DETAILED ACTION
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 .
Response to Amendment
This communication is responsive to the applicant’s amendments dated 6/2/2026. The applicant has amended claims 1, 3-5, 7-8, 10-15, and 17-20. Next, the applicant has cancelled claims 2, 9, and 16. Lastly, the applicant has added new claims 21-23.
Response to Arguments
Applicant's arguments with respect to 35 U.S.C. 101 (See Remarks pg. 9, line 25 – pg. 18, line 16) filed 6/2/2026 have been fully considered but they are not persuasive.
In terms of the obtaining limitations that was added in the claim language, the use of the first AI model doesn’t necessarily preclude human practical performance of that step. The examiner believes the whole “obtaining” limitation is abstract. A human is able to obtain a list of stored intents. Additionally, a human can mentally generate a response based off the intent that is personalized for the user knowing the context. The generating a first response limitation is not recited at such a high level where a human could not perform that step. The AI model is stepping in for a human process. Additionally, as shown in the prior art, this type of AI model/neural network/machine learning model/LLM are well-known, routine, and conventional. Therefore, the 35 U.S.C. 101 rejection is maintained.
Applicant’s arguments with respect to 35 U.S.C. 102 (See Remarks pg. 18, line 17 – pg. 20, line 2) have been considered but are moot because the new ground of rejection does not rely on any reference applied in the prior rejection of record for any teaching or matter specifically challenged in the argument. Given the amendments, a new ground of rejection is provided below.
Applicant’s arguments with respect to 35 U.S.C. 103 (See Remarks pg. 20, line 3 – pg. 22, line 10) have been considered but are moot because the new ground of rejection does not rely on any reference applied in the prior rejection of record for any teaching or matter specifically challenged in the argument. Given the amendments, a new ground of rejection is provided below.
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 1,3-8,10-15 and 17-23 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Independent claim 1 and 8 recites, “generating a first user request from a first user utterance submitted by a user”, “classifying a first user intent from the first user request and a-context information associated with the first user utterance”, “obtaining a master intent list comprising a set of stored intents for which a set of human-curated responses are available through a first Al model, wherein the first Al model is trained for predicting responses to user requests using the set of human-curated responses”, “based on at least the first user intent matching a stored intent, in the set of stored intents, of the master intent list, sending the first user request to the first Al model”, “generating, by the first Al model, a first response responsive to the first user request, wherein the first response comprises a first human-curated response that is associated with the stored intent, in the set of stored intents, of the master intent list and is personalized for the user based on the context information associated with the first user utterance”, “presenting the first response to the user”, “generating a second user request from a second user utterance submitted by the user”, “classifying a second user intent from the second user request”, “based on at least the second user intent failing to match a respective stored intent, in the set of stored intents, of the master intent list, sending the second user request to a second Al model”, “generating, by the second Al model, a second response responsive to the second user request”, and “presenting the second response to the user”.
The limitation of generating a user request from an utterance, as drafted, is a process, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components. That is, other than reciting “a memory” and “a processor”, nothing in the claim precludes the step from practically being performed in the mind. For example, “generating” in the context of this claim encompasses understanding speech, which a human can do in the mind. Next, the limitation of classifying intent, as drafted, is a process, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components. That is, other than reciting the components listed above, nothing in the claim precludes the step from practically being performed in the mind. For example, “classifying” in the context of this claim encompasses analyzing speech, which a human can do in the mind. Next, the limitation of obtaining a master intent list, as drafted, is a process, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components. That is, other than reciting the components listed above, nothing in the claim precludes the step from practically being performed in the mind. For example, “obtaining” in the context of this claim encompasses obtaining a list of human responses, which a human can do in the mind or with a pen and paper. Next, the limitation of sending a request based off a match, as drafted, is a process, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components. That is, other than reciting the components listed above, nothing in the claim precludes the step from practically being performed in the mind. For example, “sending” in the context of this claim encompasses sending a request after a certain criterion has been met, which a human can do in the mind or with a pen and paper. Next the limitation of generating a response, as drafted, is a process, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components. That is, other than reciting generic models, nothing in the claim precludes the step from practically being performed in the mind. For example, “generating” in the context of this claim encompasses producing a response to a question, which a human can do in the mind or with a pen and paper. Next, the limitation of presenting a response, as drafted, is a process, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components. That is, other than reciting the components listed above, nothing in the claim precludes the step from practically being performed in the mind. For example, “presenting” in the context of this claim encompasses transmitting information, which a human can do in the mind or with a pen and paper. Next, the limitation of generating a request, as drafted, is a process, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components. That is, other than reciting “a memory” and “a processor”, nothing in the claim precludes the step from practically being performed in the mind. For example, “generating” in the context of this claim encompasses understanding speech, which a human can do in the mind. Next, the limitation of classifying intent, as drafted, is a process, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components. That is, other than reciting the components listed above, nothing in the claim precludes the step from practically being performed in the mind. For