Patent Grant
11983673
The present application generally relates to computer systems and more particularly to computer systems that are adapted to mine data to identify risk drivers and to develop risk mitigation strategies.
Electronic insurance claim records may be stored and utilized by an enterprise, such as an insurance company. For example, an insurance company may be interested in analyzing information about risk drivers and insurance claim outcomes in each insurance claim record to model insurance claim outcomes based on different risk drivers. In some cases, the insurance company might want to advise customers how different identified risk drivers affect insurance claim outcomes and advise customers on adopting risk mitigation strategies for affecting insurance claim outcomes. Accordingly, the insurance company may add value to insurance products sold to customers by helping customers identify risk drivers that are affecting their insurance claim outcomes and their insurance costs. Further, the insurance company may add value to insurance products sold to customers by helping customers employ risk mitigation strategies that improve their insurance claim outcomes and reduce their insurance costs. Human analysis of electronic records to identify risk drivers, however, can be a time consuming, error prone and subjective process—especially where there are a substantial number of records to be analyzed (e.g., thousands of electronic records might need to be reviewed) and/or there are a lot of factors that could potentially influence insurance claim outcomes. In addition, this type of information may be spread throughout a number of different computer systems (e.g., associated with different insurance companies, a human resources department, etc.).
It would be desirable to provide systems and methods for mining data to identify risk factors and for developing risk mitigation strategies in a way that provides fast and accurate results.
According to some embodiments, systems, methods, apparatus, computer program code and means are provided for mining data to identify risk factors and for developing risk mitigation strategies in a way that provides fast and accurate results and that allow for flexibility and effectiveness when responding to those results.
The present application is directed to systems and methods adapted to automatically analyze insurance claim records, automatically identify risk drivers, automatically identify how these risk drivers affect insurance claim outcomes and automatically provide risk mitigation strategies that improve insurance claim outcomes.
In one embodiment of the present application, a data analytics system includes a data mining engine, a predictive analytics engine and a claims insight platform. The data mining engine analyzes a plurality of insurance claim files to identify flags corresponding to risk drivers. The predictive analytics engine calculates a risk score for each of the plurality of insurance claim files based on identified flags corresponding to risk drivers. The claims insight platform selects a subset of the plurality of insurance claim files, calculates an average risk score for the subset of the plurality of insurance claim files, and determines an expected claim outcome based on the calculated average risk score for the subset of the plurality of insurance claim files.
In some embodiments, a data analytics system may include a first risk relationship data store containing electronic records that represent a plurality of risk relationships between the enterprise and a first risk relationship provider. Similarly, a second risk relationship data store containing electronic records that represent a plurality of risk relationships between the enterprise and a second risk relationship provider. A back-end application computer server may include a data mining engine that analyzes a set of electronic records in the first and second risk relationship data stores to identify flags corresponding to risk drivers. A predictive analytics engine may then calculate a risk score associated with the set of electronic records based on the associated entity attribute values and the identified flags corresponding to risk drivers. An insight platform may automatically generate a recommended action for the enterprise to lower the calculated risk score.
Some embodiments comprise: means for analyzing, by a data mining engine of the back-end application computer server, a set of electronic records in a first and a second risk relationship data store to identify flags corresponding to risk drivers, wherein the first risk relationship data store contains electronic records that represent a plurality of risk relationships between the enterprise and a first risk relationship provider, and the second risk relationship data store contains electronic records that represent a plurality of risk relationships between the enterprise and a second risk relationship provider; means for calculating, by a predictive analytics engine of the back-end application computer server, a risk score associated with the set of electronic records based on associated entity attribute values and the identified flags corresponding to risk drivers; and means for automatically generating, by an insight platform of the back-end application computer server, a recommended action for the enterprise to lower the calculated risk score.
In some embodiments, a communication device associated with a back-end application computer server exchanges information with remote devices in connection with an interactive graphical user interface. The information may be exchanged, for example, via public and/or proprietary communication networks.
A technical effect of some embodiments of the invention is an improved and computerized way of mining data to identify risk factors and for developing risk mitigation strategies in a way that provides fast and accurate results. With these and other advantages and features that will become hereinafter apparent, a more complete understanding of the nature of the invention can be obtained by referring to the following detailed description and to the drawings appended hereto.
Before the various exemplary embodiments are described in further detail, it is to be understood that the present invention is not limited to the particular embodiments described. It is also to be understood that the terminology used herein is for the purpose of describing particular embodiments only, and is not intended to limit the scope of the claims of the present invention.
In the drawings, like reference numerals refer to like features of the systems and methods of the present invention. Accordingly, although certain descriptions may refer only to certain figures and reference numerals, it should be understood that such descriptions might be equally applicable to like reference numerals in other figures.
