TWI819584B - Insurance claims risk investigation methods and systems, computer-readable recording media - Google Patents

Abstract

An insurance claims risk investigation method, which uses a risk investigation device to detect the pending information of a pending claim case and the policyholder's historical data to retrieve a risk reminder content corresponding to at least one risk reminder tag, and after the detection, When a pending claim settlement case has at least one risk event, at least one corresponding content to be confirmed is generated, and at least one risk reminder label and at least one risk event are displayed. When the risk detection device determines that at least one risk reminder label is clicked, the corresponding content is displayed. Risk warning content, and when the risk detection device determines that at least one risk item is clicked, the corresponding at least one content to be confirmed is displayed for confirmation, and a confirmation result of the at least one content to be confirmed is recorded in the pending claim case. .

Description

Insurance claims risk investigation methods and systems, computer-readable recording media

The present invention relates to a method for controlling and managing insurance claim risks, and in particular, to a method for investigating insurance claim risks.

When current claims adjusters review insurance claim cases of policyholders, although they can use the computer system to execute a risk judgment model constructed by artificial intelligence (AI) in advance to determine whether the risk of the case falls into one of three categories: high, medium, or low, they still cannot It is necessary to judge the next step of the trial based on the personal experience of the claims adjuster, or it is necessary to search the system by oneself to confirm the reason why the case is judged to be high risk. However, this method of operation has no unified approach except that the judgment criteria for hearing cases vary from person to person, and it cannot accurately detect the risk of claims settlement.

Therefore, the purpose of the present invention is to provide an insurance claims risk detection method that can at least solve the above problems, as well as an insurance claims risk detection system and a computer-readable recording medium that implement the method.

Therefore, the present invention provides an insurance claims risk investigation method, including: (A) a database device that stores a pending claim case and information related to the pending claim case. A historical data related to the policyholder; (B) When a risk detection device reads the pending claim case from the database device, and when the risk detection device determines that at least one risk warning label is recorded in the historical data, it reads the historical data from the historical data. The at least one risk warning label and a risk warning content corresponding to the at least one risk warning label are retrieved from the data; (C) the risk detection device is based on a pending information included in the pending claim case, the historical data and advance According to the set abnormality judgment rules, when it is detected that the pending claim settlement case has at least one of multiple risk events, at least one content to be confirmed corresponding to the detected at least one risk event is generated; (D) the risk detection device displays the pending information , the at least one risk reminder label and at least one detected risk event; (E) when the risk detection device determines that the displayed at least one risk reminder label is clicked, display the risk reminder content corresponding to the at least one risk reminder label ; and (F) when the risk detection device determines that the at least one risk event displayed is clicked, display the at least one pending content corresponding to the selected at least one risk event for confirmation, and display the at least one pending confirmation content A confirmation result of the content is recorded in the pending claim case.

In addition, the present invention implements an insurance claims risk investigation system that implements the above method, including a database device and a risk investigation device; the database device stores a pending claim settlement case and a policy related to the pending claim settlement case. Historical data; the risk detection device is electrically connected to the database device to access the database device; wherein, when the risk detection device is operated to read the pending claim case from the database device, the risk detection device A risk detection module determines that in the historical data When at least one risk warning label is recorded, the at least one risk warning label and a risk warning content corresponding to the at least one risk warning label are retrieved from the historical data; the risk detection module is based on the information contained in the pending claim case. 1. Based on the pending information, the historical data and the pre-set exception judgment rules, when it is detected that the pending claim case has at least one of multiple risk events, at least one content to be confirmed corresponding to the detected at least one risk event will be generated; the risk The detection module causes the risk detection device to display the pending information, the at least one risk warning label and the at least one detected risk event, and the risk detection module determines that when the displayed at least one risk warning label is clicked, The risk detection device displays the risk prompt content corresponding to the at least one risk warning label; when the risk detection module determines that the displayed at least one risk item is clicked, the risk detection device displays the selected at least one risk The at least one content to be confirmed corresponding to the matter is provided for confirmation, and a confirmation result of the at least one content to be confirmed is recorded in the pending claim case.

In some implementations of the present invention, the at least one risk warning label corresponds to at least one fact record in the historical data, and the risk warning content is related to the at least one fact record.

