DETAILED ACTION
Introduction
This office action is in response to Applicant’s amendment filed on November 12, 2025.
Claims 1 and 3-8 have been amended. Claim 2 has been cancelled. Claims 1 and 3-8 are pending in the application. As such, claims 1 and 3-8 have been examined.
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Drawings
The drawings were received on November 20, 2023. These drawings have been accepted and considered by the Examiner.
Response to Amendments and Arguments
In view of the amendments to claims, the amendments to claims 1 and 3-8, have been acknowledged and entered.
In view of the amendments to claims, the objections to claims 1, 3, 4 and 6, have been withdrawn.
In view of the amendments to claims, the interpretation to the claims under 35 U.S.C. 112(f) have been withdrawn.
In view of the amendments to claims, the rejections to claims 1-4 and 6-8 under 35 U.S.C. 112(a) have been withdrawn.
In view of the amendments to claims, the rejections to claims 1-4 and 6-8 under 35 U.S.C. 112(b) have been withdrawn.
In view of the amendments to claims, the rejections to claims 1-8 under 35 U.S.C. 101 have been withdrawn.
In view of the amendments to claims, the rejections to claim 5 under 35 U.S.C. 102 have been withdrawn.
In view of the amendments to claims, the rejections to claims 1-4 and 6-8 under 35 U.S.C. 103 have been withdrawn.
In light of the amendments to the claims, new grounds for rejection for claims 1 and 3-8 under 35 U.S.C. 103 are provided in the response below. New grounds for rejection is based at least upon the following new elements:
a processor;
a program memory connected to the processor and configured to store a computer program executable by the processor;
a supposed input storage that represent possible candidate s to the dialogue apparatus; and
a causality storage
for each of said plurality of supposed inputs stored in said supposed input storage the processor of said training device is configured to execute the computer program stored in the program memory to perform functions of :
one or more s having said supposed input as an input and the extracted causality expression
storing the one or more training data samples in a prescribed storage device;
training the one or more training data samples,
wherein training the neural network used to implement the dialogue apparatus further comprises using a back propagation technique to adjust parameters of the neural network to generate a word vector sequence that forms the result expression in the causality expression of the answer of the one or more training data samples,
extracting, from said plurality of causality expressions, a causality expression having a prescribed relation as said supposed input further comprises extracting, from said plurality of causality expressions, a causality that has a noun phrase of said supposed input in its cause expression.
Applicant’s arguments regarding the prior art rejections under 35 U.S.C 103, received on November 12, 2025, have been fully considered.
Applicant’s remaining arguments with respect to claims 1 and 3-8 have been considered, are directed to the newly amended matter in the claims, are not considered to be persuasive, and are addressed accordingly in the updated rejection rationale below.
Claim Objections
Claim 3 is objected to because of the following informalities:
Claim 3, line 6, reads “wherein processor of the training device.” Examiner believes this to be a clerical error and should read “wherein the processor of the training device.” Claim 4 depends from claim 3, and therefore inherits this objection.
Appropriate correction is required.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claim 1 is rejected under 35 U.S.C. 103 as being unpatentable over Bachrach et al. (US Patent Pub. No. 20190236155 A1), hereinafter Bachrach, in view of Guo et al. (US Patent Pub. No. 20190354630 A1), hereinafter Guo, in view of Bennett (US Patent Pub. No. 20060200353 A1), in view of Zhou et al. (US Patent Pub. No. 20210319188 A1), hereinafter Zhou.