The present invention provides significant technical improvements to facilitate data analytics. The present invention is directed to more than merely a computer implementation of a routine or conventional activity previously known in the industry as it provides a specific advancement in the area of electronic record analysis by providing improvements in data leveraging to identify risk factors, identify the effect of these risk factors on outcomes, and identify risk mitigation strategies to improve outcomes. The present invention provides improvement beyond a mere generic computer implementation as it involves the novel ordered combination of system elements and processes to provide improvements in data leveraging to identify risk factors, identify the effect of these risk factors on outcomes, and identify risk mitigation strategies to improve outcomes. The present invention is directed to a claims insight factory adapted to automatically analyze insurance claim records, automatically identify risk drivers, automatically identify how these risk drivers affect insurance claim outcomes and automatically provide risk mitigation strategies that improve insurance claim outcomes. In addition, in the present invention information may be accessed, updated (e.g., with tags or other annotations), and analyzed via a back-end application server to accurately improve the exchange of information, thus improving the overall efficiency of the system associated with message storage requirements and/or bandwidth considerations (e.g., by reducing the number of messages that need to be transmitted via a network). Moreover, embodiments associated with accessing and updating accurate, pertinent information might further improve sales of risk relationships, allocations of resources, electronic record processing decisions, etc.
In the context of the present application, customer refers to an employer who purchases insurance coverage from an insurance company on behalf of its employees (i.e., the insured), the insurance company refers to the insurer that provides insurance coverage to the insured, and the claimant refers to an injured party who files an insurance claim. Further, in the context of the present application, risk factors refer to categories of risk that may affect the outcomes of insurance claims, including, e.g., social risk, psychological risk, biological risk, etc. Each of these risk factors (i.e., risk categories) includes specific risk drivers. For example, the risk drivers for social risk may include, e.g., employee skills, employer environment, employee satisfaction, etc.; risk drivers for psychological risk may include, e.g., depression, Post Traumatic Stress Disorder (“PTSD”), etc.; risk drivers for biological risk may include, e.g., obesity, diabetes, etc.
The back-end application computer server 150 and/or the other elements of the system 100 might be, for example, associated with a Personal Computer (“PC”), laptop computer, smartphone, an enterprise server, a server farm, and/or a database or similar storage devices. According to some embodiments, an “automated” back-end application computer server 150 (and/or other elements of the system 100) may facilitate the access and/or update of electronic records in the first risk relationship data store 110. As used herein, the term “automated” may refer to, for example, actions that can be performed with little (or no) intervention by a human.
As used herein, devices, including those associated with the back-end application computer server 150 and any other device described herein, may exchange information via any communication network which may be one or more of a Local Area Network (“LAN”), a Metropolitan Area Network (“MAN”), a Wide Area Network (“WAN”), a proprietary network, a Public Switched Telephone Network (“PSTN”), a Wireless Application Protocol (“WAP”) network, a Bluetooth network, a wireless LAN network, and/or an Internet Protocol (“IP”) network such as the Internet, an intranet, or an extranet. Note that any devices described herein may communicate via one or more such communication networks.
The back-end application computer server 150 may store information into and/or retrieve information from the first risk relationship data store 110 and/or a second risk relationship data store 120 (e.g., associated with a different insurance company as compared to the first risk relationship data store). The data stores 110, 120 may be locally stored or reside remote from the back-end application computer server 150. As will be described further below, the first risk relationship data store 110 may be used by the back-end application computer server 150 in connection with an interactive user interface to access and update electronic records. Although a single back-end application computer server 150 is shown in
Note that the system 100 of
At S210, a data mining engine of a back-end application computer server may analyze a set of electronic records in a first and a second risk relationship data store to identify flags corresponding to risk drivers. The first risk relationship data store may, for example, contain electronic records that represent a plurality of risk relationships between the enterprise and a first risk relationship provider (e.g., insurer). Similarly, the second risk relationship data store may contain electronic records that represent a separate plurality of risk relationships between the enterprise and a second risk relationship provider (e.g., if the first risk relationship data store is associated with pharmacy benefits, the second risk relationship data store may be associated with short or long term disability insurance claims).
At S220, a predictive analytics engine of the back-end application computer server may calculate a risk score associated with the set of electronic records based on associated entity attribute values and the identified flags corresponding to risk drivers. Note that the risk score may be based on information aggregated, combined, and/or blended from both the first risk relationship data store and the second risk relationship data store. At S230, an insight platform or solution engine of the back-end application computer server may automatically generate a recommended action for the enterprise to lower the calculated risk score (e.g., a risk mitigation strategy that relies on one or more insights that were learned by analyzing both the first and second risk relationship data stores together).
As shown in
The functions of computer server(s) 410 described herein may be implemented using computer applications comprising computer program code stored in a non-transitory computer-readable medium that is executed by a computer processor. The functions of computer server(s) 410 described herein may also be implemented in programmable hardware devices such as field programmable gate arrays, programmable array logic, programmable logic devices or the like. Further, functions of computer server(s) 410 described herein may be implemented using some combination of computer program(s) executed by a computer processor and programmable hardware devices. Thus, computer server(s) 410 of the present application comprises suitable computer hardware and software for performing the desired functions and are not limited to any specific combination of hardware and software.
