Researchers Submit Patent Application, “Self-Optimizing, Multi-channel, Cognitive Digital Insurance Rate Quoting, Comparison Shopping and Enrollment System and Method”, for Approval (USPTO 20230342860): Patent Application

2023 NOV 13 (NewsRx) -- By a News Reporter-Staff News Editor at Insurance Daily News -- From Washington, D.C., NewsRx journalists report that a patent application by the inventors COHEN, SETH (Lighthouse, FL, US); SILVESTRE, LUIS (Miramar, FL, US), filed on April 5, 2023, was made available online on October 26, 2023.

No assignee for this patent application has been made.

News editors obtained the following quote from the background information supplied by the inventors: “

“Field of the Invention

“The present invention relates to a performance internet marketing self-optimizing, anthropomorphic, artificial intelligence-based system and method to quote, compare and purchase personal lines and commercial lines property and casualty insurance or life and health insurance, or benefits products and services (hereinafter “benefits products”) from an insurance benefits provider, benefits company, healthcare discount plan provider, health care sharing ministry, or similar entities (hereinafter “benefits provider(s)”) through a cognitive virtual process via phone, mobile device, tablet, app, SMS, chat, iMessage, videoconference, or virtual reality (hereinafter “method(s) of communication”). For example, products can include personal lines policies such as auto, boat, motorcycle, and recreational vehicles, pet insurance, travel, accident, home, renters or apartments, condominium or townhouse, umbrella, inland marine, flood insurance; commercial lines policies such as business owner, abuse or molestation, fidelity, cybersecurity; and related products including, for example, life and annuity products.

“Description of the Related Art

“The property and casualty benefits product industry has changed and there is potential for disruption. Customers have easy online access to competitive rate information and options in the shopping and servicing of their policies. Often products are tied directly to other shopping events such as buying a home or automobile. For example, Carvana.com, an online used car and financing website, has changed the approach to the sale of used automobiles, including some limited forms of coverage on cars purchased from them. The conventional path to purchase insurance has transformed with the advent of technology, smartphones and digital assistants. Many benefits product purchases are made in conjunction with other purchases leveraging information and need at the point another product such as, equipment, property automobile or boat is purchased. Easy seamless ways of leveraging the point of sale information at these transactions and allowing the customer to comparison shop and have questions answered immediately through the media of their choice has become common in the marketplace.

“Consumers expect and demand convenience, speed, automation and simplicity. Any unnecessary additional effort, incremental steps or inconvenience that leads the consumers to abandon their purchase journey is defined as ‘friction’. Friction may occur offline or online, and includes subjecting consumers to unnecessarily wait, queue, click, type, tap, add information or fill-in forms. The consumers may also have to load, buffer, switch channels/windows or hold for service. It can hinder efficiency and impact the consumer experience, adversely affecting the consumer purchase journey.

“Consumers of products commonly use comparison shopping internet websites to research and compare available options. Products comparison shopping websites attempt to guide consumers through intricate product variations from multiple providers. A state of asymmetric information for products across the individual marketplace exists and potentially leads the consumer to adverse selection.

“Due to the nature of the chaotic environment, the individual marketplace is also an outlet for fraudulent products. In many cases, consumers seek the assistance of a licensed insurance agent, producer or broker (hereinafter “agent”) to select a product. It is common in the individual marketplace for a consumer to speak with an agent by phone to review available product options and complete the provider’s enrollment process. The agent typically enters the consumer’s underwriting details into a rate quoting calculator or web form to obtain rates, explains the differences between plan options from various benefits providers, and transcribes the enrollment details onto the provider’s application or enrollment portal on behalf of the consumer.

“In many cases, the agent is appointed to represent a limited number of providers thus limiting the consumer’s choices to only plans offered by those providers. Compensation arrangements for insurance agents also typically include commissions (a percentage of premium as paid by the provider) or another form of compensation (e.g., marketing fees). Agent bias, thus, is also potentially a significant factor in adverse selection.

