DETAILED ACTION
Notice of 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 .
Response to Arguments
Applicant’s amendments and arguments filed 01/10/2026, with respect to claim(s) 1-20 have been fully considered. Applicant amended claims 4, 11 and 15.
Drawings objections have been withdrawn in view of the replacement sheet of corrected drawing.
35 U.S.C. 112 Claim rejections for claims 4 and 15 have been withdrawn in view of the amended claims filed on 01/10/2026.
35 U.S.C. 101 Claim rejection for claim 11 due to software embodiments, has been withdrawn in view of the amended claims filed on 01/10/2026.
Applicant’s arguments in pages 10-20, filed 01/10/2026, with respect to 35 U.S.C 101 rejections of Claims 1-20 have been fully considered but they are not persuasive. Applicant argued that the claim limitations contain features that cannot be practically performed in human the human mind, such as “first embedding vector representing a location in an embedding graph”, “representing respective locations in the embedding graph”, ” based on a distance in the embedding graph between the location of the first embedding vector and the respective locations of the plurality of prior question embedding vectors”, “using a zero-shot confidence scoring model, a respective confidence score value for the respective prior question embedding vector”. Examiner respectfully disagrees. Embedding vector can be a numerical representation. A person can generate or draw a graph of questions and answers, where questions and answers can be represented as a number. Question answer pairs can be arranged as a set and their locations can be determined from the graph. Based on the similarity of the sets of questions, a confidence score can be calculated. All the above steps could be done with the help of pencil and paper. Applicant recited “zero shot confidence scoring model” for calculating the confidence score and which the examiner considered as additional element. Applicant argued that the use of “zero shot confidence scoring model” as an additional element, is “a first-of-its-kind application” which improves the dynamic system called switchboard. But nothing in the claim language reflects that. The claim language just mentioned “using” the “zero shot confidence scoring model” , but didn’t provide the improvement described in the specification, no features of “zero shot confidence scoring model” is mentioned which is unique for the invention and improves the system. The claim must include components or steps of the invention that provide the improvement described in the specification. Applicant also argued that “ claim 1 as a whole amounts to significantly more than the judicial exception itself” and cited section 2106.05(I)(A)(v) from MPEP to show that “the claim qualifies as significantly more when the claimed features are other than what is well understood, routine, conventional activity in the field”. But Examiner respectfully disagrees. There is no unconventional steps that confine the claim to a particular useful application as explained in section 2106.05(I)(A)(v) of MPEP, cited by applicant, while using the additional element of “zero shot confidence scoring model”. Applicant further argued that the current application is similar to USPTO example 38 and the claim 1 of the current application is eligible because it does not recite a judicial exception (i.e., abstract idea). Examiner respectfully disagrees. Example 38 was related to an analog audio mixer and the current application is related to text. There is no similarity between them. For the above reasons 35 U.S.C 101 rejections of Claims 1-20 have been maintained.
Applicant’s arguments in pages 20-27, filed 01/10/2026, with respect to 35 U.S.C 103 rejections of Claims 1-20 have been fully considered but they are not persuasive. Applicant argued that none of the prior art cited in the rejections discloses or suggests "transforming," "selecting," "generating," and "selecting" steps of the claimed computer-implemented method for clustering and answering questions and how each of the claimed "natural language processing model," "embedding graph," "embedding vectors," "zero-shot confidence scoring model," and "confidence score values" is used. Examiner respectfully disagrees. Karuppusamy in view of Yan further in view of Kermani and Getselevich teach the claimed invention. Applicant further argued that neither Karuppusamy nor the other Art cited in the Office Action discloses or suggests the specific claimed steps and specific uses with those steps of the claimed models, graphs and vectors. Examiner wanted to emphasize that it was an obviousness rejection, which was established by combining or modifying the teachings of the prior art to produce the claimed invention where there is some teaching, suggestion, or motivation to do so found either in the references themselves or in the knowledge generally available to one of ordinary skill in the art. See In re Fine, 837 F.2d 1071, 5 USPQ2d 1596 (Fed. Cir. 1988), In re Jones, 958 F.2d 347, 21 USPQ2d 1941 (Fed. Cir. 1992), and KSR International Co. v. Teleflex, Inc., 550 U.S. 398, 82 USPQ2d 1385 (2007). In this case, Karuppusamy teaches a system and method of automatically responding to request from user, where each request and/or group of requests ( prior and present) can be presented by a vector of tags ( embedding vector), a degree of similarity can be calculated between the vectors, prior question vector can be selected based on the score and answers can be generated associated with the question vectors, in para.[0039]-0041], [0044], [0050]. Karuppusamy didn’t teach transforming text to embedding vectors by using natural language processing, but Yan teaches transforming a text to embedding vectors by using NLU processing in column 2, lines 3-12, column 6, lines 33-42, Fig. 2. Karuppusamy in view of Yan didn’t teach embedding graphs with prior questions, answers and use of zero shot learning model, but Kermani teaches a community based question answering systems where embedding graphs can represent location for prior question and answer set, in para.[0039],[0045], [0051]. Getselevich teaches calculating probability score using zero shot learning for the intent of a query in para.[0026]. The rejection is made by combining the teachings of the prior art to produce the claimed invention. Applicant cited para.[0011], [0022] from specification to illustrate that their invention is an improvement over the prior art Karuppusamy. But Karuppusamy in view of Yan, Kermani and Getselevich does improve the question answering system and method with the similar benefits as stated by applicant, by using a zero shot learning model.