In some implementations of the present invention, the risk items include but are not limited to policyholders, salespersons, diseases, hospitals/physicians, and each risk item has corresponding abnormality judgment rules.

In some implementation aspects of the present invention, the at least one content to be confirmed includes Contains an agree option and a disagree option. When the agree option is clicked, the confirmation result is agree; when the disagree option is clicked, the risk detection module causes the risk detection device to wait for at least one The confirmation content further displays an input field for inputting a reason for disagreement. After the risk detection module confirms that the reason for disagreement is filled in the input field, the disagreement and the reason for disagreement are included in the confirmation result.

In some implementation aspects of the present invention, the risk detection module also causes at least one keyword in the at least one content to be confirmed to be highlighted and available for click, and when the at least one keyword is clicked, the risk The detection module causes the risk detection device to display information related to the keyword.

Furthermore, the present invention implements the above method in a computer-readable recording medium, which stores a software program including a risk detection module, and the software program is loaded and loaded by a risk detection device that can access a database device. When executed, the risk detection device can complete the insurance claim risk detection method as described above; wherein, the database device stores a pending claim case and a historical data related to the policyholder of the pending claim case.

The effect of the present invention is to: use the risk detection device to automatically detect potential claims risks of a pending claim case based on historical data related to the policyholder of the pending claim case, and display the pending claims risk of the pending claim case. While reviewing the information, the detected risk warning labels and risk matters related to the policyholder's past abnormal records or bad records are also displayed for the claims reviewers to refer to and confirm. Claims reviewers not only do not need to search the system to confirm case risks on their own, but also provide unified judgment standards for cases and enable claims reviewers to more accurately determine claims risks.

100:Insurance claims risk detection system

1: Database device

2: Risk detection device

21:Storage unit

211:Risk Detection Module

22: Processing unit

23:Display

31: Information pending review

32: Risk warning content

33, 34: Content to be confirmed

51: Details of claims related records

61: Content to be confirmed

71: Comparison table

81:Content to be confirmed

82: Statistical information on medical charges

91:Content to be confirmed

92: Statistical information on medical charges

101: Content to be confirmed

S1~S13: steps

Other features and effects of the present invention will be clearly shown in the embodiments with reference to the drawings, in which: Figure 1 shows the process steps of an embodiment of the insurance claims risk detection method of the present invention; Figure 2 shows the method for implementing Figure 1 An embodiment of the process of the insurance claims risk detection system of the present invention mainly includes hardware devices, components and software modules; Figure 3 shows the information to be reviewed, at least one risk warning label and at least one detected risk event in this embodiment and the content to be confirmed are displayed on the same screen; Figure 4 is a schematic diagram of the content to be confirmed among the "household" risk items in this embodiment; Figure 5 is a schematic diagram of the claim-related record details in this embodiment; Figure 6 is a schematic diagram of a content to be confirmed among the "disease" risk items in this embodiment; Figure 7 is a schematic diagram of a comparison table in this embodiment, which shows the hospitalization threshold days for people of different genders and different age ranges who suffer from the disease. ; Figure 8 shows one of the "hospital/physician" risk items in this embodiment for the hospital. A schematic diagram of the confirmation content; Figure 9 is a schematic diagram of the content to be confirmed for doctors among the "hospital/physician" risk items in this embodiment; and Figure 10 is a content to be confirmed among the "salesperson" risk items in this embodiment schematic diagram.

Before the present invention is described in detail, it should be noted that in the following description, similar elements are designated with the same numbering.

Refer to Figure 1, which is the main process of an embodiment of the insurance claims risk detection method of the present invention, and this method is implemented by an insurance claims risk detection system 100 shown in Figure 2. The insurance claims risk detection system 100 mainly includes A database device 1 and a risk detection device 2 capable of communicating with the database device 1 (for example, through an electrical connection or through a network) to access the database device 1 .

The database device 1 may be a remote (or cloud) server, which stores a plurality of pending claim cases and a historical data related to the policyholder of each pending claim case. The historical data includes but does not It is limited to the policyholder’s personal information, all policies, the policy’s related parties (the applicant/insured/beneficiary/first recruiter (salesperson)), the policyholder’s past claims records, and the claim settlement records of the policyholder’s family members. ..wait.