“Conventional systems have been used for extracting user data to identify available insurance plans based on responses from the consumer. U.S. Pat. No. 9,495,700 describes the use of branded virtual characters on a server computer which are accessed by an employee at a remote client computer over a network in order to engage and educate users regarding alternate benefit plans. Throughout the process, the system employs a portfolio-driven approach to identify inter-dependencies among different benefit products and plans and prioritize and recommend “portfolio-level” benefit solutions. The employee can be automatically enrolled in the selected benefit plan. This system has the drawback of needing a computer interface to access the system over the internet and can be cumbersome for a person to navigate, especially the elderly or unsophisticated computer users.

“It is desirable to provide an unbiased, anthropomorphic, artificial intelligence-based telephonic system and method to identify the consumer’s available product options based on eligibility, assist in the comparison of different options, automatically enroll the consumer into selected plans available from various insurers, and self-optimize the online performance marketing campaign for referring the consumer to the system.”

As a supplement to the background information on this patent application, NewsRx correspondents also obtained the inventors’ summary information for this patent application: “The present invention relates to an anthropomorphic, artificial intelligence-based system to optimize real-time marketing spend using historical performance data and real time buying signals and to drive consumers to the action of quoting, comparing, and purchasing personal lines and commercial lines property and casualty insurance in real time. For example, products can include personal lines policies such as auto, boat, motorcycle, and recreational vehicles, pet insurance, travel, accident, home, renters or apartments, condominium or townhouse, umbrella, inland marine, flood insurance; commercial lines policies such as business owner, abuse or molestation, fidelity, cybersecurity; and related products including, for example, life and annuity products. The process of real time optimization can use an application programming interface (“API”) to relay critical performance metrics from a cognitive virtual assistant to online advertising platforms. The online advertising platforms can be search engines, such as for example, Google Adwords, Microsoft Bing, & Yahoo Gemini. Search engines are online advertising platforms that offer advertisers the ability to purchase an ad listing which prompts the consumer to visita website or call a phone number if the consumer types a specified search query. In addition, real-time optimization of referring search marketing campaigns occurs via API by relaying critical performance metrics from the cognitive virtual assistant to an online third-party search engine or lead aggregator bidding platforms. In one embodiment, the online bidding platform uses a ping post method to allow a lead seller to send partial information, the “ping”, of a lead to their network of lead buyers. The buyer then determines whether he or she would like to purchase the lead based on this information and obtains contact information from the seller via the “post.”

“Search engines often display multiple advertisers’ ads for the same search query and give full control to the advertiser over the search queries they choose to display their ads on. The ad position for each unique search query is determined primarily by a maximum cost-per-click (“CPC”) which dictates the highest amount that an advertiser is willing to spend for a click on a website advertisement or a phone call. In addition to specifying a max CPC for each keyword query, the advertiser is also given tools to increase or decrease their max CPC based upon demographics of the consumer, such as for example gender, age, household income, parental status, and location along with environmental factors, such as for example time of day and day of week. Every unique ad listing a search engine displays to a consumer takes the above factors into consideration to determine a final ad position.

“In the present invention, if the consumer chooses to click through via an ad listing of the advertiser, a click tracking identification (ID) is generated by the search engine which is unique to the consumer’s click. The click tracking ID can be linked back to the consumer whether they choose to complete an insurance application online or call in from the website and complete an application over the phone.

“In the present invention, the process of optimization can use an API from the cognitive virtual assistant to deliver a status of the application back to the search engine using a status tracking ID generated by the search engine. By delivering the performance data, such as for example successful applications, from the cognitive virtual assistant to the search engine, real time performance of each unique keyword query on an application level can be determined along with performance factors, such as gender, age, household income, parental status, and location, along with environmental factors, such as time of day and day of the week. The system can make real time, automatic optimization using the search engines’ API to control max CPC and demographic-specific bids based upon current application performance to, in turn, optimize marketing spend and drive future applications automatically.