In response to applicant's argument that the examiner's conclusion of obviousness is based upon improper hindsight reasoning, it must be recognized that any judgment on obviousness is in a sense necessarily a reconstruction based upon hindsight reasoning. But so long as it takes into account only knowledge which was within the level of ordinary skill at the time the claimed invention was made, and does not include knowledge gleaned only from the applicant's disclosure, such a reconstruction is proper. See In re McLaughlin, 443 F.2d 1392, 170 USPQ 209 (CCPA 1971). For the above reasons, 35 U.S.C 103 rejections of Claims 1-20 have been maintained.
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claims 1-20 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an
abstract idea without significantly more.
The Independent claims 1, 11 and 12 recite “obtaining an input from a user device, wherein the input comprises a text”; “transforming, using a first natural language processing model, the text into a first embedding vector representing a location in an embedding graph, wherein the embedding graph comprises a plurality of prior question embedding vectors representing respective locations in the embedding graph and each prior question embedding vector is associated with at least one answer text”; “selecting a set of one or more prior question embedding vectors based on a distance in the embedding graph between the location of the first embedding vector and the respective locations of the plurality of prior question embedding vectors”; “for each respective prior question embedding vector in the selected set of one or more prior question embedding vectors, generating, using a zero-shot confidence scoring model, a respective confidence score value for the respective prior question embedding vector, wherein the respective confidence score value corresponds to a degree of similarity between the first embedding vector and the respective prior question embedding vector”; “selecting a first prior question embedding vector from the selected set of one or more prior question embedding vectors based on the generated respective confidence score value of the first prior question embedding vector”; “obtaining an answer text associated with the first prior question embedding vector”; “and generating a response comprising the identified answer text”. The limitations above as drafted, is a process that, under its broadest reasonable
interpretation, covers a mental process, as this could be performed in the human mind or with the aid of
pen and paper.
The limitation of " obtaining ... ", “transforming..”, "selecting ... ", "generating ... ", as drafted covers mental activities. More specifically, a human can receive a question or query, convert the question to a certain format such as representations of data point in a graph ( same as embedded vector in an embedded graph) by following natural language processing, which represent a location in a graph/representation, where previously asked questions and corresponding answers are formatted ( embedded vector) similarly and also in the same representation ( embedded graph) as the first question, human can find/generate the confidence scores of each previous formatted questions, finding out the most similar previous formatted question and generate a response from that most similar question. All the steps above are examples of observation and evaluation that could be performed in the human mind or with the aid of pencil and paper.
Claims recite the additional limitation of “natural language processing model”, “Zero shot confidence scoring model”, “user device” and additionally, claims 11, 12 recite the additional limitation of a “processor”, “processing circuitry”, “ non transitory computer readable medium” , for performing the method. “Natural language processing model”, as specified in para.[0066], [0080], [0081], is a Bidirectional Encoder Representations from Transformers (BERT) model, a well-known model available from Google, which is not sufficient to amount to significantly more than the judicial exception. “Zero shot confidence scoring model” is based on zero shot learning which is a certain training/learning method and as specified in specification para.[0016],[0038],[0082],[0083], is a well-known learning/training techniques, which is not sufficient to amount to significantly more than the judicial exception. “ User device” as specified in para.[0017] in the specification, can be any electronic computing device, such as a mobile device, laptop, computer, desktop, tablet, which is not sufficient to amount to significantly more than the judicial exception. “Processor”, “processing circuitry”, “non-transitory computer readable medium” are recited at a high level of generality and are recited as performing generic computer functions routinely used in computer applications. The current specification in paragraphs [0020],[0043],[0044] clearly specifies them as performing generic computer functions that are well-understood, routine and conventional activities amount to no more than implementing the abstract idea with a computerized system. The claims as drafted, are not patent eligible.