Regarding claim 1, Bachrach teaches a training device for a dialogue apparatus (Bachrach in [0036] teaches using a conversational agent of a system which can be trained),
comprising:
a processor (Bachrach in [0024] teaches using a processor);
a program memory connected to the processor and configured to store a computer program executable by the processor (Bachrach in [0024] teaches using a memory, a processor and executing a program code);
a supposed input storage for storing a plurality of supposed inputs that represent possible candidate inputs to the dialogue apparatus (Bachrach in [0026] teaches using a system where messages may be stored and/or managed as part of a Customer Relationship Management (CRM) platform);
and
a causality storage for storing a plurality of causality expressions (Bachrach in [0041] teaches using a template database which stores a plurality of response templates for selection by the conversational agent to send an agent response);
wherein each of said plurality of causality expressions includes a cause expression and a result expression (Bachrach in [0032] teaches using a system where messages may be grouped into pairs of messages representing a query (e.g. from a user) and a response (e.g. from an agent));
for each of said plurality of supposed inputs stored in said supposed input storage (Bachrach in [0027] teaches using a system where each conversational agent is active in one or more text dialogues at any one time, and in [0028] teaches each text dialogue comprises a sequence of messages that have been exchanged between a user and a conversational agent),
the processor of said training device is configured to execute the computer program stored in the program memory to perform functions of (Bachrach in [0024] teaches using a memory, a processor and executing a program code):
creating one or more training data samples having said supposed input as an input and the [extracted] causality expression as an answer (Bachrach in [0032] teaches using a system where messages may be grouped into pairs of messages representing a query (e.g. from a user) and a response (e.g. from an agent), and in [0036] teaches using a conversational agent of a system which can be trained, and in [0061] teaches using a mechanism to update the training data for the predictive model based on the indicated tokens to disassociate these tokens with the incorrect response template; where, the training data may be updated with text data that includes the indicated tokens, and wherein the text data is paired with the incorrect response template, and in [0041] teaches using a template database which stores a plurality of response templates for selection by the conversational agent to send an agent response),
and
storing the one or more training data samples in a prescribed storage device (Bachrach in [0032] teaches using a system where messages may be grouped into pairs of messages representing a query (e.g. from a user) and a response (e.g. from an agent), and in [0036] teaches using a conversational agent of a system which can be trained, and in [0061] teaches using a mechanism to update the training data for the predictive model based on the indicated tokens to disassociate these tokens with the incorrect response template; where, the training data may be updated with text data that includes the indicated tokens, and wherein the text data is paired with the incorrect response template, and in [0041] teaches using a template database which stores a plurality of response templates for selection by the conversational agent to send an agent response);
training the dialogue apparatus implemented by a neural network designed to generate an output sentence to an input sentence in a natural language, by using the one or more training data samples (Bachrach in [0038] teaches using predictive models which may be based, amongst others, on feed forward neural networks, convolutional neural networks or recurrent neural networks, in [0008] teaches a conversational agent comprising at least a processor and a memory to receive one or more user messages from a client device over a network and send agent messages in response to the one or more user messages, and in [0036] teaches using a conversational agent of a system which can be trained),
wherein training the neural network used to implement the dialogue apparatus further comprises (Bachrach in [0060-0061] teaches training the neural network)
using a back propagation technique to adjust parameters of the neural network to [generate a word vector sequence that] forms the result expression in the causality expression of the answer of the one or more training data samples (Bachrach in [0060-0061] teaches using the training data and back propagation to train the parameter values of the neural network).
Bachrach does not teach, however Guo teaches
extracting, from said plurality of causality expressions, a causality expression having a prescribed relation as said supposed input (Guo in [0062] teaches finding a match, from a local cache, between a query and a question/answer pair, and answering the query with the matching answer),
Guo is considered to be analogous to the claimed invention because it is in the same field of systems which answer user’s questions. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Bachrach further in view of Guo to allow for finding a match, from a local cache, between a query and a question/answer pair, and answering the query with the matching answer. Motivation to do so would allow for reducing negative effects of service waiting time in a human-machine interaction by providing a more natural interaction during the service waiting time (Guo [0021]).
Bachrach, as modified above, teaches the causality expression extracting.
Bachrach, as modified above, does not teach, however Bennett teaches
extracting, from said plurality of causality expressions, a causality that has a noun phrase of said supposed input in its cause expression (Bennett in [0237] teaches the best matching question/answer pair to the user query utterance is determined by analysis of noun-phrases, and noun-phrases are extracted from the user’s question).
Bennett is considered to be analogous to the claimed invention because it is in the same field of systems which answer user’s questions. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Bachrach, as modified above, further in view of Bennett to allow for finding a match using noun-phrases. Motivation to do so would allow for a system to be scaled to accommodate a large number of hits or customers without any corresponding need to increase the number of human resources and its attendant training issues (Bennett [0256]).
