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 § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claims 1–24 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Regarding independent claims 1, 9, and 17
Step 1 — whether the claim falls within any statutory category. See MPEP 2106.03.
Claim 1 is drawn to a computer-performed method (a process); claim 9 is drawn to a tangible, non-transitory, computer-readable media (a manufacture); and claim 17 is drawn to a computer-based chatbot server (a machine). Therefore, each of these claims falls under one of the four categories of statutory subject matter (process/method, machine/product/apparatus, manufacture, or composition of matter). (Step 1: YES.)
Step 2A Prong One — whether the claim recites a judicial exception. See MPEP 2106.04, subsection II.
Regarding independent claim 1, the claim is directed to a computer-performed method for customer assistance, the method comprising:
The limitations of "receiving a request for customer assistance from a user," "determining user intent and context information based on the received request for customer assistance," "mapping the user intent and context information, as contextual data, relative to a plurality of indicators that are suitable for display in a user interface," "setting up … for display at least one relevant smart indicator, from the plurality of indicators, based on the mapped contextual data," and "generating for display … the at least one relevant smart indicator, to provide customer assistance," under their broadest reasonable interpretation, are directed towards the abstract idea of receiving a customer request, evaluating the request to determine the customer's intent and situation, selecting from among a set of possible responsive options the option(s) relevant to that intent and situation, and presenting the selected option(s) to assist the customer.
These limitations are directed towards the abstract idea of a mental process, specifically a concept that can be performed in the human mind, or by a human using pen and paper, including observation, evaluation, judgement, and opinion (see MPEP § 2106.04(a)(2), subsection III). A human customer-service agent can receive a customer's request, mentally evaluate the request to ascertain what the customer wants and the customer's circumstances, mentally identify from known options which responses are relevant to the customer's request, and present or recommend those options to the customer.
These limitations are additionally directed towards the abstract idea of a certain method of organizing human activity, specifically managing personal behavior or relationships or interactions between people, including following rules or instructions, and commercial interactions including sales activities or behaviors and customer assistance (see MPEP § 2106.04(a)(2), subsection II). The claimed receiving of a customer-assistance request and provision of responsive indicators to assist the customer recites managing a commercial customer-service interaction between a customer and an enterprise.
Independent claim 9 is a computer-readable media claim reciting limitations corresponding to those of claim 1 and is directed towards the abstract idea for similar reasons.
Independent claim 17 is a chatbot-server claim reciting limitations corresponding to those of claim 1 and is directed towards the abstract idea for similar reasons.
Step 2A Prong Two — whether the claim as a whole integrates the recited judicial exception into a practical application. See MPEP 2106.04(d).
Regarding independent claim 1, this claim recites the additional elements of "a dynamic chatbot implemented on a chatbot server," "an AI-based configurator," and "a user interface."
These limitations amount to no more than mere instructions to apply the abstract idea using generic computer components, and generally linking the use of the judicial exception to a particular technological environment or field of use (see MPEP § 2106.05(f) and (h)). The recited "dynamic chatbot," "chatbot server," "AI-based configurator," and "user interface" are described in the specification at a high level of generality (see, e.g., specification ¶¶ [0047], [0019], [0055]–[0057]) and recite generic computer components that merely act as a tool to perform the abstract idea. The recitation that the indicators are "AI-generated" or generated by an "AI-based configurator" does not integrate the exception into a practical application, as the specification describes the artificial-intelligence based configurator generically as applying conventional machine learning, natural language processing, and neural-network techniques (see specification ¶¶ [0021], [0036]–[0040]) without any improvement to the functioning of a computer or to another technology. Accordingly, these additional elements, considered individually and in combination, fail to integrate the exception into a practical application.
Regarding independent claim 9, this claim is drawn to a computer-readable media claim reciting limitations corresponding to those of claim 1 and is rejected under the same rationale. Claim 9 also recites the additional elements of a "tangible, non-transitory, computer-readable media," "instructions," and "a processor." These limitations amount to no more than mere instructions to implement the abstract idea on a generic computer and generally linking the use of the judicial exception to a particular technological environment (see MPEP § 2106.05(f) and (h)), and thus fail to integrate the exception into a practical application.
Regarding independent claim 17, this claim is drawn to a chatbot-server claim reciting limitations corresponding to those of claim 1 and is rejected under the same rationale. Claim 17 also recites the additional elements of "a database" and "one or more computer-operable modules." These limitations amount to no more than mere instructions to implement the abstract idea on generic computer components and generally linking the use of the judicial exception to a particular technological environment (see MPEP § 2106.05(f) and (h)), and thus fail to integrate the exception into a practical application.
Step 2B — whether the claim provides an inventive concept, i.e., whether the additional elements amount to significantly more than the judicial exception. See MPEP 2106.05.
Regarding independent claims 1, 9, and 17, the additional elements, considered individually and in combination, do not amount to significantly more than the abstract idea. As discussed above with respect to Step 2A Prong Two, the additional elements of a "dynamic chatbot," "chatbot server," "AI-based configurator," "user interface," "database," "one or more computer-operable modules," "processor," and "computer-readable media" are recited at a high level of generality and represent generic computer components performing generic computer functions. Receiving a user request through a chatbot, storing and retrieving data in a database, processing the request with a processor/module, and generating a display on a user interface are functions that the courts have recognized as well-understood, routine, and conventional computer functions (see MPEP § 2106.05(d)(II), reciting receiving or transmitting data over a network, electronic recordkeeping, and presenting offers and gathering statistics as examples of well-understood, routine, and conventional functions). Applying conventional machine learning, natural language processing, and neural networks to perform the abstract idea, as described generically in the specification (¶¶ [0021], [0036]–[0040]), likewise does not supply an inventive concept. The additional elements therefore do not amount to significantly more than the judicial exception, whether considered individually or as an ordered combination. (Step 2B: NO.)
Accordingly, independent claims 1, 9, and 17 are rejected under 35 U.S.C. 101.
Regarding dependent claims 2–8, 10–16, and 18–24
Step 1 — whether the claim falls within any statutory category. See MPEP 2106.03.
Claims 2–8 depend from claim 1 and are drawn to a process; claims 10–16 depend from claim 9 and are drawn to a manufacture; claims 18–24 depend from claim 17 and are drawn to a machine. Therefore, each of these claims falls under one of the four categories of statutory subject matter. (Step 1: YES.)
