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 .
This action is in response to the amendment filed 2 September 2025. Claims 1, 17, 20 have been amended. Claim 6, 11, 15 have been canceled. Claim 23 is new. Claims 1-5, 7-10, 12-14, 16-23 are pending and have been considered below.
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 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.
This application currently names joint inventors. In considering patentability of the claims the examiner presumes that the subject matter of the various claims was commonly owned as of the effective filing date of the claimed invention(s) absent any evidence to the contrary. Applicant is advised of the obligation under 37 CFR 1.56 to point out the inventor and effective filing dates of each claim that was not commonly owned as of the effective filing date of the later invention in order for the examiner to consider the applicability of 35 U.S.C. 102(b)(2)(C) for any potential 35 U.S.C. 102(a)(2) prior art against the later invention.
Claim(s) 1-5, 7-10, 12-14, 16-23 is/are rejected under 35 U.S.C. 103 as being unpatentable over Ciano et al. (US 2019/0238487 A1) in view of Blackhurst et al. (US 2015/0032480 A1) and Punera et al. (US 2015/0134389 A1) and Allen et al. (US 2018/0011868 A1) and further in view of Cauchois et al. (US 2013/0346886 A1).
Claim 1. Ciano discloses a system for processing a request associated with an electronic record, comprising:
an artificial intelligence orchestration platform, including: an artificial intelligence orchestration communication device to receive electronic record, the computing infrastructure includes a model system interface and a model server system (P. 0035) one or more aspects of the corpus of information may be stored or referenced within data associated with the model advisor (P. 0046) which may be an automated service management system including a model advisor (P. 0038) the model advisor retrieves a corpus of information which may include records from a set of repositories (P. 0046) and configures a conversational agent learning model based on the corpus of information (P. 0047) wherein the model advisor may classify conversation content received from one or more of the plurality of clients via the at least one conversational interface, including any conversation input or any other input received from one or more of the plurality of clients (P. 0056) The received electronic record is analogous to both the retrieved corpus of information and conversational content with one or more clients,
an artificial intelligence orchestration processor coupled to the artificial intelligence orchestration communication device, the model system interface may be separate from the model server system, and include a separate processor, memory, and/or storage, or be a constituent component of the model server (P. 0037) the model server system may include a memory, storage, server I/O device interface, and a CPU (P. 0038), and
an artificial intelligence orchestration storage device in communication with said artificial intelligence orchestration processor and storing instructions adapted to be executed by said artificial intelligence orchestration processor, the model system interface includes a memory and/or storage and comprise a software application (P. 0037) to:
(ii) automatically determine one or more intents associated with the request, retrieving the corpus of information includes identifying a plurality of issues and identifying a plurality of solutions respectively corresponding to the plurality of issues, identifying client content associated with the plurality of issues or the plurality of solutions and identifying any rankings associated with the plurality of solutions (P. 0005, 0053) the model advisor may derive a plurality of intents based upon the plurality of issues (P. 0054) wherein the model advisor may classify conversation content received from one or more of the plurality of clients and classifying the intent may entail classifying at least one intent derived from the conversation content by the model advisor (P. 0056), and
(iv) transmit an indication associated with intent, updating the conversational agent learning model to address the at least one deficiency may include incorporating at least one new intent into the plurality of intents or incorporating at least one new entity into the plurality of entities (P. 0005) the client network interface may receive data from, and may transmit data to, model server system via network (P. 0036) the model system interface may receive data from, and may transmit data to, client computing system via network (P. 0037) the model advisor may accept requests sent by a client computing system to the model server system, and further may transmit data to a client computing system, via the model system interface, external sources and/or external service providers may transmit external data to, or may otherwise communicate with, the model advisor via the model system interface (P. 0039) dialog portions may include conversational text to address one or more intents (P. 0054) an intent is derived from the conversation content in a conversational interaction between at least one client of a plurality of clients and the conversational agent learning model (P. 0056);
an entity extraction platform coupled to the artificial intelligence orchestration platform, each entity in a plurality of entities may be an object class or a data type that enables selection of at least one action based upon the plurality of solutions in order to address one or more of the plurality of intents (P. 0005) the model advisor may derive a plurality of intents based upon the plurality of issues, wherein each of the plurality of intents indicates a purpose or a goal, and the model advisor may derive a plurality of entities (P. 0054) It is clear that deriving an entity in Ciano is analogous to extracting and entity as claimed, including:
an entity extraction communication device to receive the indication of intent transmitted by the artificial intelligence orchestration platform, each entity in a plurality of entities may be an object class or a data type that enables selection of at least one action based upon the plurality of solutions in order to address one or more of the plurality of intents (P. 0005) the model advisor may update the conversational agent learning model by incorporating at least one new entity into the plurality of entities derived (P. 0055) and intents and entities may be derived from conversational content (P. 0056),