Bachrach, as modified above, does not teach, however Zhou teaches
generate a word vector sequence (Zhou in [0057] teaches creating a vector of acceptable words).
Zhou is considered to be analogous to the claimed invention because it is in the same field of natural language processing using neural networks. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Bachrach, as modified above, further in view of Zhou to allow for creating a vector of acceptable words. Motivation to do so would allow for a more efficient training mechanism for training the classifier module for tasks involving generating language (Zhou [0023]).
Claim 3 is rejected under 35 U.S.C. 103 as being unpatentable over Bachrach, in view of Guo, in view of Bennett, in view of Zhou, in view of Nogueira dos Santos et al. (US Patent Pub. No. 20170308790 A1), hereinafter Nogueira dos Santos, in view of Lee et al. (US Patent No. 9666184 B2), hereinafter Lee.
Regarding claim 3, Bachrach, as modified above, teaches the training device according to claim 1.
Bachrach further teaches
wherein processor of the training device is further configured to execute the computer program stored in the program memory to perform a function of (Bachrach in [0024] teaches using a memory, a processor and executing a program code).
Bachrach, as modified above, does not teach, however Zhou teaches
specifying, for each of the causality expressions of the one or more training data samples stored in said prescribed storage device, a word having a high distribution probability for the word included in the causality expression based on outputs of the [topic word model] (Zhou in [0024] teaches identifying a word with a high probability).
Zhou is considered to be analogous to the claimed invention because it is in the same field of natural language processing using neural networks. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Bachrach, as modified above, further in view of Zhou to allow for identifying a word with a high probability. Motivation to do so would allow for a more efficient training mechanism for training the classifier module for tasks involving generating language (Zhou [0023]).
Bachrach, as modified above, does not teach, however Nogueira dos Santos teaches
further comprising:
a topic word model pre-trained such that when a word is given, context word distribution probability of the word is output for each of the words in a predefined lexicon (Nogueira dos Santos in [0020] teaches using a neural network which provides a probability distribution of each word from a given sequence of input words);
Nogueira dos Santos is considered to be analogous to the claimed invention because it is in the same field of natural language processing using neural networks. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Bachrach, as modified above, further in view of Nogueira dos Santos to allow for providing a probability distribution of each word from a given sequence of input words. Motivation to do so would allow for predicting a class label of a text string using a CNN that reduces the impact of an artificial, or none-of-the-above class, on text classification (Nogueira dos Santos [0011]).
Bachrach, as modified above, does not teach, however Lee teaches
adding the specified word to said input of said one or more training data samples to generate a new training data sample (Lee in [col 6 ln 15-25] teaches using a training data converter which replaces a word in the training data with a new word)
and
adding the new training data sample to said prescribed storage device (Lee in [col 6 ln 55-60] teaches using a training data converter which replaces a word in the training data with a new word, and inputting the new word to a neural network as a new training data and storing it in memory).
Lee is considered to be analogous to the claimed invention because it is in the same field of natural language processing using neural networks. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Bachrach, as modified above, further in view of Lee to allow for using a training data converter which replaces a word in the training data with a new word. Motivation to do so would allow for improved recognition results (Lee [col 11 ln 8-20]).
Claim 4 is rejected under 35 U.S.C. 103 as being unpatentable over Bachrach, in view of Guo, in view of Bennett, in view of Zhou, in view of Nogueira dos Santos, in view of Lee, in view of Saito et al. (US Patent Pub. No. 20230026110 A1), hereinafter Saito.
Regarding claim 4, Bachrach, as modified above, teaches the training device according to claim 3.
Bachrach further teaches
wherein processor of the training device is further configured to execute the computer program stored in the program memory to perform a function of: (Bachrach in [0024] teaches using a memory, a processor and executing a program code).