Step 2A Prong One — whether the claim recites a judicial exception. See MPEP 2106.04, subsection II.
Regarding claims 2, 10, and 18, these claims recite the limitations of "determining, for a received further request … that a trained query index is not mapped for matching contextual data" and "updating the query index for new user intent or context information." These limitations are directed towards the abstract idea of a mental process (observation, evaluation, and judgement), namely evaluating whether a further request corresponds to previously known information and updating one's knowledge to account for new intent or context.
Regarding claims 3, 11, and 19, these claims recite the limitations of "applying machine learning methods to historical data or customer interactions," "applying business rules to trigger a button set to appear based on context or intent," "creating and triggering options of smart indicators based on number of clicks … past history … or machine learning from similar users," and "using collaborative filtering or deep learning." These limitations narrow the abstract idea of a mental process and a certain method of organizing human activity (following rules or instructions to select and present options), and further recite mathematical concepts implemented generically.
Regarding claims 4, 12, and 20, these claims recite the limitations of "prioritizing, reordering, or highlighting choices," "dynamically presenting choices of a set of most relevant indicators," and "determining to display or hide each of the plurality of indicators." These limitations are directed towards the abstract idea of a mental process, namely evaluating and judging which options to present, emphasize, or omit.
Regarding claims 5, 13, and 21, these claims recite the limitation of "organizing frequently asked questions (FAQ) in smart indicators for customers who want to track orders." This limitation is directed towards the abstract idea of a certain method of organizing human activity (managing a commercial customer-service interaction) and a mental process (organizing and selecting information).
Regarding claims 6, 14, and 22, these claims recite the limitation of "setting up smart indicators based on a scenario, through a business rule that triggers a particular indicator set to appear or the order in which indicators appear." This limitation is directed towards the abstract idea of a certain method of organizing human activity, specifically following rules or instructions.
Regarding claims 7, 15, and 23, these claims recite the limitations of "mapping input data of an interactional event into an intent through an intent classification having an associated set of responses and corresponding smart indicators" and "selecting one or more of the corresponding smart indicators matching user intent." These limitations are directed towards the abstract idea of a mental process, namely classifying a request and selecting matching responses.
Regarding claims 8, 16, and 24, these claims recite the limitations of "determining a plurality of rules for a plurality of process flows through a policy builder," "determining buttons of a smart indicator group specific to each particular use case," "building analytics, in use, on clicking of indicators," and "organizing smart indicators sections for the process flows, based on the analytics." These limitations are directed towards the abstract idea of a certain method of organizing human activity (following rules or instructions and managing interactions) and a mental process (evaluating usage information and organizing options accordingly).
Step 2A Prong Two and Step 2B.
Claims 2–8, 10–16, and 18–24 merely narrow the previously cited abstract idea limitations. For the reasons described above with respect to independent claims 1, 9, and 17, these judicial exceptions are not meaningfully integrated into a practical application, nor do they amount to significantly more than the abstract idea. The dependent claims recite the same generic computer elements as the independent claims (the "AI-based configurator," "dynamic chatbot," "query index," "policy builder," and "user interface"), each described in the specification at a high level of generality and reciting generic computer components that merely act as a tool on which the abstract idea operates (see MPEP § 2106.05(f) and (h)), and reciting well-understood, routine, and conventional computer functions (see MPEP § 2106.05(d)). The additional elements do not provide anything more than the mental processes, methods of organizing human activity, and mathematical concepts identified above. Therefore, claims 2–8, 10–16, and 18–24 also recite abstract ideas that do not integrate into a practical application or amount to significantly more than the judicial exception, and are rejected under 35 U.S.C. § 101.
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.
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.
Claims 1, 2, 4-7, 9, 10, 12-15, 17, 18, and 20-23 are rejected under 35 U.S.C. 103 as being unpatentable over (Fincun) et al. (Fincun), US 11,243,991 B2, in view of (Bhardwaj) et al. (Bhardwaj), US 2022/0247700 A1, and further in view of (Barnett) et al. (Barnett), US 11,720,377 B2.
Regarding claim 1, (Fincun) teaches a computer-performed method for customer assistance using artificial intelligence (AI)-generated smart indicators, comprising ((Fincun), Title, "Contextual Help Recommendations For Conversational Interfaces Based On Interaction Patterns"; Abstract, "A mechanism is provided to implement a chatbot application for providing contextual help recommendations"; col. 7, lines 28-31, disclosing "natural language processing, machine learning, ontology-based artifact generation for the conversation system"):
(A) "receiving a request for customer assistance from a user through a dynamic chatbot implemented on a chatbot server";
(B) "determining user intent and context information based on the received request for customer assistance";
(E) "generating for display, a user interface featuring the at least one relevant smart indicator, to provide customer assistance based on the AI-based configurator set up at least one relevant smart indicator".
As to "receiving a request for customer assistance from a user through a dynamic chatbot implemented on a chatbot server," (Fincun) teaches that the conversational system implements a chatbot application on one or more computing devices operating as server computing devices, and that the chatbot application receives a query from a user in a user input zone of the chatbot user interface ((Fincun), col. 8, lines 16-20, "The conversational system 100 implements a chatbot application on one or more computing devices 104A-C (comprising one or more processors and one or more memories…)"; col. 9, lines 1-2, "104A-C, which may operate as server computing devices"; col. 12, lines 15-17, "The user asks the chatbot application a question by entering the question in user input zone 430"; FIG. 6, block 601, "RECEIVE USER INPUT"; claim 1).
As to "determining user intent and context information based on the received request for customer assistance," (Fincun) teaches applying natural language processing to the query to determine an objective (intent) of the query and to derive conversational context, wherein intents represent the purpose or goal expressed in the user query and context comprises stored information passed across dialog nodes ((Fincun), Abstract, "applies natural language processing (NLP) to the query to determine an objective of the query and a confidence of the objective"; col. 10, lines 26-35, "Intents 351 represent the purpose or goal expressed in the user input/query… the conversation engine uses Deep Learning classifiers to identify intents 351"; col. 11, lines 1-7, describing context 354; FIG. 3, elements 351 and 354).