an entity extraction processor coupled to the entity extraction communication device, the model advisor derives entities (P. 0055) based on conversational content (P. 0056),
an entity extraction storage device in communication with said entity extraction processor and storing instructions adapted to be executed by said entity extraction processor, model system interface includes a memory including a software application (P 0037) the model advisor may be stored in the model server memory, and storage may include model advisor data and model policies (P. 0038) to:
(i) based on the indication of intent … (ii) identify a type of transaction associated with the request, the model advisor may identify at least one deficiency in the conversational agent learning model which may include, e.g., subject matter not previously encountered and/or subject matter for which there is relatively little or no information to track or access in the context of the model, the model advisor may update the conversational agent learning model to address the at least one deficiency (P. 0051) entities are object classes or a data types that enable selection of an action based upon the plurality of solutions to address one or more intents, for example one intent is a bill payment (P 0054) via conventional methods or via crypto currency (P 0050) a new category may be derived from the conversation based on intent and a new entity (P 0055) a conversation content is classified based intent and an entity derived from the conversation (P 0056), The solution and intent identified in Ciano is analogous to the claimed transaction type,
… automatically extract, based on machine learning, based on information received consequent to interaction between the conversational agent learning model and the clients, including conversation content and/or general information associated with one or more of the plurality of clients, the model advisor identifies deficiencies in the conversational agent learning model, including subject matter not previously encountered and/or subject matter for which there is relatively little or no information, and updates the conversational agent learning model to address the at least one deficiency by autonomously adding new information or new categories from a set or repositories or external data by applying one or more machine learning techniques (P 0051) The model advisor updates the learning model with new categories and/or information from repositories or external data using machine learning, therefore, the new data can be interpreted to be extracted using machine learning techniques, see the next limitation for the rejection of the extracting step,
at least one requisite entity identifier from text of the electronic record in accordance with a transaction requirement for the identified type of transaction., each of the plurality of entities may be an object class or a data type (P. 0005) according to the payment example, the model advisor may derive the bill as an entity from the derived bill payment intent, which in turn may enable the model advisor to select at least one action upon the derived bill in order to address the derived bill payment intent, e.g., processing client payment of the bill (P. 0054) and an intent and an entity may be derived from conversational content (P. 0056) wherein the conversational content included conversational text (P. 0059) The “bill” is analogous to the claimed entity identifier,
wherein the at least one requisite entity identifier identifies [at least one of a name and address] adapted to receive an output of a processed request, each of the plurality of dialog portions may include conversational text to address one or more of the plurality of intents consequent to applying the at least one action to one or more of the plurality of entities (P. 0005) the conversation output may include conversational text to address the at least one intent derived from the conversation content by applying at least one action to the at least one entity derived from the conversation content (P. 0059); and
a robotic automation platform to process the request utilizing the indication of intent and the extracted requisite entity identifier, the model advisor may determine one or more dialog portions consequent to applying one or more bill payment processing actions upon the derived bill to address the derived bill payment intent, the one or more dialog portions determined by the model advisor further may involve provision of information regarding the derived entity (P. 0054) wherein the model advisor may classify conversation content received from one or more of the plurality of clients and classifying the intent may entail classifying at least one intent derived from the conversation content by the model advisor (P. 0056) a check graphic may be received by the model advisor and may apply image recognition to determine a unified text-based artifact based upon bill payment, the model advisor may complete a semantic analysis with respect to the text-based artifact related to bill payment to classify the check graphic by associating it with a bill payment intent (P. 0062) the model advisor may classify an entity according to intent (P. 0060) Intent can be derived from both the corpus of information and form conversation content and then; Applicant’s description of a robotic automation system includes: process[ing] the request utilizing the indication of intent and the extracted requisite entity identifier (P. 0041), in other words, the disclosed robotic automation system appears to disclose digital data processing and not the manipulation of mechanical robotic devices, and
an intent classifier adapted to output a … confidence value based on an analysis of the … extracted requisite entity identifier, the model advisor may identify any rankings associated with the plurality of solutions including client-provided rankings, statistically based rankings, and/or rankings from external sources, such as rankings from social media and assess such rankings upon configuring or selecting model service(s) which obtain data from the corpus of information (P 0053) each of the plurality of entities may be an object class or a data type that enables selection of at least one action based upon the plurality of solutions in order to address one or more of the plurality of intents (P 0054) An entity is associated with an intent and a solution, and the associated solutions are ranked, thereby ranking the associated entities.
