, p. 7, but rather normalize the text found
based on preset rules, e.g. para. [0099].” Appeal Br. 8. Appellant contends:
Nasri does parse or extract text based on channel, e.g. paras.
[0087]-[0094], but this is a separate process (see, e.g. Nasri
para. [0081]) from normalization, and does not teach or suggest
the claimed “normalizing, by the processor, the text entry using
the channel-specific error-type adapter framework”.
Normalizing text refers to transforming the text itself to a
standard format (e.g. translating abbreviations), and does not
include parsing, which may be considered preprocessing. In
contrast, the claims on appeal use a process where a framework
is, e.g., “optimized for a specific channel from which the text
entry originated.” Parsing in Nasri may be focused on
extracting the text according to the different formats of the
channels but normalizing in Nasri is not according to, as with
some example embodiments of the claimed invention, different
language specific models (frameworks) which are unique per
channel and with each framework including a probabilistic
model of one or more error-type priors.
Appeal Br. 8
Appeal 2019-002382
Application 14/980,056

5
Appellant argues the Examiner acknowledges the Nasri reference does
not explicitly use the term IV (in vocabulary), and the Examiner relies upon
the Beaufort reference for normalizing both IV and OOV in a single channel.
Appeal Br. 10–11.
The Examiner disagrees with Appellant’s arguments and finds
paragraph 87 of the Nasri reference discloses “generating a set of models for
a respective set of communication channels.” Ans. 4. “The Examiner notes
the SMS, MMS, video, e-mail, etc. are all deemed explicit communication
channels, the same as described by the applicant” as described in paragraph
3 of Appellant’s Specification which discloses various different
communication channels. Ans. 4–5 (emphasis omitted). The Examiner
finds the “question of the generation is inherent, without generation thereof,
they cannot be identified.” The Examiner further relies upon paragraph 58
of the Nasri reference to teach or suggest the generation of rules for a set of
communication channels. Ans. 5.
Regarding the claimed “channel-specific error-type adapter
frameworks is optimized for a respective communication channel,” the
Examiner further finds paragraphs 87–107 of the Nasri reference teach or
suggest application type and “specific defined type and corresponding
different set of rules, thus channel specific to type, the rules per type
including a specific error-type framework for correction based on
application type, see his ‘email’ vs. ‘RSS’ feed, each having respective rules
for correction.” Ans. 6 (emphasis omitted).
The Examiner further opines that the various discussions throughout
the Nasri reference clearly show the various applications “absolutely do not
all use the same general model,” and the Examiner further cites to
Appeal 2019-002382
Application 14/980,056

6
paragraphs 0094, 0097, 0120, 0053–0056 of Nasri. Ans. 7. With respect to
Appellant’s arguments that Nasri uses context opposed to channel to
determine how to normalize, the Examiner maintains that “a complete and
thorough explanation of how the channels are explicitly related to
normalization” and the Examiner further cites to paragraphs 58, 102, 120–
122 of Nasri. Ans. 7.
“‘On appeal to the Board, an applicant can overcome a rejection by
showing insufficient evidence of prima facie obviousness or by rebutting the
prima facie case with evidence of secondary indicia of nonobviousness.’” In
re Kahn, 441 F.3d 977, 985-86 (Fed. Cir. 2006) (quoting In re Rouffet, 149
F.3d 1350, 1355 (Fed. Cir. 1998)).
[T]he examiner bears the initial burden, on review of the prior
art or on any other ground, of presenting a prima facie case of
unpatentability. If that burden is met, the burden of coming
forward with evidence or argument shifts to the applicant.
After evidence or argument is submitted by the applicant
in response, patentability is determined on the totality of the
record, by a preponderance of evidence with due consideration
to persuasiveness of argument.
In re Oetiker, 977 F.2d 1443, 1445 (Fed. Cir. 1992).
“A rejection based on section 103 clearly must rest on a factual basis.
. . .” In re 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
. . . resort to speculation, unfounded assumptions or hindsight reconstruction
to supply deficiencies in its factual basis.” Id.
“Obviousness may not be established using hindsight or in view of the
teachings or suggestions of the inventor.” Para-Ordnance Mfg. v. SGS
Importers Int’l, 73 F.3d 1085, 1087 (Fed. Cir. 1995) (citing W.L. Gore &
Appeal 2019-002382
Application 14/980,056

