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 Final Rejection is filed in response to Applicant Arguments/Remarks Made in an Amendment filed 06/24/2026.
Claims 1, 4, 5, 6, 9, and 10 are amended.
In light of the amendments, the U.S.C. 112(a) rejections to claims 1 and 6 are respectfully withdrawn.
Claims 1-10 remain pending.
Response to Arguments
Argument 1, in Applicant Arguments/Remarks Made in an Amendment filed 06/24/2026 on pg. 13-15, Applicant argues that Yang does not describe a finite state machine, much less... Two serially-connected finite state machines cooperating to generate and improve a precise question message” .
Response to Augment 1, the examiner respectfully disagrees. Yang teaches state tracking framework for a chatbot is governed by a real-time probabilistic sensor tracked state of a person’s emotional state. Wherein it is noted that the state updates are discrete categories that represent the determined emotional vector of the person (e.g. happy, sad, angry). It should be noted that a Mealy machine is a computation model whose output values are determined by both its current state and its current inputs. It should be noted that the emotive service of Yang also teaches that sensor input may include analyzing a message prompt from a user by breaking up the words into tokens that may be used to infer the emotional state of the user. Thus, the BRI for “a first finite state machine” encompasses how functionally the emotive service of Yang operates as a rule-based model that takes user current emotive state and observed data of a user in order to output a current emotive state of a user.
This is supported by the following paragraphs of Yang.
[0035] The emotion service 224 can use rule-based models and/or machine-learning models to detect the emotion 226 of the user 210 based on the processed sensor data 222. For example, the emotion service 224 can include machine-learning model software that can extract and predict the affective states (e.g., emotional sentiments) of the user.
[0065], the sensors can collect signals in real time, such that the user's current state (e.g., the user's immediate reaction) can be determined in real time.
[0097] FIG. 6A shows the example original prompt from FIGS. 1A and 1B. The original prompt includes the text “he said he likes ice cream.” The words in the original prompt (which may be broken up into tokens) are fed into the emotion service along with various sensor data in order to infer the emotional states of the user.
The examiner notes that the BRI for a Moore machine encompasses a computation model whose output values is determined solely by its current state. The examiner notes that Yang teaches an empathetic prompting module that augments a prompt based on the current state of the user, which is the output of the emotion service. Thus, Moore effectively teaches the BRI of, “a first finite state machine and a second finite state machine which are connected in series”, as functionally the emotion service has a constantly updating output of a user’s emotional state which is fed to the empathetic prompting module in order to generate an appropriate emotion augmented prompt.
This is supported by the following paragraphs of Yang.
[0036], The interactive server 206 includes an empathic prompting module 228. The empathic prompting module 228 augments the prompt 214 based on the emotion 226 to generate an emotion augmented prompt 230, and then feeds the emotion augmented prompt 230 to the generative AI 212. The empathic prompting module 228 includes a prompt generation engine that can use rule-based algorithms and/or machine-learning models to generate the emotion augmented prompt 230.
The examiner further notes that the limitation, “an artificial intelligence platform, configured to receive a precise question message through an application programming interface (API) and input the precise question message to a large language model to generate an answer message … wherein the first finite state machine receives a rough question message to identify a key word in the rough question message and output the key word, and after the artificial intelligence platform generates the answer message, receives the answer message to identify an association between the rough question message and the answer message so as to adjust a state setting of the first finite state machine”, is being interpreted to cover the initial iteration cycle of the state machines where an answer is not needed for input and the step of 1st FSM identifying the association between the rough question message and the answer would not occur until, “after the artificial intelligence platform generates the answer message…”. Thus, the enablement rejection is withdrawn as the claims are now interpreted to enable the previously argued missing functionality of requiring a 1st FSM to require an input answer message that would not have been generated until a 1st FSM and 2nd FSM received and input answer message.
Argument 2, in Applicant Arguments/Remarks Made in an Amendment filed 06/24/2026 on pg. 15-16, Applicant argues that Yang “does not perform the claimed answer-message feedback into a finite state machine” .