As to "generating for display, a user interface featuring the at least one relevant smart indicator, to provide customer assistance based on the AI-based configurator set up at least one relevant smart indicator," (Fincun) teaches presenting the generated contextual help recommendations as selectable options in the content zone of the chatbot user interface ((Fincun), col. 12, line 62 to col. 13, line 1, "the contextual help recommendation engine pushes contextual help recommendations 437, 438, 439 into zone 2 420"; FIG. 4B, elements 437-439; FIG. 5B, recommendations 531-534 ["401K," "Money Market," "Roth IRA," "Contact an Agent"]; FIG. 6, block 610, "PRESENT CONTEXTUAL HELP RECOMMENDATIONS TO THE USER"; claim 1). The recited "AI-based configurator set up at least one relevant smart indicator," to which this limitation refers, is addressed in the analysis of limitation (D) below.
(Fincun) teaches, related to limitation (C), that a list of intents with associated confidence scores is returned during the intent-classification process and reduced to a set of recommendations ((Fincun), col. 14, lines 44-55). However, (Fincun) does not expressly teach:
(C) "mapping the user intent and context information, as contextual data, relative to a plurality of indicators that are suitable for display in a user interface".
In the same field of endeavor, (Bhardwaj) teaches "mapping the user intent and context information, as contextual data, relative to a plurality of indicators that are suitable for display in a user interface." Specifically, (Bhardwaj) teaches an intent classification model that determines an intent of a received chat message, wherein each intent is linked to a plurality of responses, and wherein, based on the determined intent, the chatbot service identifies a set of smart responses that are mapped to the intent and displays them as selectable options within the chat user interface ((Bhardwaj), ¶[0049], "intent classification model 300 for determining an intent of a user within a chat message… Each intent may be linked to a plurality of [responses]"; ¶[0075], "Based on the intent, the chatbot service 133 may identify a smart response (or a set of smart responses) that are mapped to the intent"; ¶[0081], "assigning a plurality of confidence scores to the plurality of recommended responses"; FIG. 2B, displaying suggested responses as selectable elements).
(Fincun) and (Bhardwaj) are analogous to the claimed invention as both are from the same field of endeavor of customer-assistance conversational interfaces (chatbots) that determine user intent and present responsive options to the user. Therefore, it would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to modify the contextual-help chatbot of (Fincun) to map the determined user intent and context, as contextual data, relative to a plurality of indicators suitable for display, as taught by (Bhardwaj). This constitutes applying a known technique (intent-classification mapping to a plurality of candidate response indicators) to a known device (a customer-assistance chatbot) to yield the predictable result of presenting the user with a plurality of contextually relevant, selectable indicators. The motivation to combine (Fincun) and (Bhardwaj) is expressly provided by (Bhardwaj), which teaches that mapping intent to a set of smart responses enables the chatbot to "provide accurate and contextually relevant smart suggestions" ((Bhardwaj), ¶[0070]) and to flexibly identify any topic from among a large number of possible topics ((Bhardwaj), ¶[0044]).
The combination of (Fincun) and (Bhardwaj), however, does not expressly teach:
(D) "setting up, by an AI-based configurator, for display at least one relevant smart indicator, from the plurality of indicators, based on the mapped contextual data".
In the same field of endeavor, (Barnett) teaches "setting up, by an AI-based configurator, for display at least one relevant smart indicator, from the plurality of indicators, based on the mapped contextual data." Specifically, (Barnett) teaches a user interface display component (an AI-based configurator) that analyzes the user's context and profile data, identifies a plurality of support topics, generates a first subset of the plurality of support topics, and modifies the user interface to include at least one user interface element (indicator) associated with a support topic in that subset, thereby dynamically setting up the relevant indicator for display based on the analyzed contextual data ((Barnett), Abstract, "The user interface display component generates a first subset of the plurality of support topics and modifies a user interface displayed by the web page to include a user interface element associated with a support topic in the first subset"; claim 1, "generating, by the user interface display component, a first subset of the plurality of support topics; modifying, by the user interface display component, a user interface displayed by the web page to include at least one user interface element associated with a support topic in the first subset of the plurality of support topics"). (Barnett) further teaches that this dynamic configuration is distinguished from conventional interfaces in which "'canned' button interfaces provide the same text" regardless of context ((Barnett), col. 1).
(Fincun), (Bhardwaj), and (Barnett) are analogous to the claimed invention as all are from the same field of endeavor of customer-assistance conversational/user interfaces that present context-dependent selectable elements to a user. Therefore, it would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to further modify the combination of (Fincun) and (Bhardwaj) such that the at least one relevant smart indicator is set up for display, from the plurality of indicators, by an AI-based configurator that dynamically generates the corresponding user-interface element based on the mapped contextual data, as taught by (Barnett). This constitutes applying a known technique (dynamic generation of contextual user-interface elements from analyzed context) to yield the predictable result of a user interface that presents indicators relevant to the user's current situation rather than a fixed set of options. The motivation to combine (Fincun), (Bhardwaj), and (Barnett) is expressly provided by (Barnett), which teaches that conventional "'canned'" button interfaces present the same content regardless of context ((Barnett), col. 1), and that dynamically generating contextual user-interface elements provides functionality tailored to "a user's needs based on the user's interactions" ((Barnett), col. 1), thereby improving the relevance and effectiveness of the presented indicators.
Regarding claim 2, the limitations of claim 1, from which claim 2 depends, are rejected under the same rationale set forth above with respect to claim 1. Claim 2 further comprises:
"determining, for a received further request for customer assistance from the user through the dynamic chatbot, that a trained query index is not mapped for matching contextual data determined from the further request"; and
"updating the query index for new user intent or context information based on the contextual data determined from the further request, for use by the AI-based configurator".
As to "determining, for a received further request for customer assistance from the user through the dynamic chatbot, that a trained query index is not mapped for matching contextual data determined from the further request," (Fincun) teaches related subject matter in that the chatbot application receives further user input and determines whether that input is understood, wherein contextual support is automatically triggered when the chatbot application does not understand the user query, for example when the chatbot application is not trained on the query ((Fincun), col. 12, lines 34-39, "Contextual support is automatically triggered when the chatbot application does not understand the user query, e.g., when the chatbot application is not trained on the query or when the chatbot application is confused between two possible intents"; FIG. 6, block 601, "RECEIVE USER INPUT," and block 602, "USER INPUT UNDERSTOOD?"). (Fincun) teaches that where the input is not understood, the process proceeds to generate contextual help recommendations ((Fincun), FIG. 6, block 602 "NO" branch to block 605, and block 609, "GENERATE CONTEXTUAL HELP RECOMMENDATIONS"). To the extent it is argued that (Fincun) does not expressly teach that "a trained query index is not mapped for matching contextual data determined from the further request," this limitation is expressly taught by (Bhardwaj), as set forth below.