Thus, taken alone, the additional elements do not amount to significantly more than the above identified judicial exception (the abstract idea). Looking at the limitations as an ordered combination adds
nothing that is not already present when looking at the elements taken individually. There is no indication
that the combination of elements improves the functioning of a computer or improves any other
technology. Their collective functions merely provide conventional computer implementation. Claims 1, 11
and 12 are therefore not drawn to eligible subject matter as they are directed to an abstract idea without
significantly more than the abstract idea.
Claims 2 and 13 recite the additional limitation of “wherein the first natural language processing model is a Bidirectional Encoder Representations from Transformers (BERT) model” , where BERT is an additional element, as shown in specification, in the cited references, para.[0080], [0081], is a well-known model available from Google, which is not sufficient to amount to significantly more than the judicial exception. The claims 2 and 13 as drafted, are not patent eligible.
Claims 3 and 14 recite “further comprising: transmitting the response towards the user device over a network”, which is an evaluation, observation that human can just determine. The claim(s) does/do not include additional elements that are sufficient to amount to significantly more than the judicial exception, as claims 3 and 14 do not recite any additional limitations. The claims as drafted, are not patent eligible.
Claims 4 and 15 recite “further comprising: outputting the first prior question embedding vector; obtaining a second input, the second input comprising an indication to use a second prior question embedding vector different than the first prior question embedding vector; and updating the zero-shot confidence scoring model based on the second input”, writing the first previous question, receiving second input/question, deciding that the second previous question is different that the first one and updating the learning or calculating the score could be done in human mind and with the help of pen and paper. Zero shot learning model as specified in specification para.[0016],[0038],[0082],[0083], is a well-known learning/training techniques, which is not sufficient to amount to significantly more than the judicial exception. The claims 4 and 15 as drafted, are not patent eligible.
Claims 5 and 16 recite “further comprising: removing a prior question embedding vector from the selected set of prior question embedding vectors based on a filter”. Removing a prior question based on some predetermined rules could be performed in the human mind or with the aid of pen and paper. The claim(s) does/do not include additional elements that are sufficient to amount to significantly more than the judicial exception, as claims 5 and 16 do not recite any additional limitations. The claims as drafted, are not patent eligible.
Claims 6 and 17 recite the additional limitations of “further comprising: receiving a third input from a second user different than the first user comprising the filter” . Receiving another question from another user which is different than the first user and comprising some restrictions, could be assessed in the human mind or with the aid of pen and paper. The claim(s) does/do not include additional elements that are sufficient to amount to significantly more than the judicial exception, as claims 6 and 17 do not recite any additional limitations. The claims as drafted, are not patent eligible.
Claims 7 and 18 recite “wherein the obtaining the answer text comprises: determining that the first prior question embedding vector is associated with a first answer text from a first data source and a second answer text from a second data source”; “and selecting at least one of the first answer text and the second answer”, where determining which question corresponds to which answer and what is the data source could be find out by evaluation, observation, assessment and could be performed in the human mind or with the aid of pen and paper. The claim(s) does/do not include additional elements that are sufficient to amount to significantly more than the judicial exception, as claims 7 and 18 do not recite any additional limitations. The claims as drafted, are not patent eligible.
Claims 8 and 19 recite “wherein the input is obtained over a predetermined time frame, and wherein the response is generated within the predetermined time frame”. Determining the time frame for both question and answer could be done by evaluation, observations and could be performed in the human mind or with the aid of pen and paper. The claim(s) does/do not include additional elements that are sufficient to amount to significantly more than the judicial exception, as claims 8 and 19 do not recite any additional limitations. The claims as drafted, are not patent eligible.