Response to Augment 2, the examiner respectfully disagrees and notes the above arguments that Yang teaches components functionally similar to a finite state machine as well as incorporating answer feedback to a chatbot generated response as ground truth labels to fine tune or personalize the model (Yang, para. [0092]).
Argument 3, Applicant argues in Applicant Arguments/Remarks Made in an Amendment filed 06/24/2026 on pg. 16-17, that “Zhang does not cure the deficiencies of Yang”.
Response to Argument 3, In response to applicant's arguments against the references individually, one cannot show nonobviousness by attacking references individually where the rejections are based on combinations of references. See In re Keller, 642 F.2d 413, 208 USPQ 871 (CCPA 1981); In re Merck & Co., 800 F.2d 1091, 231 USPQ 375 (Fed. Cir. 1986).
Argument 4, Applicant argues in Applicant Arguments/Remarks Made in an Amendment filed 06/24/2026 on pg. 17-18, that “the proffered combination is conclusory and would require impermissible hindsight”.
Response to Argument 4, the examiner respectfully disagrees. 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).
In this case lays the groundwork for logical circuity for a chatbot with an AI driven prompt refinement that leverages state data corresponding to a user’s answer and biometric sensor data. Wherein it is noted that Yang shows the capability to associate timestamp data with a user’s state. Zhang teaches a similar concept with a chatbot that refines its refinement using data in an event related session, specifically using time-based event data in real-time during an event to filter out certain chatbot’s responses. Thus, one would have been motivated to combine the chatbot response filtration using time-based conditions of Zhang and with the chatbot response augmenting of Yang as the combination may ensure freshness of responses provided in an event-related session. … and make the provided responses conform to the latest background information instead of outdated background information (Zhang, para. [0025]).
It would have been obvious to one of ordinary skill in the art at the time of filing to add wherein the on-demand conversation setting comprises a time message and a filtering parameter; storing the emotional answer message to an answer list; and automatically filtering out the emotional answer message matching the time message and the filtering parameter from the answer list as the on-demand conversation message generated based on the on-demand conversation setting, and transmitting the on-demand conversation message to the client-end host for output, to Yang’s real-time emotional considerations when generating an AI response from a server, with how generated AI responses are stored and subsequently filtered by a further time and filtering parameter in order to customize a conversation towards a user, as taught by Zhang.
Argument 5, Applicant argues in Applicant Arguments/Remarks Made in an Amendment filed 06/24/2026 on pg. 18-19, that “Even if combined, Yang and Zhang would not teach all the limitations of the independent claims”
Response to Augment 5, the examiner respectfully disagrees. See the above responses.
Claim Rejections - 35 USC § 112
The following is a quotation of the first paragraph of 35 U.S.C. 112(a):
(a) IN GENERAL.—The specification shall contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use the same, and shall set forth the best mode contemplated by the inventor or joint inventor of carrying out the invention.
The following is a quotation of the first paragraph of pre-AIA 35 U.S.C. 112:
The specification shall contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use the same, and shall set forth the best mode contemplated by the inventor of carrying out his invention.
Claim 5 rejected under 35 U.S.C. 112(a) or 35 U.S.C. 112 (pre-AIA ), first paragraph, because the specification, while being enabling for “an emotional answer message which is permitted to be received”, does not reasonably provide enablement for “an emotional answer message which is rejected to be received” in lines 4-5 of Claim 5. The specification does not enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to use the invention commensurate in scope with these claims. Applicant’s specification at most describes in para. [0034], “a trained emotion AI model to generate an emotional answer message, and stores the emotional answer message to an answer list, and automatically filters out the emotional answer message matching the time message and the filtering parameter as an on-demand conversation message from the answer list”. It is unclear what defines an emotional answer message that is “rejected to be received”. For the purposes of examination the BRI of claim limitation, “an emotional answer message which is rejected to be received”, is being interpreted as a potential emotional answer on an answer list that would have been shown but is subsequently filtered and not shown to a user due to a circumstance like the potential emotional answer matching a filtering parameter.