example, “classifying” in the context of this claim encompasses analyzing speech, which a human can do in the mind. Next, the limitation of sending a request based off a criterion, as drafted, is a process, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components. That is, other than reciting the components listed above, nothing in the claim precludes the step from practically being performed in the mind. For example, “sending” in the context of this claim encompasses sending a request after a certain criterion has been met, which a human can do in the mind or with a pen and paper. Next the limitation of generating a response, as drafted, is a process, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components. That is, other than reciting generic models, nothing in the claim precludes the step from practically being performed in the mind. For example, “generating” in the context of this claim encompasses producing a response to a question, which a human can do in the mind or with a pen and paper. Lastly, the limitation of presenting a response, as drafted, is a process, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components. That is, other than reciting the components listed above, nothing in the claim precludes the step from practically being performed in the mind. For example, “presenting” in the context of this claim encompasses transmitting information, which a human can do in the mind or with a pen and paper.
The judicial exception is not integrated into a practical application. In particular, the claim only recites the additional elements, using “a memory” and “a processor” to perform the recited limitations. These elements in these steps are recited at a high-level of generality such that is amounts no more than mere instructions to apply the exception using generic computer component. Accordingly, these additional element does not integrate the abstract idea into a practical application because it does not impose any meaningful limits on practicing the abstract idea. The claim is directed to an abstract idea.
The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception. As discussed above with respect to the integration of the abstract idea into a practical application, the additional elements of using “a memory” and “a processor” to perform the generating steps amounts to no more than mere instructions to apply the exception using generic computer components. Mere instructions to apply an exception using generic computer components cannot provide an inventive concept. The claim is not patent eligible.
Dependent claims 3-7, 10-14, and 23 are also rejected for the same reasons provided in independent claim 1 and 8 above. The dependent claim, including the further recited limitation, does not integrate the abstract idea into a practical application and the additional elements, taken individually and in combination do not contribute to an inventive concept. In other words, the dependent claim is directed to an abstract idea without significantly more.
Independent claim 15 recites, “predict responses to user requests using a set of human-curated responses”, “generate a first response to a user intent using the set of human-curated responses, the first response personalized for a user based on context information extracted from a user request”, “generate the first response to the user intent”, “selectively send a user utterance to one of the deterministic AI model or the generative AI model based on a comparison between the user intent and a master intent list comprising a set of stored intents that correspond to the set of human-curated responses”, “identify the user intent based on the user utterance and context information extracted from the user request”, “maintain a conversation list that represents a conversation with the user, and add the user intent and follow-up intents to the conversation list, the follow-up intents representing probable responses to subsequent user utterances provided by the user in reaction to the first response within the conversation”, and “receive the first response from the selected one of the deterministic AI model or the generative AI model and present the first response to the user”.
First, the limitation of predicting a response, as drafted, is a process, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components. That is, other than reciting the components listed above, nothing in the claim precludes the step from practically being performed in the mind. For example, “predict” in the context of this claim encompasses anticipated a response, which a human can do in the mind. Next, the limitation of generating a response, as drafted, is a process, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components. That is, other than reciting the components listed above, nothing in the claim precludes the step from practically being performed in the mind. For example, “generate” in the context of this claim encompasses responding to intents based off a set of rules, which a human can do in the mind or with a pen and paper. Next, the limitation of generating a response, as drafted, is a process, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components. That is, other than reciting the components listed above, nothing in the claim precludes the step from practically being performed in the mind. For example, “generate” in the context of this claim encompasses responding to intents, which a human can do in the mind or with a pen and paper. Next, the limitation of selectively direct user utterances, as drafted, is a process, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components. That is, other than reciting generic models, nothing in the claim precludes the step from practically being performed in the mind. For example, “direct” in the context of this claim encompasses classifying data, which a human can do in the mind. Next, the limitation of identifying a user’s intent from an utterance, as drafted, is a process, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components. That is, other than reciting the components listed above, nothing in the claim precludes the step from practically being performed in the mind. For example, “identify” in the context of this claim encompasses analyzing data, which a human can do in the mind. Next, the limitation of maintaining a list, as drafted, is a process, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components. That is, other than reciting the components listed above, nothing in the claim precludes the step from practically being performed in the mind. For example, “maintain” in the context of this claim encompasses managing data, which a human can do in the mind. Lastly, the limitation of receiving a response, as drafted, is a process, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components. That is, other than reciting generic models, nothing in the claim precludes the step from practically being performed in the mind. For example, “receive” in the context of this claim encompasses receiving data, which a human can do in the mind or with a pen and paper.