Bachrach, as modified above, does not teach, however Zhou teaches
further comprising
[extracting], based on an output of said topic word model, for each of the causality expressions of the one or more training data samples stored in said prescribed storage device, [a sentence] having a context word distribution probability similar to the context word distribution probability of the causality expression from a prescribed corpus (Zhou in [0024] teaches identifying a word with a high probability, and in [0073] teaches identifying each word of the probability distribution as determined by the underlying NLG logic where the candidate words of the probability distribution are listed in order of descending probability as determined by the NLG)
Zhou is considered to be analogous to the claimed invention because it is in the same field of natural language processing using neural networks. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Bachrach, as modified above, further in view of Zhou to allow for identifying a word with a high probability. Motivation to do so would allow for a more efficient training mechanism for training the classifier module for tasks involving generating language (Zhou [0023]).
Bachrach, as modified above, does not teach, however Saito teaches
extracting, [based on an output of said topic word model, for each of the causality expressions of the one or more training data sample stored in said prescribed storage device], a sentence [having a context word distribution probability similar to the context word distribution probability of the causality expression from a prescribed corpus] (Saito in [0006] teaches extracting a sentence from the text, and adding an extracted sentence as a new training data item),
adding the extracted sentence to said input of said one or more training data sample to generate a second new training data sample (Saito in [0006] teaches determining to use the extracted sentence as a new training data item)
and
adding the second new training data sample to said prescribed storage device (Saito in [0015] teaches an auxiliary storage device stores the installed program and stores necessary files, data, and the like).
Saito is considered to be analogous to the claimed invention because it is in the same field of natural language processing using neural networks. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Bachrach, as modified above, further in view of Saito to allow for extracting a sentence from the text. Motivation to do so would allow for improvement of the accuracy of the neural summarization model (Saito [0035]).
Claim 5 is rejected under 35 U.S.C. 103 as being unpatentable over Bachrach, in view of Guo, in view of Zhou.
Regarding claim 5, Bachrach teaches a natural language dialogue apparatus (Bachrach in [0036] teaches using a conversational agent of a system which can be trained, and in [0019] teaches using a natural language interface),
comprising
a neural network designed to generate an output sentence to a natural language input sentence (Bachrach in [0038] teaches using predictive models which may be based, amongst others, on feed forward neural networks, convolutional neural networks or recurrent neural networks, in [0008] teaches a conversational agent comprising at least a processor and a memory to receive one or more user messages from a client device over a network and send agent messages in response to the one or more user messages),
wherein said neural network is trained such that said output sentence represents a latent result to said input sentence (Bachrach in [0038] teaches using predictive models which may be based, amongst others, on feed forward neural networks, convolutional neural networks or recurrent neural networks, in [0008] teaches a conversational agent comprising at least a processor and a memory to receive one or more user messages from a client device over a network and send agent messages in response to the one or more user messages, and in [0036] teaches using a conversational agent of a system which can be trained),
said latent result reflecting a causality or causality chains having said input sentence as a cause (Bachrach in [0032] teaches using a system where messages may be grouped into pairs of messages representing a query (e.g. from a user) and a response (e.g. from an agent)),
and
said neural network is trained on one or more training data samples (Bachrach in [0038] teaches using predictive models which may be based, amongst others, on feed forward neural networks, convolutional neural networks or recurrent neural networks, in [0008] teaches a conversational agent comprising at least a processor and a memory to receive one or more user messages from a client device over a network and send agent messages in response to the one or more user messages, and in [0036] teaches using a conversational agent of a system which can be trained)
using a back propagation technique to adjust parameters of the neural network to [generate a word vector sequence] that forms a result expression in the causality expression of the answer of the one or more training data samples (Bachrach in [0060-0061] teaches using the training data and back propagation to train the parameter values of the neural network).
Bachrach does not teach, however Guo teaches
that include a) an input and b) a causality expression as an answer (Guo in [0062] teaches finding a match, from a local cache, between a query and a question/answer pair, and answering the query with the matching answer),
Guo is considered to be analogous to the claimed invention because it is in the same field of systems which answer user’s questions. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Bachrach further in view of Guo to allow for finding a match, from a local cache, between a query and a question/answer pair, and answering the query with the matching answer. Motivation to do so would allow for reducing negative effects of service waiting time in a human-machine interaction by providing a more natural interaction during the service waiting time (Guo [0021]).
Bachrach, as modified above, does not teach, however Zhou teaches
generate a word vector sequence (Zhou in [0057] teaches creating a vector of acceptable words).
