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 .
Claim Rejections - 35 USC § 102
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 the appropriate paragraphs of 35 U.S.C. 102 that form the basis for the rejections under this section made in this Office action:
A person shall be entitled to a patent unless –
(a)(1) the claimed invention was patented, described in a printed publication, or in public use, on sale, or otherwise available to the public before the effective filing date of the claimed invention.
Claim(s) 1-2,6-9, 13-16 and 20 is/are rejected under 35 U.S.C. 102 (a)(1) as being anticipated by D’Agostino (US 2025/0307290 A1).
Regarding claim 1, D’Agostino discloses a computer-implemented method for performing critical analysis of an online conversation (D’Agostino [0030] discloses a method directed to a host platform such as a contact center platform (which may be a call center) of a service provider which can harvest contextual information from calls, online chats, and other communications between customers and the service provider, and memorialize the communications and the context for subsequent analysis, retrieval, and use) comprising:
configuring, by an online system (fig. 1 – Host Platform 120), a user interface (user interface112/132) for performing conversations (online conversation) associated with an organization (Service provider/contact center/company’s representatives) (D’Agostino, fig. 1, [0049; 0196] a user interface 112 on a user device provides access to an in app-chatbot used by the user to provide queries about a company’s insurance policy and receive responses from a company’s representative/agent),
wherein each conversation is performed by the online system with a user of the organization (D’Agostino, fig. 1, [0049; 0196] a user interface 112 on a user device provides access to an in app-chatbot used by the user to provide queries about a company’s insurance policy and receive responses from a company’s representative/agent. The chatbot can understand natural language inputs and respond to users' queries conversationally. The chatbot is integrated directly into the company's mobile app, accessible through a designated chat interface);
performing a conversation comprising one or more interactions with a user via the user interface, each interaction comprising a natural language request received from the user (D’Agostino, fig. 1, [0049; 0196] a user interface 112 on a user device provides access to an in app-chatbot used by the user to provide queries about a company’s insurance policy and receive responses from a company’s representative/agent. The chatbot can understand natural language inputs and respond to users' queries conversationally. The chatbot is integrated directly into the company's mobile app, accessible through a designated chat interface) and
a reply to the natural language request generated using a machine learning based language model (D’Agostino [0196] Using natural language understanding capabilities, the chatbot interprets users' inquiries and contextual information from communication vectors that represents users' previous interactions, including chat transcripts, support ticket histories, policy details, and claims information. These vectors serve as valuable data sources for understanding users' insurance needs, preferences, and past inquiries. The retrieved communication vectors are processed by a sophisticated LLM embedded within the chatbot system. The LLM is trained to analyze text data and extract relevant insights, such as users' insurance coverage, policy details, claim status, and questions or concerns expressed in previous interactions. Using these vectors, the chatbot provides accurate and personalized responses or assistance to the user),
generating a prompt for input to (input prompt) a machine learning based language model (Large language model (LLM)) comprising the natural language request) (D’Agostino [0144] discloses an input prompt to an LLM during execution which enables the user to provide queries about a company’s insurance policy to a company’s representative/agent. in a natural language through in app-chatbot) and
interactions of the conversation (D’Agostino [0196] Using natural language understanding capabilities, the chatbot interprets users' inquiries and contextual information from communication vectors that represents users' previous interactions, including chat transcripts, support ticket histories, policy details, and claims information. These vectors serve as valuable data sources for understanding users' insurance needs, preferences, and past inquiries ) and
requesting a machine learning based language model (LLM) to evaluate the conversation to perform a critical analysis (contextual analysis of conversation information by trained LLM and extract relevant insights) of the conversation (D’Agostino [0196] Using natural language understanding capabilities, the chatbot interprets users' inquiries and contextual information from communication vectors that represents users' previous interactions, including chat transcripts, support ticket histories, policy details, and claims information. These vectors serve as valuable data sources for understanding users' insurance needs, preferences, and past inquiries. The retrieved communication vectors are processed by a sophisticated LLM embedded within the chatbot system. The LLM is trained to analyze text data and extract relevant insights, such as users' insurance coverage, policy details, claim status, and questions or concerns expressed in previous interactions. Using these vectors, the chatbot provides accurate and personalized responses or assistance to the user),
