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
Claims 1, 3-4, 6, 10, 13, 15, 18 are amended. Claims 9, 14, 17 and 19-20 are cancelled. Claims 21-25 are added. Claims 1-8. 10-13, 15-16, 18 and 21-25 are presented for examination.
Response to Arguments
Claim Rejections 35 U.S.C. § 101
In light of amendments rejection under 35 U.S.C. § 101 is withdrawn.
Claim Rejections 35 U.S.C. § 103
Applicant argues “In rejecting previously presented claim 1, the Office Action first acknowledges on pages 7-8 that Friio does not teach the previously recited "accessing" and "generating" steps, which include the limitations of "the first set of NLM models" and "the second set of NLM models." The Office Action then turns to Wang and asserts that the generator 204 of Fig. 2 of Wang teaches the first/second set of NLM models. In that regard, Wang teaches implementing its generator 204 via a plurality of natural language generation (NLG) templates. See paragraphs [0043] and [0045]-[0046] of Wang. However, Wang does not teach that its NLG templates constitute Natural Language Map (NLM) models, and certainly not NLM models that are established at least in part based on one or more NLP processes. According to Wang, its NLG templates simply map dialog acts to delexicalized utterances. As explained in paragraph [0025] of Wang, "For example, a dialog act may include the utterance "I had a problem with my order and I would like to know if there is an update." The template may associate the dialog act with multiple "classified_intents" including "Check_the_status_of_an_order" and "Report_an_issue." Each of the "classified_intents" may have a number of utterance paraphrases associated with them. For example, the "Check_the_status_of_and_order" may have utterances "I'm unsure if there is a update on my order," "My order got stuck, so I want to know if there's an update," and "Do you know if there is an update on my order?"Applicant respectfully submits that the NLG templates of Wang are far more simplistic than the claimed NLM models and therefore cannot teach the claimed NLM models.” However, dialog act is based on NLP process, NLP process merely means understanding the language by the computer and to respond to the dialog acts computer has to perform NLP process. And Applicant arguments that Wang is 'far more simplistic,' fails to identify a specific, necessary, and non-obvious element required to achieve the claimed results that is absent from Wang's teaching. General assertions of simplicity are not persuasive.
Applicant further asserts “Assuming arguendo (a position that the Applicant does not concede) that the NLG templates can be so broadly interpreted as the claimed NLM models, Wang still does not teach, motivate, or suggest that different sets of NLG templates-one set customized to the customers and a different set customized to the simulated service agent-are used to generate the simulated conversation” and During the Examiner Interview, Examiner Sonifrank asserted that the simulator 210 of Wang (see Fig. 2 of Wang below) teaches the NLM models for a user/customer, and that the generator 204 of Wang teaches the NLM models for a bot. In that regard, Wang teaches that the generator 204 takes a genetic bot input and generates a uniform output, which is used by the simulator 210 to generate simulation results and chatlogs. See paragraphs [0021]-[0022] and [0027]-[0029] of Wang. However, Wang is silent with respect to the generator 204 being customized to a simulated service agent, or that simulator 210 being customized to a particular customer.” However, customization in this regard is very broad. Clearly, the templates are customized for customer set and agents set. Refer to The generator 104 adopts sequence-to-sequence models to simulate lexical and syntactic variations in user queries, Para 0018; Generator 204 takes the generic bot input, which may be in a variety of forms, and produces a uniform output which may be used by simulator 210, Para 0021 – here the models are generated for the user ( user query variations) and bot output forms. Hence the generators different (customized data) for user and bot simulations.
Applicant argues that Dorogusker, Kaveti, and Jingjiao do teach the aforementioned limitation. However, examiner has not relied on these references to teach those features.
Regarding claims 13 and 18, similar response analogous to claim 1, are applicable.