Ciano does not disclose (i) determine whether the request is associated with a supported policy, (ii) automatically determine one or more intents associated with the request, in a case the request is associated with the supported policy, (iii) identify one of the one or more determined intents as a primary intent, wherein the primary intent is a main reason used to make a decision, as disclosed in the claims. However, in the same field of invention, Blackhurst discloses electronic communications are parsed to identify online purchase transaction data (P 0032) at least one insurance related transaction is identified from the one or more transactions (P 0040) based on the product description, the merchant, the transaction amount, the transaction data, or other transaction and e-receipt data, e.g. if a purchase item relates to home repair (e.g., shingles) and if the purchase is associated with a home improvement store, a home insurance related transaction can be identified based on the amount of purchase (P 0041) an insurance transaction is processed including valuing an insurance claims, submission of a proof of purchase for an insurance claim, and the like based on at least a portion of the purchase amount of the at least one insurance related transaction (P 0043) wherein whether an insurance policy has lapsed is taken into account (P 0046) the electronic communications being customer emails (P 0057) filtered by text (P 0058) parsed to extract transaction and/or shipping information using predefined templates (P 0059) the emails including purchase order confirmation, shipping confirmation, typed or handwritten notes, invoices, bills of sale, or other e-receipt data (P 0061). Blackhurst discloses that insurance customer communications in the form of emails are parsed to identify the purpose of the communications and then the extracted information is used to process the subject of the information in the email. Whether the customer’s insurance policy has lapsed or not is taken into account. Therefore, considering the teachings of Ciano and Blackhurst, one having ordinary skill in the art before the effective filing date of the invention would have been motivated to combine (i) 1 the request is associated with a supported policy, (ii) automatically determine one or more intents associated with the request, in a case the request is associated with the supported policy, (iii) identify one of the one or more determined intents as a primary intent, wherein the primary intent is a main reason used to make a decision with the teachings of Ciano with the motivation to more accurately identify and gather correct data to submit insurance claims to assist with complex and detailed agreements that can be difficult for policy holders to navigate (Blackhurst: P 0001).
Ciano does not disclose the primary intent is identified via a natural language classifier (NCL) service; an entity extraction platform including a trained natural language understanding (NLU) text extraction model, as disclosed in the claims. However, Ciano discloses the model advisor associates at least one intent derived from the conversation content with at least one intent among the plurality of intents derived by comparing of the intent derived from the conversation content with the respective intents among the plurality of intents derived during configuration of the conversational agent learning model through natural language processing (P 0060). In the same field of invention, Punera discloses a machine learning model is pretrained for natural language analysis, and a suggestion generator applies the models to extract contact information from the user's communication with other users (i.e., names, phone numbers, email addresses, job titles, affiliated companies, web sites, etc.), extract contact details from meeting communications, perform natural language analysis or other machine learning based analysis on the content of a communication (P 0046) and, for example, to locate a question posed within a block of text, a word or words indicating a timing of the request, and the contact to whom the request is directed, and various other types of calls to action (P 0068). Therefore, considering the teachings of Ciano, Blackhurst and Punera, one having ordinary skill in the art before the effective filing date of the invention would have been motivated to combine the primary intent is identified via a natural language classifier (NCL) service; an entity extraction platform including a trained natural language understanding (NLU) text extraction model with the teachings of Ciano and Blackhurst with the motivation to ensure that suggestions generated for a user are more likely to be of use to the user can based on the user profile (Punera: P 0095).
Ciano does not disclose a domain lexicon specific to intent type, as disclosed in the claims. However, in the same field of invention, Allen discloses a subsystem receives and processes questions using natural language processing to analyze and extract topic information such as named entities, phrases, urgent terms, and/or other specified terms which are stored in one or more domain entity dictionaries relating to different domains or areas to identify critical or urgent words in each question based on their presence in the domain dictionary and to derive meaning from a human-oriented question and to identify key terms and attributes in the question and compare the identified terms to the stored terms in the domain dictionary (P 0028). Therefore, considering the teachings of Ciano, Blackhurst, Punera and Allen, one having ordinary skill in the art before the effective filing date of the invention would have been motivated to combine a domain lexicon specific to intent type with the teachings of Ciano, Blackhurst and Punera with the motivation to identify the context of a user’s interactions associated with a file and a given application (Allen: 0004).