The executable computer program code may comprise one or more physical or logical blocks of computer instructions, which may be organized as an object, procedure, process or function. For example, the executable computer program code may be distributed over several different code partitions or segments, among different programs, and across several devices. Accordingly, the executable computer program need not be physically located together, but may comprise separate instructions stored in different locations which, when joined logically together, comprise the computer application and achieve the stated purpose for the computer application.
An insurance company may collect insurance claim data (e.g., insurance claim files) associated with various types of insurance (e.g., property and casualty insurance, group benefits insurance, workers' compensation insurance, etc.). The insurance company runs Extraction, Transformation, and Loading (“ETL”) processes on collected insurance claim data (e.g., insurance claim files). The processed insurance claim data (e.g., insurance claim files) is loaded into a claims data warehouse 420. Accordingly, claims insight factory 450 may comprise a claims data warehouse 420 adapted to store insurance claim data (e.g., insurance claim files). The claims data warehouse 420 may comprise one or more data marts (e.g., dimensional data mart, analytic data mart, legacy data mart) adapted for different business functions. Each insurance claim file stored in claims data warehouse 420 may include insurance claim information such as, e.g., claimant/employee, customer/employer, employer industry, employer location, employer size, type of insurance claim (e.g., property and casualty, group benefits, workers' compensation, etc.), insurance claim cost, insurance claim duration, etc. In an alternative embodiment, claims insight factory 450 does not comprise claims data warehouse 420, but instead accesses insurance claim data (e.g., insurance claim files) in a stand-alone claims data warehouse 420.
Claims insight factory 450 mines the claim data stored in claims data warehouse 420 to automatically identify flags corresponding to certain risk drivers. Flags refer to items (e.g., text, codes, structured data fields, etc.) in insurance claim data (e.g., insurance claim files) that are indicative of certain risk drivers that may affect the outcomes of insurance claims. For example, claims insight platform 430 may include a data mining engine 440, e.g., such as the semantic rules system described in U.S. Pat. No. 9,026,551, which may be used to identify text flags (e.g., semantic events) in insurance claim data (e.g., insurance claim files) that trigger semantic rules. U.S. Pat. No. 9,026,551 is herein incorporated by reference in its entirety. Accordingly, data mining engine 440 may determine certain risk drivers associated with an insurance claim file based on the triggering of corresponding flags. For example, certain semantic rules may be associated with certain risk drivers, such that the triggering of a semantic rule leads to the identification of a corresponding risk driver. Thus, claims insight platform 430 may collect metrics of risk drivers for each insurance claim file in claims data warehouse 420. In an alternative embodiment, claims insight platform 430 does not comprise data mining engine 440, but instead may collect metrics of risk drivers for each insurance claim file in claims data warehouse 420 by directing a stand-alone data mining engine 440.
The claims insight factory 450 may use the determination of certain risk drivers associated with an insurance claim file to calculate one or more risk factor scores corresponding to one or more risk factors for the insurance claim file. For example, claims insight factory 450 may first use the determination of certain risk drivers associated with an insurance claim file to calculate the likelihood of certain events occurring (e.g., events delaying recovery in disability claims, subrogation, fraud, etc.). Then, for each insurance claim file, claims insight factory 450 may assign one or more risk factor scores corresponding to one or more risk factors based on the calculated likelihood of certain events occurring on the basis of the identified risk drivers.
Accordingly, the claims insight platform 430 may include a predictive analytics engine 460 that uses as an input the risk drivers associated with an insurance claim file and produces an output of one or more risk factor scores corresponding to one or more risk factors associated with the insurance claim file. predictive analytics engine 460 may comprise a knowledge base of historical insurance claim data and predictive models that can be implemented with the knowledge base to calculate the one or more risk factor scores corresponding to one or more risk factors associated with the insurance claim file on the basis of the identified risk drivers. According to some embodiments, for each claim file the predictive analytics engine 460 may generate risk factor scores for different risk factors (e.g., social risk score, psychological risk score, biological risk score, etc.) on the basis of the risk drivers identified for the insurance claim file. In an alternative embodiment, claims insight platform 430 does not comprise predictive analytics engine 460, but instead may collect risk factor scores for different risk factors (e.g., social risk score, psychological risk score, biological risk score, etc.) on the basis of the risk drivers identified for insurance claim files in claims data warehouse 420 by directing a stand-alone predictive analytics engine 460.