Claim 10 is the method claim reciting similar limitations to claim 5 and is rejected for similar reasons.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claim(s) 1-10 is/are rejected under 35 U.S.C. 103 as being unpatentable over U.S. Patent Application Publication NO. 20240412029 “Yang” and further in light of U.S. Patent Application Publication NO. 20220210098 “Zhang”.
Claim 1:
Yang teaches an active chatbot system with composite finite state machine, comprising: an artificial intelligence platform, configured to receive a precise question message through an application programming interface (API) (i.e. para. [0038], “the emotion augmented prompt 230 includes contextual cues about the emotional state of the user 210 that were not included in the prompt 214. Accordingly, the generative AI 212 becomes aware of the context (e.g., emotion, intent, attitude, purpose, etc.) behind the words in the prompt 214”, wherein the BRI for a precise question encompasses a user’s question that is refined to be more precise by incorporating emotional context) and input the precise question message to a large language model to generate an answer message (i.e. para. [0019], “in these examples, a human user named Annie is conversing with a generative AI chat bot that is powered by an LLM via a text chat interactive mode”, wherein the BRI for an answer message encompasses an AI chat bot output), and transmit the answer message through the application programming interface (i.e. para. [0038], the response 232 returned by the generative AI 212 in response to the prompt 214); a client-end host (i.e. para. [0028], the prompt augmentation system 200 includes an interactive application 202. The interactive client 204 includes a user interface module 208 for facilitating the interaction between a user 210 and a generative AI 212), comprising: at least one sensor, configured to continuously sense at least one of a physiological state, a facial expression and a body movement, to generate a client behavior state (i.e. para. [0051], acts 304 through 314 repeat continuously (e.g., repeat multiple times between two consecutive prompts), such that context data between prompts are fed into the generative AI. For example, updated context data is received from the sensors in act 304, an updated state of the user is inferred based on the updated context data in act 306); a first non-transitory computer readable storage medium (i.e. para. [0128], Processing capability can be provided by one or more hardware processors that can execute data in the form of computer-readable instructions to provide a functionality) configured to store a plurality of first computer readable instructions; and a first hardware processor, electrically connected to the first non-transitory computer readable storage medium and the at least one sensor (i.e. para. [0124], “each of the interactive application server 704, the emotion server 705, and the generative AI server 706 includes one or more server computers. These server computers can each include one or more processor”, wherein an interactive application server connected to a sensor may have a separate processor from a generative AI server and associated memory), and configured to execute the plurality of first computer readable instructions to make the client-end host continuously transmit the client behavior state (i.e. para. [0045], act 304 is performed continuously or periodically (e.g., at regular intervals), even when the user is not providing a prompt) and on-demand conversation setting, wherein the on-demand conversation setting comprises a time message (i.e. para. [0043], “In act 304, context data is received from sensors…In one implementation, the context data is associated with timestamps”, wherein the BRI for on-demand conversation setting encompasses user context with timestamps) and
a server-end host, connected to the client-end host and configured to receive the client behavior state and the on-demand conversation setting (i.e. para. [0123], The interactive application server 704 takes the sensor data from the sensors 702 (and optionally performs pre-processing on the sensor data) and sends the sensor data to an emotion server 705 through the network 708), wherein the server-end host comprises: a question optimization circuit, comprising a plurality of registers for storing states (i.e. para. [0102], “the empathic prompting module can modify the original prompt by adding one or more emojis that correspond to the affective states output by the emotion service. For example, if the emotion service infers that the user was sad when speaking the prompt “he said he likes ice cream,” then the sad face emoji can be appended to the original prompt to generate the augmented prompt shown in FIG. 6E”, wherein the BRI for a question optimization circuit encompasses how the modules may determine and store the current state of a user in conjunction with a their prompt, thus creating a more optimal prompt for an emotional response),
a first combinational logic circuit (i.e. para. [0129], “Generally, any of the functions described herein can be implemented using software, firmware, hardware (e.g., fixed-logic circuitry), or a combination of these implementations. The term “component” or “module” as used herein generally represents software, firmware, hardware, whole devices or networks, or a combination thereof”, wherein the BRI for a combinational logic circuit and finite state machine encompasses the fixed-logic circuitry that implements a state machines that represents the step by step method) for determining a state transition (i.e. para. [0100], “the emotion service has determined that the user has expressed a strong emotion (e.g., any emotion category with a high level). The strong emotion can be positive or negative”, wherein the BRI for determining a state transition encompasses determining a user’s emotional state) and