The risk investigation device 2 is a terminal computer device, which is operated and used by claims review personnel of insurance companies to review pending claims cases. the The risk detection device 2 mainly includes a storage unit 21 (such as a computer-readable recording medium composed of a memory), a processing unit 22 (such as a central processing unit or a microprocessor) electrically connected to the storage unit 21, and a The display 23 is electrically connected to the processing unit 22, as well as related or necessary components required to perform computer operations. The storage unit 21 stores a software program including a risk detection module 211, and the processing unit 22 can access the storage unit 21 to read, load and execute the software program. Alternatively, the risk detection module 211 may be integrated into a special application integrated circuit chip or a programmable logic circuit, and the processing unit 22 is the special application integrated circuit chip or the programmable logic circuit. Alternatively, the risk detection module 211 can also be firmware burned into a microprocessor, and the processing unit 22 is the microprocessor.

Thereby, when the claim adjuster operates the risk detection device 2 and causes the processing unit 22 to execute one of the above-mentioned pre-installed software programs (such as a claims operation program), the processing unit 22 causes the display 23 to display a user ( operation) interface (not shown), and the claims adjuster instructs the processing unit 22 to read a pending claim case from the database device 1 through the user interface, the processing unit 22 reads the pending claim case At the same time, the risk detection module 211 will be executed.

First, as shown in step S1 of FIG. 1 , when the risk detection device 2 reads the pending claim case from the database device 1 , the risk detection module 211 will cause the processing unit 22 to also read the pending claim case. The historical data of the policyholder, and then, in step S2 of Figure 1, the risk detection module 211 determines whether the historical data is Whether at least one risk warning label is recorded, if so, in step S3 of Figure 1 , the risk detection module 211 retrieves the at least one risk warning label and a risk corresponding to the at least one risk warning label from the historical data Prompt content. Among them, the at least one risk warning label may be, but is not limited to, risk labels such as "Recent Complaint", "Control Policyholder", etc., and these risk warning labels are pre-determined by the system or manually based on the past facts of the policyholder or his family. Records (such as bad claims records or complaint records or professional records, etc.) are recorded in the historical data of the policyholder by labeling and recording the corresponding risk warning content.

At the same time, as shown in step S4 of Figure 1 , the risk detection module 211 detects (judges) the pending claim case based on a pending information contained in the pending claim case, the historical data and at least one preset abnormality judgment rule. Whether there is at least one of multiple risk events, if so, in step S5 of Figure 1 , the risk detection module 211 obtains at least one detected risk event and generates at least one piece of content to be confirmed, and then, in step S5 of Figure 1 S6, the risk detection module 211 causes the display 23 of the risk detection device 2 to display the pending information, the at least one risk warning label, and the detected at least one risk event and its content to be confirmed on the display screen; wherein, These risk events can be, but are not limited to, the four risk events of "household", "disease", "hospital/physician" and "clerk", and each risk event may have several or at least one content to be confirmed.

Take the screen shown in Figure 3 as an example, which shows a pending claim case. The pending information 31, a risk warning label, such as "Recent Complaint", "Household" risk item has 2 items to be confirmed, "Disease" risk item has 1 item to be confirmed, "Hospital/Physician" risk item There is 1 content to be confirmed, and there is 1 content to be confirmed for the "salesman" risk item. And as shown in step S7 of Figure 1, when the risk detection module 211 determines that the risk prompt label "Recent Complaint" is clicked, then in step S8 of Figure 1, the risk corresponding to the risk prompt label "Recent Complaint" is displayed. The prompt content 32 is as shown in Figure 3, and as shown in step S9 of Figure 1, the risk detection module 211 determines whether the risk prompt label "Recent Complaints" is still clicked, and if so, returns to step S8 and continues The risk prompt content 32 is displayed. If not, return to step S6. In this way, the risk warning content 32 can be used as a reference by claims review personnel when reviewing the pending information.