The claims insight factory 450 may further include a database 480 for storing insurance claim records 500 for associating information for each insurance claim file. As shown in
The insurance claim information collected, generated and stored may be leveraged by claims insight factory 450 to provide benchmarking information. More particularly, claims insight factory 450 may benchmark the risk factor scores for insurance claim files of an analysis group against the risk factor scores for insurance claim files of a baseline group. Also, claims insight factory 450 may benchmark the claim outcomes (e.g., claim characteristics) of insurance claim files of an analysis group against the claim outcomes (e.g., claim characteristics) of insurance claim files of a baseline group. The analysis group of insurance claim files is selected based on analysis selection criteria, and the baseline group of insurance claims is selected based on baseline selection criteria. Analysis selection criteria for selecting an analysis group of insurance claim files may be a field 510 attribute or combination of field 510 attributes in insurance claim records 500 stored in database 480. Likewise, baseline selection criteria for selecting a baseline group of insurance claim files may also be a field 510 attribute or combination of field 510 attributes in insurance claim records 500 stored in database 480. For example, the risk factor scores for the insurance claim files of a selected customer may be benchmarked against the risk factor scores for the insurance claim files of other customers in the same industry, in the same geographical region and/or of about the same size (e.g., number of employees).
The claims insight factory 450 may further include a claims insight platform 430. Once the analysis selection criteria for the analysis group are entered in claims insight platform 430, claims insight platform 430 queries database 480 for insurance claim records satisfying the analysis selection criteria. The claims insight platform 430 then determines composite risk factor scores for the insurance claim files of the analysis group according to a selected analysis composite score basis (e.g., customer-by-customer basis, claimant-by-claimant basis, etc.). The claims insight platform 430 may calculate a composite risk factor score for a selected risk factor according to the selected analysis composite score basis by taking the average of all the risk factor scores for the selected risk factor for the insurance claim files of the analysis group according to the selected analysis composite score basis.
For example, if the analysis selection criterion is a selected customer, then the database 480 query returns all the insurance claim records 500 associated with the selected customer (i.e., the analysis group of insurance claim files). Then, the claims insight platform 430 may calculate composite risk factor scores for the insurance claim files of the analysis group according to a selected analysis composite score basis (e.g., customer-by-customer basis, claimant-by-claimant basis, etc.). For instance, if the analysis composite score basis is a customer-by-customer basis, claims insight platform 430 may calculate the composite social risk factor score for the selected customer by taking the average of all the social risk factor scores for all the insurance claim files associated with the selected customer. Similarly, claims insight platform 430 may calculate the composite psychological risk factor score for the selected customer by taking the average of all the psychological risk factor scores for all the insurance claim files associated with the selected customer. Further, claims insight platform 430 may calculate the composite biological risk factor score for the selected customer by taking the average of all the biological risk factor scores for all the insurance claim files associated with the selected customer.
In another example, if the analysis composite score basis is a claimant-by-claimant basis, the claims insight platform 430 may calculate the composite social risk factor score for each claimant in the analysis group by taking the average of the social risk factor scores for the insurance claim files on a claimant-by-claimant basis. Similarly, claims insight platform 430 may calculate the composite psychological risk factor score for each claimant in the analysis group by taking the average of the psychological risk factor scores for the insurance claim files on a claimant-by-claimant basis. Further, claims insight platform 430 may calculate the composite biological risk factor score for each claimant in the analysis group by taking the average of the biological risk factor scores for the insurance claim files on a claimant-by-claimant basis.
Once the baseline selection criteria for the baseline group are entered in claims insight platform 430, the claims insight factory 450 queries database 480 for insurance claim records 500 satisfying the baseline selection criteria. The claims insight platform 430 then determines composite risk factor scores for the insurance claim files of the baseline group according to a selected baseline composite score basis (e.g., customer-by-customer basis, claimant-by-claimant basis, etc.). The claims insight platform 430 may calculate a composite risk factor score for a selected risk factor by taking the average of all the risk factor scores for the selected risk factor for the insurance claim files of the baseline group according to the selected baseline composite score basis (e.g., customer-by-customer basis, claimant-by-claimant basis, etc.).
For example, if the baseline selection criterion is a selected industry, then the database 480 query returns all the insurance claim records 500 associated with the selected industry (i.e., the baseline group of insurance claim files). If the baseline selection criteria is a selected geographic region and a selected company size range (e.g., 500-1,000 employees), then the database 480 query returns all of the insurance claim records 500 associated with the selected geographic region and company size range (i.e., the baseline group of insurance claim files). Then, claims insight platform 430 may calculate composite risk factor scores for the insurance claim files of the baseline group according to a selected baseline composite score basis (e.g., customer-by-customer basis, claimant-by-claimant basis, etc.). For instance, if the baseline composite score basis is a customer-by-customer basis, then claims insight platform 430 may calculate the composite social risk factor score for each customer in the baseline group by taking the average of the social risk factor scores for the insurance claim files on a customer-by-customer basis. Similarly, claims insight platform 430 may calculate the composite psychological risk factor score for each customer in the baseline group by taking the average of the psychological risk factor scores for the insurance claim files on a customer-by-customer basis. Further, claims insight platform 430 may calculate the composite biological risk factor score for each customer in the baseline group by taking the average of the biological risk factor scores for the insurance claim files on a customer-by-customer basis.