7
Assocs., Inc. v. Garlock, Inc., 721 F.2d 1540, 1551, 1553 (Fed. Cir. 1983)).
“The mere fact that the prior art may be modified in the manner suggested
by the Examiner does not make the modification obvious unless the prior art
suggested the desirability of the modification.” In re Fritch, 972 F.2d 1260,
1266 (Fed. Cir. 1992) (citing In re Gordon, 733 F.2d 900, 902 (Fed. Cir.
1984)). “It is impermissible to use the claimed invention as an instruction
manual or ‘template’ to piece together the teachings of the prior art so that
the claimed invention is rendered obvious.” Fritch, 972 F.2 at 1266 (citing
In re Gorman, 933 F.2d 982, 987 (Fed. Cir. 1991)).
We disagree with the Examiner as to the Nasri reference explicitly
teaching or suggesting the steps of “generating a set of models for a
respective set of communication channels” and “generating, based on the set
of models, a respective set of channel specific error-type adapter
frameworks.” We note that paragraph 58 of Nasri discloses using models to
generate the tagging module and tagging rules and paragraphs 107, 119, and
120 of Nasri disclose a single language model. Consequently, we cannot
agree with the Examiner’s expansive interpretation of the express teachings
of the Nasri reference. For essentially the same reasons argued by
Appellant, we do not agree with the Examiner’s finding that the Nasri
reference teaches or suggests “the set of models for a respective set of
communication channels, and channel specific error-type adapter
framework.” Ans. 8.
Moreover, the Examiner finds “it is clear that the application type,
TTS or email, using the explicit statistical data for that type, via the metadata
attributes, to resolve the case ambiguity, is thus channel specific.” Ans. 11.
Appeal 2019-002382
Application 14/980,056

8
The Examiner further finds that the Beaufort reference is only relied
upon for teaching in vocabulary (IV) and out of vocabulary (OOV)
determinations for the SMS channel. Final Act. 8; Ans. 12–14. The
Examiner further finds that Appellant is arguing the references individually
rather than what the combination would have taught or suggested to one of
ordinary skill in the art. Ans. 11.
We disagree with the Examiner and find that the plural models
disclosed by the Beaufort reference pertain to the sub portions of the same
SMS channel and does not teach or suggests the claimed “set of models for a
respective set of communication channels” or “generating, the based on the
set of models, a respective set of channel specific error-type adapter
frameworks,” as claimed. Consequently, we agree with Appellant that the
Examiner has not shown that the combination of Nasri and Beaufort teach or
suggest the claimed invention by a preponderance of the evidence.
Additionally, we find the Examiner speculates as to the specific
teachings and suggestions of the prior art references, as relied upon as
evidence to support the obviousness rejection.2 Therefore, we do not agree

2 The Examiner has it backwards with respect to the misplaced focus on
interpreting the teachings of the reference, instead of properly interpreting
the claims under BRI, and then reading the properly construed claim
language on the corresponding features found in the reference(s). If it is not
clear what the reference teaches, and some degree of speculation is required,
the Examiner should search for a better reference, because we cannot engage
in speculation to affirm the Examiner. Cited prior art references do not
“read on” claim limitations, rather, a particular claim limitation “reads on”
the corresponding feature(s) found in the cited prior art reference(s) after the
claim language is properly construed in accordance with the broadest
reasonable interpretation consistent with the Specification. See e.g., Atlas
Powder Co. v. IRECO, Inc., 190 F.3d 1342, 1346 (Fed Cir.
Appeal 2019-002382
Application 14/980,056

9
with the Examiner’s findings and conclusion of obviousness based upon the
speculative factual findings for the rejection of independent claim 1 and
independent claims 10 and 19 and their dependent claims 3–5, 9, 12–14,
and 18, which contain similar limitations.
With respect to dependent claims 7 and 16, the Examiner has not
identified how the teachings of the Stoytchev reference remedy the noted
deficiency above. Therefore, we similarly cannot sustain the rejection
dependent claims 7 and 16 for the same reasons.
CONCLUSION
We reverse the Examiner’s obviousness rejections.

1999)(“Anticipation of a patent claim requires a finding that the claim at
issue ‘reads on’ a prior art reference.”). We note claim construction is an
important step in a patentability determination. Medichem, S.A. v. Rolabo,
S.L., 353 F.3d 928, 933 (Fed. Cir. 2003) (“Both anticipation under § 102 and
obviousness under § 103 are two-step inquiries. The first step in both
analyses is a proper construction of the claims . . . The second step in the
analyses requires a comparison of the properly construed claim to the prior
art.” (internal citations omitted)) (emphasis added). See also the mapping
rule: 37 C.F.R. § 1.104(c)(2) (“When a reference is complex or shows or
describes inventions other than that claimed by the applicant, the particular
part relied on must be designated as nearly as practicable. The pertinence of
each reference, if not apparent, must be clearly explained and each rejected
claim specified.” (Emphasis added).
Appeal 2019-002382
Application 14/980,056

10
DECISION SUMMARY
In summary:
Claims
Rejected
35 U.S.C.
§
Reference(s)/Basis Affirmed Reversed
1, 3–5, 9,
10, 12–14,
18, 19
103 Nasri, Beaufort 1, 3–5, 9,
10, 12–14,
18, 19
7, 16 103 Nasri, Beaufort,
Stoytchev
7, 16
Overall
Outcome
1, 3–5, 7, 9,
10, 12–14,
16, 18, 19

REVERSED