KR102085814B1 - Device and method for new unfair claim pattern analysis based on artificial intelligence - Google Patents

Abstract

Disclosed is a device and method for analyzing a new invalidation pattern. The apparatus for analyzing a new invalidation pattern according to an exemplary embodiment of the present invention includes a data preprocessor for deriving characteristic variables by formulating insurance claim data and inputting the characteristic variables. An unfair billing pattern classification unit classifying a new pattern of unfair claims based on a clustering algorithm based on unsupervised learning, a discriminant model construction unit for constructing a new pattern discrimination model based on a supervised learning-based decision algorithm, and the new pattern discrimination It may include a new pattern determination unit for determining a new pattern for the new claim based on the model.

Description

DEVICE AND METHOD FOR NEW UNFAIR CLAIM PATTERN ANALYSIS BASED ON ARTIFICIAL INTELLIGENCE}

The present invention relates to a novel artificial claim pattern analysis apparatus and method based on artificial intelligence.

The existing insurance fraud prevention system calculates rules based on business rules based on the reviewer's experience and knowledge, and divides them into investigation rules and exclusion rules. And as it is advanced, it is necessary to continuously update new fraudulent patterns or fraud patterns.

In addition, insurers cannot leave a claim without taking any action on insurance fraud, and cannot spend an infinite amount of investigation on the purpose of fully revealing the fraud. Thus, insurance companies need to reduce the amount of leakage from insurance fraud and determine the level of investigation efforts on a reasonable basis without spending too much investigative expenditure.

Background art of the present application is disclosed in Korean Patent Publication No. 10-0862181.

The present invention is to solve the above-mentioned problems of the prior art, an object of the present invention to provide an artificial intelligence-based new fraud pattern analysis apparatus and method for providing a discrimination model that can determine the new pattern of claims fraudulent claims.

The present application is to solve the above-mentioned problems of the prior art, the artificial intelligence-based new fraudulent pattern analysis that can analyze the fraudulent claim data by analyzing the new fraudulent patterns, and determine the pattern of the claim pattern for a new claim It is an object to provide an apparatus and method.

However, the technical problem to be achieved by the embodiments of the present application is not limited to the technical problems as described above, and other technical problems may exist.

As a technical means for achieving the above technical problem, the new unfair claim pattern analysis apparatus according to an embodiment of the present application is a data pre-processing unit for deriving the characteristic variable by formulating the insurance claim history data, the busy busy as the input of the characteristic variable An unfair billing pattern classifying unit classifying a new pattern of unfair claims based on a group learning algorithm based on a degree learning, a discriminant model construction unit for constructing a new pattern discrimination model based on a supervised learning-based decision algorithm, and the new pattern discriminating model It may include a new pattern determination unit for determining the pattern type of the new claim based on.

According to an embodiment of the present application, the claim history data, the characteristic variable and the database further records at least one of the new pattern, wherein the invalid claim history data billing data, contract data, payment data, insurance company It may include at least one of data, customer data.

According to one embodiment of the present application, the unfair claim pattern classification unit, based on the frequency of the feature variable clusters the feature variable into a plurality of claim patterns through the clustering algorithm, based on the degree of separation between the cluster of the claim pattern The new pattern can be detected.

According to one embodiment of the present application, the unfair claim pattern classification unit may classify the new pattern into a plurality of similar pattern groups based on the similarity between the new patterns.

According to one embodiment of the present application, the unfair claim pattern classifier, the risk level of the pattern group is set on the basis of the number of the characteristic variable above the threshold value preset to each of the characteristic variables associated with each of the new pattern; Can be.

According to the exemplary embodiment of the present application, the discrimination model building unit learns a new pattern discrimination rule based on a decision algorithm for inputting the new pattern included in the pattern group for each risk level, and the new pattern discrimination rule. The new pattern discrimination model may be constructed.

According to one embodiment of the present application, the new pattern determination unit may determine whether the new claim is a pattern type of the normal claim pattern, unfair claim pattern and new pattern based on the new pattern determination rule.

