CN111724269A - Machine learning-based settlement data processing method and device - Google Patents
Machine learning-based settlement data processing method and device Download PDFInfo
- Publication number
- CN111724269A CN111724269A CN202010578066.4A CN202010578066A CN111724269A CN 111724269 A CN111724269 A CN 111724269A CN 202010578066 A CN202010578066 A CN 202010578066A CN 111724269 A CN111724269 A CN 111724269A
- Authority
- CN
- China
- Prior art keywords
- settlement
- group
- data
- grouping
- coefficient
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/08—Insurance
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/02—Reservations, e.g. for tickets, services or events
Abstract
The invention discloses a settlement data processing method and device based on machine learning, relates to the technical field of data processing, and mainly aims to train a grouping optimization model and a settlement processing model by utilizing a machine learning technology, optimize and settle settlement a settlement grouping and store the settlement grouping through a block chain technology, so that the workload of manual optimization grouping is reduced, and the settlement efficiency is improved. The method comprises the following steps: acquiring all settlement groups output by the standard grouping device, and screening to obtain a first settlement group set and a second settlement group set; performing grouping optimization on each settlement group in the second settlement group set based on the group mapping relation of the main body information in the settlement group, and merging the grouping optimization result into the first settlement group set; analyzing the settlement mode of the expense data in the merged first settlement group set, and processing the expense data according to the matched settlement mode to obtain settlement data. The invention is suitable for settlement data processing based on machine learning.
Description
Technical Field
The invention relates to the technical field of data processing, in particular to a settlement data processing method and device based on machine learning.
Background
At present, China is vigorously promoting DRGs settlement modes aiming at medical insurance settlement of hospitalization clinical diagnosis and treatment. A limited packing settlement method is obtained by carrying out a series of grouping argumentations on medical records. The settlement mode scientifically covers most of the disease treatment cost, and has good performance in the aspects of improving the service quality of medical institutions, reducing the expense cost of medical insurance funds and improving the quality of hospitalized medical records.
The inventor of the present application finds in research that, in the existing settlement process, all subjects are generally divided into a plurality of settlement groups only by the DRG standard grouper, and different settlement policies are adopted for the respective settlement groups. However, the problems of incomplete quality control, wrong association between the main body and settlement data and the like exist in the process of grouping the main body, so that the accuracy of the settlement data is low, the workload of manually adjusting the settlement grouping in the later period is increased, the settlement efficiency is reduced, and a large amount of manpower and financial resources are wasted.
Disclosure of Invention
In view of the above, the present invention provides a settlement data processing method and apparatus based on machine learning, and the main purpose of the present invention is to train a grouping optimization model and a settlement processing model by using a machine learning technique to optimize and settle a settlement grouping, and to store a processing result by using a block chain technique, thereby reducing the workload of manual optimization grouping, improving the settlement efficiency, and saving human and financial resources.
According to an aspect of the present invention, there is provided a machine learning-based settlement data processing method, including:
acquiring all settlement groups output by the standard grouping device;
screening all settlement groups to obtain a first settlement group set and a second settlement group set, wherein the first settlement group set comprises settlement groups which do not need to be optimized, and the second settlement group set comprises settlement groups to be optimized;
performing grouping optimization on each settlement group in the second settlement group set based on the group mapping relation of the main body information in the settlement group, and merging the result of the grouping optimization into the first settlement group set;
and analyzing the settlement mode of the expense data in the merged first settlement group set, and processing the expense data according to the matched settlement mode to obtain settlement data.
Further, the performing packet optimization on each settlement packet in the second settlement packet set based on the packet mapping relationship of the body information in the settlement packet, and merging the result of the packet optimization into the first settlement packet set includes:
extracting the group identification of each settlement group in the second settlement group set and the age data and the gender data of each main body in each settlement group;
matching corresponding grouping rules in a pre-established grouping rule database according to the grouping identification of each settlement grouping, wherein the grouping rules comprise the mapping relation between the current settlement grouping and age data and gender data;
judging whether the age data and the gender data of each main body in each settlement group accord with the grouping rule corresponding to the current settlement group;
if not, deleting the main body from the current settlement group;
and after deleting all the main bodies which do not accord with the grouping rule, merging the current settlement grouping into the first settlement grouping set.
Further, the screening all settlement groups to obtain a first settlement group set and a second settlement group set includes:
respectively extracting expense data in each settlement group;
respectively processing the expense data in each settlement group by using a preset variation coefficient algorithm to obtain the variation coefficient of each settlement group;
respectively processing the expense data in each settlement group by using a preset overall inter-group difference coefficient algorithm to obtain an overall inter-group difference coefficient of each settlement group;
comparing the variation coefficient and the overall inter-group difference coefficient with a preset variation coefficient threshold value and an overall inter-group difference coefficient threshold value respectively;
adding settlement packets of which the coefficient of variation exceeds a preset coefficient of variation threshold and/or the overall inter-group difference coefficient exceeds a preset overall inter-group difference coefficient threshold to the first settlement packet set;
adding settlement packets of which the coefficient of variation does not exceed a preset coefficient of variation threshold and/or the overall inter-group difference coefficient does not exceed a preset overall inter-group difference coefficient threshold to the second settlement packet set.
Further, the processing the fee data in each settlement group by using a preset coefficient of variation algorithm to obtain the coefficient of variation of each settlement group includes:
calculating the standard deviation and the mean value of the cost of each settlement group according to the cost data in each settlement group;
determining the ratio of the standard deviation of the cost to the mean value of the cost as a variation coefficient of each settlement group, wherein the characteristic of the variation coefficient algorithm is described as follows:
wherein CV is a coefficient of variation of each settlement group, σ is a standard deviation of the fee of each settlement group,
the mean value of the fees for each settlement group.