The judicial exception is not integrated into a practical application. In particular, the claim only recites the additional elements, using generic models to perform the recited limitations. These elements in these steps are recited at a high-level of generality such that is amounts no more than mere instructions to apply the exception using generic computer component. Accordingly, this additional element does not integrate the abstract idea into a practical application because it does not impose any meaningful limits on practicing the abstract idea. The claim is directed to an abstract idea.
The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception. As discussed above with respect to the integration of the abstract idea into a practical application, the additional elements of using generic models to perform the recited limitations amounts to no more than mere instructions to apply the exception using generic computer components. Mere instructions to apply an exception using generic computer components cannot provide an inventive concept. The claim is not patent eligible.
Dependent claims 17-22 are also rejected for the same reasons provided in independent claim 15 above. The dependent claim, including the further recited limitation, does not integrate the abstract idea into a practical application and the additional elements, taken individually and in combination do not contribute to an inventive concept. In other words, the dependent claim is directed to an abstract idea without significantly more.
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.
Claims 1, 3, 7-8, 10, 14, and 23 are rejected under 35 U.S.C. 103 as being unpatentable over Bhathena et al. US 20240282296 A1 (hereinafter Bhathena) in view of Durg et al. US 20250173330 A1 (hereinafter Durg).
Regarding independent claim 1 and 8, Bhathena teaches a method for implementing a composite artificial intelligence (AI) system, comprising / a processing system, comprising:
a memory comprising computer-executable instructions (FIG. 1, 106); and
a processor configured to execute the computer-executable instructions and cause the processing system to (FIG. 1, 104):
generating a first user request from a first user utterance submitted by a user (FIG. 4, S402 [0074] “the utterance includes a user query that is expressed by text using natural language in a conversational mode”);
classifying a first user intent from the first user request and context information associated with the first user utterance (FIG. 4, S404; FIG. 3, 302;);
determining to send the user request to one of a first AI model or a second AI model based on a determination that the user intent is fullfillable by one of the first AI model or the second AI model (FIG. 4, S406; [0086] “one or more domain classification models will be triggered to recognize the higher level domain(s) to which that utterance might belong”);
generating a first response by one of the first AI model or the second AI model based on the determination (FIG. 6, [0089] “A knowledge base of FAQs and their corresponding answers has been curated”); and
transmitting the first response to the user ([0086] “This Question/Answer pair either answers the user's question completely or at least lets the user know some more incrementally relevant information”).