“In the present invention, the process of optimization can use an application programming interface (API) from the cognitive virtual assistant to deliver a status of the application back to the search engine using a status tracking identification (ID) generated by the search engine. By delivering the performance data, such as for example, successful applications from the cognitive virtual assistant to the search engine, real time performance of each unique keyword query on an application level can be determined along with performance factors, such as for example, for a VIN number, license plate, state, and number of miles, and for homeowners products the address, zip code, type of dwelling, age of home, type of construction, and square footage. In similar fashion, the appropriate metadata can be collected for pet insurance, life and annuity products, mobile cellular, live TV, and streaming services. In all cases, environmental factors can be collected, such as, for example, time of day and day of week.

“The user (a buyer, consumer, or system herein referred to as “user”) initiates a session via various methods of communication. In one embodiment, a session is initiated by a user using one of the selected methods of communication. For voice sessions, audio interfaces of the system allow multi-language bidirectional speech-based conversations between the cognitive virtual assistant and the user. As the user speaks to the cognitive assistant, the process of automated speech recognition (ASR) digitally converts audible speech into transcribed text.

“Through natural language processing, including sentiment and tone analysis, the system evaluates the meaning and context of the transcribed text and adjusts the language and tone of the cognitive virtual assistant’s responses accordingly to accommodate the user. Context-switching capability of the system allows the user to interrupt and restart any embedded process while retaining the user’s information. The system can include anepisodic memory to allow the cognitive virtual assistant to recall details from previous segments of the current conversation or previous conversations altogether. The cognitive virtual assistant can be implemented in a conversational manner for receiving information from a user and generating responses using cognitive learning abilities during the conversation. The cognitive learning abilities of the cognitive virtual assistant can also include analytic memory for understanding trend of data, affective memory for understanding emotion and deep back projection networks (“DBPN”) for learning process flows via empirical learning. The cognitive virtual assistant can be considered to be a “trusted” virtual producer for the user as compared to a human motivated to make any sale.

“Once a session is initiated, the user is introduced to the cognitive virtual assistant trained to generate rate quotes, compare available options, and enroll the consumer into selected plans available from various insurers. The cognitive virtual assistant follows a roadmap embedded in a semantic memory for guiding the user through the session. It allows the cognitive virtual assistant to engage in conversation with the user while restricting the conversation to insurance rate quoting, comparison shopping and enrollment topics and processes. In some embodiments, the user can be referred to the cognitive virtual assistant by an internet search engine, social media ad, or other internet-based performance marketing platforms. The cognitive virtual assistant captures the user’s referral metadata including referring website URL, search phrase or ad clicked, geolocation, internet protocol (“IP”) address, and other values for analytics and optimization.

“In one embodiment, once a session is initiated, the user is introduced to a cognitive virtual assistant trained to generate rate quotes, compare available products options, and enroll the consumer into selected plans available from various insurers. The cognitive virtual assistant follows a roadmap embedded in the system’s semantic memory for guiding the user through the session. It allows the virtual assistant to engage in conversation with the user while restricting the conversation to insurance rate quoting, comparison shopping and enrollment topics and processes.

“In one embodiment, once a session is initiated, the user is introduced to the cognitive virtual assistant trained to generate rate quotes, prices and packages, compare available coverage options, and enroll the consumer into selected plans available from various benefits providers.

“The cognitive virtual assistant follows a roadmap embedded in a semantic memory for guiding the user through the session. It allows the cognitive virtual assistant to engage in conversation with the user while restricting the conversation to benefits product options, pricing, comparison shopping and enrollment topics and processes. In some embodiments, the user can be referred to the cognitive virtual assistant by an internet search engine, social media ad, or other internet-based performance marketing platforms. The system captures the user’s referral metadata including referring website URL, search phrase or ad clicked, geolocation, Internet Protocol (IP) address, and other values for analytics and optimization. Once a session is initiated, the user is introduced to a cognitive virtual assistant trained to generate rate quotes, compare available coverage options, and enroll the consumer into selected plans available from various providers.”

