2022 JUL 21 (NewsRx) — By a News Reporter-Staff News Editor at Insurance Daily News — From Washington, D.C., NewsRx journalists report {that a} patent software by the inventors Frankowiak, Sara (Bloomington, IL, US); Isaacs, Craig Dean (Bloomington, IL, US); McCarty, Jeff (Bloomington, IL, US); Roll, Leif Agerholm (Bloomington, IL, US); Taylor, Kelli (Bloomington, IL, US), filed on June 24, 2020, was made accessible on-line on July 7, 2022.
No assignee for this patent software has been made.
News editors obtained the next quote from the background info provided by the inventors: “Individuals who search insurance coverage protection and are delicate to pricing and product options (e.g., protection sorts and/or limits, deductibles, and so on.), or “frequent customers,” typically expend appreciable effort and time find insurance coverage suppliers that greatest meet their wants. Conventionally, a frequent shopper finds an insurance coverage supplier by the use of an agent/dealer, an aggregator, a comparability site, basic internet shopping, and so on. Once the frequent shopper obtains an insurance coverage coverage from the specified supplier, the frequent shopper is often tied to that supplier, and to the speed and product options of the coverage provided by the supplier, till and until she or he proactively outlets round for a brand new supplier providing a coverage with a greater fee and/or product options. For instance, a frequent shopper may resolve to look into the choices of different insurance coverage suppliers when the frequent shopper’s present coverage is up for renewal. Thus, a frequent shopper usually should both spend effort and time wanting for a better-priced insurance coverage providing on a recurring foundation (e.g., as soon as each six months or yearly), or just renew his or her present coverage no matter whether or not that coverage supplies the very best fee and/or product options. Conventional agency-based insurance coverage fashions could not suffice to satisfy a frequent shopper’s wants, because of the perceived extra value related to having an agent.”
As a complement to the background info on this patent software, NewsRx correspondents additionally obtained the inventors’ abstract info for this patent software: “The current embodiments could, inter alia, routinely present frequent customers with insurance coverage insurance policies that supply superior charges and/or product options on a seamless foundation (e.g., throughout a number of coverage phrases), thereby decreasing or eliminating the time and/or effort that frequent customers should spend researching the choices of various insurance coverage suppliers, in addition to offering frequent customers with insurance coverage insurance policies which have decrease value and/or are extra reflective of a threat rating, traits, and/or preferences of the frequent shopper as they alter over time. The phrases “frequent shopper,” “shopper,” and “buyer” are utilized interchangeably herein, and usually discuss with an individual who’s an insured celebration or a possible insured celebration, no matter how regularly that particular person the truth is wish to store for insurance coverage protection or has shopped for insurance coverage protection prior to now. A frequent shopper could also be represented by himself or herself, or could also be represented by an agent (e.g., by a partner, an individual who has energy of lawyer for the frequent shopper, an administrative assistant, and so on.).
“An middleman entity could act on behalf of frequent customers and/or their brokers (i.e., with the consent of the frequent customers and/or brokers) to search out coverage charges and/or different options that greatest meet the frequent customers’ insurance coverage necessities and/or preferences. Based upon an evaluation of particular person frequent shopper traits and/or insurance coverage preferences, every particular person frequent shopper could also be grouped with different insurance coverage frequent customers which have the identical or comparable traits and/or insurance coverage preferences. These “affinity groupings” (or “affinity teams”) could also be primarily based upon demographic info for the frequent customers (e.g., gender, start date, and so on.), details about property of the frequent customers (e.g., a make, mannequin and 12 months of an car, and so on.), declare and/or accident historical past of the frequent customers, threat (or lack thereof) traits of the frequent customers, insurance coverage declare expectations of the frequent customers, insurance coverage firm scores, the content material and/or availability of telematics knowledge obtained from autos and/or cellular units of the frequent customers, driving behaviors of the frequent customers, and so on.
“The proper to offer insurance coverage protection for the affinity groupings (both on a per-member foundation or to every affinity group as a complete) could also be provided for sale to varied insurance coverage suppliers, similar to by way of a web-based public sale. In some embodiments, different entities might also take part within the on-line public sale. For instance, reinsurers and/or entities that handle funding funds (e.g., hedge fund corporations searching for arbitrage alternatives) could take part within the on-line public sale, e.g., if these entities have agreements with insurance coverage suppliers are licensed to jot down insurance coverage and may legally service claims, and so on., for the members of the affinity group.
