in ,

Progressive Casualty Insurance Company – InsuranceNewsNet


2022 OCT 06 (NewsRx) — By a News Reporter-Staff News Editor at Insurance Daily NewsProgressive Casualty Insurance Company (Mayfield Village, Ohio, United States) has been issued patent number 11449726, according to news reporting originating out of Alexandria, Virginia, by NewsRx editors.

The patent’s inventors are McCormack, Geoffrey S. (Shaker Heights, OH, US), Panguluri, Rama Rao (Aurora, OH, US), Sesnowitz, Craig S. (Chagrin Falls, OH, US), Wagner, Robert R. (Fairview Park, OH, US).

This patent was filed on July 23, 2020 and was published online on September 20, 2022.

From the background information supplied by the inventors, news correspondents obtained the following quote:

“Technical Field

“This disclosure relates to linguistic analysis and specifically to biometric systems that recognize characteristics conveyed through an input to render automated classifications.

“Related Art

“Text messages and speech are common forms of communication. A text message can express opinions and thoughts through short messages; share ideas and expressions through email, and provide access to services through web pages. Both text messages and speech are rich sources of information that may provide insights into the personality, characteristics, and needs of the communicators. Deconstructing text messages and speech and extracting information from them is challenging because neither is structured nor organized. Text messages takes many forms like speech; it is difficult to classify, and thus difficult to automate, especially when transacting services.”

Supplementing the background information on this patent, NewsRx reporters also obtained the inventors’ summary information for this patent: “An automatic recognition system uses biometrics to recognize characteristics conveyed in messages and/or speech. Speech is a spoken input to the automatic recognition system in an aural range. It may be a single word, an entire phrase, a sentence or several sentences. The systems and methods (referred to as systems or services) derive insights from free-form input and generate and train deep-learning and/or machine-learning models such as classification models. Using an application programing interface (API) and a translator, the system translates natural and/or synthesized text (e.g., via a speech-to-text synthesis) into feature groups that are processed by statistical, functional, and/or neural network models in a distributed architecture. Recognitions are made even when resources are limited. In some systems, recognition events are refined by evolutionary models, such as one or more exemplary classification evolutionary models. The recognition may comprise a recognition of a request or a response to an inquiry, such as a request for a composite business classification that is based on consumer classifications and/or industrial classifications. An industrial classification may be based on the North American Industry Classification System Index Classification or NAICS system.

“The automatic recognition systems simulate human dialogue. The systems learn user preferences and provide users with suggestions such as to how to navigate self-service channels on a user device or through a processing node. Segmentations generate the predictions that provide access to restricted content or select underwriting and deliver targeted messages such as a predicted business classification that provides access to one or many insurance carriers, for example. The systems use natural language processing models and/or evolutionary models to render one or more intents and sub-entities. An intent generally refers to an interpreted aim or purpose of a communication. Its sub-entities provide context of that communication often in a graduated or hierarchical order. In an insurance quoting application, an intent may represent a detailed classification of a business requesting insurance coverage.

“In FIG. 1, an automatic recognition system receives speech and/or text input through an API 102. A speech-to-text synthesis translates speech into text through a translator 104. The text is translated into feature sets that are represented by vectors. The vectors represent the frequency of words and phrases within a text or textual frame or segment. It is processed by a natural language processing model 108 (also referred to as a natural language processing engine) and stored in a local or a distributed database (e.g., a memory). In FIG. 1, select semantically relevant words and phrases (e.g., that express a user’s intent and convey contextual meaning) detected in the input vector by the natural language processing model 108 are chosen and marked as designation words and/or designation phrases. Designation words and/or designation phrases uniquely identify evolutionary models that are selected by the controller 100 to validate, distill, and/or further process the input vector and/or natural language model’s intents and sub-entity recognition results. An evolutionary model is a model that processes the natural language model’s 108 intents and sub-entity recognition results and/or the speech and/or text input, which may be vectorized. An evolutionary model generates species’ predictions and processes intents and sub-entities that are particular or unique to a sub-category. A species ranks below a genus or subgenus in a particular domain. A genus ranks below a family but above a species. It consists of a group exhibiting similar characteristics. In a risk mitigation context, for example, a family may comprise insurance carrier products and services, a genus may comprise an on-line insurance quoting domain, an on-line insurance servicing domain, an on-line claims processing domain, and/or etc. In a risk mitigation context, a sub-genus may comprise a general business classification that best describes the business, and a species may comprise an NAICS classification that best describes that business.