Further, the processing the fee data in each settlement group by using a preset overall inter-group difference coefficient algorithm to obtain an overall inter-group difference coefficient of each settlement group includes:
respectively calculating the sum of the squares of the cost differences of each settlement group and the sum of the squares of the cost differences of all the groups according to the cost data in each settlement group;
determining a ratio of the sum of squared cost deviations of each settlement group to the sum of squared cost deviations of all groups as an overall inter-group difference coefficient for each settlement group, wherein the overall inter-group difference coefficient algorithm is characterized by:
wherein RIV is the overall inter-group difference coefficient of each settlement group, SS is the sum of squared deviations of each settlement group, and SS is the sum of squared deviations of all settlement groups.
Further, the analyzing a settlement manner of the fee data in the merged first settlement packet set, and processing the fee data according to a matched settlement manner to obtain settlement data includes:
extracting the group identification of each settlement group in the first settlement group set and the charge data of each main body in each settlement group;
searching a corresponding settlement formula and a settlement coefficient in a pre-established settlement database according to the group identifier of each settlement group;
and processing the settlement coefficient and the expense data of each main body by using the settlement formula to obtain settlement data of each main body, wherein the settlement data is stored in a designated block of a block chain network.
Further, before the screening all settlement packets to obtain the first settlement packet set and the second settlement packet set, the method further includes:
and training the grouping optimization model according to the machine learning model, the grouping training sample data and the grouping demonstration label.
According to two aspects of the present invention, there is provided a machine learning-based settlement data processing apparatus, comprising:
an acquisition unit for acquiring all settlement packets output by the standard packetizer;
the screening unit is used for screening all settlement groups according to a preset screening algorithm to obtain a first settlement group set and a second settlement group set, wherein the first settlement group set comprises settlement groups which do not need to be optimized, and the second settlement group set comprises settlement groups to be optimized;
an optimization unit, configured to perform packet optimization on each settlement packet in the second settlement packet set based on a packet mapping relationship of main information in the settlement packet, and merge a result of the packet optimization into the first settlement packet set;
and the settlement unit is used for analyzing the settlement mode of the expense data in the merged first settlement group set and processing the expense data according to the matched settlement mode to obtain the settlement data.
Further, the optimization unit includes:
the second extraction module is used for extracting the group identification of each settlement group in the second settlement group set and the age data and the gender data of each main body in each settlement group;
the first matching module is used for matching corresponding grouping rules in a pre-established grouping rule database according to the grouping identification of each settlement grouping, wherein the grouping rules comprise the corresponding relation between the current settlement grouping and age data and gender data;
the judging module is used for judging whether the age data and the gender data of each main body in each settlement group accord with the grouping rule corresponding to the current settlement group or not;
a deleting module, configured to delete the subject from the current settlement group if the current settlement group is not the same as the current settlement group;
and the merging module is used for merging the current settlement groups into the first settlement group set after deleting all the main bodies which do not accord with the grouping rules corresponding to the current settlement groups.
Further, the screening unit includes:
the first extraction module is used for respectively extracting the expense data in each settlement group;
the first processing module is used for respectively processing the expense data in each settlement group by using a preset variation coefficient algorithm to obtain the variation coefficient of each settlement group;
the second processing module is used for respectively processing the expense data in each settlement group by using a preset overall inter-group difference coefficient algorithm to obtain an overall inter-group difference coefficient of each settlement group;
the comparison module is used for comparing the variation coefficient and the overall inter-group difference coefficient with a preset variation coefficient threshold and an overall inter-group difference threshold respectively;
a first adding module, configured to add a settlement group in which the coefficient of variation exceeds a preset coefficient of variation threshold and/or the overall inter-group difference coefficient exceeds a preset overall inter-group difference coefficient threshold to the first settlement group set;
a second adding module, configured to add the settlement packets whose coefficient of variation does not exceed a preset coefficient of variation threshold and/or whose overall inter-group difference coefficient does not exceed a preset overall inter-group difference coefficient threshold to the second settlement packet set.
Further, the first processing module is specifically configured to calculate a standard deviation and a mean value of the charges of each settlement group according to the charge data in each settlement group;
determining the ratio of the standard deviation of the cost to the mean value of the cost as a variation coefficient of each settlement group, wherein the characteristic of the variation coefficient algorithm is described as follows:
wherein CV is a coefficient of variation of each settlement group, σ is a standard deviation of the fee of each settlement group,
the mean value of the fees for each settlement group.
Further, the second processing module is specifically configured to calculate, according to the fee data in each settlement group, a sum of squared difference of the fees of each settlement group and a sum of squared difference of the fees of all the groups, respectively;
determining a ratio of the sum of squared cost deviations of each settlement group to the sum of squared cost deviations of all groups as an overall inter-group difference coefficient for each settlement group, wherein the overall inter-group difference coefficient algorithm is characterized by:
wherein RIV is the overall inter-group difference coefficient of each settlement group, SS is the sum of squared deviations of each settlement group, and SS is the sum of squared deviations of all settlement groups.
Further, the settlement unit includes:
the third extraction module is used for extracting the group identification of each settlement group in the first settlement group set and the charge data of each main body in each settlement group;
the search module is used for searching a corresponding settlement formula and a settlement coefficient in a pre-established settlement database according to the grouping identifier of each settlement group;
and the settlement module is used for processing the settlement coefficient and the expense data of each main body by using the settlement formula to obtain settlement data of each main body, wherein the settlement data is stored in a specified block of a block chain network. (ii) a
Further, the apparatus further comprises:
and the training unit is used for training the grouping optimization model according to the machine learning model, the grouping training sample data and the grouping demonstration label.