“Once a profitable bid is accepted, any present insurance coverage insurance policies of the frequent customers affiliated with the auctioned group could (or could not) be up to date to mirror new insurance coverage coverage phrases or parameters (e.g., premiums, charges, and so on.), reductions, refunds, and so on. In some circumstances, new insurance coverage insurance policies could also be offered to a number of frequent customers (similar to when a frequent shopper is an insurance coverage applicant, or when an present insurance coverage coverage is canceled and a brand new coverage is issued in its stead). The affinity teams could also be up to date (and/or new affinity teams could also be created) over time as new or more moderen frequent shopper attribute knowledge and/or desire info is collected and/or up to date. The insurance coverage insurance policies related to the up to date (or new) affinity teams could then be re-auctioned (or auctioned).
“Additionally or alternatively, insurance coverage suppliers could mitigate the dangers related to insurance coverage insurance policies which are already in impact (or will quickly be in impact), by grouping/segmenting these insurance policies and auctioning the chance to reinsure these coverage teams to different entities (e.g., reinsurers). The grouping for these auctions could correspond to the affinity teams mentioned above (e.g., with a selected group of insurance policies that’s being auctioned consisting of the insurance policies of all members of a selected affinity group), or could also be an impartial/subsequent grouping of the insurance coverage insurance policies, for instance.
“Various machine studying applied sciences described herein could improve the effectivity of any or all the grouping and/or auctioning methods mentioned above, in some embodiments. For instance, machine studying fashions could also be used to judge dangers (e.g., decide threat scores/classifications and/or infer risk-related traits) related to totally different frequent customers for a selected sort of insurance coverage (e.g., dangers of vehicular accidents and/or theft for auto insurance coverage), previous to segmenting these frequent customers into totally different affinity teams primarily based upon these dangers. Machine studying might also be used to outline and/or replace/refine standards for totally different affinity teams, e.g., through the use of regression fashions to find out which groupings of frequent customers have traditionally attracted extra curiosity (e.g., extra frequent and/or larger bids) from insurance coverage suppliers, and/or have traditionally had extra steady group membership, and so on.
“Machine studying methods might also, or as a substitute, be used to arrange an public sale, and/or to facilitate the public sale itself. For instance, machine studying fashions could assist decide which insurance coverage suppliers to ask to take part in an public sale, by predicting which suppliers usually tend to be inquisitive about (e.g., extra more likely to submit bids for) offering insurance coverage protection to a selected affinity group, and/or could decide an acceptable beginning bid (e.g., “reserve”) quantity, and so on.
“Many or all aspects of the public sale course of, and/or different procedures previous to and/or after the public sale course of, could also be automated. For instance, communications with public sale members (e.g., insurance coverage suppliers and/or reinsurers, and so on.), and/or communications with customers that happen earlier than, throughout, and/or after the public sale course of, could also be automated. For instance, notifying insurance coverage suppliers and/or reinsurers relating to an upcoming public sale, speaking bid quantities amongst suppliers and/or reinsurers throughout an public sale, notifying public sale winners, corresponding with customers relating to insurance coverage supplier placements, billings, sending insurance coverage playing cards, and so on., and/or different communications could also be automated. In some embodiments, information related to customers, insurance coverage suppliers (and/or reinsurers or different public sale members), affinity teams, and/or auctions could also be securely saved using blockchain methods and/or methods.
“In one side, a computer-implemented technique includes: (1) dividing, by a number of processors, a plurality of customers into a number of affinity teams primarily based upon a number of traits and a number of preferences of the plurality of customers, a minimum of partly by analyzing the a number of traits and/or the a number of preferences of the plurality of customers utilizing a machine studying mannequin; (2) auctioning, by the a number of processors and by way of a communications community, a possibility to offer insurance coverage for a number of of the a number of affinity teams; (3) receiving, by the a number of processors and by way of the communications community, a number of bids for buy and/or affords of insurance coverage for the a number of of the a number of affinity teams; (4) accepting, by the a number of processors, a profitable bid of the a number of bids; and/or (5) inflicting, by the a number of processors, particular person insurance coverage insurance policies or a gaggle insurance coverage coverage to be offered to or up to date for customers related to a selected affinity group similar to the profitable bid, thereby offering decrease value insurance coverage and/or insurance coverage that’s extra reflective of precise threat, or lack thereof, to the customers related to the actual affinity group.
“In one other side, a system includes a persistent reminiscence storing a shopper profile database, a communication interface configured to speak with distant units by way of a communications community, a number of processors, and/or a number of non-transitory, computer-readable media storing directions. The directions, when executed by the a number of processors, trigger the system to: (1) divide a plurality of customers into a number of affinity teams primarily based upon a number of traits and a number of preferences of the plurality of customers, a minimum of partly by analyzing the a number of traits and/or the a number of preferences of the plurality of customers utilizing a machine studying mannequin; (2) public sale, by way of the communication interface and the communications community, a possibility to offer insurance coverage for a number of of the a number of affinity teams; (3) obtain, by way of the communication interface and the communications community, a number of bids for buy and/or affords of insurance coverage for the a number of of the a number of affinity teams; (4) settle for a profitable bid of the a number of bids; and/or (5) trigger particular person insurance coverage insurance policies or a gaggle insurance coverage coverage to be offered to or up to date for customers related to a selected affinity group similar to the profitable bid, thereby offering decrease value insurance coverage and/or insurance coverage that’s extra reflective of precise threat, or lack thereof, to the customers related to the actual affinity group.