Zhou is considered to be analogous to the claimed invention because it is in the same field of natural language processing using neural networks. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Bachrach, as modified above, further in view of Zhou to allow for creating a vector of acceptable words. Motivation to do so would allow for a more efficient training mechanism for training the classifier module for tasks involving generating language (Zhou [0023]).
Claim 6 is rejected under 35 U.S.C. 103 as being unpatentable over Bachrach, in view of Guo, in view of Zhou, in view of Su et al. (US Patent No. 9996523 B1), hereinafter Su.
Regarding claim 6, Bachrach, as modified above, teaches the dialogue apparatus according to claim 5.
Bachrach further teaches
further comprising
a processor (Bachrach in [0024] teaches using a processor);
a program memory connected to the processor and configured to store a computer program executable by the processor (Bachrach in [0024] teaches using a memory, a processor and executing a program code);
wherein the processor of the dialogue apparatus executes the computer program stored in the program memory to perform a function of (Bachrach in [0024] teaches using a memory, a processor and executing a program code).
Bachrach, as modified above, does not teach, however Su teaches
adding to an input sentence a related expression, which includes a word or sentence related to said input sentence, to said input sentence and inputting the modified input sentence to said neural network (Su in [col 10 ln 15-45] teaches using a model which is a neural network to autocomplete a user input followed by feeding the combined autocompleted input to the model, and in [col 8 ln 50-65] teaches the autocomplete portion can include words or phrases) .
Su is considered to be analogous to the claimed invention because it is in the same field of systems which answer user’s questions. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Bachrach, as modified above, further in view of Su to allow for autocompletion. Motivation to do so would allow for identifying exact matches and/or autocomplete suggests that are more akin to the user input (Su [col 6 ln 55-60]).
Claims 7 and 8 are rejected under 35 U.S.C. 103 as being unpatentable over Bachrach, in view of Guo, in view of Zhou, in view of Nogueira dos Santos.
Regarding claim 7, Bachrach teaches a dialogue apparatus (Bachrach in [0036] teaches using a conversational agent of a system),
comprising:
an utterance storage storing a past utterance of a user (Bachrach in [0026] teaches using a system where messages may be stored and/or managed as part of a Customer Relationship Management (CRM) platform);
and
a processor configured to execute a computer program stored in a program memory connected to the processor and to perform a function of (Bachrach in [0024] teaches using a memory, a processor and executing a program code)
receiving a user utterance as an input, for generating a response to the user utterance by using user utterances stored in said utterance storage [and the topic model] (Bachrach in [0008] teaches a conversational agent comprising at least a processor and a memory to receive one or more user messages from a client device over a network and send agent messages in response to the one or more user messages),
wherein the response is generated by a neural network trained on one or more training data samples (Bachrach in [0060-0061] teaches training the neural network on data)
using a back propagation technique to adjust parameters of the neural network to [generate a word vector sequence that] forms the result expression in the causality expression of the answer of the one or more training data samples (Bachrach in [0060-0061] teaches using the training data and back propagation to train the parameter values of the neural network).
Bachrach does not teach, however Guo teaches
that include a) an input and b) a causality expression as an answer (Guo in [0062] teaches finding a match, from a local cache, between a query and a question/answer pair, and answering the query with the matching answer),
Guo is considered to be analogous to the claimed invention because it is in the same field of systems which answer user’s questions. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Bachrach further in view of Guo to allow for finding a match, from a local cache, between a query and a question/answer pair, and answering the query with the matching answer. Motivation to do so would allow for reducing negative effects of service waiting time in a human-machine interaction by providing a more natural interaction during the service waiting time (Guo [0021]).
Bachrach, as modified above, does not teach, however Zhou teaches
generate a word vector sequence (Zhou in [0057] teaches creating a vector of acceptable words).
Zhou is considered to be analogous to the claimed invention because it is in the same field of natural language processing using neural networks. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Bachrach, as modified above, further in view of Zhou to allow for creating a vector of acceptable words. Motivation to do so would allow for a more efficient training mechanism for training the classifier module for tasks involving generating language (Zhou [0023]).