the critical analysis (contextual analysis of conversation information by trained LLM and extract relevant insights) determining one or more conversation attributes (users' previous interactions, policy details, insurance needs and preferences), wherein a conversation attribute describes the conversation (D’Agostino [0196] Using natural language understanding capabilities, the chatbot interprets users' inquiries and contextual information from communication vectors that represents users' previous interactions, including chat transcripts, support ticket histories, policy details, insurance needs, preferences and claims information);
providing the prompt to the machine learning based language model for execution (D’Agostino [0141], fig. 6C, a process 600C of a retriever 642 generates a prompt 670 which includes the subset of vectors including vector 662, vector 664, and vector 666. Prompt 670 includes additional text 672 used to request the LLM 646 to “generate a product offer based on the item of interest and the user mood.” In response, the LLM 646 may ingest the prompt 670 when generating the response(s) that is output during the active communication session between the source device 610 and the service provider device 620);
receiving a response generated (generated output response) by the machine learning based language model (LLM) based on the prompt, the response including values for the one or more conversation attributes (generate a product offer based on the item of interest and the user mood) describing the conversation (item of interest) (D’Agostino [0141], fig. 6C, prompt 670 the process 600C includes additional text 672 that may request/prompt the LLM 646 to “generate a product offer based on the item of interest, preferences and the user mood.” In response, the LLM 646 may ingest the prompt 670 when generating the response(s) that is output during the active communication session between the source device 610 and the service provider device 620); and
modifying prompts (change of user’s interest over time) generated for responding to one or more subsequent natural language requests received from the user to cause the machine learning based language model to generate responses that cause the one or more conversation attributes describing the conversation to change (provide vectors associated with user’s new interest and current mood) (D’Agostino fig. 8C, [0183-0185], a process 800C, monitors a user’s interaction with a chatbot to identify an item which the user has a changed interest over time and generating content about the item based on the change in interest. The LLM 820 may identify an aggregate of vectors about an item such as “item G”. The LLM 820 may detect a “change” in interest with respect to the item that happens over time based on the contextual attributes within the vectors. Here, the user has changed their mind with respect to “item G”. The LLM 820 may provide the vectors associated to a second LLM 824, which can determine a current mood of the user with respect to “item G”).
Regarding claim 2, D’Agostino discloses the computer-implemented method of claim 1, wherein the machine learning based language model is a large language model (D’Agostino, fig. 1 [0038] discloses a host platform 120 which include an artificial intelligence (AI) framework or model that includes one or more large language models (LLMs) that can extract context from a communication session between a user and a contact center (or chatbot), and generate a vectorized representation of the communication session which includes the context).
Regarding claim 6, D’Agostino discloses the computer-implemented method of claim 1, wherein the one or more conversation attributes comprise a measure of a mismatch (standard response deviation from a tailored/personalized response) between a personality of the user (user profile/ emotions such as stress, anxiety, optimism, or frustration) and a personality of the online system (standard machine model response) in responding to natural language requests from the user (D’Agostino [0170-0171;0207] an emotionally intelligent chatbot, using natural language understanding capabilities, may receive a request from a user and provide tailored/personalized responses based on a user’s behavioral profile, where the behavioral data may include emotions such as stress, anxiety, optimism, or frustration by analyzing linguistic cues, tone of voice, and contextual information. Leveraging its understanding of users' emotions, the chatbot delivers empathetic and supportive responses tailored to users' emotional needs instead of a standard machine response which will be a mismatch from the users’ emotional need).