The new fraud pattern analysis method according to an embodiment of the present invention comprises the steps of deriving a feature variable by formulating insurance claim history data, and based on an unsupervised learning-based clustering algorithm using the feature variable as a new method of fraudulent claim. The method may include classifying a pattern, constructing a new pattern discrimination model based on a supervised learning-based decision algorithm, and determining a new pattern for a new claim based on the new pattern discrimination model.

According to an embodiment of the present application, the method further comprises the step of recording at least one of the claim data, the characteristic variable and the new pattern, wherein the claim data, claim data, contract data, payment data, insurance agent It may include at least one of data, customer data.

According to an embodiment of the present disclosure, the step of classifying the new pattern of the unfair claims, clustering the feature variable into a plurality of claim patterns through the clustering algorithm based on the frequency of the feature variable, clustering of the claim pattern The novel pattern can be detected based on liver separation.

According to an embodiment of the present disclosure, in the step of classifying the new pattern of the illegal claim, the new pattern may be classified into a plurality of similar pattern groups based on the similarity between the new patterns.

According to an embodiment of the present disclosure, the step of classifying the new pattern of the unfair claims, the pattern group based on the number of feature variables above the threshold value preset to each of the feature variables associated with each of the new patterns You can set the risk level.

According to an embodiment of the present disclosure, the building of the new pattern determination model may include learning a new pattern determination rule based on a decision algorithm that inputs the new pattern included in the pattern group for each risk level, The new pattern discrimination model including the new pattern discrimination rule may be constructed.

According to an exemplary embodiment of the present disclosure, the determining of the new pattern may include determining whether the new claim is a pattern type of a normal claim pattern, an invalid claim pattern, and a new pattern based on the new pattern determination rule.

The above-mentioned means for solving the problems are merely exemplary, and should not be construed as limiting the present application. In addition to the above-described exemplary embodiments, additional embodiments may exist in the drawings and detailed description of the invention.

According to the aforementioned problem solving means of the present application, it is possible to provide an artificial intelligence-based new fraud pattern analysis apparatus and method for providing a discrimination model that can determine a new pattern of claims fraudulent claims.

According to the above-described problem solving means of the present invention, it provides a novel artificial claim pattern analysis apparatus and method based on artificial intelligence capable of learning the new claim data to analyze the new claim pattern, and determine the pattern type for the new claim can do.

1 is a view showing the configuration of a new unfair claim pattern analysis apparatus according to an embodiment of the present application.
2 is a diagram illustrating an example of a new pattern detection of the new invalid claim pattern analysis apparatus according to an embodiment of the present application.
3 is a diagram illustrating an example of similar pattern group classification of the apparatus for analyzing a new invalid pattern according to an embodiment of the present application.
Figure 4 is a diagram showing the flow of a new invalid claim pattern analysis method according to an embodiment of the present application.

DETAILED DESCRIPTION Hereinafter, exemplary embodiments of the present disclosure will be described in detail with reference to the accompanying drawings so that those skilled in the art may easily implement the present disclosure. As those skilled in the art would realize, the described embodiments may be modified in various different ways, all without departing from the spirit or scope of the present invention. In the drawings, parts irrelevant to the description are omitted for simplicity of explanation, and like reference numerals designate like parts throughout the specification.

Throughout this specification, when a part is "connected" to another part, this includes not only "directly connected" but also "electrically connected" with another element in between. do.

Throughout this specification, when a member is said to be located on another member "on", "upper", "top", "bottom", "bottom", "bottom", this means that any member This includes not only the contact but also the case where another member exists between the two members.

Throughout this specification, when a part is said to "include" a certain component, it means that it can further include other components, without excluding the other components unless otherwise stated.

1 is a view showing the configuration of a new unfair claim pattern analysis apparatus according to an embodiment of the present application. Referring to FIG. 1, the new fraudulent billing pattern analysis apparatus 100 includes a data preprocessor 110, a fraudulent billing pattern classification unit 120, a discriminant model construction unit 130, a new pattern discrimination unit 140, and a database ( 150). The data preprocessor 110 may derive the characteristic variable by shaping the insurance claim history data, which is a pattern analysis target of the new unfair claim.