Ciano does not disclose access and analyze other forms for data extraction on intent type element extractions per the domain lexicon specific to intent type, wherein the other forms are at least one of an e-form, a scanned form and an online web form, as disclosed in the claims. However, Ciano unstructured content may include unstructured forms (P 0061). Blackhurst discloses the communications may take the form of purchase order confirmations provided as a web page (P 0055). Punera discloses text-based communications include email communications, although other forms text-based communications (e.g., SMS messages, MMS messages, OCR text extracted from a physical communication, etc.) (P 0067) (meeting) communications can include a meeting invitation sent via electronic mail, an existing calendar appointment stored in an exchange system or a web-based calendar system, etc. (P 0073) machine learning model analysis techniques, such as natural language analysis, analyze the content, format, and placement of content within the electronic communication in order to locate a message signature (P 0079) the electronic communication is obtained from one of a web-based electronic mail system (Claim 19). Allen discloses a user may be using a laptop or tablet computer to access a web site to purchase an item, access their bank account to pay bills, or plan upcoming business travel (P 0058). Therefore, considering the teachings of Ciano, Blackhurst, Punera and Allen, one having ordinary skill in the art before the effective filing date of the invention would have been motivated to combine access and analyze other forms for data extraction on intent type element extractions per the domain lexicon specific to intent type, wherein the other forms are at least one of an e-form, a scanned form and an online web form with the teachings of Ciano, Blackhurst, Punera and Allen with the motivation to identify the context of a user’s interactions associated with a file and a given application (Allen: 0004).
Ciano does not disclose based on the analysis of other forms, automatically extract, per the NLU extraction model, as disclosed in the claims. However, However, Allen discloses when a file is selected its associated metadata, such as the file's name, author, last modification date, location within a directory structure, frequency of use, type (e.g., document, spreadsheet, audio file etc.), association with a particular application, and so forth, is collected and then analyzed (P 0050) the information used to perform the NLP analysis operations may likewise include file types stored in various folders, file path segments, directory and file names, or any combination thereof to generate augmented file classification information and likelihood and probability scoring approaches are implemented to determine which identities are a match and what, if any, non-obvious relationships exist between them (P 0053) a document is analyzed to identify sentences associated with one or more knowledge domains for intent determination (P 0054) and generate intent classification (P 0055). Therefore, considering the teachings of Ciano, Blackhurst, Punera and Allen, one having ordinary skill in the art before the effective filing date of the invention would have been motivated to combine based on the analysis of other forms, automatically extract, per the NLU extraction model with the teachings of Ciano, Blackhurst, Punera and Allen with the motivation to identify the context of a user’s interactions associated with a file and a given application (Allen: 0004).
Ciano does not disclose the at least one requisite entity identifier identifies at least one of a name and an address, as disclosed in the claims. However, Ciano discloses the model advisor may access such profile data from the client data repository and/or from external sources, e.g., which may provide social networking data associated with the client to be more capable of accurately deriving intents and/or entities from the conservation content involving the client, as the profile data may enable the model advisor to detect client preferences (P 0056). While Ciano discloses that accessing the client profile data allows the model advisor to more accurately derive intents and/or entities as the profile data may enable the model advisor to detect client preferences, Ciano does not explicitly disclose, where the entity is a bill, that the entity identifier identifies at least one of a name and an address associated with the entity (bill). In the same field of invention, Punera discloses identifying a name and address within a text-based communication (P 0079) determining a match between a contact based on name and/or address (P 0086). Therefore, considering the teachings of Ciano, Blackhurst, Punera and Allen, one having ordinary skill in the art before the effective filing date of the invention would have been motivated to combine the at least one requisite entity identifier identifies at least one of a name and an address with the teachings of Ciano, Blackhurst, Punera and Allen with the motivation to more accurately derive the intent of the entity (bill) by identifying the customer and a profile of the customer as disclosed in Ciano (P 0056).