In another example, if the baseline composite score basis is a claimant-by-claimant basis, the claims insight platform 430 may calculate the composite social risk factor score for each claimant in the baseline group by taking the average of the social risk factor scores for the insurance claim files on a claimant-by-claimant basis. Similarly, claims insight platform 430 may calculate the composite psychological risk factor score for each claimant in the analysis group by taking the average of the psychological risk factor scores for the insurance claim files on a claimant-by-claimant basis. Further, claims insight platform 430 may calculate the composite biological risk factor score for each claimant in the analysis group by taking the average of the biological risk factor scores for the insurance claim files on a claimant-by-claimant basis.
Further, the claims insight platform 430 may provide users access to claims insight factory 450 via GUIs 610 rendered on remote computing devices 600 in communication with claims insight platform 430. For instance, a user may enter the analysis selection criteria, analysis composite score basis, baseline selection criteria and baseline composite score basis via a GUI 610 rendered on a computing device 600 in communication with claims insight platform 430. Further, claims insight platform 430 provides users benchmarking analysis according to the selected analysis selection criteria, analysis composite score basis, baseline selection criteria and baseline composite score basis received from the user via the GUI 610 rendered on the computing device 600.
For example, the claims insight platform 430 may execute benchmark analysis of the risk factor scores for insurance claim files of an analysis group against the risk factor scores for insurance claim files of a baseline group in accordance with user specified analysis selection criteria, analysis composite score basis, baseline selection criteria and baseline composite score basis.
Additionally, the benchmark analysis may include an indication of how the risk factor scores for the insurance claim files of the analysis group affect the claim outcomes (e.g., claim characteristics) of different types of claims compared to the risk factor scores for the insurance claim files of the baseline group. For example, for a given benchmark analysis for the risk factor scores of an analysis group compared to the risk factor scores of a baseline group, claims insight platform 430 may calculate how the risk factor scores for the insurance claim files of the analysis group affect the claim outcomes (e.g., claim characteristics) of a selected type of insurance claim (e.g., property and casualty, group benefits, workers' compensation, etc.) compared to the risk factor scores for the insurance claim files of the baseline group.
The benchmark analysis of
The benchmark analysis of
The claims insight platform 430 may execute the claim characteristics analysis for a selected type of insurance claim based on stored historical data of insurance claim records 500 for the selected type of insurance claim. The claims insight platform 430 may query database 480 for insurance claim records 500 for the selected type of insurance claim that have the same risk factor scores as the analysis group and the same risk factor scores as some specified risk factor scores of the baseline group. Then, claims insight platform 430 may calculate the average claim characteristic values for the insurance claim records 500 having the same risk factor scores as the analysis group and the same risk factor scores as some specified risk factor scores of the baseline group. In the context of the risk factor scores, “same” does not necessarily mean identical and may mean substantially the same within a specified range (e.g., +/−5%).
For instance, for the benchmark analysis of
Then, based on the insurance claim information associated with the insurance claim records 500 returned by the query, the claims insight platform 430 may calculate the average claim characteristic values (e.g., medical expenses, other expenses and duration) corresponding to each risk factor score (e.g., social risk score, psychological risk score, biological risk score) of the analysis group (e.g., the selected customer). For example, for the benchmark analysis of
Also, for the benchmark analysis of
Then, based on the insurance claim information associated with the insurance claim records 500 returned by the query, the claims insight platform 430 may calculate the average claim characteristic values (e.g., medical expenses, other expenses and duration) corresponding to each specified risk factor score (e.g., social risk score, psychological risk score, biological risk score) of the baseline group (e.g., other customers in the same industry). For example, for the benchmark analysis of
Accordingly, based on the claims insight platform's 430 calculations described above, claims insight platform 430 may determine how claim characteristic values for insurance claims having the same risk factor score as the analysis group compare to the claim characteristic values of insurance claims having some specified value of the risk factor score (e.g., average, highest 10%, lowest 10%, etc.) of the baseline group. As shown in the benchmark analysis of
Also, the claims insight platform 430 may provide account level data aggregations or claimant level data aggregations by aggregating information for all insurance claim records 500 associated with a selected account or a selected claimant, respectively. Thus, claims insight platform 430 may provide insurance claim information on an account wide basis or a claimant-by-claimant basis. Also, the metrics of risk drivers for insurance claim files associated with a particular account may be aggregated to generate an account profile. Account profile may define different risk factors (i.e., risk categories), including, e.g., social risk, psychological risk, biological risk, etc., and may identify an account's specific risk drivers for each risk factor. Similarly, the metrics of risk drivers for insurance claim files associated with a particular claimant may be aggregated to generate a claimant profile. Claimant profile may define different risk factors (i.e., risk categories), including, e.g., social risk, psychological risk, biological risk, etc., and may identify a claimant's specific risk drivers for each risk factor.