Moreover, the risk detection module 211 detects whether there are various risk events in the pending information and the historical data based on an abnormality judgment rule corresponding to various risk events; for example, for the "household" risk event, the The abnormality judgment rules may include but are not limited to, for example, judging whether the policyholder's (insured's) household's recent (for example, within 1 year) surgery claims record has reached a threshold (for example, 3 times) based on the pending information and the historical data. If so, it is determined that the pending claim case has "household" risk matters, and corresponding content to be confirmed related to "household" risks will be generated. Therefore, in step S10 of FIG. 1 , the risk detection module 211 determines that when the option “Household” risk event is selected, as shown in step S11 of FIG. 1 and FIG. 3 , the risk detection module 211 makes the The display 23 displays one of the unconfirmed contents 33 of the "household" risk event.

In addition, as shown in Figure 5, the risk detection module 211 will highlight the keywords in the content to be confirmed 33, such as the number of operations "4 ", for click, and when " 4 " is clicked, the The risk detection module 211 will sort out a claim-related record detail 51 of the family member who performed the surgery from the historical data and cause the display 23 to display it, as shown in Figure 5 . In this way, the claims reviewer can determine whether there is a problem of excessive applications from each household based on the claim-related record details 51.

As shown in Figures 3 and 4, the content 33 to be confirmed will provide an agree option and a disagree option for clicking, and as shown in step S12 of Figure 1, if the agree option is clicked, then the The risk detection module 211 generates a confirmation result containing "agree" and records it in the pending claim case; and if the disagree option is clicked, as shown in Figure 4, the risk detection module 211 will The confirmation content further displays an input field 41 for inputting a reason for disagreement. After confirming that the reason for disagreement is filled in the input field 41, the risk detection module 211 adds "disagree" and the reason for disagreement. included in the confirmation result, and then record the confirmation result in the pending claim case; then, in step S13 of Figure 1 , the risk detection module 211 determines whether all risk matters have been confirmed, and if so, then End the risk investigation process. If not, repeat the above steps S10~S13.

For "disease" risk matters, the abnormality judgment rules may include but are not limited to, for example, judging whether the diagnosed disease name is a single diagnosis, whether it is a non-severe or chronic disease, and whether the number of hospitalization days exceeds a threshold based on the pending information and the historical data. number of days, if the above conditions are met (yes), the risk detection module 211 determines that the pending claim case has a "disease" risk event, and correspondingly generates unconfirmed content related to the "disease"risk; therefore, as shown in Figure 1 In step S10, the risk detection module 211 determines that when the "disease" risk event option is selected, as shown in step S11 of Figure 1 and Figure 6, the risk detection module 211 causes the display 23 to display "disease" One of the risk events to be confirmed is 61. Among them, the risk detection module 211 will use the disease name "486 pneumonia " and the age range "20-64 " as keywords for clicks. When the disease name "486 pneumonia " is clicked, the risk detection module 211 will cause the display 23 to display a dashboard of days for the disease (not shown in the figure), which displays the general distribution of hospitalization days for the disease, and when the age range "20-64 " is clicked, the risk detection module 211 The display 23 will display a comparison table 71 as shown in Figure 7, which displays the hospitalization threshold days for people of different genders and different age ranges who have the disease; thereby, the claim reviewer can judge whether the patient is related to the disease. "Risk-related content to be confirmed 61, and click the agree option or disagree option to make a decision of "agree" or "disagree".

For the hospital part of the "hospital/physician" risk matter, the abnormal judgment rules can include but are not limited to, for example, the current disease treatment at clinic A , the receipt fee is more than 20,000, the current insurance within 2 years, and the insured amount Accumulating 3,000 (yuan) or more and other conditions, when all the above conditions are met, the risk investigation module 211 determines that the pending claim case has "hospital/physician" risk matters, and correspondingly generates claims related to "hospital/physician" risk Content to be confirmed; therefore, as shown in step S10 of Figure 1 , the risk detection module 211 determines that when the option "hospital/physician" risk event is selected, as shown in step S11 of Figure 1 and Figure 8 , the risk detection module 211 The module 211 causes the display 23 to display a content to be confirmed 81 of the "hospital/physician" risk event; and in the content to be confirmed 81 shown in Figure 8, the risk detection module 211 makes " Clinic A" a keyword and Available for click. When " Clinic A" is clicked, as shown in Figure 8, the risk detection module 211 will cause the display 23 to display a statistical information 82 on medical treatment charges related to Clinic A for the reference of claims reviewers.