“The accuracy of a recognition event is measured by a first confidence score generated by the natural language processing model 108 that quantitatively represents the correctness of the prediction made by the natural language processing model 108. When the first confidence score equals or exceeds a predetermined threshold, a controller 110 transfers the input vector and/or natural language recognition results to the one of more recognition engines 112-116 that host the evolutionary models 118-122. There are one or multiple recognition engines 112-116 and evolutionary models 118-122 (that is shown by the N designation in the drawings) via controller 100. If the first confidence score is below a predetermined threshold, the controller 110 may direct the natural language processing model 108 to re-execute the recognition event, or select another natural language processing model or vocabulary in alternate systems, to re-execute the recognition event. If the deficiency persists, controller 110 reports the deficiency by transmitting an alert to a user or IT specialist, sets a flag, and/or may prompt the user to re-enter the input via an alert device 602 (shown in FIGS. 6 and 7) before re-executing the recognition event. A flag is a state or a signal indicating the existence or status of a particular condition. A controller is a hardware device or processor that actuates and/or directs the devices it controls, or is capable of making decisions with respect to the operation or actuation of the devices it controls, including being operable to selectively sequence their operation, enable and disable and/or select prediction models and/or vocabularies and active grammars, and/or delay recognition and/or training operations. In FIG. 1, controller 100 is a hardware device or processor that actuates and/or directs the operation of the recognition engines 112-116, or is capable of making decisions with respect to the operation or actuation of the recognition engines 112-116 including being operable to selectively sequence their operation, enable and disable the evolutionary models 118-122 they serve or host, and/or delay recognition and/or training operations.

“Some recognition engines load two or more or all of the instances of the evolutionary models 118-122 that controller 110 independently enables or disables to modify or execute recognition event sequences and behavior. The loading and parallel execution of two or more evolutionary models 118-122 improves recognition capacity, by among other things enabling the controller 110 to shift work and balance processing loads. By enabling one or more recognition engines to take over for one or more or all of the recognition engines 112-116, the system enhances network stability, minimizes or eliminates downtime caused by recognition engine or evolutionary model failures, and ensures that the recognition engines 112-116 work in a unified way that is seamless to the user. Since the enablement and disablement of the evolutionary models 118-122 occurs independently of their loading, the recognition system can preserve processing resources by enabling its evolutionary models 118-122 only when they are needed.”

The claims supplied by the inventors are:

“1. A system that determines a classification by simulating a human user comprising; a translator that translates an input segment to an output segment and represents a frequency of words and phrases in the output segment as an input vector; a natural language processing engine that processes the input vector and generates a plurality of intents and a plurality of sub-entities; a plurality of recognition engines hosting one or more evolutionary models that process the plurality of intents and the plurality of sub-entities to generate a second plurality of intents and a second plurality of sub-entities that represent a species classification; a plurality of application engines that process the plurality of intents, the plurality of sub-entities, the second plurality of intents, and the second plurality of sub-entities, and render an identifying output; where the plurality of application engines automatically render a plurality of vectors and a plurality of textual associations and associate the plurality of vectors and the plurality of textual associations with a free-form input to the one or more evolutionary models when the identifying output is rejected; and where the natural language processing engine selects an instance of an evolutionary model of the one or more evolutionary models as a result of a recognition of one or more predefined semantically relevant words and phrases that the natural language processing engine detected in the input vector.

“2. The system of claim 1 where the plurality of recognition engines loads two or more instances of the one or more evolutionary models actuated by a controller as a result of a recognition engine failure.

“3. The system of claim 1 where the plurality of recognition engines loads two or more instances of the one or more evolutionary models actuated by a controller.

“4. The system of claim 1 where the one or more evolutionary models is actuated as a result of a plurality post-processing applications.

“5. The system of claim 1 where the one or more evolutionary models is actuated as a result of a plurality of sub-classifications made from a hierarchical classification model.

“6. The system of claim 1 where the one or more evolutionary models is iteratively trained on a supplemental training dataset automatically generated by a plurality of post-processing application engines.

“7. The system of claim 6 where the supplemental training dataset is vectorized and a plurality of scalar functions of a training vector are adjusted by a weighting function.

“8. The system of claim 6 where the plurality of post-processing application engines comprises an insurance quoting system, an insurance claims processing system, or an on-line insurance servicing system.