The claims insight platform 430 may automatically generate a report for any benchmark analysis performed and may automatically send the benchmark analysis report to the customer/employer and/or claimant/employee. Also, claims insight platform 430 may automatically generate reports for account profiles and claimant profiles and may automatically send the account profile reports and claimant profile reports to their respective customer/employer and claimant/employee. Additionally, benchmark analysis reports, account profile reports and claimant profile reports may be stored by claims insight platform 430 and accessed and viewed by users via the GUI 610 rendered on a computer device.
Additionally, the information identified in benchmark analysis reports account profile reports and claimant profile reports (e.g., identification of key risk drivers, benchmarking of risk factors) may be leveraged by claims insight factory 450 to provide risk mitigation recommendations to claimants/employees and/or customers/employers. For example, claims insight platform 430 may automatically generate recommendations for claimants/employees and/or customers/employers based on information identified in benchmark analysis reports, account profile reports and claimant profile reports (e.g., identification of key risk drivers, benchmarking of risk factors). For example, if the claims insight platform's 430 data analysis reveal that there are certain risk drivers that are negatively affecting insurance claim outcomes, claims insight platform 430 may automatically generate recommendations to help claimants/employees or customers/employers address some of the identified key risk drivers.
The nature of the recommendations will depend on the nature of the risk drivers. For example, if an identified risk driver is missed medical appointments, then claims insight platform 430 may recommend providing appointment reminders, transportation services and/or more conveniently located medical service providers. In another example, if an identified risk driver is a bad relationship with medical service provider, then claims insight platform 430 may recommend switching medical service providers. The claims insight platform 430 provides actionable items in the form of risk mitigation services and/or risk mitigation strategies that are aimed at preventing loss and/or decreasing the duration of loss. To that end, claims insight platform 430 may automatically follow up with claimants/employees and/or customers/employers via automated electronic communication (e.g., email, text message, phone call, etc.) to confirm that recommended actionable items were carried out.
Note that an aggregation and analytics platform may collect information from various data sources, such as health/medical data, prescription pharmacy data, leave of absence data, workers' compensation data, disability data, etc. The platform may then aggregate the information and apply analytic techniques to generate dashboard displays (e.g., to keep managers up-to-date about risk relationships), generate insights (e.g., recommended actions for the enterprise), etc.
Insurance claim data (e.g., insurance claim files) 410 associated with various types of insurance (e.g., property and casualty insurance, group benefits insurance, workers' compensation insurance, etc.) is collected. The ETL processes 820 are run on the collected insurance claim data 410 (e.g., insurance claim files). The processed insurance claim data (e.g., insurance claim files) is loaded into a claims data warehouse 420. The claims data warehouse 420 is adapted to store insurance claim data (e.g., insurance claim files) and may comprise one or more data marts (e.g., dimensional data mart, analytic data mart, legacy data mart) adapted for different business functions. Each insurance claim file stored in claims data warehouse 420 may include insurance claim information such as, e.g., claimant/employee, customer/employer, employer industry, employer location, employer size, type of insurance claim (e.g., property and casualty, group benefits, workers' compensation, etc.), insurance claim cost, insurance claim duration, etc.
The claims insight platform 430 mines the claim data stored in claims data warehouse 420 to automatically identify flags corresponding to certain risk drivers. Flags refer to items (e.g., text, codes, structured data fields, etc.) in insurance claim data (e.g., insurance claim files) that are indicative of certain risk drivers that may affect the outcomes of insurance claims. For example, claims insight platform 430 may include a data mining engine 440, e.g., such as the semantic rules system described in U.S. Pat. No. 9,026,551, which may be used to identify text flags (e.g., semantic events) in insurance claim data (e.g., insurance claim files) that trigger semantic rules. U.S. Pat. No. 9,026,551 is herein incorporated by reference in its entirety. Thus, claims insight platform 430 may collect metrics of risk drivers for each insurance claim file in claims data warehouse 420.
The claims insight platform 430 uses the determination of certain risk drivers associated with an insurance claim file to calculate one or more risk factor scores corresponding to one or more risk factors for the insurance claim file. For example, claims insight platform 430 may include a predictive analytics engine 460 that uses as an input the risk drivers associated with an insurance claim file and produces an output of one or more risk factor scores corresponding to one or more risk factors associated with the insurance claim file. The predictive analytics engine 460 may comprise a knowledge base of historical insurance claim data and predictive models that can be implemented with the knowledge base to calculate the one or more risk factor scores corresponding to one or more risk factors associated with the insurance claim file on the basis of the identified risk drivers. Accordingly, for each claim file, predictive analytics engine 460 may generate risk factor scores for different risk factors (e.g., social risk score, psychological risk score, biological risk score, etc.) on the basis of the risk drivers identified for the insurance claim file.