According to a third aspect of the present invention, there is provided a storage medium having at least one executable instruction stored therein, the executable instruction causing a processor to perform the steps of: acquiring all settlement groups output by the standard grouping device; screening all settlement groups according to a preset screening algorithm to obtain a first settlement group set and a second settlement group set, wherein the first settlement group set comprises settlement groups which do not need to be optimized, and the second settlement group set comprises settlement groups to be optimized; optimizing the second settlement packet set by using a pre-trained packet optimization model, and merging the optimization processing result into the first settlement packet set; and processing the first settlement packet set according to a pre-trained settlement processing model, and saving the output settlement data into a designated block of the block chain network.
According to a fourth aspect of the present invention, there is provided a computer device comprising a processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface communicate with each other via the communication bus, and the memory is used for storing at least one executable instruction, and the executable instruction causes the processor to perform the following steps: acquiring all settlement groups output by the standard grouping device; screening all settlement groups according to a preset screening algorithm to obtain a first settlement group set and a second settlement group set, wherein the first settlement group set comprises settlement groups which do not need to be optimized, and the second settlement group set comprises settlement groups to be optimized; optimizing the second settlement packet set by using a pre-trained packet optimization model, and merging the optimization processing result into the first settlement packet set; and processing the first settlement packet set according to a pre-trained settlement processing model, and saving the output settlement data into a designated block of the block chain network.
The invention provides a settlement data processing method and device based on machine learning, compared with the prior art that all settlement main bodies are divided into a plurality of settlement groups only through a standard grouping device, and different settlement strategies are adopted for each settlement group, the invention obtains all settlement groups output by the standard grouping device; screening the settlement groups according to a preset screening algorithm to obtain a first settlement group set and a second settlement group set, wherein the first settlement group set comprises settlement groups which do not need to be optimized, and the second settlement group set comprises settlement groups to be optimized; optimizing the second result packet set by using a pre-trained packet optimization model, and merging the optimization processing result into the first result packet set; the first settlement packet set is processed according to a pre-trained settlement processing model, and the output settlement data is stored in a designated block of the block chain network. Therefore, the grouping optimization model and the settlement processing model can be trained by utilizing the machine learning technology to optimize and settle the settlement packets, and the processing results are stored by utilizing the block chain technology, so that the workload of manual grouping optimization is reduced, the settlement efficiency is improved, and the manpower and financial resources are saved.
The foregoing description is only an overview of the technical solutions of the present invention, and the embodiments of the present invention are described below in order to make the technical means of the present invention more clearly understood and to make the above and other objects, features, and advantages of the present invention more clearly understandable.
Drawings
Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiments. The drawings are only for purposes of illustrating the preferred embodiments and are not to be construed as limiting the invention. Also, like reference numerals are used to refer to like parts throughout the drawings. In the drawings:
fig. 1 is a flowchart illustrating a settlement data processing method based on machine learning according to an embodiment of the present invention;
FIG. 2 is a schematic diagram illustrating a machine learning-based settlement data processing technique according to an embodiment of the present invention;
fig. 3 is a schematic structural diagram illustrating a machine learning-based settlement data processing apparatus according to an embodiment of the present invention;
fig. 4 shows a physical structure diagram of a computer device according to an embodiment of the present invention.
Detailed Description
Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be embodied in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.
As described in the background, the inventor of the present application has found in research that, in the existing settlement process, all subjects are generally divided into a plurality of settlement groups only by the DRG standard grouper, and different settlement policies are adopted for the respective settlement groups. However, the problems of incomplete quality control, wrong association between the main body and settlement data and the like exist in the process of grouping the main body, so that the accuracy of the settlement data is low, the workload of manually adjusting the settlement grouping in the later period is increased, the settlement efficiency is reduced, and a large amount of manpower and financial resources are wasted.
In order to solve the above problem, an embodiment of the present invention provides a settlement data processing method based on machine learning, as shown in fig. 1, the method including:
101. all settlement packets output by the standard packetizer are acquired.
The standard grouper may be specifically a CN-DRGs grouper, and the DRGs (diagnostic formed groups) is a disease diagnosis related classification, and the standard grouper can classify patients into 500 and 600 settlement groups according to the age, sex, hospital stay number, clinical diagnosis, disease symptoms, surgery, disease severity, complications, and outcome of the patients, and can be used for settling the compensation amount to be paid to the hospital for each DRG settlement group. The grouping process of the standard grouping device may specifically include: for example, patient a, the primary diagnosis: gastric ulcer with bleeding, code K25.401, surgery and operation diagnosis: none. The grouping step may specifically be: 1) the grouping device classifies the patients into MDCG digestive system diseases and functional disorder categories according to the diagnosis codes; 2) the grouping device identifies that no operation and operation codes exist, and further enters a medical group GS for non-operation treatment; 3) the grouper identifies the absence of comorbidities, concomitant conditions, and ultimately assigns the patient to a DRG settlement group, GS 15.
Specifically, the classification subject according to the embodiment of the present invention may be different medical record subjects, and the standard grouping device is used to classify all medical record subjects, output a plurality of settlement groups, and obtain all the settlement groups, so as to perform group optimization and settlement policy making using the settlement groups in the following.