Regarding claim 7, D’Agostino discloses the computer-implemented method of claim 6, wherein modifying prompts generated for responding (Dynamically adjusting responses based on real-time engagement with a user and user’s changing emotions/sentiments) to one or more subsequent natural language requests received from the user (D’Agostino [0172] the system employs sentiment analysis algorithms to analyze customers' natural language communication content and dynamically prompt the system to modify/adjust the systems’ responses for subsequent conversation by drawing insights from sentiment analysis and contextual understanding, to deliver personalized product recommendations, promotional offers, and service suggestions tailored to customers' preferences and sentiments) comprises:
receiving a natural language request from the user (D’Agostino, fig. 1 [0049 0196] a user interface 112 on a source device 110 provides user access to an in app-chatbot used by the user to provide queries to the app-chatbot in natural language);
generating a second prompt (dynamic prompts for user request) for input to a machine learning based language model comprising the natural language request and a request to generate a response that causes the personality (personalized/tailored responses) of the online system to match the personality (preferences and sentiments) of the user (D’Agostino [0172] the system employs sentiment analysis algorithms to analyze customers' natural language communication content and dynamically prompt the system to modify/adjust the systems’ responses for subsequent conversation by drawing insights from sentiment analysis and contextual understanding, to deliver personalized product recommendations, promotional offers, and service suggestions tailored to customers' preferences and sentiments);
receiving a response obtained by executing the machine learning based language
model using the second prompt (D’Agostino [0172] the system employs sentiment analysis algorithms to analyze customers' natural language communication content and dynamically prompt the system to modify/adjust the systems’ responses for subsequent conversation by drawing insights from sentiment analysis and contextual understanding, to deliver personalized product recommendations, promotional offers, and service suggestions tailored to customers' preferences and sentiments); and
generating a reply to the natural language request based on the response generated by the machine learning based language model based on the second prompt (D’Agostino [0170-0171;0207] an emotionally intelligent chatbot, using natural language understanding capabilities, may receive a request from a user and provide tailored/personalized responses based on a user’s behavioral profile, where the behavioral data may include emotions such as stress, anxiety, optimism, or frustration by analyzing linguistic cues, tone of voice, and contextual information. Leveraging its understanding of users' emotions, the chatbot delivers empathetic and supportive responses tailored to users' emotional needs instead of a standard machine response which will be a mismatch from the users’ emotional need. By understanding customers' emotional tone, attitudes, and sentiments, the platform categorizes interactions into positive, neutral, or negative sentiment categories. Leveraging LLMs with multiple attention heads, the system dynamically adjusts its real-time engagement strategies based on customers' sentiments and preferences. For instance, during a conversation session with a customer, the system analyzes the sentiment of the interaction and tailors its responses, offers, or recommendations accordingly to align with the customer's emotional state and needs).
Regarding claim 8, D’Agostino discloses a non-transitory computer readable storage medium storing instructions that when executed by one or more computer processors cause the one or more computer processors to perform steps for performing critical analysis of an online conversation, (D’Agostino [0002] discloses an apparatus that may include a memory and a processor coupled to the memory, the processor configured to store first interaction content with a service provider, to implement certain steps) the steps comprising:
The rest of the limitations of claim 8, are rejected with ration similar to that of claim 1.
Regarding claim(s) 9,13 and 14, the claim(s) is/are rejected with rational similar to that of claim(s) 2, 6 and 7, respectively.
Regarding claim 15, D’Agostino discloses a computer system comprising:
one or more computer processors; and a non-transitory computer readable storage medium storing instructions that when executed by one or more computer processors cause the one or more computer processors to perform steps for performing critical analysis of an online conversation, (D’Agostino [0002] discloses an apparatus that may include a memory and a processor coupled to the memory, the processor configured to store first interaction content with a service provider, to implement certain steps) comprising:
The rest of the limitations of claim 15, are rejected with ration similar to that of claim 1. comprising:
Regarding claim(s) 16 and 20, the claim(s) is/are rejected with rational similar to that of claim(s) 2 and 7, respectively.
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.
The text of those sections of Title 35, U.S. Code not included in this action can be found in a prior Office action.
Claim(s) 3 – 5, 10-12 and 17- 19 is/are rejected under 35 U.S.C. 103 as being unpatentable over D’Agostino (US 2025/0307290 A1), in view of Wyss et al. (US 2022/0400091 A1).
Regarding claim 3, D’Agostino discloses the computer-implemented method of claim 1 but did not explicitly disclose wherein the one or more conversation attributes describing the conversation comprise a measure of pacing of the conversation.
Wyss discloses wherein the one or more conversation attributes describing the conversation comprise a measure of pacing of the conversation (Wyss, fig. 4, [0078] in block 410 of process 400 implemented on system 100, the system determines whether to perform bot-driven interaction pacing in one or more of the conversations between a chat bot 114 and corresponding human user. If the system 100 determines, in block 412, to provide such pacing, then the method 400 advances to block 414 in which the system 100 provides bot-driven content filler via the relevant chat bot(s) 114).
One of ordinary skill in the art would have been motivated to combine D’Agostino and Wyss because these teachings are from the same field of endeavor with respect to disclosing techniques for related to monitoring and improving human conversations chatbots.
Therefore, before the effective filing date of the invention, it would have been obvious to a person of ordinary skill in the art to incorporate the strategies by Wyss into the invention of D’Agostino. The motivation would have been to leverage bot-driven interaction pacing when the cognitive load on a particular agent is too high, when the agent is backlogged (e.g., to keep the user occupied), or if it is otherwise prudent to automatically engage with the user, Wyss, [0078].