As to "updating the query index for new user intent or context information based on the contextual data determined from the further request, for use by the AI-based configurator," (Fincun) does not expressly teach updating a query index for new user intent or context information.
In the same field of endeavor, (Bhardwaj) teaches:
"determining, for a received further request for customer assistance from the user through the dynamic chatbot, that a trained query index is not mapped for matching contextual data determined from the further request"; and
"updating the query index for new user intent or context information based on the contextual data determined from the further request, for use by the AI-based configurator".
As to "determining, for a received further request for customer assistance from the user through the dynamic chatbot, that a trained query index is not mapped for matching contextual data determined from the further request," (Bhardwaj) teaches a trained intent classification model that receives a chat message during a live conversation and predicts an intent of the chat message, and teaches that the model may lack a mapping for a given intent such that the configuration must be modified to account for it, thereby determining that the trained model (query index) is not mapped for the contextual data of the further message ((Bhardwaj), ¶[0075], "the trained intent classification model 300 may receive a chat message during a live conversation between two users and predict an intent of the chat message using the trained intent classification model 300"; ¶[0033]-[0034], describing that a user's question may be related to a topic and that "This configuration can be easily modified to add a new intent or… update the behavior… of any existing intent," i.e., the trained model is not already mapped for the new intent).
As to "updating the query index for new user intent or context information based on the contextual data determined from the further request, for use by the AI-based configurator," (Bhardwaj) teaches that the intent classification model configuration can be modified to add a new intent, and that the input layer of the intent classification model is updated with additional embeddings identified from the chat messages, either by the user or automatically by the service training the model, whereupon the updated trained model is stored on the host platform for use by the chatbot service in generating smart suggestions ((Bhardwaj), ¶[0034], "This configuration can be easily modified to add a new intent or… update the behavior… of any existing intent"; ¶[0074], "As the model is trained, additional embeddings (unigrams, bigrams, etc.) may be identified for use in intent mining and added to the input layer by the user (e.g., based on inputs via a user interface) or automatically by the service training the model… The trained intent classification model may be stored on the host platform for use by the chatbot service 133"; FIG. 4, "lifecycle 400 of the intent classification model").
(Fincun) and (Bhardwaj) are analogous to the claimed invention as both are from the same field of endeavor of customer-assistance conversational interfaces (chatbots) that determine user intent from a received message and present responsive options to the user. Therefore, it would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to modify the contextual-help chatbot of (Fincun), which triggers recommendation generation when the chatbot is not trained on the further query, such that the system determines that the trained query index is not mapped for the matching contextual data and updates the query index for the new user intent or context information for use by the AI-based configurator, as taught by (Bhardwaj). This constitutes applying a known technique (updating a trained intent-classification model to add a new intent identified from a live message) to a known device (a customer-assistance chatbot that generates recommendations when a query is not understood) to yield the predictable result of a chatbot that expands its coverage over time and correctly maps subsequently received requests. The motivation to combine (Fincun) and (Bhardwaj) is expressly provided by (Bhardwaj), which teaches that this updating enables the model to "easily be extended" ((Bhardwaj), ¶[0071]) and to "provide accurate and contextually relevant smart suggestions" ((Bhardwaj), ¶[0070]) for intents not previously mapped, thereby improving the relevance of the indicators generated for the user.
Regarding claim 4, the limitations of claim 1, from which claim 4 depends, are rejected under the same rationale set forth above with respect to claim 1.
Claim 4 recites the added limitation in the alternative, as a group of actions of which the AI-based configurator performs "at least one." Because the limitation is drawn to a Markush-type group requiring only one of the recited alternatives, the prior art need teach only a single alternative to render the limitation obvious.
Regarding claim 4, (Fincun) teaches:
"prioritizing, reordering, or highlighting choices available to the user through at least a subset of the plurality of indicators" ((Fincun) teaches that a list of intents with associated confidence scores is returned during the intent-classification process, and that the list is reduced by removing options that are similar to the highest-confidence intent, duplicate intents, and intents lacking a human-readable identifier, whereby the resulting recommendation subset is arranged for presentation to the user; (Fincun), col. 14, lines 44-55, "A possible list of contextual help recommendations is obtained during the chatbot's intent classification process where a list of intents with various confidence scores is returned. The list is reduced by removing options: a) that are similar to the intent classified with the highest confidence based on closeness of confidence scores, b) duplicate intents… and c) removing intents that do not have a human readable identifier… The resulting recommendation list may be sorted…").
"determining to display or hide each of the plurality of indicators" ((Fincun) teaches that the contextual help recommendation button and the contextual help recommendations are displayed conditionally—being triggered and presented when the chatbot application does not understand the user query or when the confidence score is below a predetermined threshold, and otherwise not presented—thereby determining to display or hide the indicators; (Fincun), col. 12, lines 34-44, "Contextual support is automatically triggered when the chatbot application does not understand the user query… the chatbot application may detect when a confidence score for an entity or intent is below a predetermined threshold"; FIG. 6, block 602 and block 605).
(Fincun) teaches something related to dynamically presenting a set of most relevant indicators, in that it reduces the returned intent list to a subset of contextual help recommendations for presentation ((Fincun), col. 14, lines 44-55). However, (Fincun) does not expressly teach "dynamically presenting choices of a set of most relevant indicators."
In the same field of endeavor, (Bhardwaj) teaches "dynamically presenting choices of a set of most relevant indicators." Specifically, (Bhardwaj) teaches that the suggested responses are sorted by confidence score in descending order and that a top subset of the suggestions is presented to the user, and that the chat service identifies a top number of predicted responses for presentation, thereby dynamically presenting a set of the most relevant indicators ((Bhardwaj), ¶[0072], "the final suggestion is sorted with the confidence scores in descending order and a top subset (e.g., top 3, 5, etc.) of suggestions"; ¶[0043], the chat service "identifies a top number (e.g., top 3, etc.) of predicted [responses]"; ¶[0081], "selecting the recommended response from the plurality of recommended responses based on the plurality of confidence scores").