Examiner’ Note:
To overcome the outstanding rejections, Applicant is advised to amend Claim 1 to incorporate the concepts detailed in Paragraphs 0065–0068 of the specification. Specifically, the amendment must explicitly detail how the models are customized for a particular user based on user authentication or an identifier. The claim should further specify that the customization utilizes a "propensity-to-contact score" based on affinity score. Finally, the amendment must incorporate specific details regarding the "mother" and "child" NLM (Natural Language Model) models as described in the specification. Claims 13 and 18 are likely still too broad. Applicant is advised to incorporate the all the limitation from claim 1 and the aforementioned suggestion.
Claim Rejections - 35 USC § 103
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.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
And
KSR, 550 U.S. at 418, 82 USPQ2d at 1396. Exemplary rationales that may support a conclusion of obviousness include:
(A) Combining prior art elements according to known methods to yield predictable results;
(B) Simple substitution of one known element for another to obtain predictable results;
(C) Use of known technique to improve similar devices (methods, or products) in the same way;
(D) Applying a known technique to a known device (method, or product) ready for improvement to yield predictable results;
(E) "Obvious to try" – choosing from a finite number of identified, predictable solutions, with a reasonable expectation of success;
(F) Known work in one field of endeavor may prompt variations of it for use in either the same field or a different one based on design incentives or other market forces if the variations are predictable to one of ordinary skill in the art;
(G) Some teaching, suggestion, or motivation in the prior art that would have led one of ordinary skill to modify the prior art reference or to combine prior art reference teachings to arrive at the claimed invention.
See MPEP § 2143 for a discussion of the rationales listed above along with examples illustrating how the cited rationales may be used to support a finding of obviousness. See also MPEP § 2144 - § 2144.09 for additional guidance regarding support for obviousness determination.
Claims 1-3, 6-8 and 13, 15-16, 18 and 21-25 are rejected under 35 U.S.C. 103 as being unpatentable over Friio ( US 20220070296) and further in view of Wang (US 20230237275 )
Regarding claim 1, Friio teaches a method, comprising: determining that a customer event has been initiated by a customer, the customer event being associated with a plurality of words ( input from a customer, Fig 3, Para 0074) ; parsing the plurality of words ( parsing the natural language input, Para 0075) ; based on a result of the parsing of the plurality of words, predicting an intent of the customer corresponding to the customer event (determining intent, Fig 7; interaction predictions, Para 0097-0098) ;; associating, based on the predicted intent, a first simulated service agent of a plurality of simulated service agents with the customer event ( After the customer's intent is determined, the flowchart 350 proceeds to an operation 365 where the customer automation system 300 loads a script associated with the given intent, Para 0076-0077) ; accessing, based on the predicted intent, a models for the customer first simulated service agent customized to the customer ( determine the profile or other person profile, datasets etc.) , Para 0136, 0147 ) and generating, based on the models, conversation between the customer and the first service agent, the conversation involving the predicted intent ( based on the predictors, predict how the interaction would go, Para 0137-0142) generating conversation, Fig 9 and Fig 10, Para 0162) receiving additional input from the customer after the simulated conversation has been generated (subsequent or incoming interaction, Para 0137); altering, based on the received additional input, one or more models in the first set of models or in the second set of models ( the step of modifying, in accordance with the behavioral factor, a manner in which services are delivered to the customer in an incoming interaction, Para 0138) ; and re-generating the simulated conversation based on the altered one or more NLM models ( an incoming interaction instigated by the first customer may be detected as being the same as the first interaction type. In response this detection, the derived interaction predictor may be retrieved from the customer profile of the first customer, and, upon being retrieved, the relevant behavioral factor can be identified. The manner in which services are delivered to the first customer in the incoming interaction may be modified pursuant to the behavior factor. More specifically, once identified, the behavior factor may be transmitted to the contact center involved in the incoming interaction. The contact center may then use the insight provided by the behavior factor to modify the way it delivers services to the first customer in the incoming interaction, Para 0137-0138)
Friio does not explicitly teach accessing, based on the predicted intent, a first set of Natural Language Map (NLM) models to the customer customized for the customer and a second set of NLM models customized to the first simulated service agent, wherein the NLM models in the second set of NLM models are different from the NLM models in the first set of NLM models, and wherein at least one of the NLM models in the first set or the second set of the NLM models is established at least in part based on one or more natural language processing (NLP) processes; and the altering information in the set of NLM models.