102. And screening all settlement groups to obtain a first settlement group set and a second settlement group set.
Wherein the first set of settlement packets comprises settlement packets that do not require optimization and the second set of settlement packets comprises settlement packets to be optimized. The first settlement group set may be a set formed by all standard settlement groups, and the second settlement group set may be a set formed by all settlement groups to be optimized, and the preset screening algorithm may be configured to screen the settlement groups output by the standard grouping device, and screen out settlement groups having defects by using a preset index, so as to optimize the settlement groups subsequently, thereby improving the accuracy of the settlement groups and reducing the workload of manual grouping optimization. Specifically, all settlement groups are screened according to a preset screening algorithm, settlement groups which do not need to be optimized are added to a first settlement group set, and settlement groups to be optimized are added to a second settlement group set.
It should be noted that, at present, all the subjects are grouped only by the standard grouping device, because the grouping process has subjectivity and some subjects satisfy multiple grouping standards, the grouping result is not perfect, and a large amount of extreme data exists, which causes inconvenience to the subsequent settlement process. The embodiment of the invention firstly optimizes the grouping result of the standard grouping device, basically ensures the homogeneity of the main body in each group, so that batch processing can be uniformly carried out when the subsequent settlement strategy is customized, and the working efficiency is improved.
103. And performing grouping optimization on each settlement group in the second settlement group set based on the grouping mapping relation of the main body information in the settlement group, and merging the result of the grouping optimization into the first settlement group set.
The group optimization model may be a machine learning model trained in advance, and each subject in the second settlement group is judged and classified by a built-in classifier. Specifically, after the second settlement group set is input into the group optimization model, the group optimization model may determine each settlement group by using a group rule, and delete a subject that does not satisfy the group rule from the settlement group, so as to ensure the homogeneity of the subject in each settlement group, so that a settlement policy can be formulated for all subjects in each settlement group during settlement processing, thereby shortening the settlement time and improving the settlement efficiency.
104. And analyzing the settlement mode of the expense data in the merged first settlement group set, and processing the expense data according to the matched settlement mode to obtain settlement data.
The settlement processing model may be specifically configured to process the first settlement group set, and obtain settlement data by searching a settlement formula and a settlement parameter corresponding to each settlement group. The settlement parameter may be a preset settlement coefficient for each DRG settlement group, for example, the settlement parameter may be preset to be 1.2 for digestive system diseases and dysfunction categories, and the final settlement amount information for this group may be: in an actual application scenario, because an actual settlement dialing strategy is very complex, the embodiment of the present invention may also obtain final settlement amount information through a preset settlement formula, for example, the preset settlement formula is a settlement amount 1.2+ correction coefficient, the specific settlement parameters and the settlement formula may be set by itself in combination with the actual settlement condition, and the embodiment of the present invention is not specifically specified. Specifically, the settlement amount information of each group can be obtained and fed back according to the optimal grouping result and preset settlement parameters.
It should be noted that, for the embodiment of the present invention, the output settlement data may be stored in a designated block of the blockchain network. In practical application, the corresponding relation between each settlement group and each block can be established, so that after settlement data is obtained, the settlement data can be stored in a specified block according to the corresponding relation, and the safety of settlement data storage is improved.
Further, in order to better explain the process of the above settlement data processing method based on machine learning, as a refinement and extension of the above embodiment, the embodiment of the present invention provides several alternative embodiments, but is not limited thereto, and specifically, the following embodiments are provided:
in an optional embodiment of the present invention, the step 102 may specifically include: extracting the group identification of each settlement group in the second settlement group set and the age data and the gender data of each main body in each settlement group; matching corresponding grouping rules in a pre-established grouping rule database according to the grouping identification of each settlement grouping, wherein the grouping rules comprise the mapping relation between the current settlement grouping and age data and gender data; judging whether the age data and the gender data of each main body in each settlement group accord with the grouping rule corresponding to the current settlement group; if not, deleting the main body from the current settlement group; and after deleting all the main bodies which do not accord with the grouping rule, merging the current settlement grouping into the first settlement grouping set.
The settlement packet may include a packet identifier of the settlement packet, and age data and gender data of each subject in the settlement packet. The group identifier may be used to search a pre-established database for a group rule corresponding to the settlement group, and the age data and the gender data may be used as evaluation thresholds of the group rule, so as to determine whether the subject should belong to the current settlement group.
Specifically, the grouping rule corresponding to the settlement grouping is searched in a pre-established database by using the grouping identifier. For example, the settlement group 1 may correspond to a grouping rule of 20-30 years old, male. Judging whether the age data and the gender data of each main body in the settlement group accord with the grouping rule of the current settlement group, if the age data and the gender data corresponding to the main body of Zhang III are 23 years old and male, determining that the main body accords with the grouping rule of the current settlement group without optimization adjustment; if the age data and the gender data corresponding to the lee four main body are 38 years old and male, the main body can be determined not to be in accordance with the grouping rule of the current settlement group, and the main body can be deleted from the current settlement group. In an actual application scenario, the deleted main body may be manually subjected to grouping demonstration, or a settlement policy may be individually specified for the deleted main body, which is not specifically specified in the embodiment of the present invention.
After all the subjects in the settlement group that do not comply with the group rule are deleted, the group optimization of the settlement group is completed, and the settlement group may be added to the first settlement group set, so as to facilitate the subsequent settlement policy formulation.
According to the embodiment of the invention, the second settlement group set is optimized through the group optimization model, the terminal data of each settlement group in the second settlement group can be removed, the data quality control is improved, and only the deleted main body needs to be manually subjected to group demonstration or independent settlement, so that the manual workload is greatly reduced, and the settlement efficiency is improved.