However, for the physician part of the "hospital/physician" risk item, the abnormal judgment rules may include but are not limited to, for example, for this disease diagnosis , Dr. Wang OO of the Urology Department of Hospital B performed YY surgery, and the receipt of the surgery fee was more than 20,000 yuan. Or the medical expenses are more than 10,000, the accident happened within 2 years, and there are paid medical insurance and other conditions. When all the above conditions are met, the risk detection module 211 determines that the pending claim case has "hospital/physician""Risk matters, and corresponding content to be confirmed related to the "Hospital/Physician" risk is generated; therefore, as shown in step S10 in Figure 1, the risk detection module 211 determines when the option of the "Hospital/Physician" risk event is clicked , as shown in step S11 of FIG. 1 and FIG. 9 , the risk detection module 211 causes the display 23 to display a pending confirmation content 91 of the "hospital/physician" risk event, for example, as shown in FIG. 9 . And in the content to be confirmed 91 shown in Figure 9, the risk detection module 211 makes " Dr. Wang OO, Department of Urology, Hospital B" as a keyword and available for clicking. When " Dr. Wang OO, Department of Urology, Hospital B" is clicked At this time, the risk detection module 211 will cause the display 23 to display a medical fee statistical information 92 related to Dr. Wang OO of Department XX of Hospital B for the reference of the claim review personnel.

For "salesperson" risk matters, the abnormal judgment rules may include but are not limited to, for example, the customer's XX surgery risk probability is greater than a threshold or N customers have applied for XX surgery claims recently (for example, within 1 year). If it meets Based on the aforementioned conditions, the risk investigation module 211 determines that the pending claims case has a "salesperson" risk event, and correspondingly generates content to be confirmed related to the "salesperson" risk; therefore, as shown in step S10 of Figure 1 , the risk When the detection module 211 determines that the option of the "salesperson" risk event is clicked, as shown in steps S11 and 10 of Figure 1, the risk detection module 211 causes the display 23 to display the "salesperson" risk event. A to-be-confirmed content 101 is provided for confirmation by the claims reviewer.

And in the above-mentioned step S12, one of the consent options and disagree options in the above-mentioned content to be confirmed 33, 34, 61, 81, 91, 101 must be clicked by the (claims reviewer) to make "agree" or "Disagree" decision, and after clicking the Disagree option, the reason for the disagreement needs to be entered to generate the corresponding confirmation result for the risk detection module 211 to record in the pending claim case.

In addition, when the judgment in step S2 is yes but the judgment in step S4 is no, only the pending information and the at least one risk warning label will be displayed in step S6; and when the judgment in step S2 and step S4 is If both are negative, only the pending information will be displayed in step S6, and steps S7 to S13 will not be executed.

To sum up, in the above embodiment, the risk detection device 2 is based on a Historical data related to the policyholder of the pending claim case, automatically detects the potential claims risks of the pending claim case, and displays the detected risk warning label and the pending information of the pending claim case at the same time. Risk matters related to the policyholder's past abnormal records or bad records are provided for claims reviewers to refer to and confirm. This not only eliminates the need for claims reviewers to search the system themselves to confirm case risks, but also provides unified judgment standards for hearing cases. This enables claims reviewers to more accurately determine the risk of claim settlement, thereby achieving the effects and purposes of the present invention.

However, the above are only examples of the present invention, and should not be used to limit the scope of the present invention. All simple equivalent changes and modifications made based on the patent scope of the present invention and the content of the patent specification are still within the scope of the present invention. Within the scope covered by the patent of this invention.