However, Wang teaches accessing, a first set of Natural Language Map (NLM) models to the customer customized for the customer ( At step 820, the generator (e.g., generator 204 of FIG. 2) determines a plurality of natural language generation (NLG) templates based on the task-oriented dialog data. The NLG templates serve as the language generator for system 200. The templates may be maintained as a JSON file to map from dialog acts to delexicalized utterances, Para 0043-0046; additionally users can customize or override these rules to simulate the behavior of their customers, Para 0030) and a second set of NLM models customized to the first simulated service agent (a generator (e.g., generator 204 in FIG. 2) determines a plurality of natural language understanding pairs including bot dialog acts and respective bot messages based on the task-oriented dialog data. The data may be in a form that is easily readable by humans, as compared to the source dialog data from the dialog agent. For example, the bot dialog act may take a form of “I had a problem with my order and I would like to know if there is an update”, and the respective bot message may take a form of “I'm unsure if there is an update on my order”, Para 0043), wherein the NLM models in the second set of NLM models are different from the NLM models in the first set of NLM models (The generator 104 adopts sequence-to-sequence models to simulate lexical and syntactic variations in user queries, Para 0018; Generator 204 takes the generic bot input, which may be in a variety of forms, and produces a uniform output which may be used by simulator 210, Para 0021 – here the models are generated for the user ( user query variations) and bot output forms) , and wherein at least one of the NLM models in the first set or the second set of the NLM models is established at least in part based on one or more natural language processing (NLP) processes (NLP models, Fig 2) ; and generating, based on the first set of NLM models and the second set of NLM models, a simulated conversation between the customer and the first simulated service agent, the simulated conversation involving the predicted intent (simulated conversation, Fig 8, S830) and re-generating the simulated conversation based on the altered one or more NLM models in the first set of NLM models or in the second set of NLM models (The plug-and play user response templates can be constantly updated to include more variations as encountered in real world use cases, Para 0027)
Friio has a base concept of generating the suggested insights based on predictions how user will react/respond. Friio different by the claimed invention on the concept of simulating conversation. The results of the combination would be predictable and it would have been obvious having the teachings of Friio to further include the concept of Wang before effective filing date because by doing so a more appropriate actionable item can be suggested (Para 0048, Wang)
Regarding claim 2, Friio as above in claim 1 teaches, accessing previous electronic chat transcripts involving the customer, wherein the determining is further based on the previous electronic chat transcripts ( historical interactions, Para 0075)
Regarding claim 3, Friio modified by Wang as above in claim 2, teaches , wherein the first set of NLM models are established at least in part based on performing the one or more processes on the previous electronic chat transcripts ( load the script (bot ) and relevant information ( historical transcripts), Fig 8, Friio; previously stored information, Para 0036, Wang)
Regarding claim 6, Friio as above in claim 1, teaches predicting, based on the simulated conversation, that the customer will a request an action to be taken at a future point in time (That is, such communication may include any follow-up actions that the customer may need to complete to make that happen. When necessary, the personal bot assistant also may follow-up with the customer during the process to collect any additional information that, along the way, is determined to be necessary for a resolving the customer's request, Para 0156-0158; the resolution facilitator compiles a case related to the customer request with all the relevant details, including, for example, the workflow that is to follow. Once this is done, the resolution package is routed to an appropriate agent, Para 0161)
Regarding claim 7, Friio as above in claim 6, teaches performing the action before the future point in time; and informing the customer about the performing of the action ( follow up with the customer, Para 0158)
Regarding claim 8, Wang as above in claim 1, teaches wherein the generating the simulated conversation comprises: generating, for each of the customer and the first simulated service agent, a plurality of sentences; and sequencing the plurality of sentences ( simulated conversation, Para 0031)