a second combinational logic circuit for determining an output to form
a first finite state machine and a second finite state machine which are connected in series (i.e. para. [0046, 0048], “In act 306, a state of the user is determined based on the context data. For example, machine-learning models can predict the user's physiological, cognitive, and environmental states… In act 310, the augmented prompt is input into a generative AI”, wherein the BRI for a first finite state machine encompasses the combinational rules logic used to determine a user’s emotional state, which feeds into a second finite state machine for passing an emotional state and user prompt to a generative AI and that recursively updates in real-time),
wherein the first finite state machine (i.e. para. [0035], The emotion service 224 can use rule-based models and/or machine-learning models to detect the emotion 226 of the user 210 based on the processed sensor data 222. For example, the emotion service 224 can include machine-learning model software that can extract and predict the affective states (e.g., emotional sentiments) of the user) receives a rough question message to identify a key word in the rough question message and output the key word ,(i.e. para. [0101], “If the generative AI is capable of accepting and interpreting rich text, then the original prompt can be modified to include formatting. For example, if the emotion service determines that the user expressed a strong emotion while speaking the word “likes,” which can be determined using timestamps, then the empathic prompting module can highlight the word “likes,” wherein it is noted that the emotive service is capable of identifying a key word of “likes” and outputs the key word with an associated emotion. Wherein the BRI for a rough question message encompasses the information about the emotive state of the user), and after the artificial intelligence platform generates the answer message, receives the answer message to identify an association between the rough question message and the answer message so as to adjust a state setting of the first finite state machine (i.e. para. [0110], “if the response is causing the user to be bored, angry, disappointed, less attentive, etc., then the generative AI can modify the response, cut the conversation short, and/or interrupt the response to ask a question seeking feedback (e.g., “Is my answer helpful?”)”, Wherein it is noted that a case exists wherein the generative AI has already generated an answer message to a user and may further refine future responses by taking further real-time input and observations about the user’s emotive state. Wherein the BRI to identify an association between the rough question message and the answer message encompasses how the generative AI can recursively analyze the user response to a rough initial response characterizing the user’s emotive state by finding the association on if the response was helpful or not. Thus, the emotive service may update the state of the user, which is then used to further refine logical circuit for generating further responses to a user), an output of the first finite state machine is used as an input of the second finite state machine (i.e. para. [0095], “The specific technique employed for augmenting the prompt can depend on the content and format of the emotional states that are output from the emotion service and the content and format of the prompts (and of meta-prompts)”, Wherein the BRI for the second finite state machine encompasses empathetic module that uses the output of the emotive service (e.g. the current emotive state of the user) to refine a generated prompt), and
the second finite state machine outputs the precise question message to the artificial intelligence platform through the application programming interface (i.e. para. [0098], “the emotion service interpreted the sensor data and predicted that the user is experiencing a happy emotion. Accordingly, consistent with one implementation of the present concepts, an additional token consisting of the word “happy” is appended to the original prompt to generate the augmented prompt shown in FIG. 6B”, wherein the emotive state is refined into a more precise prompt which is output as a prompt to the generative AI platform as an augmented prompt)
a second non-transitory computer readable storage medium, configured to store a plurality of second computer readable instructions; and a second hardware processor, electrically connected to the second non-transitory computer readable storage medium and the question optimization circuit, and configured to execute the plurality of second computer readable instructions (i.e. para. [0129], “Generally, any of the functions described herein can be implemented using software, firmware, hardware (e.g., fixed-logic circuitry), or a combination of these implementations. The term “component” or “module” as used herein generally represents software, firmware, hardware, whole devices or networks, or a combination thereof”, wherein the BRI for a second medium and processor encompasses remote hardware that may make up the cloud servers of the emotion service) to make the server-end host execute: generating a rough question message having a natural language structure based on the received client behavior state (i.e. para. [0099], “, if the emotion service outputs a value representing a different emotion, such as “sad” or “excited,” then the corresponding word can be appended to the original prompt”, wherein it is noted that sad and excited are words that follow a natural language structure) and on-demand conversation setting, and inputting the rough question message to the question optimization circuit (i.e. para. [0103], “FIG. 6F shows an example augmented prompt that includes an emoji inserted into the original prompt. In this example, an angry face emoji has been inserted at a point in time when the angry emotion was the strongest, as determined by the timestamps associated with the original prompt, the sensor data, and/or the emotions output by the emotion service”, wherein the rough observed emotional data is incorporated into a more refined prompt for the generative AI platform) ;