Claims 9 and 20 recite “further comprising: identifying a location of a cluster of prior question embedding vectors nearest to the location of the first embedding vector, wherein the selected set of prior question embedding vectors comprises one or more prior question embedding vectors of the cluster”, where identifying a cluster of previous questions close to the current questions is observations, assessments, performed in the human mind or with the aid of pen and paper. The claim(s) does/do not include additional elements that are sufficient to amount to significantly more than the judicial exception, as claims 9 and 20 do not recite any additional limitations. The claims as drafted, are not patent eligible.
Claim 10 recites “wherein the selected set of prior embedding vectors comprises a predetermined number of prior question embedding vectors of the cluster”. Determining the number of prior questions in the selected previous group of questions are predetermined, could be done by evaluation, observations and could be performed in the human mind or with the aid of pen and paper. The claim(s) does/do not include additional elements that are sufficient to amount to significantly more than the judicial exception, as claim 10 does not recite any additional limitations. The claim as drafted, is not patent eligible.
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 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 of this title, 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.
Claims 1-4, 7, 9-14, 15, 18 and 20 are rejected under 35 U.S.C. 103 as being unpatentable over Karuppusamy et al. ( US 20190349320 A1), hereinafter referenced as Karuppusamy, in view of Yan et al. ( US 11842738 B1), hereinafter referenced as Yan, further in view of Zihayat Kermani et al. (US 20230030624 A1), hereinafter referenced as Zihayat Kermani, further in view of Getselevich et al. (US 20230142339 A1), hereinafter referenced as Getselevich.
Regarding Claim 1, Karuppusamy teaches a computer-implemented method for clustering and answering questions, the method comprising:
obtaining an input from a user device, wherein the input comprises a text ( Karuppusamy: Para. [0026],[0050], Figs. 2, 7 illustrates a method 700 of providing automatic responses to requests received from users. At step 704, a new request (e.g., a text message) is received from a user of the online messaging system from user device 206);
for each respective prior question embedding vector in the selected set of one or more prior question embedding vectors, generating, [using a zero-shot confidence scoring model,] a respective confidence score value for the respective prior question embedding vector, wherein the respective confidence score value corresponds to a degree of similarity between the first embedding vector and the respective prior question embedding vector ( Karuppusamy: Para.[0040], each request and/or group of requests can be presented by a vector of tags. Para.[0039],[0041], [0045], Fig. 4, suggestion module 418, by using relevance model, which can be of various types, can implement
a ranking function such as, for example, cosine similarity. (a degree of similarity) for the previous request vectors and the vector of new request);
selecting a first prior question embedding vector from the selected set of one or more prior question embedding vectors based on the generated respective confidence score value of the first prior question embedding vector ( Karuppusamy: Para. [0041], Fig. 4, to find a match between a vector of new request and the vector of one or a group of previous requests, the one or the group of previous requests having the largest cosine similarity with the new request can be considered to be the closest match (first prior question embedding vector) );
obtaining an answer text associated with the first prior question embedding vector ( Karuppusamy: Para. [0041], Fig. 4, Once a match has been found between the new request and one or more groups of previous requests, one or more automatic responses 420 mapped to the one or more groups of previous requests can be identified ( e.g., based on response IDs) by the suggestion module 418 );
and generating a response comprising the identified answer text ( Karuppusamy: Para. [0041],[0044], Table 4, the suggestion module 418 can rank any matches between the new request and the one or multiple groups of previous requests (e.g., based on calculated cosine similarities. Automatic responses that most closely matches the new request can be given a highest ranking. Para.[0050], Fig. 7 illustrates a method 700 of providing automatic responses to requests received from users. At step 710 retrieved previous automatic response is provided to the user of the online messaging system).
Karuppusamy while teaching the method of claim 1, fails to explicitly teach the claimed, transforming, using a first natural language processing model, the text into a first embedding vector representing a location in an embedding graph, wherein the embedding graph comprises a plurality of prior question embedding vectors representing respective locations in the embedding graph and each prior question embedding vector is associated with at least one answer text; selecting a set of one or more prior question embedding vectors based on a distance in the embedding graph between the location of the first embedding vector and the respective locations of the plurality of prior question embedding vectors; generating, using a zero-shot confidence scoring model, a respective confidence score value for the respective prior question embedding vector.