Bhathena fails to teach obtaining a master intent list comprising a set of stored intents for which a set of human-curated responses are available through a first Al model, wherein the first Al model is trained for predicting responses to user requests using the set of human-curated responses; based on at least the first user intent matching a stored intent, in the set of stored intents, of the master intent list, sending the first user request to the first Al model; generating, by the first Al model, a first response responsive to the first user request, wherein the first response comprises a first human-curated response that is associated with the stored intent, in the set of stored intents, of the master intent list and is personalized for the user based on the context information associated with the first user utterance; presenting the first response to the user; generating a second user request from a second user utterance submitted by the user; classifying a second user intent from the second user request; based on at least the second user intent failing to match a respective stored intent, in the set of stored intents, of the master intent list, sending the second user request to a second Al model; generating, by the second Al model, a second response responsive to the second user request; presenting the second response to the user;
However, Durg teaches obtaining a master intent list comprising a set of stored intents for which a set of human-curated responses are available through a first Al model, wherein the first Al model is trained for predicting responses to user requests using the set of human-curated responses ([0021] “The orchestrator 160 can compare the intent prediction confidence to a confidence limit 162 configured therein. If the confidence of the intent prediction is higher than the confidence limit 162, the intent 112 and the slot 114 are provided to the structured data manager 170 for information retrieval. The data responsive to the user query 110 can be obtained from an appropriate one of the data sources 120”, examiner interprets 120 as the master list; [0016] “The LLM is trained to automatically generate a structured query (e.g., Structured Language (SQL) Query) for the intent or to carry out the task conveyed in the user query”)
based on at least the first user intent matching a stored intent, in the set of stored intents, of the master intent list, sending the first user request to the first Al model (FIG. 7, 704, 710);
generating, by the first Al model, a first response responsive to the first user request, wherein the first response comprises a first human-curated response that is associated with the stored intent, in the set of stored intents, of the master intent list and is personalized for the user based on the context information associated with the first user utterance (FIG. 7, 712, examiner interprets currently selected LLM as the first AI model),
presenting the first response to the user ([0017] “The answer thus generated can be provided to the user via an output screen of the chatbot interface”);
generating a second user request from a second user utterance submitted by the user (FIG. 5, 502);
classifying a second user intent from the second user request (FIG. 5, 504, 506)
based on at least the second user intent failing to match a respective stored intent, in the set of stored intents, of the master intent list (FIG. 5, 508), sending the second user request to a second Al model (FIG. 5, 516)
generating, by the second Al model, a second response responsive to the second user request (FIG. 5, 516)
presenting the second response to the user (FIG. 5, 514)
Bhathena in view of Durg are considered to be analogous to the claimed invention because both are the same field of virtual assistants responding to a user queries. 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 the techniques performing hierarchical domain routing and intent classification on user queries in order to improve accuracy in responding to such queries and to create smoother conversations between virtual voice assistants and users of Bhathena with the technique of providing queries to certain generative models based off a match taught by Durg in order to improve a Generative Artificial Intelligence (Gen. AI) based chatbot apparatus implements classical AI models and Generative AI models to answer user queries with results retrieved from relational databases and unstructured knowledge bases (see Durg [Abstract]).
Regarding claims 3 and 10, Bhathena in view of Durg teaches all of the limitations of claim 1 and 8, upon which claims 3 and 10 depend.
Additionally, Bhathena teaches wherein the first user request is generated based on at least the first user utterance, customer information, and experience information, where the customer information, and the experience information provide the context information of the first user utterance ([0080] “the user experience would be improved if the VA could respond with a more targeted response, at least showing that the VA understands the general idea of the query, but needs more information to drill down to the exact user intent”).
Regarding claims 7 and 14, Bhathena in view of Durg teaches all of the limitations of claim 1 and 10, upon which claims 7 and 14 depend.
Additionally, Bhathena teaches wherein the first AI model is a natural language understanding (NLU) model using a database of the set of human-curated responses, and the second AI model is a generative AI model referencing a predefined set of information ([0081] “the present inventive concept is designed to provide a conditional bottom-up hierarchical NLU model which can always provide some form of an answer to a user's original question, either in the form of an in-domain frequently asked question (FAQ) or explaining the possible intents to the user within the relevant domain”; [0017] “an artificial intelligence (AI) model that is configured to assign the received utterance to at least one domain from among a predetermined plurality of domains”; FIG. 2, 206(1)-206(n)).
Regarding claim 23, Bhathena in view of Durg teaches all of the limitations of claim 1, upon which claim 23 depends.
Additionally, Bhathena teaches the first user intent is associated with a request for assistance with a financial product, the first human-curated response comprises financial advice verified by a financial advisor, and the first response comprises a first answer related to the financial product ([0010] The predetermined plurality of domains may include a first domain group that relates to products associated with a financial institution and a second domain group that relates to activity associated with the financial institution.).
Claims 4-6, 11-13, and 15, 17-21 are rejected under 35 U.S.C. 103 as being unpatentable over Bhathena in view of Durg in view of Rodriguez Garcia et al. US 20250307564 A1 (hereinafter Rodriguez Garcia).
Regarding claims 4 and 11, Bhathena in view of Durg teaches all of the limitations of claim 3 and 10, upon which claim 4 and 11 depend.