The description below will include not only claims unfair claims, but also claims, subsidies, security deposits, etc., but will be described based on claims unfair claims for convenience of explanation. For example, the insurance claim data may include at least one of claim data, insurance or subsidy-related contract data, insurance payment data, insurance company data, customer data, and insurance fraud detection result data. The data preprocessor 110 may derive the characteristic variable by formalizing the insurance claim data. For example, in the case of the customer data, the standardization of the insurance claim data means to quantify the income level of the customer, the number of hospitals visited by the customer, the number of reasons for the disease applied by the customer, and the like. In addition, the characteristic variable derived by the data preprocessing unit 110 is a variable that may have a digitized value in relation to a claim, for example, a customer ID number, whether or not an insurance fraud, the number of duplicate applications under the same disease name, contract The number of months signed, the maximum number of contracts per day, the number of securities approved for payment, the number of securities applied for payment, the number of securities entered by the customer, the number of claims for insurance coverage, the change in credit rating, the type of contracts signed up, the total number of hospital visits, and customers The number of reasons for the illness, the number of doctors the client met, the number of hospitals the client visited, the total number of days of effective hospitalization, the number of treatment subjects, the level of the client's income, the number of financial planner changes, the number of deficits, and the fraud It may include at least one of the number of FP contracts. The database 150 may record at least one of claims history data, the characteristic variable, and the new pattern. In addition, the database 150 may store a normal billing pattern.

The unfair billing pattern classification unit 120 may classify a new pattern of unfair billing based on an unsupervised learning-based clustering algorithm using a feature variable as an input. Unsupervised learning refers to an algorithm that learns by analyzing or clustering data itself, rather than constructing learning data. Since this is a known matter, a detailed description thereof will be omitted. The unfair claim pattern classifying unit 120 may cluster the feature variables into a plurality of claim patterns based on the frequency of the claim details data. For example, the unfair claim pattern classifying unit 120 may classify a new pattern of unfair claims based on at least one of a K-means clustering algorithm, a self-organizing-maps (SOM) algorithm, and an EM & Canopy algorithm. . The K-means clustering algorithm is a traditional classification technique that repeats the target group into K clusters based on the mean value (similarity) of the distance, and the SOM algorithm learns the input patterns of training sets by weight based on artificial neural networks. Clustering technique. In addition, EM & Canopy algorithm refers to a technique for clustering by updating the parameter value through the iterative process from the maximum probability to the given initial value.

The invalid claim pattern classification unit 120 may cluster the feature variables into a plurality of claim patterns through the clustering algorithm based on the frequency of the feature variables. In other words, the claim pattern may be a cluster of homogeneous feature variables. In addition, the new pattern may be detected based on the degree of separation between clusters of the claimed pattern, that is, the distance between the clusters.

2 is a diagram illustrating an example of a new pattern detection of the new invalid claim pattern analysis apparatus according to an embodiment of the present application. FIG. 2 shows the billing patterns clustered according to the frequency of the feature variables, and if the frequencies of the feature variables belonging to different claim patterns are similar, they may be divided into the same to similar colors (blue) as shown in FIG. 2. In other words, when the colors of the claims are the same or similar to each other, it can be said that the degree of separation between groups is low. In addition, the claim pattern clustered at similar frequencies between the feature variables may be determined to be a clustered claim pattern, that is, a normal claim pattern rather than an unfair claim, based on the frequency of the feature variables resulting from the normal insurance claim.

Meanwhile, in the case of a claim pattern clustered with feature variables having a similar frequency and feature variables having a different frequency, the claim pattern may be divided into a different color (red) from the above-described claim pattern as shown in FIG. 2. Since this claim pattern is different from the frequency of the feature variable of the normal claim pattern (for example, when the number of FP changes is relatively higher than the feature variable of the normal claim pattern), it may be a new pattern of claim, It may be detected by the pattern classifier 120.

3 is a diagram illustrating an example of similar pattern group classification of the apparatus for analyzing a new invalid pattern according to an embodiment of the present application.