(Fincun) and (Bhardwaj) are analogous to the claimed invention as both are from the same field of endeavor of customer-assistance conversational interfaces (chatbots) that determine user intent and present a set of responsive options to the user. Therefore, it would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to configure the AI-based configurator of (Fincun) to dynamically present choices of a set of most relevant indicators, as taught by (Bhardwaj). The motivation to combine (Fincun) and (Bhardwaj) is that doing so applies a known technique (confidence-based ranking and presentation of a top subset of suggestions) to a known customer-assistance chatbot to yield the predictable result of presenting the user with the most relevant indicators, as (Bhardwaj) teaches that this ranking and presentation enables the chatbot to "provide accurate and contextually relevant smart suggestions" ((Bhardwaj), ¶[0070]), consistent with the motivation established in the rejection of claim 1 above.
Regarding claim 5, the limitations of claim 1, from which claim 5 depends, are rejected under the same rationale set forth above with respect to claim 1.
Regarding claim 5, (Fincun) teaches:
"organizing frequently asked questions (FAQ) in smart indicators" ((Fincun) teaches that the contextual help recommendation engine organizes help articles and topics on which the chatbot application has been trained, and presents them as selectable contextual help recommendations (smart indicators) in the user interface; (Fincun), col. 11, lines 3-12, "the contextual help recommendation engine may suggest speaking to or contacting a human agent, searching for help articles, or viewing topics on which the chatbot application has been trained"; FIG. 6, block 612, "SUGGEST HUMAN AGENT, HELP ARTICLES, OR TRAINED TOPICS"; FIG. 5B, elements 531-534, presenting the organized topics as selectable indicators; (Fincun), claim 2, "the presented first set of recommendations further comprise contacting a human agent, searching help articles, and viewing topics").
(Fincun) teaches something related to the recited limitation, in that it organizes help articles and trained topics (FAQ) and presents them as smart indicators in the user interface. However, (Fincun) does not expressly teach organizing such smart indicators "for customers who want to track orders."
In the same field of endeavor, (Barnett) teaches organizing smart indicators "for customers who want to track orders." Specifically, (Barnett) teaches that the user interface display component accesses the user's profile data, determines that the profile data identifies an existing travel reservation of the user, generates a subset of support topics associated with that existing reservation, and modifies the user interface to include user-interface elements (indicators) associated with those support topics, and further teaches determining the status of the existing reservation and filtering the support-topic subset based upon the identified status—thereby organizing topic indicators for a customer seeking to track an existing order/reservation ((Barnett), Abstract, "determines whether the profile data identifies an existing travel reservation… generates a first subset of the plurality of support topics and modifies a user interface displayed by the web page to include a user interface element associated with a support topic in the first subset"; claim 1, "determining, by the user interface display component, that the profile data associated with the user identifies an existing travel reservation of the user; generating, by the user interface display component, a first subset of the plurality of support topics; modifying, by the user interface display component, a user interface displayed by the web page to include at least one user interface element associated with a support topic in the first subset"; claim 2, "status of the existing travel reservation; and filtering, by the user interface display component, the first subset of the plurality of support topics based upon the identified status"). Under the broadest reasonable interpretation, the existing travel reservation of (Barnett) constitutes an "order," and the organizing of support-topic indicators enabling the user to view and track the status of that reservation reads on organizing smart indicators "for customers who want to track orders."
(Fincun) and (Barnett) are analogous to the claimed invention as both are from the same field of endeavor of customer-assistance conversational/user interfaces that present context-dependent selectable elements to a user. Therefore, it would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to modify the organizing of FAQ/help-topic smart indicators of (Fincun) to organize such smart indicators for customers who want to track orders, as taught by (Barnett). This constitutes applying a known technique (organizing support-topic indicators tied to a user's existing order/reservation to enable status tracking) to a known customer-assistance chatbot to yield the predictable result of surfacing FAQ/help indicators relevant to customers tracking their orders. The motivation to combine (Fincun) and (Barnett) is expressly provided by (Barnett), which teaches that dynamically generating contextual user-interface elements tailored to the user's transaction, rather than presenting "canned" elements that "provide the same text" regardless of context ((Barnett), col. 1), provides functionality responsive to "a user's needs based on the user's interactions" ((Barnett), col. 1), thereby improving the relevance of the indicators presented to customers seeking to track their orders.
Regarding claim 6, the limitations of claim 1, from which claim 6 depends, are rejected under the same rationale set forth above with respect to claim 1.
Regarding claim 6, (Fincun) teaches:
"setting up smart indicators based on a scenario, through a business rule that triggers a particular indicator set to appear" ((Fincun) teaches that the contextual help recommendation button set is triggered to appear based on a defined scenario condition, namely when the chatbot application does not understand the user query or when a confidence score for an entity or intent is below a predetermined threshold, such that a particular indicator set is triggered to appear based on that scenario; (Fincun), col. 12, lines 34-44, "Contextual support is automatically triggered when the chatbot application does not understand the user query, e.g., when the chatbot application is not trained on the query or when the chatbot application is confused between two possible intents. More specifically, the chatbot application may detect when a confidence score for an entity or intent is below a predetermined threshold"; FIG. 6, blocks 602, 605, and 609).
(Fincun) teaches something related to the recited limitation, in that it triggers a particular contextual help recommendation set to appear based on a defined confidence-threshold scenario. However, (Fincun) does not expressly teach a business rule governing "the order in which indicators appear."
In the same field of endeavor, (Barnett) teaches setting up smart indicators through a business rule governing "the order in which indicators appear." Specifically, (Barnett) teaches an administrative mapping between attributes and user-interface elements, wherein attributes assigned to each user-interface element available for dynamic generation govern which user-interface element is generated and how it is presented, and wherein an administrative user of the system defines these attribute-to-element mappings and rules that determine the presentation of the user-interface elements—thereby setting up, through a business rule, the manner and order in which the indicators appear ((Barnett), col. 17, "a mapping between a user interface element and an attribute… attributes assigned to each user interface element available for dynamic generation"; col. 17, describing that "an administrative user of the system" defines the mappings governing generation and presentation of the user-interface elements; claim 1, "generating, by the user interface display component, a first subset of the plurality of support topics; modifying, by the user interface display component, a user interface displayed by the web page to include at least one user interface element associated with a support topic in the first subset"). Under the broadest reasonable interpretation, the administratively defined attribute-to-element mapping of (Barnett), which governs which user-interface elements are generated and how they are presented, constitutes a "business rule" governing "the order in which indicators appear."