S1~S13: steps

Claims (11)

An insurance claims risk investigation method, including: (A) A database device stores a pending claim case and a historical data related to the policyholder of the pending claim case; (B) When a risk detection device reads the pending claim case from the database device, and when the risk detection device determines that at least one risk warning label is recorded in the historical data, it retrieves the at least one risk from the historical data. A warning label and a risk warning content corresponding to the at least one risk warning label; (C) When the risk detection device detects that the pending claim case has at least one of a variety of risk matters based on the pending information contained in the pending claim case, the historical data and the preset abnormality judgment rules, it will target the detected At least one risk event generates at least one corresponding content to be confirmed; (D) The risk detection device displays the pending information, the at least one risk warning label and the at least one detected risk event; (E) When the risk detection device determines that the displayed at least one risk warning label is clicked, display the risk warning content corresponding to the at least one risk warning label; and (F) When the risk detection device determines that the at least one risk item displayed is clicked, the at least one content to be confirmed corresponding to the at least one risk item that is clicked is displayed for confirmation, and the at least one content to be confirmed is displayed. A confirmation result is recorded in the pending claim case. The insurance claims risk investigation method as described in claim 1, wherein the at least one risk warning label corresponds to at least one fact record in the historical data, and the risk warning content is related to the at least one fact record. The insurance claims risk investigation method as described in claim 1, wherein the risk items include policyholders, salespersons, diseases, hospitals/physicians, and each risk item has corresponding abnormality judgment rules. The insurance claims risk investigation method as described in any one of claims 1 to 3, wherein the at least one content to be confirmed includes an agree option and a disagree option. When the agree option is clicked, the confirmation result is Agree; when the disagree option is clicked, the risk detection device further displays an input field in the content to be confirmed for inputting a reason for disagreement, and the risk detection device confirms that the reason for disagreement is filled in the input field. Afterwards, the disagreement and the reasons for the disagreement will be included in the confirmation result. The insurance claims risk detection method as described in any one of claims 1 to 3, wherein the risk detection device also causes at least one keyword in the at least one content to be confirmed to be highlighted and available for click, and the at least one keyword is highlighted and available for click. When a keyword is clicked, the risk detection device displays information related to the keyword. An insurance claims risk investigation system, including: A database device that stores a pending claim case and a historical data related to the policyholder of the pending claim case; and A risk detection device electrically connected to the database device to access the database device; wherein When the risk detection device is operated to read the pending claim case from the database device, when a risk detection module of the risk detection device determines that at least one risk warning label is recorded in the historical data, it reads the pending claim case from the historical data. Retrieve the at least one risk warning label and a risk warning content corresponding to the at least one risk warning label; The risk detection module detects that the pending claim case has at least one of multiple risk matters based on the pending information, the historical data and the preset abnormality judgment rules contained in the pending claim case, and targets at least one of the detected The risk event generates at least one content to be confirmed corresponding to the risk event; The risk detection module causes the risk detection device to display the pending information, the at least one risk warning label and the at least one detected risk event, and the risk detection module determines that the displayed at least one risk warning label is clicked. , causing the risk detection device to display the risk warning content corresponding to the at least one risk warning label; when the risk detection module determines that the displayed at least one risk item is clicked, causing the risk detection device to display the selected at least one risk item. The at least one content to be confirmed corresponding to a risk event is provided for confirmation, and a confirmation result of the at least one content to be confirmed is recorded in the pending claim case. The insurance claims risk investigation system as described in claim 6, wherein the at least one risk warning label corresponds to at least one fact record in the historical data, and the risk warning content is related to the at least one fact record. An insurance claims risk investigation system as described in claim 6, wherein the risk items are policyholders, salespersons, diseases, hospitals/doctors, and each risk item has corresponding abnormality judgment rules. The insurance claims risk investigation system as described in any one of claims 6 to 8, wherein the at least one content to be confirmed includes an agree option and a disagree option. When the agree option is clicked, the confirmation result is Agree; when the disagree option is clicked, the risk detection module causes the risk detection device to further display an input field for inputting a reason for disagreement in the at least one content to be confirmed, and the risk detection module confirms the disagreement. After the reason for consent is filled in the input field, the disagreement and the reason for the disagreement will be included in the confirmation result. The insurance claims risk investigation system as described in any one of claims 6 to 8, wherein the risk investigation module also causes at least one keyword in the at least one content to be confirmed to be highlighted and available for click, and the When at least one keyword is clicked, the risk detection module causes the risk detection device to display information related to the keyword. A computer-readable recording medium that stores a software program including a risk detection module. When the software program is loaded and executed by a risk detection device that can access a database device, the risk detection device can complete The insurance claim risk investigation method as described in any one of claims 1 to 5; wherein the database device stores a pending claim case and a historical data related to the policyholder of the pending claim case.