Regarding claim 4, D’Agostino modified by Wyss disclose the computer-implemented method of claim 3, wherein the measure of pacing of the conversation is determined based on a number of interactions with the user for a particular conversation topic (Wyss, fig. 4, [0078;0081;0088-0089] in block 406, the system 100 determines a metric associated with a topic of the conversation between the human user and the chat bot 114. System 100 may determine a set of topics and each of the topics may have a particular scalar value assigned to it that is representative of its importance or priority and a topic that is more important than another will be assigned a greater scalar value. Whether or not bot-driven interaction pacing is provided (e.g., as content filler) in block 414, depends on the priority score of the conversation and the system 100 may be configured to refresh each of the priority scores periodically (e.g., every second or other suitable period).
The motivation to combine is similar to that of claim 3.
Regarding claim 5, D’Agostino modified by Wyss disclose the computer-implemented method of claim 3, wherein the prompt is a first prompt, wherein modifying prompts generated for responding to one or more subsequent natural language requests received from the user (Wyss [0086] the system 100 may leverage a natural language processing algorithm to receive input prompt for requesting responses based on analysis of a user’s sentiment. Responses to a user may be modified according to results of the sentimental analysis), comprises
receiving a natural language request from the user (Wyss, fig. 1, [0041] accessing a user interface on a user’s device 108, the user may send a request through a natural language to chatbots 114);
generating a second prompt for input (dynamic prompts/request from a user during a natural language conversation between a human user and a chatbot) to a machine learning based language model comprising the natural language request (Wyss, fig. 1, [0018;0041;0097] accessing a user interface on a user’s device 108, the user may send a request through a natural language to chatbots 114. The system dynamically prompt the user to provide a request, where the request includes user’s sentiments about a topic. Each topic associated with a conversation between a human user and a chatbot 114 is assigned to a scalar value which represents the importance and priority of the topic. The topics may be arranged by order of importance, and a topic that is more important than another will be assigned to a greater scalar value) and
a request to generate a response that causes the pacing of the conversation to change receiving a response obtained by executing the machine learning based language
model using the second prompt (Wyss, fig. 4, [0078] in block 410 of process 400 implemented on system 100, the system determines whether to perform bot-driven interaction pacing in one or more of the natural language conversations between a chat bot 114 and corresponding human user. Whether or not bot-driven interaction pacing is provided (e.g., as content filler) in block 414, is provided for a given conversation depending on the priority score associated with the topic of the conversation which may be configured to refresh each of the priority scores periodically (e.g., every second or other suitable period. That is, the priority score of the topic may be used to request pacing request/response associated with conversation of a topic); and
generating a reply to the natural language request based on the response generated by the machine learning based language model based on the second prompt (generating a prompt to adjust the packing of a conversation based on a change in the priority score of the topic of the conversation) (Wyss, [0097] the chat bot 114, using a natural language may simply have indicated that it is requesting an agent as a mechanism to pace the conversation (e.g., as content filler) while the priority score is adjusted to reflect the current state and priority of the conversation. In other contexts, a message similar to the message 606 may be a result of the human agent recognizing that the priority score reflects that intervention has a high priority and therefore seizing control of the conversation from the chat bot 114, which automatically prompted a message similar to the message 606 in order for the conversation to be seamless).
The motivation to combine is similar to that of claim 3.
Regarding claim(s) 10-12, the claim(s) is/are rejected with rational similar to that of claim(s) 3-5, respectively.
Regarding claim(s) 17-19, the claim(s) is/are rejected with rational similar to that of claim(s) 3-5, respectively.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant’s disclosure. The following publications show the state of the art related to techniques for improving online conversations with a chatbot.
Casper (US 2023/0186034 A1)
Bi et al. (US 2025/0335716 A1)
Any inquiry concerning this communication or earlier communications from the examiner should be directed to DIXON F DABIPI whose telephone number is (571)270-3673. The examiner can normally be reached on Monday - Friday from 9:00 am – 5:00 pm.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Christopher L Parry, can be reached at telephone number 571-272-8328. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of an application may be obtained from Patent Center. Status information for published applications may be obtained from Patent Center. Status information for unpublished applications is available through Patent Center to authorized users only. Should you have questions about access to the USPTO patent electronic filing system, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free).
Examiner interviews are available via a variety of formats. See MPEP § 713.01. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) Form at https://www.uspto.gov/InterviewPractice.
/D.F.D/ Examiner, Art Unit 2451
/Chris Parry/Supervisory Patent Examiner, Art Unit 2451
Prosecution Insights