“9. The system of claim 1 further comprising a plurality of post-processing application engines that consume the second plurality of intents and the second plurality of sub-entities when selected by a controller and where the controller selects the plurality of post-processing application engines in response to an analysis of a plurality of contexts of the input vector.

“10. The system of claim 9 where the plurality of post-processing application engines identifies a business classification associated with a free-form input that is translated into a textual segment.

“11. A method that determines a classification by simulating a human user comprising; translating an input segment by a translator to an output segment; representing how often words and phrases appear in the output segment by an input vector; processing by a natural language processing engine the input vector to generate a plurality of intents and a plurality of sub-entities; processing the plurality of intents and the plurality of sub-entities by a plurality of recognition engines hosting the one or more evolutionary models; generating a second plurality of intents and a second plurality of sub-entities by the plurality of recognition engines that represent a species classification; selecting an instance of an evolutionary model of the one or more evolutionary models as a result of a recognition of one or more predefined semantically relevant words and phrases that the natural language processing engine detected in the input vector; processing the plurality of intents, the plurality of sub-entities, the second plurality of intents, and the second plurality of sub-entities by a plurality of application engines render an identifying output; rendering a plurality of vectors and a plurality of textual associations and associating the plurality of vectors and the plurality of textual associations with a free-form input to the one or more evolutionary models when the identifying output is rejected.

“12. The method of claim 11 further comprising loading two or more instances of the one or more evolutionary models actuated by a controller as a result of a recognition engine failure.

“13. The method of claim 11 further comprising loading two or more instances of the one or more evolutionary models actuated by a controller as a result of an evolutionary model failure.

“14. The method of claim 11 further comprising actuating the one or more evolutionary models in response to an availability of a plurality of post-processing applications.

“15. The method of claim 11 further comprising actuating the one or more evolutionary models in response to a plurality of sub-classifications made in a hierarchical classification model.

“16. The method of claim 11 further comprising iteratively training the one or more evolutionary models on a supplemental training dataset automatically generated by a plurality of post-processing application engines.

“17. The method of claim 16 further comprising vectorizing the supplemental training dataset and modifying a plurality of scalar functions of a training vector by a weighting function.

“18. The method of claim 16 where the plurality of post-processing application engines comprises an insurance quoting system, an insurance claims processing system, and an on-line insurance servicing system.

“19. The method of claim 16 further comprising consuming the second plurality of intents and the second plurality of sub-entities by a plurality of post-processing application engines in response to an analysis of a plurality of contexts of the input vector.

“20. The method of claim 19 where the plurality of post-processing application engines identifies a business classification associated with a free-form input that is translated into a textual segment.

“21. A system that determines a classification by simulating a human user comprising; a translator that translates an input segment to an output segment and represents a frequency of words and phrases in the output segment as an input vector; a natural language processing engine that processes the input vector and generates a plurality of intents and a plurality of sub-entities; a plurality of recognition engines hosting one or more evolutionary models that process the plurality of intents and the plurality of sub-entities to generate a second plurality of intents and a second plurality of sub-entities that represent a species classification; a plurality of application engines that process the plurality of intents, the plurality of sub-entities, the second plurality of intents, and the second plurality of sub-entities, and render an identifying output; where the plurality of application engines automatically render a plurality of vectors and a plurality of textual associations and associate the plurality of vectors and the plurality of textual associations with a free-form input to the one or more evolutionary models when the identifying output is rejected; and where the natural language processing engine selects an instance of an evolutionary model of the one or more evolutionary models as a result of a recognition of one or more predefined semantically relevant words and phrases that the natural language processing engine detected in the input vector; where the species classification ranks below a genus classification or a subgenus classification in a domain.”

For the URL and additional information on this patent, see: McCormack, Geoffrey S. Tailored artificial intelligence. U.S. Patent Number 11449726, filed July 23, 2020, and published online on September 20, 2022. Patent URL: http://patft.uspto.gov/netacgi/nph-Parser?Sect1=PTO1&Sect2=HITOFF&d=PALL&p=1&u=%2Fnetahtml%2FPTO%2Fsrchnum.htm&r=1&f=G&l=50&s1=11449726.PN.&OS=PN/11449726RS=PN/11449726

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





Source link

What do you think?

Artificial Intelligence systems increasingly important for M&A due diligence: corporate lawyer

The White House unveils a “Bill of Rights” for artificial intelligence