Bachrach does not teach, however Nogueira dos Santos teaches
a topic model for outputting context word occurrence probability distribution with respect to an input word (Nogueira dos Santos in [0020] teaches using a neural network (topic model) which provides a probability distribution of each word from a given sequence of input words).
Nogueira dos Santos is considered to be analogous to the claimed invention because it is in the same field of natural language processing using neural networks. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Bachrach, as modified above, further in view of Nogueira dos Santos to allow for providing a probability distribution of each word from a given sequence of input words. Motivation to do so would allow for predicting a class label of a text string using a CNN that reduces the impact of an artificial, or none-of-the-above class, on text classification (Nogueira dos Santos [0011]).
Regarding claim 8, Bachrach teaches a dialogue apparatus (Bachrach in [0036] teaches using a conversational agent of a system),
comprising:
an utterance storage storing a past utterance of a user (Bachrach in [0026] teaches using a system where messages may be stored and/or managed as part of a Customer Relationship Management (CRM) platform);
and
a processor configured to execute a computer program stored in a program memory connected to the processor and to perform a function of: (Bachrach in [0024] teaches using a memory, a processor and executing a program code)
receiving a user utterance as an input, for generating a response to the user utterance (Bachrach in [0008] teaches a conversational agent comprising at least a processor and a memory to receive one or more user messages from a client device over a network and send agent messages in response to the one or more user messages);
and
adjusting generation of said response in accordance with an output of [said topic model] in response to said user utterance (Bachrach in [0036] teaches using a conversational agent of a system which can be trained [here training maps to response adjuster adjusting generation of said response]),
wherein the response is generated by a neural network trained on one or more training data samples (Bachrach in [0060-0061] teaches training the neural network on data)
using a back propagation technique to adjust parameters of the neural network to [generate a word vector sequence that] forms the result expression in the causality expression of the answer of the one or more training data samples (Bachrach in [0060-0061] teaches using the training data and back propagation to train the parameter values of the neural network).
Bachrach does not teach, however Guo teaches
that include a) an input and b) a causality expression as an answer (Guo in [0062] teaches finding a match, from a local cache, between a query and a question/answer pair, and answering the query with the matching answer),
Guo is considered to be analogous to the claimed invention because it is in the same field of systems which answer user’s questions. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Bachrach further in view of Guo to allow for finding a match, from a local cache, between a query and a question/answer pair, and answering the query with the matching answer. Motivation to do so would allow for reducing negative effects of service waiting time in a human-machine interaction by providing a more natural interaction during the service waiting time (Guo [0021]).
Bachrach, as modified above, does not teach, however Zhou teaches
generate a word vector sequence (Zhou in [0057] teaches creating a vector of acceptable words).
Zhou is considered to be analogous to the claimed invention because it is in the same field of natural language processing using neural networks. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Bachrach, as modified above, further in view of Zhou to allow for creating a vector of acceptable words. Motivation to do so would allow for a more efficient training mechanism for training the classifier module for tasks involving generating language (Zhou [0023]).
Bachrach does not teach, however Nogueira dos Santos teaches
a topic model for outputting context word occurrence probability distribution with respect to an input word (Nogueira dos Santos in [0020] teaches using a neural network (topic model) which provides a probability distribution of each word from a given sequence of input words).
Nogueira dos Santos is considered to be analogous to the claimed invention because it is in the same field of natural language processing using neural networks. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Bachrach, as modified above, further in view of Nogueira dos Santos to allow for providing a probability distribution of each word from a given sequence of input words. Motivation to do so would allow for predicting a class label of a text string using a CNN that reduces the impact of an artificial, or none-of-the-above class, on text classification (Nogueira dos Santos [0011]).
Conclusion
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to PAUL J. MUELLER whose telephone number is (571)272-1875. The examiner can normally be reached M-F 9:00am-5:00pm (Eastern).
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Daniel C. Washburn can be reached at 571-272-5551. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
PAUL MUELLER
Examiner
Art Unit 2657
/PAUL J. MUELLER/Examiner, Art Unit 2657
/DANIEL C WASHBURN/Supervisory Patent Examiner, Art Unit 2657
Prosecution Insights