Bhathena in view of Durg fails to teach generating the first response using a natural language understanding (NLU) model as the first AI model, generating the first response including: assigning an intent identifier associated with at least one selected response from among the set of human-curated responses, the at least one selected response corresponding to the first user intent; adding the first user intent to a conversation list that represents a conversation with the user; and adding follow-up intents to the conversation list.
However, Rodriguez Garcia teaches generating the first response using a natural language understanding (NLU) model as the first AI model, generating the first response including: assigning an intent identifier associated with at least one selected response from among the set of human-curated responses, the at least one selected response corresponding to the user intent ([0096] “the utterance and the corresponding label is randomly selected from a training dataset, where the training dataset includes a plurality of utterance-intent pairs of a particular domain”, examiner interprets training dataset to be the human-curated responses);
adding the first user intent to a conversation list that represents a conversation with the user (FIG. 6, 602, 610, [0074] “, the processing logic may add any such identified new intents to the list of known intents”); and
adding follow-up intents to the conversation list ([0074] “, the processing logic may add any such identified new intents to the list of known intents”).
Bhathena in view of Durg in view of Rodriguez Garcia are considered to be analogous to the claimed invention because both are the same field of speech processing. 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 the techniques of providing responses to users through virtual assistants of Bhathena in view of Durg with the technique of modifying intents in a conversation list taught by Rodriguez Garcia in order to improve identifying intents in utterances to update a known list of intents (see Rodriguez Garcia [0001]).
Regarding claims 5 and 12, Bhathena in view of Durg in view of Rodriguez Garcia teaches all of the limitations of claims 4 and 11, upon which claims 5 and 12 depend.
Additionally, Bhathena teaches wherein the follow-up intents represent probable responses to subsequent utterances provided by the user in reaction to the first response within the conversation ([0078] “the outputting may include displaying, to the user, a predetermined list of items that corresponds to possible intentions associated with the domain(s) to which the utterance has been assigned, together with a prompt that acts as an invitation to the user to provide a response by which one or more of the possible intentions is selected by the user”).
Regarding claims 6 and 13, Bhathena in view of Durg in view of Rodriguez Garcia teaches all of the limitations of claims 4 and 11 upon which claims 6 and 13 depend.
Additionally, Rodriguez Garcia teaches applying response rules and response templates, by a dialog manager, to the set of human-curated responses ([0045] “the prompt 122 may be constructed by the prompt generator 106 based on a template that precisely defines its content and format”); and
personalizing, by the dialog manager, the first response using the customer information ([0090] “one or more of the data centers 916 can be configured using a multi-instance cloud architecture to provide every customer with its own unique customer instance or instances”).
Regarding independent claim 15, Bhathena teaches a composite artificial intelligence system (AI), comprising: a deterministic AI model configured to:
predict responses to user requests using a set of human-curated responses ([0008] “analyzing, by the at least one processor, the received utterance in order to make an initial determination of an intent of the user and a confidence level that relates to the initial determination of the intent”)
generate a first response to a user intent using a set of human-curated responses, the first response personalized for a user based on context information extracted from a user request; (FIG. 6; [0089] “the incoming query is encoded using the same embeddings model, and then the most similar FAQ questions are retrieved, based on cosine similarity. To provide contextual answers, the answers are not directly presented in the retrieved FAQs to the user; instead, a prompt is constructed to instruct the LLM to utilize the top-k most relevant FAQ answers, together with the available conversation history, to generate appropriate responses to user questions.”);
a generative AI model configured to generate the first response to the user intent (FIG. 6, [0089] “A knowledge base of FAQs and their corresponding answers has been curated”); and
a classifier configured to identify the user intent based on the user utterance and context information extracted from the user request (FIG. 4, S404; FIG. 3, 302;),
a responder configured to receive the first response from one of the deterministic AI model or the generative AI model and present the first response to the user (FIG. 6, [0089] “A knowledge base of FAQs and their corresponding answers has been curated”; [0086] “This Question/Answer pair either answers the user's question completely or at least lets the user know some more incrementally relevant information”).