However, Yan does teach the claimed, transforming, using a first natural language processing model, the text into a first embedding vector representing a location in an embedding graph ( Yan: Column 2, lines 3-12, column 6, lines 33-42, Fig. 2, Upon receiving the text 212, after NLU processing at 220, the ML transformer 230 generates embedding vectors. Column 9, lines 54-63, Fig. 4 illustrates an example of generating embedding vectors from text. Each embedding vector represents a word embedding a set of words from the text, such as a learned representation of these words in an embedding space) ,
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate Yan’s teaching of providing computing services using embeddings of a transformer-based encoder, into the system and method of automatically responding to requests from users in an online messaging system, taught by Karuppusamy, because, this would improve the user perceived latency and allow flexibility and scalability in the deployment which involve pre-computing embedding vectors by natural language processing and an ML task. (Yan [ Column 2, lines 17-29]).
Karuppusamy in view of Yan, while teaching the method of claim 1, fail to explicitly teach the claimed, wherein the embedding graph comprises a plurality of prior question embedding vectors representing respective locations in the embedding graph and each prior question embedding vector is associated with at least one answer text; selecting a set of one or more prior question embedding vectors based on a distance in the embedding graph between the location of the first embedding vector and the respective locations of the plurality of prior question embedding vectors; generating, using a zero-shot confidence scoring model, a respective confidence score value for the respective prior question embedding vector.
However, Zihayat Kermani does teach the claimed, wherein the embedding graph comprises a plurality of prior question embedding vectors representing respective locations in the embedding graph and each prior question embedding vector is associated with at least one answer text (Zihayat Kermani: Para.[0039], Fig. 3, Existing question/answer data 320 includes previous questions and answers mapping, where "n past" questions denoted as Q={ q_l, q_2, . . , q_n}, A_i={ a_l, a_2, ... ,a_ni} as the set of ni answers to question q_i ) ;
selecting a set of one or more prior question embedding vectors based on a distance in the embedding graph between the location of the first embedding vector and the respective locations of the plurality of prior question embedding vectors (Zihayat Kermani: Para.[0045], [0051], Fig. 4, selecting top k most similar question to new question 370 from existing question/answer data 320 by computing distances between the embedding vectors of the existing teams in the embedding space and the embedding vector of the new question. Team discovery stage 340 then uses embedding vectors of the top k most similar questions to map new question 370 into embedding space 350);
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate Zihayat Kermani’s teaching of community-based question answering (CQA) system that connect information seekers with expert’s knowledge and are dedicated platforms for users to respond to other user’s questions, into the system and method, taught by Karuppusamy in view of Yan, because, this would allow the user to receive highly specific answers to their questions by utilizing the experts in the CQA systems. (Zihayat Kermani [ Para.[0002]-[0004]).
Karuppusamy in view of Yan further in view of Zihayat Kermani, while teaching the method of claim 1, fail to explicitly teach the claimed, generating, using a zero-shot confidence scoring model, a respective confidence score value,
However, Getselevich does teach the claimed, generating, using a zero-shot confidence scoring model, a respective confidence score value (Getselevich: Para.[0026], Fig. 3, the classifier 302 utilizes one or more zero-shot approaches (e.g., zero-shot learning) to determine the probability of the intent. The intent classifier 302 receives, as an input, one or more words or word sequences, which may have undergone one or more preprocessing steps, and then determines a probability that the word or word sequence belongs to one or more intents classifications (e.g., labels). A highest probability score ( confidence score) may then be selected for classifying the word or word sequences).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate Getselevich’s teaching of systems and methods for zero-shot approaches to interaction environments, into the system and method, taught by Karuppusamy in view of Yan further in view of Zihayat Kermani, because, this would improve the flexibility to the system, where specific training examples are not used for a specific environment, but rather, the zero-shot approach allows a trained system to then adapt to various different user-provided labels. (Getselevich [ Para.[0017],[0027]).
Claim 11 is a computer program product comprising a non-transitory computer readable medium storing a computer program comprising claim comprising instructions which when executed by processing circuity of a device causes the device ( Karuppusamy: Para.[0053], [0055], Implementations of the subject matter and the operations described in this specification can be implemented in digital electronic circuitry, or in computer software, firmware, or hardware, including the structures disclosed in this specification and their structural equivalents, or in combinations of one or more of them. Implementations of the subject matter described in this specification can be implemented as one or more computer programs, i.e., one or more modules of computer program instructions, encoded on computer storage medium for execution by, or to control the operation of, data processing apparatus), to perform the steps in method claim 1 above and as such, claim 11 is similar in scope and content to claim 1 and therefore, claim 11 is rejected under similar rationale as presented against claim 1 above.