after the question optimization circuit inputs the precise question message to the artificial intelligence platform (i.e. para. [0104], the generative AI is capable of accepting metadata (e.g., meta-prompts) along with the original prompt. Therefore, the contextual information (e.g., emotion words, emojis, emotion vectors, etc.) can be input as metadata to the generative AI), receiving the answer message corresponding to the precise question message from the artificial intelligence platform (i.e. para. [0048-0049], “Consistent with the present concepts, the augmented prompt, which additionally includes non-verbal communication, is input into the generative AI. In act 312, a response is received from the generative AI. Because the state of the user determined in act 306 has been fed into the generative AI via the augmented prompt, the generative AI is context aware and the response is context appropriate. In act 314, the response is presented to the user”, wherein the BRI for a precise question encompasses the augmented prompt), and inputting the answer message to a trained emotion AI model to generate an emotional answer message (i.e. para. [0114], the generative AI (i.e., the LLM) is specifically trained to be context-aware (i.e., trained to accept and process the emotional context provided in the augmented prompts). That is, the training dataset used to develop the generative AI includes emotion indicators (e.g., emotion vectors, emotion emojis, emotion metadata, etc.)), and storing the emotional answer message to an answer list; and automatically filtering out the emotional answer message matching the time message and the filtering parameter from the answer list as the on-demand conversation message generated based on the on-demand conversation setting, and transmitting the on-demand conversation message to the client-end host for output.
While Yang teaches a prompt refinement to an AI server that relays back an answer to a user prompt that accounts for sensor data and on-demand conversation settings data that includes time stamped sensor data, Yang may not explicitly teach that
wherein the on-demand conversation setting comprises a time message and a filtering parameter;
storing the emotional answer message to an answer list; and automatically filtering out the emotional answer message matching the time message and the filtering parameter from the answer list as the on-demand conversation message generated based on the on-demand conversation setting, and transmitting the on-demand conversation message to the client-end host for output.
However, Zhang teaches
wherein the on-demand conversation setting comprises a time message and a filtering parameter (i.e. para. [0046], “The chatbot may obtain real-time information 310 about the event. The real-time information 310 may comprise various types of information about the latest progress of the event”, wherein the BRI for an on-demand conversation setting comprises a time stamp real-time in which a user has sent a message to the chatbot and the BRI for a filtering parameter encompasses a filtering parameter for a time period of recency);
storing the emotional answer message to an answer list (i.e. para. [0047], “The chatbot may comprise an event content generating module 320, which may be used for generating real-time event content 330 according to the real-time information 310. In an implementation, the event content generating module 320 may generate the real-time event content 330”, wherein the generated candidate answer responses may be generated based on the on-demand conversation setting in that real time events related to a time of a message may be generated, stored, and then filtered based on a time recency threshold to a user’s prompt. Wherein the BRI for an emotional answer message encompasses a generated answer message with the goal of providing a more human like response from a generative AI); and automatically filtering out the emotional answer message matching the time message and the filtering parameter from the answer list as the on-demand conversation message generated based on the on-demand conversation setting (i.e. para. [0065], “candidate responses related to layer N having time stamps within 7 days in the set of candidate responses may be retained, and candidate responses having time stamps 7 days before and being probably related to performance of player N in team C may be filtered out”, wherein from a list of candidate response messages, candidate responses matching a time stamp of recency to a user message may be kept and candidate responses not matching a time filtering parameter for recency may be filtered), and transmitting the on-demand conversation message to the client-end host for output (i.e. para. [0068], “the process 400 may select the top-ranked candidate response as the final response 470 to be provided in the session”, wherein the message may be displayed on a client terminal device).