The unfair claim pattern classification unit 120 may classify the new pattern into a plurality of similar pattern groups based on the similarity between the new patterns. Since the similarity between the new patterns can be calculated based on the clustering algorithm described above, redundant descriptions are omitted. 3 shows an example classified into seven similar pattern groups based on the similarity between new patterns. It is apparent that the number of similar pattern groups may vary depending on the similarity between new patterns. In addition, the unfair claim pattern classification unit 120 may set the risk level of the pattern group based on the number of feature variables above a threshold set in each of the feature variables associated with each of the new patterns. For example, the risk level may be classified into a attention group, a low risk group, a medium risk group, and a high risk group as shown in FIG. 3, but is not limited thereto. That is, in the case of a pattern group classified as a high risk group, it may be said that a pattern group including many new patterns having many characteristic variables above a threshold is highly likely.

The discrimination model building unit 130 may construct a new pattern discrimination model based on a supervised learning-based decision algorithm. Supervised learning means learning a model using pre-built training data. In addition, the decision algorithm may include, for example, a decision tree algorithm, but is not limited thereto.

The discrimination model building unit 130 may learn a new pattern discrimination rule based on a decision algorithm that takes the claim pattern classified into a pattern group in which a risk level is set. In addition, the discrimination model building unit 130 may construct a new unfair billing pattern discrimination model including the new pattern discrimination rule. The above-described new pattern classification based on unsupervised learning can simply classify the new pattern based on the frequency, but it is not known by which variable. Therefore, new pattern discrimination rules can be learned based on supervised learning-based decision algorithms.

The new pattern discrimination rule means a rule capable of discriminating a claim pattern as a new pattern based on a characteristic variable of a threshold value or more. In addition, the new pattern discrimination rule may be integrated into a new pattern discrimination model by the discrimination model constructing unit 130. It is apparent that the new patterns of the pattern group set to the high risk group have a relatively high probability of having a characteristic variable above a predetermined threshold value compared to other risk groups. However, even a group of low-risk groups does not have a probability of having an overcharged feature variable (that is, a feature variable above a predetermined threshold), so that new patterns included in all risk level pattern groups can be used as input. New pattern determination rules can be learned. In addition, the learning of the new pattern discrimination rule may be considered the importance of the feature variable. That is, among the various characteristic variables, the characteristic variables with high risk of fraudulent claims may have a relatively high importance, and the importance of these characteristic variables may be considered when learning new pattern discrimination rules. For example, the total number of significant hospital visits among the characteristic variables may be given a higher importance than the number of doctors the client met. The discrimination model building unit 130 may construct a new pattern discrimination model more accurate by considering the importance of the feature variables.

The new pattern determination unit 140 may determine the pattern type of the new claim based on the new pattern determination model. The new pattern determination unit 140 may determine whether the new claim is a pattern type among the normal claim pattern, the unfair claim pattern, and the new pattern based on the new pattern determination rule. In other words, by inputting the characteristic variable of the new claim into the constructed new pattern discrimination model, it is possible to determine the type of the pattern whether the claim pattern determined to be normal, the pattern judged to be an illegal claim, or the new pattern of the new invalid claim. In addition, when it is determined that the new claim is a pattern judged as an illegal claim and a new pattern of an invalid claim, it is possible to detect a characteristic variable that acted as a factor (ie, a characteristic variable) of the judgment of an invalid claim.

For example, if there is a large distribution of health insurance among the new claims, the age of the examinees in their mid-20s, and the number of females with a characteristic variable, the initial and re-examination examinations show a high distribution. It may be determined as a pattern. In addition, when the ratio of the second visit is relatively higher than that of the first visit, it can be judged as a pattern of unfair claim before treatment, procedure and surgery.

Figure 4 is a diagram showing the flow of a new invalid claim pattern analysis method according to an embodiment of the present application.

The new fraudulent claim pattern analysis method illustrated in FIG. 4 may be performed by the novel fraudulent claim pattern analysis apparatus described with reference to FIGS. 1 to 3. Therefore, even if omitted below, the descriptions of the apparatus for analyzing new unfair claims through FIGS. 1 to 3 may be equally applicable to FIG. 4.

Referring to FIG. 4, in step S410, the data preprocessor 110 may derive characteristic variables by shaping insurance claim details data. According to an exemplary embodiment of the present application, the database 150 may record at least one of the billing history data, the characteristic variable, and the new pattern. In addition, the database 150 may store a normal billing pattern. The invalid claim history data may include at least one of billing data, contract data, payment data, insurance company data, customer data. For example, in the case of the customer data, the standardization of the insurance claim data means to quantify the income level of the customer, the number of hospitals visited by the customer, the number of reasons for the disease applied by the customer, and the like. In addition, the characteristic variable derived by the data preprocessor 110 refers to a variable that may have a numerical value associated with the claim.