Ciano does not disclose (iv) determine, via a validate data process, the data is missing or invalid in the request; (v) receive missing data or valid data in response to the determination that data is missing or invalid in the request, as disclosed in the claims. However, Blackhurst discloses the system determines if date or time information is missing or cannot be used to match a specific piece of transaction data to a specific piece of e-receipt data (P 0039) the original value of the insurance claim may be based on outdated data, incorrect data, or faulty calculations (P 0047). Furthermore, Punera discloses update existing contacts with new contact data, add missing contact details (P 0029). Therefore, considering the teachings of Ciano, Blackhurst, Punera and Allen, one having ordinary skill in the art before the effective filing date of the invention would have been motivated to combine (iv) determine, via a validate data process, the data is missing or invalid in the request; (v) receive missing data or valid data in response to the determination that data is missing or invalid in the request with the teachings of Ciano, Blackhurst, Punera and Allen with the motivation to more accurately derive the intent of the entity (bill) by identifying the customer and a profile of the customer as disclosed in Ciano (P 0056).
Ciano does not disclose wherein the robotic automation platform processes the request by automatically determining additional information received after the request and associated with the request, as disclosed in the claims. However, Ciano discloses based upon information received consequent to the interaction between the conversational agent learning model and the plurality of clients, the model advisor identifies at least one deficiency in the conversational agent learning model, including conversation content and/or general information associated with one or more of the plurality of clients, the model advisor may updates the conversational agent learning model to address the at least one deficiency by adding new information to existing categories within the model structure or by determining one or more new categories and may embed such new categories into the model structure (P 0051). Punera discloses an electronic text communication is analyzed to determine if a signature is present in the body of the communication, if a signature is present, then machine learning is applied to the content of the communication (P 0079) and contact information is extracted (P 0080) processing logic determines a confidence level associated with the contact details extracted from the electronic communication during the machine learning model analysis relating to the machine learning model analysis results, and whether the specifics of a contact detail can be identified (i.e., a contact detail is likely to be the contact's email address, title, company, etc.) (P 0081). That is, while both Ciano and Punera disclose analyzing and identifying information in a communication, the claim requires that additional information is automatically determined that is received after the request and associated with the request. Ciano discloses limitations that are close to the amended limitations, but does not explicitly disclose that information received consequent to the interaction between the conversational agent learning model and the plurality of clients is additional information. Blackhurst discloses once the purchase transaction data has been extracted, additional data may be further enriched with additional and/or updated information associated with products or services within the data (P 0066). Therefore, considering the teachings of Ciano, Blackhurst, Punera and Allen, one having ordinary skill in the art before the effective filing date of the invention would have been motivated to combine wherein the robotic automation platform processes the request by automatically determining additional information received after the request and associated with the request with the teachings of Ciano, Blackhurst, Punera and Allen with the motivation to more accurately identify and gather correct data to submit insurance claims to assist with complex and detailed agreements that can be difficult for policy holders to navigate (Blackhurst: P 0001).
Ciano further discloses an analytics platform to identify type I and type II errors in classification, and generate final decision processing logs, the model advisor may identify at least one deficiency in the conversational agent learning model which may include, e.g., subject matter not previously encountered and/or subject matter for which there is relatively little or no information to track or access in the context of the model, the model advisor may update the conversational agent learning model to address the at least one deficiency (P. 0051) entities are object classes or a data types that enable selection of an action based upon the plurality of solutions to address one or more intents, for example one intent is a bill payment (P 0054) via conventional methods or via crypto currency (P 0050) a new category may be derived from the conversation based on intent and a new entity (P 0055) a conversation content is classified based intent and an entity derived from the conversation (P 0056). Ciano does not disclose maintain statistics on historic decisions, as disclosed in the claims. However, in the same field of invention, Cauchois discloses enhancing the virtual assistant may include clustering and other statistical techniques (P. 0052). Therefore, considering the teachings of Ciano, Blackhurst, Punera, Allen and Cauchois, one having ordinary skill in the art before the effective filing date of the invention would have been motivated to combine maintain statistics on historic decisions with the teachings of Ciano, Blackhurst, Punera and Allen with the motivation to provide a more effective method for accurately tracking the deficiencies in the conversational agent learning model.