It would have been obvious to one of ordinary skill in the art at the time of filing to add wherein the on-demand conversation setting comprises a time message and a filtering parameter; storing the emotional answer message to an answer list; and automatically filtering out the emotional answer message matching the time message and the filtering parameter from the answer list as the on-demand conversation message generated based on the on-demand conversation setting, and transmitting the on-demand conversation message to the client-end host for output, to Yang’s real-time emotional considerations when generating an AI response from a server, with how generated AI responses are stored and subsequently filtered by a further time and filtering parameter in order to customize a conversation towards a user, as taught by Zhang. one would have been motivated to combine the chatbot response filtration using time-based conditions of Zhang and with the chatbot response augmenting of Yang as the combination may ensure freshness of responses provided in an event-related session. … and make the provided responses conform to the latest background information instead of outdated background information (Zhang, para. [0025]).
Claim 2:
Yang and Zhang teach the active chatbot system with composite finite state machine according to claim 1.
Yang further teaches wherein the server-end host selects at least one of a natural language processing (NLP), a generative model and a template matching to generate the rough question message having the natural language structure (i.e. para. [0084-0085], “The emotion service can receive many different types of sensor data (including the prompt text) as inputs. The emotion service interacts with one or more machine-learning models, which can function as services. Each machine-learning model has a set of inputs that it takes and a set of outputs that it returns … the emotion service outputs a set of real number values (e.g., normalized between 0 and 1) representing various levels of different emotion categories, for example, as a JavaScript Object Notation (JSON) array: {‘neutral’:0.1, ‘calm’:0.0, ‘happy’:0.6, ‘sad’: 0.0, ‘angry’: 0.2, ‘fearful’: 0.2, ‘disgust’:0.0, ‘surprised’:0.4}”, wherein the interactive server selects appropriate models matching sensor data, may assign an emotional template score to the sensor data, and output a most prominent emotion to augment a user’s prompt. Wherein it is noted that the emotions selected have a natural language characteristic such as happy or sad).
Claim 3:
Yang and Zhang teach the active chatbot system with composite finite state machine according to Claim 1.
Yang further teaches wherein the first finite state machine and the second finite state machine perform parsing on the rough question message to generate a key word and a syntax structure (i.e. para. [0098], , consistent with one implementation of the present concepts, an additional token consisting of the word “happy” is appended to the original prompt to generate the augmented prompt shown), and transit the states thereof to determine a question type based on a parsing result (i.e. para. [0101], FIG. 6D shows an example augmented prompt that includes rich text. If the generative AI is capable of accepting and interpreting rich text, then the original prompt can be modified to include formatting. For example, if the emotion service determines that the user expressed a strong emotion while speaking the word “likes,” which can be determined using timestamps, then the empathic prompting module can highlight the word “likes,” as shown in FIG. 6D), and use a pre-defined template or a syntax rule to generate the precise question message which is more specific and clearer than the rough question message (i.e. para. [0101], “Highlighting can involve bolding, italicizing, underlining, coloring, capitalizing, enlarging, etc. In one implementation, rich text formatting can be added using a markup language, for example, “<bold>likes</bold>”, wherein the BRI for a predefined template or syntax rule encompasses augmenting the prompt with pre-defined highlighting or other text formatting to generate an augmented prompt that is more specific than an augmented prompt or an isolated/rough emotion derived from real-time sensor data).
Claim 4:
Yang and Zhang teach the active chatbot system with composite finite state machine according to Claim 1.