In operation S420, the invalid claim pattern classification unit 120 may classify the new pattern of the invalid claim based on an unsupervised learning-based clustering algorithm using the characteristic variable as an input. The unfair claim pattern classifying unit 120 may cluster the feature variables into a plurality of claim patterns based on the frequency of the claim details data. For example, the unfair claim pattern classifying unit 120 may classify a new pattern of unfair claims based on at least one of a K-means clustering algorithm, a self-organizing-maps (SOM) algorithm, and an EM & Canopy algorithm. . The invalid claim pattern classification unit 120 may cluster the feature variables into a plurality of claim patterns through the clustering algorithm based on the frequency of the feature variables. In other words, the claim pattern may be a cluster of homogeneous feature variables. In addition, the new pattern may be detected based on the degree of separation between clusters of the claimed pattern, that is, the distance between the clusters.

In addition, the unfair claim pattern classification unit 120 may classify the new pattern into a plurality of similar pattern groups based on the similarity between the new patterns. Since the similarity between the new patterns can be calculated based on the clustering algorithm described above, redundant descriptions are omitted.

In operation S430, the discrimination model building unit 130 may build a new pattern discrimination model based on the supervised learning-based decision algorithm. The decision algorithm may include, but is not limited to, a Decision Tree algorithm, for example. The discrimination model construction unit 130 may learn a new pattern discrimination rule based on a decision algorithm that takes the claim pattern classified into a pattern group in which a risk level is set. In addition, the discrimination model building unit 130 may construct a new unfair billing pattern discrimination model including the new pattern discrimination rule. Learning new pattern discrimination rules can take into account the importance of the feature variables. That is, among the various characteristic variables, the characteristic variables with high risk of fraudulent claims may have a relatively high importance, and the importance of these characteristic variables may be considered when learning new pattern discrimination rules. For example, the total number of significant hospital visits among the characteristic variables may be given a higher importance than the number of doctors the client met. The discrimination model building unit 130 may construct a new pattern discrimination model more accurate by considering the importance of the feature variables.

In operation S440, the new pattern determination unit 140 may determine the pattern type of the new claim based on the new pattern determination model. The new pattern determination unit 140 may determine whether the new claim is a pattern type among the normal claim pattern, the unfair claim pattern, and the new pattern based on the new pattern determination rule. In other words, by inputting the characteristic variable of the new claim into the constructed new pattern discrimination model, it is possible to determine the type of the pattern whether the claim pattern determined to be normal, the pattern judged to be an illegal claim, or the new pattern of the new invalid claim. In addition, when it is determined that the new claim is a pattern judged as an illegal claim and a new pattern of an invalid claim, it is possible to detect a characteristic variable that acted as a factor (ie, a characteristic variable) of the judgment of an invalid claim.

According to an embodiment of the present disclosure, the method for analyzing a new unfair claim pattern may be implemented in the form of program instructions that can be executed by various computer means and recorded in a computer readable medium. The computer readable medium may include program instructions, data files, data structures, and the like, alone or in combination. Program instructions recorded on the media may be those specially designed and constructed for the purposes of the present invention, or they may be of the kind well-known and available to those having skill in the computer software arts. Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks, and magnetic tape, optical media such as CD-ROMs, DVDs, and magnetic disks, such as floppy disks. Magneto-optical media, and hardware devices specifically configured to store and execute program instructions, such as ROM, RAM, flash memory, and the like. Examples of program instructions include not only machine code generated by a compiler, but also high-level language code that can be executed by a computer using an interpreter or the like. The hardware device described above may be configured to operate as one or more software modules to perform the operations of the present invention, and vice versa.

The above description of the present application is intended for illustration, and it will be understood by those skilled in the art that the present invention may be easily modified in other specific forms without changing the technical spirit or essential features of the present application. Therefore, it should be understood that the embodiments described above are exemplary in all respects and not restrictive. For example, each component described as a single type may be implemented in a distributed manner, and similarly, components described as distributed may be implemented in a combined form.