Claim 12 is a system claim for clustering and answering questions, the system comprising: a processor ( Karuppusamy: Para.[0009],[0012], system having one or more computer processors programmed to perform operations); and a non-transitory computer readable memory coupled to the processor, wherein the system is configured ( Karuppusamy: Para.[0012], non-transitory computer-readable medium having instructions stored thereon that, when executed by one or more computer processors, cause the computer processors to perform operations), to perform the steps in method claim 1 above and as such, claim 12 is similar in scope and content to claim 1 and therefore, claim 12 is rejected under similar rationale as presented against claim 1 above.
Regarding Claim 2, Karuppusamy in view of Yan further in view of Zihayat Kermani, further in view of Getselevich teach the method of claim 1.Yan further teaches, wherein the first natural language processing model is a Bidirectional Encoder Representations from Transformers (BERT) model ( Yan: Column 8, lines 42-48, Fig. 3, ML transformer 310 includes a deep learning model trained for natural language processing and which can be a pre-trained system, such as a Bidirectional Encoder Representations from Transformers (BERT)).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate Yan’s teaching of providing computing services using embeddings of a transformer-based encoder, into the system and method, taught by Karuppusamy in view of Zihayat Kermani, further in view of Getselevich, because, this would improve the user perceived latency and allow flexibility and scalability in the deployment which involve pre-computing embedding vectors by natural language processing and an ML task. (Yan [ Column 2, lines 17-29]).
Claim 13 is a system claim performing the steps in method claim 2 above and as such, claim 13 is similar in scope and content to claim 2 and therefore, claim 13 is rejected under similar rationale as presented against claim 2 above.
Regarding Claim 3, Karuppusamy in view of Yan further in view of Zihayat Kermani, further in view of Getselevich teach the method of claim 1. Karuppusamy further teaches, further comprising: transmitting the response towards the user device over a network ( Karuppusamy: Para.[0023],[0026], Figs. 1, 2, The identified responses can be provided (step 212) to the application module 118, which can forward (step 214) the responses to the client device 206 through a network 134).
Claim 14 is a system claim performing the steps in method claim 3 above and as such, claim 14 is similar in scope and content to claim 3 and therefore, claim 14 is rejected under similar rationale as presented against claim 3 above.
Regarding Claim 4, Karuppusamy in view of Yan further in view of Zihayat Kermani, further in view of Getselevich teach the method of claim 1. Karuppusamy further teaches, further comprising: outputting the first prior question embedding vector ( Karuppusamy: Para.[0044], Table 4 illustrates existing requests. The request with response ID=1 ( second prior question));
obtaining a second input, the second input comprising an indication to use a second prior question embedding vector different than the first prior question embedding vector ( Karuppusamy: Para.[0044], Table 4 illustrates another existing request with response ID=2 ( first prior question). When a new request is received ( second input), the response ranker model can rank Response ID=1 higher than Response ID=2, and the automatic response associated with Response ID=1 can be provided to the user who submitted the new request);
Getselevich further teaches, and updating the zero-shot confidence scoring model based on the second input ( Getselevich: Para.[0018], [0021], the system use conversational AI using zero shot approach and can provide the operator flexibility to add new command/intent/request. Each of these intents may have a corresponding question or follow on action, which may then be used to select a value to fill a slot. As an example, a user intent may relate to changing a car color, the corresponding question would be “which color” and the values to fill that slot (e.g., answer the question) could be any number of colors. Para.[0026], [0027], Fig. 3, the classifier 302 utilizes one or more zero-shot approaches (e.g., zero-shot learning) to determine a probability that the word or word sequence belongs to one or more intents classifications (e.g., labels). Labels may be associated with changing a color, changing a camera angle,( as second input/request) and the like. Systems and methods may be utilized to establish specific labels for specific actions based, on the interaction environment. Zero-shot approach allows a trained system to then adapt to various different user-provided labels).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate Getselevich’s teaching of systems and methods for zero-shot approaches to interaction environments, into the system and method, taught by Karuppusamy in view of Yan further in view of Zihayat Kermani, because, this would improve the flexibility to the system, where specific training examples are not used for a specific environment, but rather, the zero-shot approach allows a trained system to then adapt to various different user-provided labels. (Getselevich [ Para.[0017],[0027]).