“Advantages will grow to be extra obvious to these expert within the artwork from the next description of the popular embodiments which have been proven and described by the use of illustration. As can be realized, the current embodiments could also be able to different and totally different embodiments, and their particulars are able to modification in varied respects. Accordingly, the drawings and outline are to be considered illustrative in nature and never as restrictive.”
There is extra abstract info. Please go to full patent to learn additional.”
The claims provided by the inventors are:
“1. A pc-implemented technique comprising: dividing, by a number of processors, a plurality of customers into a number of affinity teams primarily based a minimum of upon a number of traits and a number of preferences of the plurality of customers, a minimum of partly by analyzing the a number of traits and/or the a number of preferences of the plurality of customers utilizing a machine studying mannequin; auctioning, by the a number of processors and by way of a communications community, a possibility to offer insurance coverage for a number of of the a number of affinity teams; receiving, by the a number of processors and by way of the communications community, a number of bids for buy and/or affords of insurance coverage for the a number of of the a number of affinity teams; accepting, by the a number of processors, a profitable bid of the a number of bids; and inflicting, by the a number of processors, particular person insurance coverage insurance policies or a gaggle insurance coverage coverage to be offered to or up to date for customers related to a selected affinity group similar to the profitable bid, thereby offering decrease value insurance coverage and/or insurance coverage that’s extra reflective of precise threat, or lack thereof, for the customers related to the actual affinity group.
“2. The computer-implemented technique of declare 1, whereby dividing the plurality of customers into a number of affinity teams consists of: figuring out threat scores for the plurality of customers by analyzing the a number of traits of the plurality of customers utilizing the machine studying mannequin; and dividing the plurality of customers into the a number of affinity teams primarily based a minimum of upon the danger scores and the a number of preferences of the plurality of customers.
“3. The computer-implemented technique of declare 1, whereby dividing the plurality of customers into a number of affinity teams consists of: figuring out desire classifications for the plurality of customers by analyzing the a number of preferences of the plurality of customers utilizing the machine studying mannequin; and dividing the plurality of customers into the a number of affinity teams primarily based a minimum of upon the desire classifications and the a number of traits of the plurality of customers.
“4. The computer-implemented technique of declare 1, whereby dividing the plurality of customers into a number of affinity teams consists of: figuring out classifications for the plurality of customers by analyzing the a number of traits and the a number of preferences of the plurality of customers utilizing the machine studying mannequin; and dividing the plurality of customers into the a number of affinity teams primarily based a minimum of upon the classifications.
“5. The computer-implemented technique of declare 1, whereby dividing the plurality of customers into a number of affinity teams consists of: utilizing the machine studying mannequin to deduce a minimum of one extra attribute and/or a minimum of one extra desire for a minimum of a few of the plurality of customers; and dividing the plurality of customers into the a number of affinity teams primarily based a minimum of partly upon the a minimum of one extra attribute and/or the a minimum of one extra desire.
“6. The computer-implemented technique of declare 1, additional comprising, previous to dividing the plurality of customers into the a number of affinity teams: figuring out, by the a number of processors analyzing historic knowledge indicative of (i) shopper traits and/or preferences for totally different affinity teams and (ii) bidding exercise for the totally different affinity teams, necessities for membership in every of the a number of affinity teams.
“7. The computer-implemented technique of declare 1, whereby the machine studying mannequin is a neural community, and additional comprising, previous to dividing the plurality of customers into the a number of affinity teams: coaching the neural community utilizing historic knowledge indicative of (i) shopper traits and/or preferences for totally different customers and (ii) risk-related outcomes for the totally different customers.
“8. The computer-implemented technique of declare 1, whereby auctioning the chance to offer insurance coverage for a number of of the a number of affinity teams consists of: for every affinity group of the a number of of the a number of affinity teams, auctioning the chance to offer particular person insurance coverage insurance policies for every shopper inside the affinity group.
“9. The computer-implemented technique of declare 1, whereby auctioning the chance to offer insurance coverage for a number of of the a number of affinity teams consists of: for every affinity group of the a number of of the a number of affinity teams, auctioning the chance to offer a gaggle insurance coverage coverage for all customers inside the affinity group.
“10. The computer-implemented technique of declare 1, additional comprising, previous to dividing the plurality of customers into the a number of affinity teams: receiving, by the a number of processors, car telematics knowledge for the plurality of customers, the car telematics knowledge being indicative of a number of driving behaviors, and the a number of traits of the plurality of customers together with the a number of driving behaviors.