Regarding claim 13, Friio teaches a system, comprising: a non-transitory memory; and one or more hardware processors coupled to the non-transitory memory and configured to read instructions from the non-transitory memory to cause the system to perform operations comprising: determining that an event has been initiated by a user (receive input from a customer, Para 0074) ; accessing a plurality of previous electronic chat transcripts associated with the user ( historical information from the user, Para 0075) ; analyzing the event and the plurality of previous electronic chat transcripts ( intents are mined, Para 0075) ; predicting, based on the analyzing, an intent of the user corresponding to the event ( intent of the user, Para 0075/ (interaction predictions, Para 0097-0098)) ; assigning, based on the predicted intent, a service agent of a plurality of service agents to the user (After the customer's intent is determined, the flowchart 350 proceeds to an operation 365 where the customer automation system 300 loads a script associated with the given intent, Para 0076-0077 ) , the service agent having a skill associated with the predicted intent ( script of a particular bot, Fig 8-9) ; determining models based on intent ( based on intent determine relevant information ( determine the profile or other person profile, datasets etc.) , Para 0136, 0147) ; and automatically generating, based on the models an electronic conversation between the user and the service agent, the electronic conversation comprising a plurality of sentences exchanged between the user and the service agent, wherein at least one of the sentences is associated with the predicted intent ( based on the predictors, predict how the interaction would go, Para 0137-0142) generating conversation, Fig 9 and Fig 10, Para 0162) receiving additional information from the user after the simulated electronic conversation has been automatically generated (subsequent or incoming interaction, Para 0137); altering, based on the received additional information, one or more models ( the step of modifying, in accordance with the behavioral factor, a manner in which services are delivered to the customer in an incoming interaction, Para 0138); and re-generating the simulated electronic conversation based on the altered one or more models ( an incoming interaction instigated by the first customer may be detected as being the same as the first interaction type. In response this detection, the derived interaction predictor may be retrieved from the customer profile of the first customer, and, upon being retrieved, the relevant behavioral factor can be identified. The manner in which services are delivered to the first customer in the incoming interaction may be modified pursuant to the behavior factor. More specifically, once identified, the behavior factor may be transmitted to the contact center involved in the incoming interaction. The contact center may then use the insight provided by the behavior factor to modify the way it delivers services to the first customer in the incoming interaction, Para 0137-0138; additionally from Fig 6-10, Para 0094, 0103-0104, 0118-0119)
Friio does not explicitly teach determining a first set of Natural Language Map (NLM) models for the user that are associated with the user and a second set of NLM models that are associated with the service agent; wherein at least one NLM model in the first set or the second set of NLM models is established at least in part based on one or more natural language processing (NLP) techniques; and automatically generating, based on the first set of NLM models and the second set of NLM models, a simulated electronic conversation between the user and the service agent, the simulated electronic conversation comprising a plurality of sentences exchanged between the user and the service agent, wherein at least one of the sentences is associated with the predicted intent; and altering one or more set of NLM models
However, Wang teaches determining a first set of Natural Language Map (NLM) models for the user that are associated with the user ( At step 820, the generator (e.g., generator 204 of FIG. 2) determines a plurality of natural language generation (NLG) templates based on the task-oriented dialog data. The NLG templates serve as the language generator for system 200. The templates may be maintained as a JSON file to map from dialog acts to delexicalized utterances, Para 0043-0046; additionally users can customize or override these rules to simulate the behavior of their customers, Para 0030) and a second set of NLM models that are associated with the service agent (a generator (e.g., generator 204 in FIG. 2) determines a plurality of natural language understanding pairs including bot dialog