(Fincun) and (Barnett) are analogous to the claimed invention as both are from the same field of endeavor of customer-assistance conversational/user interfaces that present context-dependent selectable elements to a user. Therefore, it would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to modify the scenario-triggered setting up of smart indicators of (Fincun) such that a business rule governs the order in which the indicators appear, as taught by (Barnett). This constitutes applying a known technique (administratively defined attribute-to-element mapping rules governing which user-interface elements are generated and how they are presented) to a known customer-assistance chatbot to yield the predictable result of presenting a scenario-appropriate indicator set in a rule-governed order. The motivation to combine (Fincun) and (Barnett) is expressly provided by (Barnett), which teaches that dynamically generating contextual user-interface elements according to such mappings, rather than presenting "canned" elements that "provide the same text" regardless of context ((Barnett), col. 1), provides functionality responsive to "a user's needs based on the user's interactions" ((Barnett), col. 1), thereby improving the relevance and presentation of the indicators.
Regarding claim 7, the limitations of claim 1, from which claim 7 depends, are rejected under the same rationale set forth above with respect to claim 1.
Regarding claim 7, (Fincun) teaches the "mapping input data of an interactional event into an intent through an intent classification" portion of sub-limitation (Q), in that the chatbot application applies natural language processing to the received user input and determines an intent through an intent classification process that returns a list of intents from the user query ((Fincun), col. 14, lines 44-47, "A possible list of contextual help recommendations is obtained during the chatbot's intent classification process where a list of intents with various confidence scores is returned"; col. 10, lines 30-35, "the conversation engine uses Deep Learning classifiers to identify intents 351"; FIG. 6, block 601, "RECEIVE USER INPUT," and block 603).
(Fincun) teaches something related to the remainder of the added limitation, in that it identifies an intent through intent classification and generates corresponding contextual help recommendations. However, (Fincun) does not expressly teach an intent classification "having an associated set of responses and corresponding smart indicators," nor "selecting one or more of the corresponding smart indicators matching user intent."
In the same field of endeavor, (Bhardwaj) teaches the intent classification "having an associated set of responses and corresponding smart indicators" of sub-limitation (Q), in that the intent classification model receives a chat message of a live conversation (an interactional event) and predicts an intent, wherein each intent is linked to a plurality of responses, such that the intent classification has an associated set of responses and corresponding smart indicators mapped to the intent ((Bhardwaj), ¶[0049], "FIG. 3A illustrates an intent classification model 300 for determining an intent of a user within a chat message… Each intent may be linked to a plurality of [responses]"; ¶[0075], "the trained intent classification model 300 may receive a chat message during a live conversation… and predict an intent of the chat message… Based on the intent, the chatbot service 133 may identify a smart response (or a set of smart responses) that are mapped to the intent").
(Bhardwaj) further teaches sub-limitation (R), in that, based on the predicted intent, the chatbot service selects one or more of the smart responses mapped to that intent for presentation to the user, selecting the recommended response from the plurality of recommended responses based on the confidence scores ((Bhardwaj), ¶[0075], "Based on the intent, the chatbot service 133 may identify a smart response (or a set of smart responses) that are mapped to the intent"; ¶[0081], "selecting the recommended response from the plurality of recommended responses based on the plurality of confidence scores"; ¶[0043], the chat service "identifies a top number (e.g., top 3, etc.) of predicted [responses]").
(Fincun) and (Bhardwaj) are analogous to the claimed invention as both are from the same field of endeavor of customer-assistance conversational interfaces (chatbots) that determine user intent through intent classification and present responsive options to the user. Therefore, it would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to configure the AI-based configurator of (Fincun) such that the intent classification has an associated set of responses and corresponding smart indicators, and to select one or more of the corresponding smart indicators matching user intent, as taught by (Bhardwaj). This constitutes applying a known technique (intent classification with intents mapped to associated response sets, and confidence-based selection of the matching responses) to a known customer-assistance chatbot to yield the predictable result of presenting the user with the smart indicators corresponding to the determined intent. The motivation to combine (Fincun) and (Bhardwaj) is expressly provided by (Bhardwaj), which teaches that mapping intent to an associated set of responses and selecting the matching responses enables the chatbot to "provide accurate and contextually relevant smart suggestions" ((Bhardwaj), ¶[0070]), consistent with the motivation established in the rejection of claim 1 above.
Regarding claim 9, claim 9 recites a tangible, non-transitory, computer-readable media having instructions that cause a processor to perform a method comprising limitations corresponding to those of claim 1. Claim 9 is rejected under the same rationale set forth above with respect to claim 1.
Regarding claim 10, claim 10 recites limitations corresponding to those of claim 2. Claim 10 is rejected under the same rationale set forth above with respect to claim 2.
Regarding claim 12, claim 12 recites limitations corresponding to those of claim 4. Claim 12 is rejected under the same rationale set forth above with respect to claim 4.
Regarding claim 13, claim 13 recites limitations corresponding to those of claim 5. Claim 13 is rejected under the same rationale set forth above with respect to claim 5.
Regarding claim 14, claim 14 recites limitations corresponding to those of claim 6. Claim 14 is rejected under the same rationale set forth above with respect to claim 6.
Regarding claim 15, claim 15 recites limitations corresponding to those of claim 7. Claim 15 is rejected under the same rationale set forth above with respect to claim 7.
Regarding claim 17, claim 17 recites a computer-based chatbot server comprising a database and one or more computer-operable modules configured to perform operations corresponding to the limitations of claim 1. Claim 17 is rejected under the same rationale set forth above with respect to claim 1.
Regarding claim 18, claim 18 recites limitations corresponding to those of claim 2. Claim 18 is rejected under the same rationale set forth above with respect to claim 2.
Regarding claim 20, claim 20 recites limitations corresponding to those of claim 4. Claim 20 is rejected under the same rationale set forth above with respect to claim 4.
Regarding claim 21, claim 21 recites limitations corresponding to those of claim 5. Claim 21 is rejected under the same rationale set forth above with respect to claim 5.
Regarding claim 22, claim 22 recites limitations corresponding to those of claim 6. Claim 22 is rejected under the same rationale set forth above with respect to claim 6.
Regarding claim 23, claim 23 recites limitations corresponding to those of claim 7. Claim 23 is rejected under the same rationale set forth above with respect to claim 7.
Claims 3, 8, 11, 16, 19, and 24 are rejected under 35 U.S.C. 103 as being unpatentable over (Fincun) et al. (Fincun), US 11,243,991 B2, in view of (Bhardwaj) et al. (Bhardwaj), US 2022/0247700 A1, in view of (Barnett) et al. (Barnett), US 11,720,377 B2, and further in view of (Makar) et al. (Makar), US 2014/0279050 A1.