In another alternative embodiment of the present invention, the method may further comprise: respectively extracting expense data in each settlement group; respectively processing the expense data in each settlement group by using a preset variation coefficient algorithm to obtain the variation coefficient of each settlement group; respectively processing the expense data in each settlement group by using a preset overall inter-group difference coefficient algorithm to obtain an overall inter-group difference coefficient of each settlement group; comparing the variation coefficient and the overall inter-group difference coefficient with a preset variation coefficient threshold value and an overall inter-group difference coefficient threshold value respectively; adding settlement packets of which the coefficient of variation exceeds a preset coefficient of variation threshold and/or the overall inter-group difference coefficient exceeds a preset overall inter-group difference coefficient threshold to the first settlement packet set; adding settlement packets of which the coefficient of variation does not exceed a preset coefficient of variation threshold and/or the overall inter-group difference coefficient does not exceed a preset overall inter-group difference coefficient threshold to the second settlement packet set.
The coefficient of variation algorithm may be configured to calculate a coefficient of variation of each settlement group, the overall inter-group coefficient of difference algorithm may be configured to calculate an overall inter-group coefficient of difference of each settlement group, and the coefficient of variation and the overall inter-group coefficient of difference may be used to filter settlement groups. Specifically, if the coefficient of variation and the overall inter-group difference coefficient are greater than any one of the preset values, the settlement group may be determined to need to be subjected to group optimization; if the variation coefficient and the overall inter-group difference coefficient are both smaller than a preset threshold, it can be determined that the settlement group does not need to be subjected to group optimization. According to the embodiment of the invention, all settlement groups are screened through two indexes of the variation coefficient and the total inter-group difference coefficient, so that the settlement groups needing to be optimized can be separately divided for optimization, and the accuracy of the settlement groups is improved conveniently.
For the embodiment of the present invention, the processing the fee data in each settlement group by using a preset variation coefficient algorithm to obtain the variation coefficient of each settlement group specifically includes: and calculating the standard deviation and the mean value of the cost of each settlement group according to the cost data in each settlement group.
Determining the ratio of the standard deviation of the cost to the mean value of the cost as a variation coefficient of each settlement group, wherein the characteristic of the variation coefficient algorithm is described as follows:
wherein CV is a coefficient of variation of each settlement group, σ is a standard deviation of the fee of each settlement group,
the mean value of the fees for each settlement group.
Wherein, the variation coefficient is a statistic for measuring variation degree of each observation value in the data. When two or more data variation degrees are compared, if the measurement unit is the same as the average, the standard deviation can be directly used for comparison, but if the measurement unit and/or the average are different, the variation degree cannot be used for comparison, and the ratio (relative value) of the standard deviation and the average is used for comparison. The ratio of the standard deviation to the mean is called the coefficient of variation and is denoted as CV. The coefficient of variation may eliminate the effect of differences in units and/or averages on the comparison of the degree of variation of two or more data. The set of standard deviations σ for the cost may be calculated by the following equation:
wherein: n may represent the number of bodies of the settlement packet, "x1，x2，x3，…xn"may represent the cost of each principal in the settlement package,
may represent the average of the fees of all subjects of the settlement group. For example: the settlement group has a main body number of 5, and the fees are 1000, 1500, 1000, 1000, 1000 respectively, then the average value
The standard deviation σ was 200 at 1500, and the coefficient of variation CV was 0.14.
For the embodiment of the present invention, the processing the fee data in each settlement group by using a preset overall inter-group difference coefficient algorithm to obtain the overall inter-group difference coefficient of each settlement group specifically includes:
respectively calculating the sum of the squares of the cost differences of each settlement group and the sum of the squares of the cost differences of all the groups according to the cost data in each settlement group; determining a ratio of the sum of squared cost deviations of each settlement group to the sum of squared cost deviations of all groups as an overall inter-group difference coefficient for each settlement group, wherein the overall inter-group difference coefficient algorithm is characterized by:
wherein RIV is the overall inter-group difference coefficient of each settlement group, SS is the sum of squared deviations of each settlement group, and SS is the sum of squared deviations of all settlement groups.
Specifically, the sum of squared differences from the distance of a certain settlement group i is calculated by the formula ss:
ss＝σ2×n
wherein, σ may represent the standard deviation of the settlement grouping fees, and the specific calculation formula is described above, which is not described herein, and n may represent the number of the settlement grouping main bodies. For example, if the standard deviation σ of the settlement group charge is 200 and the number n of subjects is 5, the sum of squared deviations of the settlement group subjects is 8000.
Similarly, the calculation method of the sum of squared deviations from the mean square and SS of all the grouped subjects is consistent with SS, and only the data range is expanded to all the groups, and the specific formula is as follows:
SS＝σ2×n
where σ may represent the standard deviation of the cost for all settlement packets and n may represent the number of subjects for all settlement packets. In an actual application scenario, the obtained RIV value is in direct proportion to the accuracy of the settlement grouping result, the higher the RIV value is, the better the grouping is, specifically, after the RIV value is obtained, the RIV value can be compared with a preset RIV threshold, and if the RIV value is greater than the preset RIV threshold, it is determined that the grouping result is high in accuracy and does not need to be optimized; if the RIV value is smaller than a preset RIV threshold value, the grouping result needs to be further optimized.
In yet another alternative embodiment of the present invention, the step 104 may specifically include: extracting the group identification of each settlement group in the first settlement group set and the charge data of each main body in each settlement group; searching a corresponding settlement formula and a settlement coefficient in a pre-established settlement database according to the group identifier of each settlement group; and processing the settlement coefficient and the expense data of each main body by using the settlement formula to obtain settlement data of each main body, wherein the settlement data is stored in a designated block of a block chain network.