There is additional summary information. Please visit full patent to read further.”

The claims supplied by the inventors are:

“1. A computer-implemented method comprising the steps of: a. initiating an interface to a cognitive virtual assistant and receiving by the cognitive virtual assistant, audible speech from a user interface during a conversation of a user with the cognitive virtual assistant, digitally converting, by the cognitive virtual assistant, the audible speech into transcribed text, generating data and responses by interpreting the transcribed text with artificial intelligence using cognitive learning abilities at a processor of the computer, forwarding the responses to the user interface during the conversation; b. prompting, by the cognitive virtual assistant, the user during the conversation for information and a description of one or more products which are of interest to the user for purchase or enrollment, the one or more products being selected from personal lines insurance, commercial lines insurance, property insurance, casualty insurance, benefits products, benefits services and residential services; c. receiving, by the cognitive virtual assistant via one or more of provider application interfaces, data from an application server of one or more providers for the one or more products; d. identifying, by the cognitive virtual assistant, eligibility of the user for purchase or enrollment in the one or more products which are of interest by the user based on eligibility determined from the data collected from the user by the cognitive virtual assistant during the conversation and the data collected from the one or more provider application interfaces e. presenting, by the cognitive virtual assistant, the one or more products identified in step d. to the user during the conversation; f. completing, by the cognitive virtual assistant, application requirements of the one or more products; and g. communicating, by the cognitive virtual assistant, the application requirements to a respective one or more providers.

“2. The method of claim 1 wherein the step of prompting a user for information includes prompting the user to answer one or more pre-qualifying questions and the identification of the relevant products being determined from eligibility of the user based on answers to the one or more pre-qualifying questions.

“3. The method of claim 1 wherein the step of initiating an interface to a cognitive virtual assistant comprises matching one or more user search terms entered in an online advertising platform.

“4. The method of claim 1 further comprising adjusting parameters of the virtual assistant to accommodate the user.

“5. The method of claim 1 further comprising the step of receiving enrolled policy information after enrollment and presenting the enrolled policy information to the user with the cognitive virtual assistant.

“6. The method of claim 1 wherein the interface to the cognitive virtual assistant is via audio, phone, mobile device, tablet, app, SMS, chat, iMessage, videoconference, virtual reality game or game system.

“7. The method of claim 1 wherein the step of initiating an interface to a cognitive virtual assistant includes displaying a link to connect with the cognitive virtual assistant.

“8. The method of claim 3 further comprising the steps of collecting information from the online advertising platform and the cognitive virtual assistant during one or more of steps a. through e and returning the collected information to the online advertising platform.

“9. The method of claim 8 further comprising the step of improving the online advertising platform by improving ad position for each unique search query of the one or more search terms entered in the online advertising platform to increase or decrease a max cost-per click (CPC) based upon demographics of the user determined by using the collected information from the online advertising platform.

“10. The method of claim 9 wherein the collected information includes tracking user click information using the online advertising platform and the step of improving the online advertising platform uses the tracking user click information to generate a click tracking identification (ID) by the online advertising platform during step a. and the click tracking identification (ID) being used during steps b. through f.

“11. The method of claim 9 wherein the step of improving the online advertising platform process provides a status of the step of completing application requirements of the identified one or more relevant products to the online advertising platform.