The claims insight platform 430 may further include a database 480 for storing insurance claim records 500 for associating information for each insurance claim file, including information determined/generated by data mining engine 440 and predictive analytics engine 460. The claims insight platform 430 may execute the claim characteristics analysis for a selected type of insurance claim based on stored historical data of insurance claim records 500 for the selected type of insurance claim. Also, claims insight platform 430 may provide account level data aggregations or claimant level data aggregations by aggregating information for all insurance claim records 500 associated with a selected account or a selected claimant, respectively. Thus, claims insight platform 430 may provide insurance claim information on an account wide basis or a claimant-by-claimant basis. Also, the metrics of risk drivers for insurance claim files associated with a particular account may be aggregated to generate an account profile. Similarly, the metrics of risk drivers for insurance claim files associated with a particular claimant may be aggregated to generate a claimant profile.
The claims insight platform 430 may automatically generate a report for any benchmark analysis performed and may automatically send the benchmark analysis report to the customer/employer and/or claimant/employee. Also, claims insight platform 430 may automatically generate reports for account profiles and claimant profiles and may automatically send the account profile reports and claimant profile reports to their respective customer/employer and claimant/employee.
Additionally, benchmark analysis reports, account profile reports and claimant profile reports may be stored by claims insight platform 430 and accessed and viewed by users via the GUI 610 rendered on a computer device 600. Additionally, the information identified in benchmark analysis reports account profile reports and claimant profile reports (e.g., identification of key risk drivers, benchmarking of risk factors) may be leveraged by claims insight platform 430 to provide risk mitigation recommendations to claimants/employees and/or customers/employers. For example, claims insight platform 430 may automatically generate recommendations for claimants/employees and/or customers/employers based on information identified in benchmark analysis reports, account profile reports and claimant profile reports (e.g., identification of key risk drivers, benchmarking of risk factors). For example, if the claims insight platform's 430 data analysis reveal that there are certain risk drivers that are negatively affecting insurance claim outcomes, claims insight platform 430 may automatically generate recommendations to help claimants/employees or customers/employers address some of the identified key risk drivers.
According to some embodiments, the back-end application computer server 950 may also receive external information, such as third-party data 930 (e.g., providing Occupational Safety and Health Administration (“OSHA”) information), payroll data 932, governmental data 934, credit score data (e.g., associated with a level of risk), map or search result data, and social media data (e.g., a number of friends or likes on a web site). This data might be used, for example, to pre-populate fields in the first insurance policy data store 910. A user may then review the information via remote user devices 960, 962 and transmit updated information to the back-end application computer server 950 (e.g., by tagging an entity). Based on the updated information, the back-end application computer server 950 may adjust data in the first insurance policy data store 910 and make that information available to other employees of an enterprise as appropriate. According to some embodiments, the back-end application computer server 950 may transmit information to an email server, workflow application, a chatbot text interface, a streaming video interface, a voice recognition application, or a calendar function (e.g., to generate reminders that recommendation should be implemented). This information might be used by the system 900, for example, to automatically establish a channel of communication with an entity, automatically transmit a message to an entity, etc. Similarly, the back-end application computer server 950 might transmit updated electronic records 912 to an underwriter device for manual review and a determination of a proposed or approximate insurance premium.
The embodiments described herein may be implemented using any number of different hardware configurations. For example,
The processor 1310 also communicates with a storage device 1330. The storage device 1330 may comprise any appropriate information storage device, including combinations of magnetic storage devices (e.g., a hard disk drive), optical storage devices, mobile telephones, and/or semiconductor memory devices. The storage device 1330 stores a program 1315 and/or a risk evaluation tool or application for controlling the processor 1310. The processor 1310 performs instructions of the program 1315, and thereby operates in accordance with any of the embodiments described herein. For example, the processor 1310 may analyze a set of electronic records in multiple risk relationship data stores to identify flags corresponding to risk drivers. The processor 1310 may then calculate a risk score associated with the set of electronic records based on the associated entity attribute values and the identified flags corresponding to risk drivers. The processor 1310 may also automatically generate a recommended action for an enterprise to lower the calculated risk score.
The program 1315 may be stored in a compressed, uncompiled and/or encrypted format. The program 1315 may furthermore include other program elements, such as an operating system, a database management system, and/or device drivers used by the processor 1310 to interface with peripheral devices.
As used herein, information may be “received” by or “transmitted” to, for example: (i) the apparatus 1300 from another device; or (ii) a software application or module within the apparatus 1300 from another software application, module, or any other source.
In some embodiments (such as shown in
Referring to
The customer identifier 1402 may be, for example, a unique alphanumeric code identifying a customer who has purchased an insurance policy. The customer name 1404 may be associated with the insured, and the date and time 1406 might indicate when an insurance policy was purchased. The database 1400 may store various attributes associated with the customer, such as a number of employees, income information, overtime information, etc. The risk relationship database 1400 may also store risk factors 1408 (e.g., associated with an expected cost of claims filed in connection with insurance policies). The recommended actions 1410 might comprise text, links, pointers, executable code, etc. about changes an enterprise might make to reduce risk and/or costs.