Ciano does not explicitly disclose an intent classifier adapted to output a machine learning confidence value, as disclosed in the claims. However, Ciano discloses the model advisor may identify deficiencies in the learning model and may update the learning model to address the at least one deficiency by applying one or more machine learning techniques to obtain new data or categories form data repositories or external data (P 0051). However, in determining data and/or categories to resolve the deficiencies of the learning model, Ciano does not disclose outputting a machine learning confidence factor for the new data or categories. Punera discloses a system for automatic and intelligent relationship management, including automatic suggestion generation that captures, analyses, and reports communications between a single user, or multiple users, within an organization, and various contacts outside of the organization for a specific purpose (e.g., sales, business, recruiting, funding, etc.) (P 0022) and automatically generates suggestions for users and collaborators in response to communications from a process, suggesting team members to add as new collaborators to a process, suggesting external contacts to add as relationships to a process, etc, and intelligently interprets the relationship management data (P 0023) a confidence level associated with the contact details extracted from the electronic communication during the machine learning model analysis, wherein the confidence level relates to the machine learning model analysis results, and whether the specifics of a contact detail can be identified (i.e., a contact detail is likely to be the contact's email address, title, company, etc.) (P 0081). Punera identifies relationship data in messages exchanged between users and, and if the data is interpreted to indicate a relationship, e.g. sales, business, recruiting, funding, etc., based on a machine learning analysis, then a relationship is formed. Therefore, considering the teachings of Ciano, Blackhurst, Punera, Allen and Cauchois, one having ordinary skill in the art before the effective filing date of the invention would have been motivated to combine an intent classifier adapted to output a machine learning confidence value with the teachings of Ciano, Blackhurst, Punera, Allen and Cauchois with the motivation to ensure that suggestions generated for a user are more likely to be of use to the user can based on the user profile (Punera: P 0095).
Ciano does not explicitly disclose … based on an analysis of the intent including the intent type and the at least one of the name and the address of the extracted requisite entity identifier, wherein both the intent and the extracted requisite entity identifier includes both freeform data and pre-defined text, as disclosed in the claims. However, Ciano discloses the model advisor may identify deficiencies in the learning model and may update the learning model to address the at least one deficiency by applying one or more machine learning techniques to obtain new data or categories form data repositories or external data (P 0051). However, in determining data and/or categories to resolve the deficiencies of the learning model, Ciano does not disclose outputting a machine learning confidence factor for the new data or categories. In the same field of invention, Punera discloses a confidence level associated with the contact details extracted from the electronic communication during the machine learning model analysis, wherein the confidence level relates to the machine learning model analysis results, and whether the specifics of a contact detail can be identified (i.e., a contact detail is likely to be the contact's email address, title, company, etc.) (P 0081) processing logic can infer that a co-worker is likely to be added as a collaborator of the user within a new process, and therefore automatically generates the collaborator addition suggestion (P 0092). It’s clear that not only does Punera generate a confidence level for a determined relationship, but also to the results of the machine learning analysis, i.e. did the machine learning correctly interpret the data in the communication. Therefore, a combination of both the confidence level of the details extracted from the electronic communication and a confidence level of the machine learning model analysis itself is used to interpret the relationship data to infer a relationship. Furthermore, Allen discloses intents are classified and ranked (P 0055). Therefore, considering the teachings of Ciano, Blackhurst, Punera, Allen and Cauchois, one having ordinary skill in the art before the effective filing date of the invention would have been motivated to combine … based on an analysis of the intent including the intent type and the at least one of the name and the address of the extracted requisite entity identifier, wherein both the intent and the extracted requisite entity identifier includes both freeform data and pre-defined text with the teachings of Ciano, Blackhurst, Punera, Allen and Cauchois with the motivation to ensure that suggestions generated for a user are more likely to be of use to the user can based on the user profile (Punera: P 0095).
Claim 2. Ciano, Blackhurst, Punera, Allen and Cauchois disclose the system of claim 1, and Ciano further discloses said automatic determination of the intent associated with the request comprises at least one of: (i) communicating with a classification platform service, and (ii) analyzing the electronic record to determine the intent associated with the request, wherein the model advisor may classify conversation content received from one or more of the plurality of clients and classifying the intent may entail classifying at least one intent derived from the conversation content by the model advisor (P. 0056).
Claim 3. Ciano, Blackhurst, Punera, Allen and Cauchois disclose the system of claim 1, and Ciano further discloses the electronic record is associated with at least one of: (i) an email message, (ii) a text message, (iii) a voice channel request that has been translated into text, (iv) a fax, (v) a video request, (vi) a web site submission, (vii) a mobile application, and (viii) a messaging application, the conversational interaction between a plurality of clients and respective instances of the conversational agent learning model in accordance with the various embodiments may include unstructured content (e.g., audiovisual aspects) as well as structured textual content (P. 0022) the at least one model instance includes at least one conversational interface accessible to at least one of the plurality of clients (P. 0056).