acts and respective bot messages based on the task-oriented dialog data. The data may be in a form that is easily readable by humans, as compared to the source dialog data from the dialog agent. For example, the bot dialog act may take a form of “I had a problem with my order and I would like to know if there is an update”, and the respective bot message may take a form of “I'm unsure if there is an update on my order”, Para 0043; The generator 104 adopts sequence-to-sequence models to simulate lexical and syntactic variations in user queries, Para 0018; Generator 204 takes the generic bot input, which may be in a variety of forms, and produces a uniform output which may be used by simulator 210, Para 0021 – here the models are generated for the user ( user query variations) and bot output forms) , wherein at least one NLM model in the first set or the second set of NLM models is established at least in part based on one or more natural language processing (NLP) techniques; (NLP models, Fig 2); and automatically generating, based on the first set of NLM models and the second set of NLM models, a simulated electronic conversation between the user and the service agent, the simulated electronic conversation comprising a plurality of sentences exchanged between the user and the service agent, wherein at least one of the sentences is associated with the predicted intent (simulated conversation, Fig 8, S830) altering, based on the received additional information, one or more NLM models in the first set of NLM models or in the second set of NLM models; and re-generating the simulated electronic conversation based on the altered one or more NLM models in the first set of NLM models or in the second set of NLM models (The plug-and play user response templates can be constantly updated to include more variations as encountered in real world use cases, Para 0027)
Friio has a base concept of generating the suggested insights based on predictions how user will react/respond. Friio different by the claimed invention on the concept of simulating conversation. The results of the combination would be predictable and it would have been obvious having the teachings of Friio to further include the concept of Wang before effective filing date because by doing so a more appropriate actionable item can be suggested (Para 0048, Wang)
Regarding claim 15, Friio as above in claim 13, teaches wherein the analyzing further comprises performing an NLP process on a textual content of the event or the plurality of previous electronic chat transcripts ( parsing, Para 0075)
Regarding claim 16, Friio as above in claim 13, teaches wherein the operations further comprise: predicting a future point in time at which the user will submit a request associated with the intent, wherein the simulated electronic conversation is automatically generated before the predicted future point in time (That is, such communication may include any follow-up actions that the customer may need to complete to make that happen. When necessary, the personal bot assistant also may follow-up with the customer during the process to collect any additional information that, along the way, is determined to be necessary for a resolving the customer's request, Para 0156-0158; the resolution facilitator compiles a case related to the customer request with all the relevant details, including, for example, the workflow that is to follow. Once this is done, the resolution package is routed to an appropriate agent, Para 0161)
Regarding claim 18, Friio teaches a non-transitory machine-readable medium having stored thereon machine-readable instructions executable to cause a machine to perform operations comprising: accessing textual content associated with an event initiated by a user of a platform and textual content of historical electronic chat records involving the user ( based on input and historical information determine the intent, Para 0074-0075) ; predicting, based on the textual content of the event and the historical electronic chat records, a concern of the user that has not been specifically raised in association with the event ( intent and predicted intent/behavior or a type of interaction , Para 0109) ; determining, based on the predicting, a service agent of a plurality of service agents for engaging in a potential interaction with the user, wherein the service agent is determined at least in part based on a match between a skill of the service agent and a skill for addressing the concern ( (After the customer's intent is determined, the flowchart 350 proceeds to an operation 365 where the customer automation system 300 loads a script associated with the given intent, Para 0076-0077, script of a particular bot, Fig 8-10) ) ; determining models based on intent ( based on intent determine relevant information ( determine