The pre-trained grouping optimization model can be a machine learning model, as shown in fig. 2, for DRG groups which do not accord with preset standards, the DRG groups can be labeled manually through field experts, the machine learning model is utilized, and the manually labeled data are trained to obtain a grouping optimization model, so that the grouping optimization model can be input into the subsequent grouping results which do not accord with the preset standards, and then the optimized grouping results are automatically output, the accuracy of the grouping results is guaranteed, and the settlement efficiency is improved.
For example, patient name zhang, fee information: hospital fee 1300, medicine fee 5300. And processing the expense information to obtain a variation coefficient and a total inter-group difference coefficient, comparing the variation coefficient and the total inter-group difference coefficient with a preset threshold value respectively, and determining the medical record information exceeding the threshold value as the medical record information not meeting the preset standard. For example, if the obtained coefficient of variation is 0.2 and the set coefficient of variation threshold is 0.5, it may be determined that the grouping result meets the preset standard, and if the obtained coefficient of variation is 0.6, it may be determined that the grouping result does not meet the preset standard; similarly, if the overall inter-group difference coefficient is 10 and the set overall inter-group difference coefficient threshold is 100, it may be determined that the grouping result meets the preset standard, and if the obtained variation coefficient is 200, it may be determined that the grouping result does not meet the preset standard.
For the embodiment of the invention, after the settlement groups are optimized, the fees of each settlement group can be automatically settled by combining with self-defined settlement configuration parameters and formulas. Specifically, the optimized settlement grouping medical record information is input into a pre-trained settlement dialing control model, and different settlement parameters and settlement formulas are configured according to different groups in a user-defined manner, so that settlement information of different groups can be obtained. The configuration parameters may specifically be parameters freely configured according to local actual situation requirements through a standard and easy-to-operate parameter configuration interface provided by the system, for example, the configuration coefficient of the Xinjiang area may be dynamically configured to be 1.0, the configuration coefficient of the Beijing area may be 1.5, that is, the business settlement amounts of the Xinjiang area and the Beijing area are 1 time and 1.5 times of the basic settlement amount, respectively, and the configuration coefficients may be configured according to specific settlement information requirements, which is not specifically specified in the embodiment of the present invention. The allocation configuration formula may specifically be that after parameter configuration is completed, a medical insurance decision-making person sets an allocation rule/formula according to configured parameters, a manual setting system at each step performs memory processing, and finally, an allocation rule may be generated according to different dimensions and different granularities, for example, a settlement amount is a basic settlement amount, a configuration coefficient + a regional subsidy, and the allocation configuration consensus may be configured according to specific settlement information requirements.
For the embodiment of the present invention, before processing the optimal grouping result according to a pre-trained settlement control model and preset settlement parameters to obtain settlement amount information of each group, the method may further include: and training a settlement processing model by using the machine learning model and the sample grouping data.
The sample grouping data can be specifically a pre-established optimized grouping medical record information data set, and a settlement control model is trained by using the sample data set machine learning model. In the obtained training settlement control model, user-defined configuration parameters and a user-defined pay-per-dial settlement formula can be input, so that different grouping information is processed, and settlement information is obtained. As shown in fig. 2, after the optimized grouping result is obtained, the grouping information is processed by using the settlement dialing control model through the user-defined parameter configuration and the user-defined dialing measurement formula setting, and the final settlement information is output.
It should be noted that, the embodiment of the present invention can not only generate the grouping devices suitable for various regions in a customized manner, so as to obtain an extremely accurate grouping result, thereby improving the accuracy of the settlement grouping and the efficiency of the settlement.
In yet another alternative embodiment of the present invention, in order to obtain the grouping optimization model, the method may further include: and training the grouping optimization model according to the machine learning model, the grouping training sample data and the grouping demonstration label.
The grouped training sample data in the application can be specifically selected from the existing medical record grouped data, and because the existing medical record grouped data has the defect of poor grouping accuracy, the selected medical record grouped data needs to be marked and processed in a manual adjustment mode to form a grouped demonstration label. Specifically, through an AI self-learning function, the splitting/merging process of manual grouping demonstration is recorded, meanwhile, specific reason fields are analyzed, finally, rules of manual operation are learned into computer code rules, the computer code rules are stored in the shell of an original grouping device, and when similar medical cases subsequently participate in grouping, the grouping models can be used for processing, so that the grouping meeting the rules is achieved. As shown in FIG. 2, this AI learning process is illustrated, for example, in default grouping A, medical records of gender of women and age greater than 60 are classified into grouping C by manual adjustment so that both groups have CV and RIV values that meet preset threshold criteria; the grouping rule can be memorized through machine learning, and in subsequent sample data, all data which do not meet the grouping A are divided into a grouping C.
After the optimization of the grouping information, the subject information in each group belongs to the subject information with the same property, and the settlement can be performed according to the same settlement standard. For example, before optimizing the grouping, there may be {30 years old, 31 years old, 60 years old } in one settlement grouping, and after grouping, the data which cannot be subjected to the same settlement is divided into other groups, so that the same settlement can be performed in the group of data, thereby avoiding performing multiple settlement works for one grouping, and improving the settlement efficiency.