Claim 4. Ciano, Blackhurst, Punera, Allen and Cauchois disclose the system of claim 1, and Ciano further discloses the robotic automation platform processes the request by automatically transmitting a complete response to the request, one or more dialog portions may be presented to one or more of the plurality of clients in the context of the interaction between the conversational agent learning model and the plurality of clients including "Your bill has been processed" and may further involve provision of information regarding the derived entity, and may further provide a dialog portion including information regarding the bill, for instance: "The top portion of the bill reveals a payment balance of $50.00" the model advisor may update the at least one intent derived from the conversation content based upon the at least one modification (P. 0058).
Claim 5. Ciano, Blackhurst, Punera, Allen and Cauchois disclose the system of claim 1, and Ciano further discloses the robotic automation platform processes the request by automatically pre-populating data in a template provided to a human knowledge worker, upon determining that Client A is a teenager, the model advisor may present to Client A dialog portions consistent with teenage speech, e.g., "That is cool." Accordingly, the model advisor may increase client familiarity and comfort in the context of model interaction (P. 0049) the model advisor may receive input from a domain administrator (e.g., a domain expert) to refine the interaction between the conversational agent learning model and the plurality of clients with respect to one or more model instances, and may add or update categories within the model structure to facilitate access to new sources for purposes of populating the model (P. 0052).
Claim 6. Canceled.
Claim 7. Ciano, Blackhurst, Punera, Allen and Cauchois disclose the system of claim 1, and Ciano further discloses said determining is associated with both: (i) asking an originator of the request for the additional information, once a unified text-based artifact is determined for the unstructured content in the request, the model adviser completes a semantic analysis on the text-based artifact to determine an intent, and then creates a conversation output asking the requestor to confirm the determined intent (P 0062), and (ii) receiving the additional information from a third-party device, the conversation agent learning model is integrated with social media or by otherwise accessing social networks or applications (e.g., by communicating with external sources), the model advisor may dynamically adjust the model based upon new social/cultural tastes by adjusting dialog portions presented in a model instance to a client among the plurality of clients based upon one or more language trends, one or more cultural trends, and/or one or more behavioral trends manifested in social networks or social applications (P. 0049).
Claim 8. Ciano, Blackhurst, Punera, Allen and Cauchois disclose the system of claim 1, and Ciano further discloses an intent library storing, for each of a plurality of customers of an enterprise, historic customer request and intent definition information, the model advisor may access profile data associated with the client from the client data repository and/or from external sources, which may provide social networking data associated with the client to be more capable of accurately deriving intents and/or entities from the conservation content involving the client, as the profile data may enable the model advisor to detect client preferences (P. 0056) the model advisor may update the at least one intent derived from the conversation content based upon the at least one modification (P. 0058).
Claim 9. Ciano, Blackhurst, Punera, Allen and Cauchois disclose the system of claim 1, and Ciano further discloses an artificial intelligence customer service terminal to facilitate interactions with a customer, the client computing system includes a model instance comprising a conversational agent instance, a bot instance, or a chatbot instance) for interacting with one or more other aspects of a conversational agent learning model constructed and configured via model server system (P. 0036).
Claim 10. Ciano, Blackhurst, Punera, Allen and Cauchois disclose the system of claim 1, and Ciano further discloses a parsing platform to decompose large blob text into smaller sentence structures for finite intent understanding and handling multiple intent requests within one blob of text, the model advisor may perform data extraction of user content, e.g. user comments, from a social network or social application such that the content may be transmitted to model advisor for analysis and/or processing (P. 0041) and from a set of client data repositories including knowledge base repositories to identify a number of issues (P. 0042) where the model advisor may derive a plurality of intents based upon the plurality of issues and determine the plurality of dialog portions based upon the plurality of intents (P. 0054).
Claim 11. (Canceled).
Claim 12. Ciano, Blackhurst, Punera, Allen and Cauchois disclose the system of claim 1, and Ciano further discloses the entity extraction platform accesses at least one of: (i) e-forms, (ii) scanned forms, and (iii) online web forms for data extraction on intent type element extractions, the conversational interaction between a plurality of clients and respective instances of the conversational agent learning model in accordance with the various embodiments may include unstructured content (e.g., audiovisual aspects) as well as structured textual content (P. 0022) the model advisor may complete such incorporations consequent to identifying potential new model aspects via interaction with both structured channels (e.g., formal records) and unstructured channels (e.g., social media data) (P. 0055) the model advisor may determine whether the conversation content received via the at least one conversational interface from one or more of the plurality of clients includes unstructured content including unstructured documents or forms (P. 0061).