the profile or other person profile, datasets etc., Fig 7-8) , an electronic conversation between the user and the service agent in which the concern of the user is addressed ( based on the predictors, predict how the interaction would go, Para 0137-0142) generating conversation, Fig 9 and Fig 10, Para 0162) receiving additional input from the user ( subsequent or incoming interaction, Para 0137); adjusting the models ( the step of modifying, in accordance with the behavioral factor, a manner in which services are delivered to the customer in an incoming interaction, Para 0138); and re-generating the simulated electronic conversation based on the adjusted models or the adjusted second set of models ( an incoming interaction instigated by the first customer may be detected as being the same as the first interaction type. In response this detection, the derived interaction predictor may be retrieved from the customer profile of the first customer, and, upon being retrieved, the relevant behavioral factor can be identified. The manner in which services are delivered to the first customer in the incoming interaction may be modified pursuant to the behavior factor. More specifically, once identified, the behavior factor may be transmitted to the contact center involved in the incoming interaction. The contact center may then use the insight provided by the behavior factor to modify the way it delivers services to the first customer in the incoming interaction, Para 0137-0138)
Friio does not explicitly teach accessing a first set of Natural Language Map (NLM) model customized to the user and a second set of NLM models customized to the service agent, wherein at least one NLM model of the first set or the second set of NLM models was generated at least in part via one or more natural language processing (NLP) techniques and simulating, based on the first set of NLM models and the second set of NLM models
However, Wang teaches accessing a first set of Natural Language Map (NLM) model customized to the user r ( At step 820, the generator (e.g., generator 204 of FIG. 2) determines a plurality of natural language generation (NLG) templates based on the task-oriented dialog data. The NLG templates serve as the language generator for system 200. The templates may be maintained as a JSON file to map from dialog acts to delexicalized utterances, Para 0043-0046; additionally users can customize or override these rules to simulate the behavior of their customers, Para 0030) and a second set of NLM models customized to the service agent (a generator (e.g., generator 204 in FIG. 2) determines a plurality of natural language understanding pairs including bot dialog acts and respective bot messages based on the task-oriented dialog data. The data may be in a form that is easily readable by humans, as compared to the source dialog data from the dialog agent. For example, the bot dialog act may take a form of “I had a problem with my order and I would like to know if there is an update”, and the respective bot message may take a form of “I'm unsure if there is an update on my order”, Para 0043; The generator 104 adopts sequence-to-sequence models to simulate lexical and syntactic variations in user queries, Para 0018; Generator 204 takes the generic bot input, which may be in a variety of forms, and produces a uniform output which may be used by simulator 210, Para 0021 – here the models are generated for the user ( user query variations) and bot output forms) ,, wherein at least one NLM model of the first set or the second set of NLM models was generated at least in part via one or more natural language processing (NLP) techniques (NLP models, Fig 2);and simulating, based on the first set of NLM models and the second set of NLM models (simulated conversation, Fig 8, S830)
Friio has a base concept of generating the suggested insights based on predictions how user will react/respond. Friio different by the claimed invention on the concept of simulating conversation. The results of the combination would be predictable and it would have been obvious having the teachings of Friio to further include the concept of Wang before effective filing date because by doing so a more appropriate actionable item can be suggested (Para 0048, Wang)
Regarding claim 21, Friio modified by Wang as above in claim 18, teaches , wherein at least one NLM model of the first set of NLM models was generated at least in part via performing the one or more NLP techniques on the historical electronic chat records involving the user ( load the script (bot ) and relevant information ( historical transcripts), Fig 8, Para 0075Friio; previously stored information, Para 0036, Wang)
Regarding claim 22, arguments analogous to claim 7, are applicable.
Regarding claim 23, arguments analogous to claim 8, are applicable.