Yang wherein the first finite state machine is a Mealy-machine finite state machine, and the output of the first finite state machine is affected by a current state, the rough question message and the answer message (i.e. para. [0066], “the sensor data collected by the sensors is fed into an emotion service to determine the state of the user”, wherein it is noted that functionally the emotion service of Yang operates as a Mealy-machine whose NPL emotion outputs are dependent on the sensor inputs), wherein the second finite state machine is a Moore-machine finite state machine, and an output of the second finite state machine is affected by a current state (i.e. para. [0099], “if the emotion service outputs a value representing a different emotion, such as “sad” or “excited,” then the corresponding word can be appended to the original prompt. If the emotion service outputs a set of integers representing multiple emotions, then additional words for those emotions can be appended to the original prompt. If the emotion service outputs an emotion vector, then one or more emotion categories having a degree above a certain threshold can be appended to the original prompt”, wherein it is noted that the empathetic prompting module functionally acts as a Moore-machine as the system will augment a prompt with a specific detected state unless a certain threshold representing a new emotional state is detected).
Claim 5:
Yang and Zhang teach the active chatbot system with composite finite state machine according to claim 1.
Zhang further teaches wherein the filtering parameter is configured to set according to a user preference (i.e. [0065], “In the criteria 457 for ensuring freshness, the filtering 450 may obtain the latest background information 459 related to a specific entity in the current event, and retain candidate responses in the set of candidate responses that conform to the latest background information 459 of the specific entity”, wherein the BRI for a user preference encompasses a user preference for a response containing fresh information that conforms to the latest background information), an emotional answer message which is permitted to be received and an emotional answer message which is rejected to be receive (i.e. para. [0065], “candidate responses being related to the entity and having time labels later than the time point in the set of candidate responses may be retained”, wherein the BRI for answer messages permitted and rejected encompasses the candidate responses that meet a freshness criteria to a specific entity and candidate responses that did not meet the freshness criteria), such that the server-end host selects, from the answer list, the emotional answer message which is permitted to be received and transmits the emotional answer message to the client-end host, wherein the time message is used as a basis of determining the an emotional answer message associated with time (i.e. para. [0065], “Taking an event related to a football game between team A and team B as an example, if it is known that player N was transferred from team C to team B 7 days before the game, candidate responses related to layer N having time stamps within 7 days in the set of candidate responses may be retained, and candidate responses having time stamps 7 days before and being probably related to performance of player N in team C may be filtered out”, wherein answers are filtered based on associated timestamps with regards to the candidate response meeting a freshness criteria).
Claim 6:
Claim 6 is the method claim reciting similar limitations to claim 1 and is rejected for similar reasons.
Claim 7:
Claim 7 is the method claim reciting similar limitations to claim 2 and is rejected for similar reasons.
Claim 8:
Claim 8 is the method claim reciting similar limitations to claim 3 and is rejected for similar reasons.
Claim 9:
Claim 9 is the method claim reciting similar limitations to claim 4 and is rejected for similar reasons.
Claim 10:
Claim 10 is the method claim reciting similar limitations to claim 5 and is rejected for similar reasons.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
U.S. Patent Application Publication NO. 20180357286 “Wang”, which teaches in para. [0032], The emotional intelligence determination server 106 then uses the labeled emotion data to determine one or more potential emotional states of the user 108 based on a provided query and/or command. Using a similar supervised machine learning algorithm, the emotional intelligence determination server 106 may also determine relevant responses to the provided user query and/or command. Although not germane to this disclosure, one of ordinary skill in the art will appreciate that the potential responses to the user query and/or command may be determined similarly (e.g., via labeled and/or unlabeled user data and a supervised machine learning algorithm) as the emotional state of the user 108.
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 DAVID H TAN whose telephone number is (571)272-7433. The examiner can normally be reached M-F 7:30-4:30.
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, Cesar Paula can be reached at (571) 272-4128. 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.
/D.T./ Examiner, Art Unit 2145
/CESAR B PAULA/ Supervisory Patent Examiner, Art Unit 2145
Prosecution Insights