Regarding claim 3, the limitations of claim 1, from which claim 3 depends, are rejected under the same rationale set forth above with respect to claim 1.
Claim 3 recites the added limitation in the alternative, as a group of actions of which the AI-based configurator performs "at least one." Because the limitation is drawn to a Markush-type group requiring only one of the recited alternatives, the prior art need teach only a single alternative to render the limitation obvious. Each alternative is nonetheless addressed below.
Regarding claim 3, (Fincun) teaches:
"applying machine learning methods to historical data or customer interactions to initiate dynamic smart indicator generation" ((Fincun) teaches that the conversation service uses machine learning and Deep Learning classifiers to identify intents from the user query, and that the contextual help recommendations are generated based on interaction patterns, such that machine learning applied to the conversational interaction data initiates the generation of the contextual help recommendations (smart indicators); (Fincun), Title, "…Based On Interaction Patterns"; col. 10, lines 30-35, "Conversational services typically use a Deep Learning network to identify intents from a given user query… the conversation engine uses Deep Learning classifiers to identify intents 351"; col. 8, lines 15-20; FIG. 6, block 609, "GENERATE CONTEXTUAL HELP RECOMMENDATIONS"; col. 7, lines 28-31, "natural language processing, machine learning, ontology-based artifact generation for the conversation system").
"applying business rules to trigger a button set to appear based on context or intent of a question" ((Fincun) teaches that contextual support is automatically triggered—i.e., the contextual help recommendation button set is triggered to appear—based on the context or intent state of the conversation, specifically when the chatbot application detects that a confidence score for an entity or intent is below a predetermined threshold; (Fincun), col. 12, lines 34-44, "Contextual support is automatically triggered when the chatbot application does not understand the user query, e.g., when the chatbot application is not trained on the query or when the chatbot application is confused between two possible intents. More specifically, the chatbot application may detect when a confidence score for an entity or intent is below a predetermined threshold"; FIG. 6, blocks 602 and 605).
(Fincun) teaches something related to using deep learning based on the messages of users, in that it applies Deep Learning classifiers to the user query to identify intents ((Fincun), col. 10, lines 30-35). However, (Fincun) does not expressly teach "using collaborative filtering or deep learning, based on clicks or messages of users."
In the same field of endeavor, (Bhardwaj) teaches "using collaborative filtering or deep learning, based on clicks or messages of users." Specifically, (Bhardwaj) teaches that the intent classification model comprises a bidirectional long short-term memory (Bi-LSTM) deep-learning architecture that is trained on the messages of users and used to generate the smart suggestions, thereby using deep learning based on messages of users ((Bhardwaj), ¶[0053], describing training of "the Bi-LSTM architecture" on messages; ¶[0044], "trained on historical chat data between users engaged in the [conversation]"; ¶[0054], "the Bi-LSTM architecture… predict an intent of the user within the message").
(Fincun) and (Bhardwaj) are analogous to the claimed invention as both are from the same field of endeavor of customer-assistance conversational interfaces (chatbots) that apply machine-learning and deep-learning methods to user interactions and messages to generate responsive options for the user. Therefore, it would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to configure the AI-based configurator of (Fincun) to use deep learning based on messages of users, as taught by (Bhardwaj). The motivation to combine (Fincun) and (Bhardwaj) is that doing so applies a known deep-learning technique to a known customer-assistance chatbot to yield the predictable result of dynamically generated, contextually relevant smart indicators, as (Bhardwaj) teaches that the deep-learning intent model enables the chatbot to "provide accurate and contextually relevant smart suggestions" ((Bhardwaj), ¶[0070]), consistent with the motivation established in the rejection of claim 1 above.
The combination of (Fincun) and (Bhardwaj) teaches machine learning from user messages, and (Bhardwaj) further teaches the "machine learning from similar users with similar content of message" sub-alternative of the recited action. However, the combination of (Fincun) and (Bhardwaj) does not expressly teach the remaining sub-alternatives of "creating and triggering options of smart indicators based on number of clicks counted across other users, past history of a particular user, or machine learning from similar users with similar content of message."
In the same field of endeavor, (Makar) teaches "creating and triggering options of smart indicators based on number of clicks counted across other users, past history of a particular user, or machine learning from similar users with similar content of message." Specifically, (Makar) teaches a chatbot in which responses and options presented to the user are selected based on behavioral information of the user, wherein the behavioral information includes a source identifier of how the end user was directed to the web page, including a selectable hyperlink selected (clicked) by the end user, and includes information derived from previous interactions with the chatbot (past history of the particular user), and wherein a neural-network predictive engine is used to select and generate the responses ((Makar), ¶[0021], "The behavioral information may include a source identifier of how the end user was directed to the web page… The source identifier may include any of a referring website from which the end user navigated, a selectable hyperlink selected by the end user"; ¶[0022], "Information about the end-user can be derived through a variety of sources such as previous registrations with the retailer's site, the end-user's IP address, previous interactions with the chatbot or through other sources… the greetings, the sales pitch and/or the replies are also customized to a specific user"; ¶[0071], "The Engine 570 in one example is a neural network engine" used to "find a response").
(Fincun), (Bhardwaj), and (Makar) are analogous to the claimed invention as all are from the same field of endeavor of chatbot-based customer-assistance systems that present responsive options to a user based on user data and interactions. Therefore, it would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to configure the AI-based configurator of the (Fincun)–(Bhardwaj) combination to create and trigger options of smart indicators based on the number of clicks and past history of the user, as taught by (Makar). The motivation to combine (Fincun), (Bhardwaj), and (Makar) is that doing so applies a known technique (selecting chatbot options based on a user's click-source and prior-interaction behavioral data via a predictive engine) to a known customer-assistance chatbot to yield the predictable result of options tailored to the user's behavior, as (Makar) teaches that customizing the chatbot replies to the specific user based on such behavioral information "enables highly customizable chatbots specifically tailored to the specific user" ((Makar), ¶[0022]).
Regarding claim 8, the limitations of claim 1, from which claim 8 depends, are rejected under the same rationale set forth above with respect to claim 1.