The invention provides a settlement data processing method based on machine learning, which is characterized in that all settlement groups output by a standard grouping device are obtained; screening the settlement groups according to a preset screening algorithm to obtain a first settlement group set and a second settlement group set, wherein the first settlement group set comprises settlement groups which do not need to be optimized, and the second settlement group set comprises settlement groups to be optimized; optimizing the second result packet set by using a pre-trained packet optimization model, and merging the optimization processing result into the first result packet set; the first settlement packet set is processed according to a pre-trained settlement processing model, and the output settlement data is stored in a designated block of the block chain network. Therefore, the grouping optimization model and the settlement processing model can be trained by utilizing the machine learning technology to optimize and settle the settlement packets, and the processing results are stored by utilizing the block chain technology, so that the workload of manual grouping optimization is reduced, the settlement efficiency is improved, and the manpower and financial resources are saved.
Further, as a specific implementation of fig. 1, an embodiment of the present invention provides a settlement data processing apparatus based on machine learning, as shown in fig. 3, the apparatus including: an acquisition unit 21, a screening unit 22, an optimization unit 23 and a settlement unit 24.
An acquisition unit 21, which may be used to acquire all settlement packets output by the standard packetizer;
a screening unit 22, configured to perform a screening process on all settlement packets to obtain a first settlement packet set and a second settlement packet set, where the first settlement packet set includes settlement packets that do not need to be optimized, and the second settlement packet set includes settlement packets to be optimized;
an optimizing unit 23, configured to perform packet optimization on each settlement packet in the second settlement packet set based on a packet mapping relationship of body information in the settlement packet, and merge a result of the packet optimization into the first settlement packet set;
the settlement unit 24 may be configured to analyze a settlement manner of the fee data in the merged first settlement packet set, and process the fee data according to a matching settlement manner to obtain settlement data.
Further, the optimization unit 23 includes:
a second extraction module 231, configured to extract the group identifier of each settlement group in the second settlement group set, and the age data and gender data of each subject in each settlement group;
a matching module 232, configured to match a corresponding grouping rule in a pre-established grouping rule database according to the grouping identifier of each settlement group, where the grouping rule includes a correspondence between a current settlement group and age data and gender data;
the determining module 233 may be configured to determine whether the age data and the gender data of each subject in each settlement group both conform to the grouping rule corresponding to the current settlement group;
a deleting module 234, which may be configured to delete the subject from the current settlement group if no;
the merging module 235 may be configured to merge the current settlement packet into the first settlement packet set after deleting all the subjects that do not comply with the packet rule.
Further, the screening unit 22 includes:
a first extraction module 221, which may be configured to extract the fee data in each settlement group respectively;
the first processing module 222 may be configured to process the fee data in each settlement group by using a preset coefficient of variation algorithm, to obtain a coefficient of variation of each settlement group;
the second processing module 223 may be configured to process the fee data in each settlement group by using a preset overall inter-group difference coefficient algorithm, so as to obtain an overall inter-group difference coefficient of each settlement group;
a comparing module 224, configured to compare the coefficient of variation and the overall inter-group difference coefficient with a preset coefficient of variation threshold and an overall inter-group difference coefficient threshold, respectively;
a first adding module 225, configured to add settlement packets with the coefficient of variation exceeding a preset coefficient of variation threshold and/or with the overall inter-group difference coefficient exceeding a preset overall inter-group difference coefficient threshold to the first settlement packet set;
a second adding module 226, configured to add the settlement packets with the coefficient of variation not exceeding a preset coefficient of variation threshold and/or with the overall inter-group difference coefficient not exceeding a preset overall inter-group difference coefficient threshold to the second settlement packet set.
Further, the first processing module 222 is specifically configured to calculate a standard deviation and a mean value of the charges of each settlement group according to the charge data in each settlement group;
determining the ratio of the standard deviation of the cost to the mean value of the cost as a variation coefficient of each settlement group, wherein the characteristic of the variation coefficient algorithm is described as follows:
wherein CV is a coefficient of variation of each settlement group, σ is a standard deviation of the fee of each settlement group,
the mean value of the fees for each settlement group.
Further, the second processing module 223 is specifically configured to calculate, according to the fee data in each settlement group, a sum of squared difference of the fees of each settlement group and a sum of squared differences of the fees of all settlement groups, respectively;
determining a ratio of the sum of squared cost deviations of each settlement group to the sum of squared cost deviations of all groups as an overall inter-group difference coefficient for each settlement group, wherein the overall inter-group difference coefficient algorithm is characterized by:
wherein RIV is the overall inter-group difference coefficient of each settlement group, SS is the sum of squared deviations of each settlement group, and SS is the sum of squared deviations of all settlement groups.
Further, the settlement unit 24 includes:
a third extracting module 241, configured to extract a group identifier of each settlement group in the first settlement group set and charge data of each main body in each settlement group;
the searching module 242 may be configured to search a settlement formula and a settlement coefficient corresponding to each settlement group in a pre-established settlement database according to the group identifier of each settlement group;
the settlement module 243 may be configured to process the settlement coefficient and the fee data of each subject by using the settlement formula to obtain settlement data of each subject, where the settlement data is stored in a designated block of a block chain network.
Further, the apparatus further comprises:
the training unit 25 may be configured to train the grouping optimization model according to the machine learning model, the grouping training sample data, and the grouping demonstration tag.
It should be noted that, for other corresponding descriptions of the functional modules involved in the settlement data processing device based on machine learning according to the embodiment of the present invention, reference may be made to the corresponding description of the method shown in fig. 1, and details are not described herein again.