Claim 15 is a system claim performing the steps in method claim 4 above and as such, claim 15 is similar in scope and content to claim 4 and therefore, claim 15 is rejected under similar rationale as presented against claim 4 above.
Regarding Claim 7, Karuppusamy in view of Yan further in view of Zihayat Kermani, further in view of Getselevich teach the method of method of claim 1. Karuppusamy further teaches, wherein the obtaining the answer text comprises: determining that the first prior question embedding vector is associated with a first answer text from a first data source and a second answer text from a second data source ( Karuppusamy: Para.[0049], Fig. 6, the suggestion module 418 can receive a user request 606 as input and provide, based on the training data which is a collection of previous user requests and corresponding automatic responses ( first and second answer), as output. The data can be received from response data 130 database( first data source). Para.[0028], manual response can be generated from customer service representative ( second data source) and data base can be updated with the manual response );
and selecting at least one of the first answer text and the second answer ( Karuppusamy: Para.[0049], Fig. 6, the suggestion module 418 can rank the automatic responses 420 according to a predicted relevance. Para.[0050], at step 708, text responses are provided);
Claim 18 is a system claim performing the steps in method claim 7 above and as such, claim 18 is similar in scope and content to claim 7 and therefore, claim 18 is rejected under similar rationale as presented against claim 7 above.
Regarding Claim 9, Karuppusamy in view of Yan further in view of Zihayat Kermani, further in view of Getselevich teach the method of claim 1. Zihayat Kermani further teaches, further comprising: identifying a location of a cluster of prior question embedding vectors nearest to the location of the first embedding vector, wherein the selected set of prior question embedding vectors comprises one or more prior question embedding vectors of the cluster ( Zihayat Kermani: Para.[0047], [0048], Fig. 4 illustrates a CQA system that maps existing question expert pairs to an embedding space; maps a new question to the embedding space; and forms a new team based on the relative location of the new question to existing questions in the embedding space).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate Zihayat Kermani’s teaching of community-based question answering (CQA) system that connect information seekers with expert’s knowledge and are dedicated platforms for users to respond to other user’s questions, into the system and method, taught by Karuppusamy in view of Yan, because, this would allow the user to receive highly specific answers to their questions by utilizing the experts in the CQA systems. (Zihayat Kermani [ Para.[0002]-[0004]).
Claim 20 is a system claim performing the steps in method claim 9 above and as such, claim 20 is similar in scope and content to claim 9 and therefore, claim 20 is rejected under similar rationale as presented against claim 9 above.
Regarding Claim 10, Karuppusamy in view of Yan further in view of Zihayat Kermani, further in view of Getselevich teach the method of claim 9. Zihayat Kermani further teaches, wherein the selected set of prior embedding vectors comprises a predetermined number of prior question embedding vectors of the cluster ( Zihayat Kermani: Para.[0044], Fig. 4, CQA system 300 uses learn to rank model 450 to obtain the top k most similar question to new question 370 from existing question/answer data 320).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate Zihayat Kermani’s teaching of community-based question answering (CQA) system that connect information seekers with expert’s knowledge and are dedicated platforms for users to respond to other user’s questions, into the system and method, taught by Karuppusamy in view of Yan, because, this would allow the user to receive highly specific answers to their questions by utilizing the experts in the CQA systems. (Zihayat Kermani [ Para.[0002]-[0004]).
Claims 5, 6, 16 and 17 are rejected under 35 U.S.C. 103 as being unpatentable over Karuppusamy et al. ( US 20190349320 A1), hereinafter referenced as Karuppusamy, in view of Yan et al. ( US 11842738 B1), hereinafter referenced as Yan, further in view of Zihayat Kermani et al. (US 20230030624 A1), hereinafter referenced as Zihayat Kermani, further in view of Getselevich et al. (US 20230142339 A1), hereinafter referenced as Getselevich, further in view of Liu et al. (US 20080294637 A1), hereinafter referenced as Liu.