“11. A system comprising: a persistent reminiscence storing a shopper profile database; a communication interface configured to speak with distant units by way of a communications community; a number of processors; and a number of non-transitory, computer-readable media storing directions that, when executed by the a number of processors, trigger the system to divide a plurality of customers into a number of affinity teams primarily based a minimum of upon a number of traits and a number of preferences of the plurality of customers which are included within the shopper profile database, a minimum of partly by analyzing the a number of traits and/or the a number of preferences of the plurality of customers utilizing a machine studying mannequin, public sale, by way of the communication interface and the communications community, a possibility to offer insurance coverage for a number of of the a number of affinity teams, obtain, by way of the communication interface and the communications community, a number of bids for buy and/or affords of insurance coverage for the a number of of the a number of affinity teams, settle for a profitable bid of the a number of bids, and trigger particular person insurance coverage insurance policies or a gaggle insurance coverage coverage to be offered to or up to date for customers related to a selected affinity group similar to the profitable bid, thereby offering decrease value insurance coverage and/or insurance coverage that’s extra reflective of precise threat, or lack thereof, for the customers related to the actual affinity group.
“12. The system of declare 11, whereby dividing the plurality of customers into a number of affinity teams consists of: figuring out threat scores for the plurality of customers by analyzing the a number of traits of the plurality of customers utilizing the machine studying mannequin; and dividing the plurality of customers into the a number of affinity teams primarily based a minimum of upon the danger scores and the a number of preferences of the plurality of customers.
“13. The system of declare 11, whereby dividing the plurality of customers into a number of affinity teams consists of: figuring out desire classifications for the plurality of customers by analyzing the a number of preferences of the plurality of customers utilizing the machine studying mannequin; and dividing the plurality of customers into the a number of affinity teams primarily based a minimum of upon the desire classifications and the a number of traits of the plurality of customers.
“14. The system of declare 11, whereby dividing the plurality of customers into a number of affinity teams consists of: figuring out classifications for the plurality of customers by analyzing the a number of traits and the a number of preferences of the plurality of customers utilizing the machine studying mannequin; and dividing the plurality of customers into the a number of affinity teams primarily based a minimum of upon the classifications.
“15. The system of declare 11, whereby dividing the plurality of customers into a number of affinity teams consists of: utilizing the machine studying mannequin to deduce a minimum of one extra attribute and/or a minimum of one extra desire for a minimum of a few of the plurality of customers; and dividing the plurality of customers into the a number of affinity teams primarily based a minimum of partly upon the a minimum of one extra attribute and/or the a minimum of one extra desire.
“16. The system of declare 11, whereby the directions additional trigger the system to, previous to dividing the plurality of customers into the a number of affinity teams: decide, by analyzing historic knowledge indicative of (i) shopper traits and/or preferences for totally different affinity teams and (ii) bidding exercise for the totally different affinity teams, necessities for membership in every of the a number of affinity teams.
“17. The system of declare 11, whereby the machine studying mannequin is a neural community, and whereby the directions additional trigger the system to, previous to dividing the plurality of customers into the a number of affinity teams: prepare the neural community utilizing historic knowledge indicative of (i) shopper traits and/or preferences for totally different customers and (ii) risk-related outcomes for the totally different customers.
“18. The system of declare 11, whereby auctioning the chance to offer insurance coverage for a number of of the a number of affinity teams consists of: for every affinity group of the a number of of the a number of affinity teams, auctioning the chance to offer particular person insurance coverage insurance policies for every shopper inside the affinity group.
“19. The system of declare 11, whereby auctioning the chance to offer insurance coverage for a number of of the a number of affinity teams consists of: for every affinity group of the a number of of the a number of affinity teams, auctioning the chance to offer a gaggle insurance coverage coverage for all customers inside the affinity group.”
There are extra claims. Please go to full patent to learn additional.
For extra info on this patent software, see: Frankowiak, Sara; Isaacs, Craig Dean; McCarty, Jeff; Roll, Leif Agerholm; Taylor, Kelli. Machine Learning Technologies for Efficiently Obtaining Insurance Coverage. Filed June 24, 2020 and posted July 7, 2022. Patent URL: https://appft.uspto.gov/netacgi/nph-Parser?Sect1=PTO1&Sect2=HITOFF&d=PG01&p=1&u=%2Fnetahtml%2FPTO%2Fsrchnum.html&r=1&f=G&l=50&s1=%2220220215476%22.PGNR.&OS=DN/20220215476&RS=DN/20220215476
(Our studies ship fact-based information of analysis and discoveries from all over the world.)
https://insurancenewsnet.com/oarticle/researchers-submit-patent-application-machine-learning-technologies-for-efficiently-obtaining-insurance-coverage-for-approval-uspto-20220215476-patent-application