Regarding claim 24, Friio modified by Wang as above in claim 1, teaches wherein the first set of NLM models comprises at least a first NLM model that is associated with a first customer question and a second NLM model that is associated with a second customer question different from the first customer question ( generator generates query variations, Para 0018; simulate behaviours of their customers, Para 0030, Wang; users have personalized profile and hence models will be different and the models are stored in a template ( set of models), Para 0013-0014, 0060, 0067)
Regarding claim 25, Friio modified by Wang as above in claim 1, wherein the second set of NLM models comprises multiple NLM models that are specialized in addressing a particular customer question in different ways ( personalized responses, Para 0060, 0067)
Claim 4 is rejected under 35 U.S.C. 103 as being unpatentable over Friio ( US 20220070296) and further in view of Wang (US 20230237275 ) and further in view of Dorogusker (US 20220337898)
Regarding claim 4, Friio as above in claim 1, teaches wherein one or more of the determining, the parsing, the predicting, the associating, the accessing, the generating, the receiving, the altering or the re-generating is performed by one or more hardware processors of a platform on which the customer is a user ( fig 1-2) , Friio modified by Wang does not explicitly teach wherein the method further comprises: recommending an action for incentivizing the customer to stay with the platform
However, Dorogusker teaches wherein the method further comprises: recommending an action for incentivizing the customer to stay with the platform (the multi-media platform 108 can recommend actions to incentivize a user or other users to engage more with the media/content based on an audience member 236 attending a virtual or in-person live experience(s), Para 0135)
It would have been obvious having the teachings of Friio and Wang to further include the concept of Dorogusker before the effective date to improve user experience
Claim 5 is rejected under 35 U.S.C. 103 as being unpatentable over Friio ( US 20220070296) and further in view of Wang (US 20230237275 ) and further in view of Kaveti ( US 20230031111)
Regarding claim 5, Friio modified by Wang as above in claim 1, does not explicitly teach calculating a propensity-to-contact score based at least in part on the parsing of the plurality of words, wherein the predicting the intent is performed in response to the propensity-to-contact score exceeding a predefined threshold.
However, Kaveti teaches calculating a propensity-to-contact score based at least in part on the parsing of the plurality of words, wherein the predicting the intent is performed in response to the propensity-to-contact score exceeding a predefined threshold (the confidence score indicates a likelihood that recommending the predicted next action will result in the user listed in the row continuing to use the system 100., Para 0113, 0120)
It would have been obvious having the teachings of Friio and Wang to further include the concept of Kaveti before effective filing date to it is desirable to generate more effective user guidance that can better help users with discovering, accessing, and using various system features ( Para 0005, Kaveti )
Claims 10-12 are rejected under 35 U.S.C. 103 as being unpatentable over Friio ( US 20220070296) and further in view of Wang (US 20230237275 ) and further in view of Lingjiao ( FrugalGPT: How to Use Large Language Models While Reducing Cost and Improving Performance)
Regarding claim 10, Friio modified by Wang as above in claim 1, does not explicitly teaches before the altering, the first set of NLM models comprises a mother NLM model and a first child NLM model having a first affinity score with the mother NLM model that exceeds a specified threshold; and after the altering, the altered first set of NLM models comprises the mother NLM model and a second child NLM model having a second affinity score with the mother NLM model that exceeds the first affinity score
However, Lingiao teaches before the altering, the first set of NLM models comprises a mother NLM model and a first child NLM model having a first affinity score with the mother NLM model that exceeds a specified threshold; and after the altering, the altered first set of NLM models comprises the mother NLM model and a second child NLM model having a second affinity score with the mother NLM model that exceeds the first affinity score ( cascading the model based on different input based on the cost function (scores), Fig 2, Under 3 How to Use LLMs Affordably and Accurately)
It would have been obvious having the teachings of Friio and Wang to further include the teachings of Lingiao before effective filing date to reduce the computational cost ( 3 How to Use LLMs Affordably and Accurately—conclusion)
Regarding claim 10, Friio as above in claim 10, teaches causing the simulated conversation to be displayed via a user interface of a mobile device ( mobile device, Para 0138, S780, Fig 7; )
Regarding claim 10, Friio modified by Wang as above in claim 10, teaches wherein the first simulated service agent is a computerized chatbot ( bot, Fig 6-7).
Conclusion
THIS ACTION IS MADE FINAL. 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 Richa Sonifrank whose telephone number is (571)272-5357. The examiner can normally be reached M-T 7AM - 5:30PM.
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, Phan Hai can be reached at (571)272-6338. 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.
/Richa Sonifrank/Primary Examiner, Art Unit 2654
Prosecution Insights