Bhathena fails to teach a dispatcher configured to selectively send a user utterance to one of the deterministic AI model or the generative AI model based on a comparison between the user intent and a master intent list comprising a set of stored intents that correspond to the set of human-curated responses, the dispatcher including; a conversation tracker configured to cooperate with the deterministic AI model to maintain a conversation list that represents a conversation with the user; and add the user intent and follow-up intents to the conversation list, the follow-up intents representing probable responses to subsequent user utterances provided by the user in reaction to the first response within the conversation
However, Durg teaches a dispatcher configured to selectively send a user utterance to one of the deterministic AI model or the generative AI model based on a comparison between the user intent and a master intent list comprising a set of stored intents that correspond to the set of human-curated responses ([0021] “The user query 110 along with the predicted intent 112 and slot 114 with corresponding confidences are provided to the orchestrator 160. The orchestrator 160 can compare the intent prediction confidence to a confidence limit 162 configured therein. If the confidence of the intent prediction is higher than the confidence limit 162, the intent 112 and the slot 114 are provided to the structured data manager 170 for information retrieval”; FIG. 7, 710;)
Bhathena in view of Durg are considered to be analogous to the claimed invention because both are the same field of virtual assistants responding to a user queries. 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 the techniques performing hierarchical domain routing and intent classification on user queries in order to improve accuracy in responding to such queries and to create smoother conversations between virtual voice assistants and users of Bhathena with the technique of providing queries to certain generative models based off a match taught by Durg in order to improve a Generative Artificial Intelligence (Gen. AI) based chatbot apparatus implements classical AI models and Generative AI models to answer user queries with results retrieved from relational databases and unstructured knowledge bases (see Durg [Abstract]).
Bhathena in view of Durg fails to teach a conversation tracker configured to cooperate with the deterministic AI model to maintain a conversation list that represents a conversation with the user; and add the user intent and follow-up intents to the conversation list, the follow-up intents representing probable responses to subsequent user utterances provided by the user in reaction to the first response within the conversation
However, Rodriguez Garcia teaches a conversation tracker configured to cooperate with the deterministic AI model to maintain a conversation list that represents a conversation with the user; and add the user intent and follow-up intents to the conversation list, the follow-up intents representing probable responses to subsequent user utterances provided by the user in reaction to the first response within the conversation ([0003] “The method may further include generating, using a second large language model on the modified prompt, a list of predicted intents”; [0005] “the operations may further comprise determining that a particular intent in the list of predicted intents is not in the list of known intents; and updating the list of known intents with the particular intent”; [0074] “the processing logic may add any such identified new intents to the list of known intents”)
Bhathena in view of Durg in view of Rodriguez Garcia are considered to be analogous to the claimed invention because both are the same field of speech processing. 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 the techniques of providing responses to users through virtual assistants of Bhathena in view of Durg with the technique of modifying intents in a conversation list taught by Rodriguez Garcia in order to improve identifying intents in utterances to update a known list of intents (see Rodriguez Garcia [0001]).
Regarding claim 17, Bhathena in view of Durg in view of Rodriguez Garcia teaches all of the limitations of claim 15, upon which claim 17 depends.
Additionally, Durg teaches wherein the dispatcher is further configured to send the user utterance to the generative AI model based on a failure of the dispatcher to match the user intent to stored intent, of the set of stored intents, in the master intent list, the generative AI model being a large language model (LLM) ([0021] “The orchestrator 160 can compare the intent prediction confidence to a confidence limit 162 configured therein”; FIG. 5, 516).
Regarding claim 18, Bhathena in view of Durg in view of Rodriguez Garcia teaches all of the limitations of claim 15, upon which claim 18 depends.
Additionally, Bhathena teaches wherein the user request is generated based on at least the user utterance, customer information, and experience information, where the customer information, and the experience information provide the context information of the user utterance ([0080] “the user experience would be improved if the VA could respond with a more targeted response, at least showing that the VA understands the general idea of the query, but needs more information to drill down to the exact user intent”).
Regarding claim 19, Bhathena in view of Durg in view of Rodriguez Garcia teaches all of the limitations of claim 18, upon which claim 19 depends.
Additionally, Rodriguez Garcia teaches wherein the deterministic AI model is a natural language understanding (NLU) model, the NLU model further comprising a dialog manager configured to:
apply response rules and response templates to the set of human-curated responses ([0045] “the prompt 122 may be constructed by the prompt generator 106 based on a template that precisely defines its content and format”);
and personalize the first response using the customer information ([0090] “one or more of the data centers 916 can be configured using a multi-instance cloud architecture to provide every customer with its own unique customer instance or instances”).
Regarding claim 20, Bhathena in view of Durg in view of Rodriguez Garcia teaches all of the limitations of claim 15, upon which claim 20 depends.