Claim 13. Ciano, Blackhurst, Punera, Allen and Cauchois disclose the system of claim 1, and Ciano further discloses the model advisor may receive input from a domain administrator (e.g., a domain expert) to refine the interaction between the conversational agent learning model and the plurality of clients with respect to one or more model instances, and may add or update categories within the model structure to facilitate access to new sources for purposes of populating the model (P. 0052) and Punera discloses various pre-learned machine learning models, such as one or more natural language analysis, decision tree, neural network, support vector machines, conditional random fields, unsupervised learning (e.g. clustering), etc. techniques can be deployed by the relationship analyzer (P 0028). It is clear that Punera does not restrict the type of machine learning model used. Therefore, considering the teachings of Ciano, Blackhurst, Punera, Allen and Cauchois, one having ordinary skill in the art before the effective filing date of the invention would have been motivated to combine a supervised learning platform for low machine learning confidence transactions to facilitate model training and ongoing systematic model enhancements with the teachings of Ciano, Blackhurst, Punera, Allen and Cauchois with the motivation to ensure that suggestions generated for a user are more likely to be of use to the user can based on the user profile (Punera: P 0095).
Claim 14. Ciano, Blackhurst, Punera, Allen and Cauchois disclose the system of claim 1, and Ciano further discloses updating the conversational agent learning model to address the at least one deficiency may include incorporating at least one new intent into the plurality of intents or incorporating at least one new entity into the plurality of entities (P. 0005) in an automated service management system including the model advisor (P. 0038) the model advisor may update the conversational agent learning model to address the at least one deficiency by incorporating at least one new intent into the plurality of intents or by incorporating at least one new entity into the plurality of entities (P. 0055) and Punera discloses various pre-learned machine learning models, such as one or more natural language analysis, decision tree, neural network, support vector machines, conditional random fields, unsupervised learning (e.g. clustering), etc. techniques can be deployed by the relationship analyzer (P 0028). It is clear that Punera does not restrict the type of machine learning model used. Therefore, considering the teachings of Ciano, Blackhurst, Punera, Allen and Cauchois, one having ordinary skill in the art before the effective filing date of the invention would have been motivated to combine an unsupervised learning platform for low machine learning confidence transactions to facilitate model training and ongoing systematic model enhancements to systematically leverage knowledge worker request processing with the teachings of Ciano, Blackhurst, Punera, Allen and Cauchois with the motivation to ensure that suggestions generated for a user are more likely to be of use to the user can based on the user profile (Punera: P 0095).
Claim 15. (Canceled).
Claim 16. Ciano, Blackhurst, Punera, Allen and Cauchois disclose the system of claim 1, but Ciano does not disclose the robotic automation platform interacts with downstream systems through Application Programming Interface (“API”) and/or front end Robotic Processing Automation (“RPA”) to transact customer requests, as disclosed in the claims. However, in the same field of invention, Cauchois discloses as a connector is developed for a particular live chat server, the connector may expose the same Application Programming Interface (API) to the virtual assistant server, such that a change comes at almost no additional cost and without the user noticing any change of the user interface in use (P. 0055). Therefore, considering the teachings of Ciano, Blackhurst, Punera, Allen and Cauchois, one having ordinary skill in the art before the effective filing date of the invention would have been motivated to combine the robotic automation platform interacts with downstream systems through Application Programming Interface (“API”) and/or front end Robotic Processing Automation (“RPA”) to transact customer requests with the teachings of Ciano, Blackhurst, Punera, Allen and Cauchois with the motivation to provide a low cost and easy to change method for modified and updated implementations to the conversational model advisor.
Claim 17. Ciano discloses a computer-implemented method for processing a request associated with an electronic record, comprising:
automatically determining, by an artificial intelligence orchestration platform, one or more intents of the electronic record associated with the request, retrieving the corpus of information includes identifying a plurality of issues and identifying a plurality of solutions respectively corresponding to the plurality of issues, identifying client content associated with the plurality of issues or the plurality of solutions and identifying any rankings associated with the plurality of solutions (P. 0005, 0053) the model advisor may derive a plurality of intents based upon the plurality of issues (P. 0054) wherein the model advisor may classify conversation content received from one or more of the plurality of clients and classifying the intent may entail classifying at least one intent derived from the conversation content by the model advisor (P. 0056);
identifying a type of transaction as