Thus, embodiments may provide an automated and efficient way of mining data (e.g., associated with various insurers, third-parties, human resource departments, etc.) to identify risk factors and for developing risk mitigation strategies in a way that provides fast and accurate results. Embodiments may also provide an ability to access and interpret data in a holistic, tactical fashion. According to some embodiments, the system may be agnostic regarding particular web browsers, sources of information, etc. For example, information from multiple sources (e.g., an internal insurance policy database and an external data store) might be blended and combined (with respect to reading and/or writing operations) so as to appear as a single “pool” of information to a user at a remote device. Moreover, embodiments may be implemented with a modular, flexible approach such that deployment of a new system for an enterprise might be possible relatively quickly. Embodiments may improve cost avoidance, improve employee wellness, reduce gaps in healthcare, lead to higher retention, smarter plan designs, and overall better health outcomes.
The following illustrates various additional embodiments of the invention. These do not constitute a definition of all possible embodiments, and those skilled in the art will understand that the present invention is applicable to many other embodiments. Further, although the following embodiments are briefly described for clarity, those skilled in the art will understand how to make any changes, if necessary, to the above-described apparatus and methods to accommodate these and other embodiments and applications.
Although specific hardware and data configurations have been described herein, note that any number of other configurations may be provided in accordance with embodiments of the present invention (e.g., some of the information associated with the displays described herein might be implemented as a virtual or augmented reality display and/or the databases described herein may be combined or stored in external systems). Moreover, although embodiments have been described with respect to particular types of insurance policies, embodiments may instead be associated with other types of insurance policies in addition to and/or instead of the policies described herein (e.g., business insurance policies, automobile insurance policies, etc.). Similarly, although certain attributes were described in connection some embodiments herein, other types of attributes might be used instead. Still further, the displays and devices illustrated herein are only provided as examples, and embodiments may be associated with any other types of user interfaces. For example,
The present invention has been described in terms of several embodiments solely for the purpose of illustration. Persons skilled in the art will recognize from this description that the invention is not limited to the embodiments described, but may be practiced with modifications and alterations limited only by the spirit and scope of the appended claims.
The present application is a continuation of U.S. patent application Ser. No. 17/847,339 entitled “DATA ANALYTICS SYSTEM TO AUTOMATICALLY RECOMMEND RISK MITIGATION STRATEGIES FOR AN ENTERPRISE” and filed on Jun. 23, 2022, which was a continuation of U.S. patent application Ser. No. 17/061,919 entitled “DATA ANALYTICS SYSTEM TO AUTOMATICALLY RECOMMEND RISK MITIGATION STRATEGIES FOR AN ENTERPRISE” and filed on Oct. 2, 2020 which claimed priority to U.S. Patent Application No. 62/923,759 entitled “DATA ANALYTICS SYSTEM TO AUTOMATICALLY RECOMMEND RISK MITIGATION STRATEGIES FOR AN ENTERPRISE” and filed on Oct. 21, 2019. The entire contents of those applications are incorporated herein by reference.
| Number | Name | Date | Kind |
|---|---|---|---|
| 6023578 | Birsan | Feb 2000 | A |
| 7003482 | Margoscin | Feb 2006 | B1 |
| 7856617 | Szlam | Dec 2010 | B2 |
| 8065168 | Szlam | Nov 2011 | B2 |
| 8453138 | Meller | May 2013 | B2 |
| 8478769 | Goldfarb | Jul 2013 | B2 |
| 8904349 | Hudson, Jr. | Dec 2014 | B2 |
| 20030158760 | Kannenberg | Aug 2003 | A1 |
| 20040103124 | Kupkova | May 2004 | A1 |
| 20060010032 | Eicher | Jan 2006 | A1 |
| 20070150319 | Menendez | Jun 2007 | A1 |
| 20080235064 | Gulko | Sep 2008 | A1 |
| 20110072416 | Hudson, Jr. | Mar 2011 | A1 |
| 20130151631 | Jensen | Jun 2013 | A1 |
| Entry |
|---|
| Going Digital: A Survey on Digitalization and Large-Scale Data Analytics in Healthcare; Proceedings of the IEEE (vol. 104, Issue: 11, pp. 2180-2206); Volker Tresp, J. Marc Overhage, Markus Bundschus, Shahrooz Rabizadeh, Peter A. Fasching, Shipeng Yu, Oct. 19, 2016. (Year: 2016). |
| Deep analytics for workplace risk and disaster management; IBM Journal of Research and Development (vol. 64, Issue: 1/2, pp. 14:1-14:9); S. Dalal, D. Bassu; Oct. 11, 2019. (Year: 2019). |
| Number | Date | Country | |
|---|---|---|---|
| 20230325780 A1 | Oct 2023 | US |
| Number | Date | Country | |
|---|---|---|---|
| 62923759 | Oct 2019 | US |
| Number | Date | Country | |
|---|---|---|---|
| Parent | 17847339 | Jun 2022 | US |
| Child | 18329054 | US | |
| Parent | 17061919 | Oct 2020 | US |
| Child | 17847339 | US |