Additionally, Rodriguez Garcia teaches wherein the conversation tracker is further configured to update the conversation list based on subsequent user utterances received in reaction to the first response within the conversation ([0005] “the operations may further comprise determining that a particular intent in the list of predicted intents is not in the list of known intents; and updating the list of known intents with the particular intent”).
Regarding claim 21, Bhathena in view of Durg in view of Rodriguez Garcia teaches all of the limitations of claim 15, upon which claim 20 depends.
Additionally, Bhathena teaches wherein: the deterministic AI model is a natural language understanding (NLU) model ([0080] “the present inventive concept provides a method that is implemented in a Natural Language Understanding (NLU) component of a pipeline-based task oriented agent, where the method is designed to gracefully handle such cases in order to make smoother conversations between digital assistants and customers”), and
Additionally, Durg teaches the dispatcher is further configured to send the user utterance to the NLU model based on a success of the dispatcher to match the user intent to a stored intent, of the set of stored intents, in the master intent list ([0021] “The user query 110 along with the predicted intent 112 and slot 114 with corresponding confidences are provided to the orchestrator 160. The orchestrator 160 can compare the intent prediction confidence to a confidence limit 162 configured therein. If the confidence of the intent prediction is higher than the confidence limit 162, the intent 112 and the slot 114 are provided to the structured data manager 170 for information retrieval”; FIG. 7, 704, 710).
Claim 22 is rejected under 35 U.S.C. 103 as being unpatentable over Bhathena in view of Durg in view of Rodriguez Garcia in view of Zhang et al. US 20250190449 A1 (hereinafter Zhang).
Regarding claim 22, Bhathena in view of Durg in view of Rodriguez Garcia teaches all of the limitations of 15, upon which claim 22 depends.
Bhathena in view of Durg in view of Rodriguez Garcia fails to teach wherein the classifier comprises at least one of: an intent matching model configured to use a set of grammar matching rules to identify the user intent; or a machine learning transformer model configured to classify text of the user utterance to identify the user intent.
However, Zhang teaches wherein the classifier comprises at least one of: an intent matching model configured to use a set of grammar matching rules to identify the user intent; or a machine learning transformer model configured to classify text of the user utterance to identify the user intent ([0073] “The process flow 600 includes correcting basic errors in the user prompt, at 604. For example, spellcheck and grammar check may be run on the user prompt, missing punctuation may be added to the user prompt, or the like.”).
Bhathena in view of Durg in view of Rodriguez Garcia in view of Zhang are considered to be analogous to the claimed invention because all are the same field of speech processing. 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 the techniques of providing responses to users through virtual assistants of Bhathena in view of Durg in view of Rodriguez Garcia with the technique of using grammar matching rules taught by Zhang in order to determine intent of user prompts and to intelligently select generative artificial intelligence agents to perform analytics tasks to generate responses to the user prompts. (see Zhang [0001]).
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Mahmood et al. (US 20210090572 A1) teaches techniques for a natural language processing (NLP) system to implement more than one assistant are described. The NLP system may receive a natural language input from a device. The NLP system may also receive one or more signals representing one or more assistants to be implemented with respect to the natural language input. The NLP system may intelligently select an assistant to be invoked with respect to the natural language input. Once the assistant is selected, the NLP system may cause content, output to a user, to have characteristics specific to the assistant.
Reyes et al. (US 12573398 B2) teaches a computer system implementing a conversational artificial intelligence platform. The computer system may receive a request that includes text data specifying an utterance of a user. The computer system determines, form the utterance, an intent of the user. The computer system selects a first virtual agent to handle the request based on the determined intent. The computer system routes the text data and subsequent text data of the request to the first virtual agent for handling. The computer system receives, from the first virtual agent, a message that the determined intent is incorrect, the message including state information indicative of actions taken by the first virtual agent relative to the request. The computer system selects a second virtual agent to handle the request based on the state information. The computer system may receive input from an interactive voice response system or another channel.
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 ZEESHAN SHAIKH whose telephone number is (703)756-1730. The examiner can normally be reached Monday-Friday 7:30AM-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, Richemond Dorvil can be reached at (571) 272-7602. 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.
/ZEESHAN MAHMOOD SHAIKH/Examiner, Art Unit 2658
/RICHEMOND DORVIL/Supervisory Patent Examiner, Art Unit 2658
Prosecution Insights