“12. A computer implemented system comprising: a user interface communicating with a cognitive virtual assistant, and the cognitive virtual assistant receives audio from the user interface which is interpreted with artificial intelligence at a processor to generate data and responses, the responses being forwarded to the user interface; a system server communicating over a network to the cognitive virtual assistant; and one or more products application interfaces communicating with the cognitive virtual assistant, wherein the cognitive virtual assistant prompts a user for information and a description of one or more products which are of interest to the user for purchase, the one or more products being selected from personal lines insurance, commercial lines insurance, property insurance, casualty insurance, benefits products, benefits services and residential services, the system server identifies relevant products based on eligibility determined from data collected from the user by the cognitive virtual assistant, presents the relevant products to the user with the cognitive virtual assistant, identifies one or more relevant products based on eligibility determined from data collected from the user by the cognitive virtual assistant and data collected from a product application interface, presents the identified one or more relevant products to the user using the cognitive virtual assistant; identifies application requirements of the identified one or more relevant products which are selected by the user and communicates the application requirements to a respective one or more providers for enrollment of the user.

“13. The system of claim 12 wherein the interface to the cognitive virtual assistant is via audio, mobile device, tablet, app, wearable computer device, SMS, chat, iMessage, videoconference, virtual reality, game or game system.

“14. The system of claim 12 wherein the cognitive virtual assistant has cognitive learning abilities.

“15. The system of claim 12 wherein the cognitive virtual assistant is determined based on a natural language parser.

“16. The system of claim 12 wherein parameters of the cognitive virtual assistant are adjusted to accommodate the user.

“17. The system of claim 12 wherein the interface to a cognitive virtual assistant matches one or more user search terms entered in an online advertising platform and displays a link to connect with the cognitive virtual assistant.

“18. The system of claim 17 wherein the system collects information from the online advertising platform and the cognitive virtual assistant and returns the collected information to the online advertising platform for improvement of the online advertising platform.

“19. The system of claim 18 wherein the improvement includes optimizing ad position for each unique search query of the one or more search terms entered in the online advertising platform of to increase or decrease a max cost-per click (CPC) based upon demographics of the user.

“20. The system of claim 12 wherein a API on the server or the cognitive virtual assistant is activated to establish a connection with a API of the product application interface.

“21. The system of claim 12 wherein the data collected from the user includes one or more of VIN number, license plate, state, number of miles driven, address, zip code, type of dwelling, age of home, type of construction, and square footage.

“22. A non-transitory computer-readable medium having stored thereon a plurality of sequences of instructions, said instructions including sequences of instructions which, when executed by at least one processor, cause said processor to receive audio from a user interface in communication with a cognitive virtual assistant, and the cognitive virtual assistant receives the audio from the user interface, the audio is interpreted with artificial intelligence at a processor to generate data and responses, the responses being forwarded to the user interface; the cognitive virtual assistant communicating with a providers interface, wherein the cognitive virtual assistant prompts a user for information and a description of one or more products which are of interest to the user for purchase, the one or more products being selected from personal lines insurance, commercial lines insurance, property insurance, casualty insurance, benefits products, benefits services and residential services, the processor identifies relevant products based on eligibility determined from data collected from the user by the cognitive virtual assistant, presents the relevant products to the user with the cognitive virtual assistant, identifies one or more relevant products based on eligibility determined from data collected from the user by the cognitive virtual assistant and data collected from a product application interface, presents the identified one or more relevant products to the user using the cognitive virtual assistant; identifies application requirements of the identified one or more relevant products which are selected by the user and communicates the application requirements to a respective one or more providers for enrollment of the user.”

There are additional claims. Please visit full patent to read further.

For additional information on this patent application, see: COHEN, SETH; SILVESTRE, LUIS. Self-Optimizing, Multi-channel, Cognitive Digital Insurance Rate Quoting, Comparison Shopping and Enrollment System and Method. U.S. Patent Application Number 20230342860, filed April 5, 2023 and posted October 26, 2023. Patent URL (for desktop use only): https://ppubs.uspto.gov/pubwebapp/external.html?q=(20230342860)&db=US-PGPUB&type=ids

(Our reports deliver fact-based news of research and discoveries from around the world.)