Regarding Claim 5, Karuppusamy in view of Yan further in view of Zihayat Kermani, further in view of Getselevich teach the method of claim 1. Liu further teaches, further comprising: removing a prior question embedding vector from the selected set of prior question embedding vectors based on a filter ( Liu: Para.[0059], filtering out those previous questions whose filling words/phrases in the corresponding blanks are not among the synonyms).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate Liu’s teaching of web-based user-interactive question-answering method and system, into the system and method, taught by Karuppusamy in view of Yan further in view of Zihayat Kermani, further in view of Getselevich , because, this would improve the efficiency of user’s asking and answering, the computer's processing accuracy of questions and/or answers, and facilitate knowledge acquisition. (Liu [ Para.[0004],[0005]).
Claim 16 is a system claim performing the steps in method claim 5 above and as such, claim 16 is similar in scope and content to claim 5 and therefore, claim 16 is rejected under similar rationale as presented against claim 5 above.
Regarding Claim 6, Karuppusamy in view of Yan further in view of Zihayat Kermani, further in view of Getselevich, further in view of Liu teach the method of claim 5. Liu further teaches, further comprising: receiving a third input from a second user different than the first user comprising the filter ( Liu: Para.[0217], [0223], [0225], in response to the input/question from another user, distance based method of filtering is used, which is different than the previous filtering, which was based on synonyms) .
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate Liu’s teaching of web-based user-interactive question-answering method and system, into the system and method, taught by Karuppusamy in view of Yan further in view of Zihayat Kermani, further in view of Getselevich , because, this would improve the efficiency of user’s asking and answering, the computer's processing accuracy of questions and/or answers, and facilitate knowledge acquisition. (Liu [ Para.[0004],[0005]).
Claim 17 is a system claim performing the steps in method claim 6 above and as such, claim 17 is similar in scope and content to claim 6 and therefore, claim 17 is rejected under similar rationale as presented against claim 6 above.
Claims 8 and 19 are rejected under 35 U.S.C. 103 as being unpatentable over Karuppusamy et al. ( US 20190349320 A1), hereinafter referenced as Karuppusamy, in view of Yan et al. ( US 11842738 B1), hereinafter referenced as Yan, further in view of Zihayat Kermani et al. (US 20230030624 A1), hereinafter referenced as Zihayat Kermani, further in view of Getselevich et al. (US 20230142339 A1), hereinafter referenced as Getselevich, further in view of Doerflinger et al. (US 20190180871 A1), hereinafter referenced as Doerflinger.
Regarding Claim 8, Karuppusamy in view of Yan further in view of Zihayat Kermani, further in view of Getselevich teach the method of claim 1. Karuppusamy in view of Yan further in view of Zihayat Kermani, further in view of Getselevich fail to explicitly teach the claimed, wherein the input is obtained over a predetermined time frame, and wherein the response is generated within the predetermined time frame.
However, Doerflinger does teach the claimed, wherein the input is obtained over a predetermined time frame, and wherein the response is generated within the predetermined time frame ( Doerflinger: Para.[0033], Figs. 1, 2, In step 204, evaluation function 112 prompts the patient to produce a predetermined quantity of requirements, such as of answers to prompts. The evaluation function 112 may require these requirements input within a predetermined time period. Para.[0038], Figs. 1, 2, In decision 210, evaluation function 112 determines if the patient responds to the alert. The patient has a predetermined time frame to respond to the alert).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate Doerflinger’s teaching of a method for determining and assisting in a patient's cognitive or emotional condition by analyzing data sets, into the system and method, taught by Karuppusamy in view of Yan, further in view of Zihayat Kermani, further in view of Getselevich, because, this would allow continuous communication between the providers and the client so that the provider can properly monitor and assist the client. (Doerflinger [ Para.[0004]).
Claim 19 is a system claim performing the steps in method claim 8 above and as such, claim 19 is similar in scope and content to claim 8 and therefore, claim 19 is rejected under similar rationale as presented against claim 8 above.
Conclusion
10. 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 extension fee 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 date of this final action.
11. Any inquiry concerning this communication or earlier communications from the examiner should be directed to NADIRA SULTANA whose telephone number is (571)272-4048. The examiner can normally be reached M-F,7:30 am-5:00pm.
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, Paras D. Shah can be reached on (571) 270-1650. 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.
/NADIRA SULTANA/
Examiner, Art Unit 2653
/Paras D Shah/Supervisory Patent Examiner, Art Unit 2653
04/09/2026
Prosecution Insights