The scope of the present application is indicated by the following claims rather than the above detailed description, and all changes or modifications derived from the meaning and scope of the claims and their equivalents should be construed as being included in the scope of the present application.

100: new unfair claim pattern analysis device
110: data preprocessor
120: unfair claim pattern classification unit
130: Discrimination Model Builder
140: new pattern determination unit
150: database

Claims (15)

In the new unfair claim pattern analysis apparatus,
A data preprocessing unit for deriving characteristic variables by formulating insurance claim data;
An illegal claim pattern classification unit classifying a new pattern of an illegal claim based on an unsupervised learning-based clustering algorithm using the feature variable as an input;
Discrimination model construction unit for constructing a new pattern discrimination model based on supervised learning-based decision algorithm; And
Including a new pattern determination unit for determining a pattern type of a new claim based on the new pattern determination model,
The invalid claim pattern classification unit,
Cluster the feature variables into a plurality of claimed patterns through the clustering algorithm based on the frequency of the feature variables,
Detecting the new pattern based on the degree of separation between the clusters of the claimed patterns;
Classify the new pattern into a plurality of similar pattern groups based on the similarity between the new patterns;
A risk level of the pattern group may be set based on the number of feature variables above a threshold set in each of the feature variables among the feature variables associated with each of the new patterns.
The characteristic variable may be a variable that may have a numerical value in relation to a claim, wherein the characteristic variable includes a number of duplicate applications for the same disease name and a number of reasons for a disease applied by a customer. Device. The method of claim 1,
And a database for recording at least one of the claim history data, the feature variable, and the new pattern.
The claim details data includes at least one of claim data, contract data, payment data, insurance company data, customer data, new unfair claim pattern analysis device delete delete delete The method of claim 1,
The discrimination model construction unit,
Learning a new pattern discrimination rule based on a decision algorithm for inputting the new pattern included in the pattern group for each risk level, and constructing the new pattern discrimination model including the new pattern discrimination rule; New unfair billing pattern analysis device. The method of claim 6,
The new pattern determination unit,
And determining whether the new claim is a pattern type among a normal claim pattern, an illegal claim pattern, and a new pattern based on the new pattern discrimination rule. In the novel fraud pattern analysis method by the novel fraud pattern analysis apparatus,
Deriving a feature variable by formulating insurance claim data;
Classifying a new pattern of unfair claims based on an unsupervised learning based clustering algorithm using the feature variable as an input;
Constructing a new pattern discrimination model based on supervised learning based decision algorithm; And
Determining a pattern type of a new claim based on the new pattern determination model,
Classifying the new pattern,
Cluster the feature variables into a plurality of claimed patterns through the clustering algorithm based on the frequency of the feature variables,
Detecting the new pattern based on the degree of separation between the clusters of the claimed patterns;
Classify the new pattern into a plurality of similar pattern groups based on the similarity between the new patterns;
The risk level of the pattern group may be set based on the number of feature variables above a threshold set in each of the feature variables among the feature variables associated with each of the new patterns.
The characteristic variable may be a variable that may have a numerical value in relation to the claim, wherein the characteristic variable includes the number of duplicate applications for the same disease name and the number of reasons for the disease for which the customer has applied for an unfair claim pattern analysis. Way. The method of claim 8,
Recording at least one of the claim history data, the feature variable, and the new pattern;
The claim details data includes at least one of billing data, contract data, payment data, insurance company data, customer data, new fraud pattern analysis method. delete delete delete The method of claim 8,
Building the new pattern discrimination model,
Learning a new pattern discrimination rule based on a decision algorithm for inputting the new pattern included in the pattern group for each risk level, and constructing the new pattern discrimination model including the new pattern discrimination rule; New fraud pattern analysis method. The method of claim 13,
The determining of the pattern type may include:
And determining whether the new claim is a pattern type of a normal claim pattern, an invalid claim pattern, and a new pattern based on the new pattern discrimination rule. A computer-readable recording medium having recorded thereon a program for executing the method of any one of claims 8, 9, 13 and 14 on a computer.

Images