(Fincun) teaches something related to sub-limitation (U) "building analytics, in use, on clicking of indicators," in that the contextual help recommendations are provided based on interaction patterns of the conversation ((Fincun), Title, "Contextual Help Recommendations For Conversational Interfaces Based On Interaction Patterns"; col. 14, lines 44-55, describing that the recommendation list is derived and reduced based on the intent-classification results of the interaction). However, (Fincun) does not teach:
"determining a plurality of rules for a plurality of process flows through a policy builder";
"determining buttons of a smart indicator group specific to each particular use case";
"building analytics, in use, on clicking of indicators"; and
"organizing smart indicators sections for the process flows, based on the analytics, for use by the AI-based configurator".
In the same field of endeavor, (Barnett) teaches:
"determining a plurality of rules for a plurality of process flows through a policy builder" ((Barnett) teaches an administrative user of the system that defines a mapping between user-interface elements and attributes, wherein the attributes assigned to each user-interface element govern which user-interface element is generated for a given flow, thereby determining a plurality of rules for a plurality of process flows through a policy builder; (Barnett), col. 17, "a mapping between a user interface element and an attribute… attributes assigned to each user interface element available for dynamic generation"; col. 17, "An administrative user of the system… the system 100 may provide a user interface to such a user with which the user may 'tag' contact center agents with skills… [including] actions (e.g., flight changes, train changes, etc.)").
"determining buttons of a smart indicator group specific to each particular use case" ((Barnett) teaches that user-interface elements, including buttons, are generated as a subset specific to the particular use case identified from the user's context—for example, a subset of support-topic buttons associated with a particular booking type or reservation status—such that the buttons of the indicator group are determined specific to each particular use case; (Barnett), col. 16-17, "User interface elements may include buttons… context buttons"; claim 1, "generating, by the user interface display component, a first subset of the plurality of support topics; modifying… to include at least one user interface element associated with a support topic in the first subset"; claim 2, "filtering… the first subset of the plurality of support topics based upon the identified status").
(Fincun) and (Barnett) are analogous to the claimed invention as both are from the same field of endeavor of customer-assistance conversational/user interfaces that present context-dependent selectable elements to a user. Therefore, it would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to modify the contextual-help chatbot of (Fincun) to determine a plurality of rules for a plurality of process flows through a policy builder and to determine buttons of a smart indicator group specific to each particular use case, as taught by (Barnett). This constitutes applying a known technique (administratively defined attribute-to-element mapping rules governing which user-interface element buttons are generated for a given flow/use case) to a known customer-assistance chatbot to yield the predictable result of use-case-specific indicator groups presented according to defined rules. The motivation to combine (Fincun) and (Barnett) is expressly provided by (Barnett), which teaches that dynamically generating contextual user-interface elements according to such mappings, rather than presenting "canned" elements that "provide the same text" regardless of context ((Barnett), col. 1), provides functionality responsive to "a user's needs based on the user's interactions" ((Barnett), col. 1).
The combination of (Fincun) and (Barnett), however, does not teach:
"building analytics, in use, on clicking of indicators"; and
"organizing smart indicators sections for the process flows, based on the analytics, for use by the AI-based configurator".
In the same field of endeavor, (Makar) teaches:
"building analytics, in use, on clicking of indicators" ((Makar) teaches that the end user's interaction with the chatbot indicators—including clicking through on a selectable link or interactive button—is captured as explicit feedback and stored in a database, thereby building analytics, in use, on the clicking of the indicators; (Makar), ¶[0088], "This interaction with the direct response from the end-user via chat window 756 is used as explicit feedback and this context is updated in database 760. For example, filling out a form, placing an order, supplying a credit card number, completing a survey… are all forms of transactions"; ¶[0021], the behavioral information includes "a selectable hyperlink selected by the end user"; ¶[0088], "The feedback information is used by the predictive model 758").
"organizing smart indicators sections for the process flows, based on the analytics, for use by the AI-based configurator" ((Makar) teaches that the captured feedback/analytics is transmitted back to the predictive model, which uses that feedback information to refine future predictions of the responses and offers to deliver to the end user through the chat window, and further teaches organizing the responses into a rule-based hierarchy of campaigns searched and presented for the applicable process flow—thereby organizing the smart-indicator sections for the process flows based on the analytics for use by the AI-based configurator (predictive model); (Makar), ¶[0083], "The feedback information is used by the predictive model 758 to further refine future predictions about the optimal responses or offers to deliver from the chat server 752 to the end-user through chat window 756"; ¶[0075], "the process flows direct to search the hierarchy in step 618… This precedence-in-time creates a hierarchy… searches campaigns based on each campaign's start date"; ¶[0071], "The Engine 570 in one example is a neural network engine" used to select the response).
(Fincun), (Barnett), and (Makar) are analogous to the claimed invention as all are from the same field of endeavor of chatbot-based customer-assistance systems that present responsive options to a user and adapt those options based on user interaction data. Therefore, it would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to modify the combination of (Fincun) and (Barnett) to build analytics, in use, on the clicking of indicators, and to organize the smart-indicator sections for the process flows based on the analytics for use by the AI-based configurator, as taught by (Makar). This constitutes applying a known technique (capturing user click/interaction feedback and using it in a predictive model to refine and organize the options presented for subsequent flows) to a known customer-assistance chatbot to yield the predictable result of indicator sections organized according to observed interaction analytics. The motivation to combine (Fincun), (Barnett), and (Makar) is expressly provided by (Makar), which teaches that using the interaction feedback in the predictive model serves "to further refine future predictions about the optimal responses or offers to deliver" to the user ((Makar), ¶[0083]), thereby improving the relevance of the smart indicators organized and presented for each process flow.
Regarding claim 11, claim 11 recites limitations corresponding to those of claim 3. Claim 11 is rejected under the same rationale set forth above with respect to claim 3.
Regarding claim 16, claim 16 recites limitations corresponding to those of claim 8. Claim 16 is rejected under the same rationale set forth above with respect to claim 8.
Regarding claim 19, claim 19 recites limitations corresponding to those of claim 3. Claim 19 is rejected under the same rationale set forth above with respect to claim 3.
Regarding claim 24, claim 24 recites limitations corresponding to those of claim 8. Claim 24 is rejected under the same rationale set forth above with respect to claim 8.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to HUNG VAN LE whose telephone number is (571)270-0164. The examiner can normally be reached 8 a.m. - 5 p.m..
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.
/HUNG VAN LE/Examiner, Art Unit 2145
/CESAR B PAULA/Supervisory Patent Examiner, Art Unit 2145
Prosecution Insights