The invention provides a settlement data processing device based on machine learning, which can obtain all settlement groups output by a standard grouping device; screening the settlement groups according to a preset screening algorithm to obtain a first settlement group set and a second settlement group set, wherein the first settlement group set comprises settlement groups which do not need to be optimized, and the second settlement group set comprises settlement groups to be optimized; optimizing the second result packet set by using a pre-trained packet optimization model, and merging the optimization processing result into the first result packet set; the first settlement packet set is processed according to a pre-trained settlement processing model, and the output settlement data is stored in a designated block of the block chain network. Therefore, the grouping optimization model and the settlement processing model can be trained by utilizing the machine learning technology to optimize and settle the settlement packets, and the processing results are stored by utilizing the block chain technology, so that the workload of manual grouping optimization is reduced, the settlement efficiency is improved, and the manpower and financial resources are saved.
Based on the method shown in fig. 1, correspondingly, an embodiment of the present invention further provides a storage medium, where at least one executable instruction is stored in the storage medium, and the executable instruction causes a processor to perform the following steps: acquiring all settlement groups output by the standard grouping device; screening the settlement groups according to a preset screening algorithm to obtain a first settlement group set and a second settlement group set, wherein the first settlement group set comprises settlement groups which do not need to be optimized, and the second settlement group set comprises settlement groups to be optimized; optimizing the second result packet set by using a pre-trained packet optimization model, and merging the optimization processing result into the first result packet set; the first settlement packet set is processed according to a pre-trained settlement processing model, and the output settlement data is stored in a designated block of the block chain network.
Based on the above embodiments of the method shown in fig. 1 and the apparatus shown in fig. 3, the embodiment of the present invention further provides a computer device, as shown in fig. 4, including a processor (processor)31, a communication Interface (communication Interface)32, a memory (memory)33, and a communication bus 34. Wherein: the processor 31, the communication interface 32, and the memory 33 communicate with each other via a communication bus 34. A communication interface 34 for communicating with network elements of other devices, such as clients or other servers. The processor 31 is configured to execute a program, and may specifically execute relevant steps in the above-described settlement data processing method based on machine learning. In particular, the program may include program code comprising computer operating instructions. The processor 31 may be a central processing unit CPU or an application specific integrated circuit asic or one or more integrated circuits configured to implement embodiments of the present invention.
The terminal comprises one or more processors, which can be the same type of processor, such as one or more CPUs; or may be different types of processors such as one or more CPUs and one or more ASICs. And a memory 33 for storing a program. The memory 33 may comprise a high-speed RAM memory, and may further include a non-volatile memory (non-volatile memory), such as at least one disk memory. The program may specifically be adapted to cause the processor 31 to perform the following operations: acquiring all settlement groups output by the standard grouping device; screening the settlement groups according to a preset screening algorithm to obtain a first settlement group set and a second settlement group set, wherein the first settlement group set comprises settlement groups which do not need to be optimized, and the second settlement group set comprises settlement groups to be optimized; optimizing the second result packet set by using a pre-trained packet optimization model, and merging the optimization processing result into the first result packet set; the first settlement packet set is processed according to a pre-trained settlement processing model, and the output settlement data is stored in a designated block of the block chain network.
By the technical scheme of the invention, all settlement groups output by the standard grouping device can be obtained; screening the settlement groups according to a preset screening algorithm to obtain a first settlement group set and a second settlement group set, wherein the first settlement group set comprises settlement groups which do not need to be optimized, and the second settlement group set comprises settlement groups to be optimized; optimizing the second result packet set by using a pre-trained packet optimization model, and merging the optimization processing result into the first result packet set; the first settlement packet set is processed according to a pre-trained settlement processing model, and the output settlement data is stored in a designated block of the block chain network. Therefore, the grouping optimization model and the settlement processing model can be trained by utilizing the machine learning technology to optimize and settle the settlement packets, and the processing results are stored by utilizing the block chain technology, so that the workload of manual grouping optimization is reduced, the settlement efficiency is improved, and the manpower and financial resources are saved.
It will be apparent to those skilled in the art that the modules or steps of the present invention described above may be implemented by a general purpose computing device, they may be centralized on a single computing device or distributed across a network of multiple computing devices, and alternatively, they may be implemented by program code executable by a computing device, such that they may be stored in a storage device and executed by a computing device, and in some cases, the steps shown or described may be performed in an order different than that described herein, or they may be separately fabricated into individual integrated circuit modules, or multiple ones of them may be fabricated into a single integrated circuit module. Thus, the present invention is not limited to any specific combination of hardware and software.
The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (10)
1. A settlement data processing method based on machine learning is characterized by comprising the following steps:
acquiring all settlement groups output by the standard grouping device;
screening all settlement groups to obtain a first settlement group set and a second settlement group set, wherein the first settlement group set comprises settlement groups which do not need to be optimized, and the second settlement group set comprises settlement groups to be optimized;
performing grouping optimization on each settlement group in the second settlement group set based on the group mapping relation of the main body information in the settlement group, and merging the result of the grouping optimization into the first settlement group set;
and analyzing the settlement mode of the expense data in the merged first settlement group set, and processing the expense data according to the matched settlement mode to obtain settlement data.
2. The method of claim 1, wherein the performing packet optimization on each settlement packet in the second settlement packet set based on the packet mapping relationship of the body information in the settlement packet and merging the result of the packet optimization into the first settlement packet set comprises:
extracting the group identification of each settlement group in the second settlement group set and the age data and the gender data of each main body in each settlement group;
matching corresponding grouping rules in a pre-established grouping rule database according to the grouping identification of each settlement grouping, wherein the grouping rules comprise the mapping relation between the current settlement grouping and age data and gender data;
judging whether the age data and the gender data of each main body in each settlement group accord with the grouping rule corresponding to the current settlement group;
if not, deleting the main body from the current settlement group;
and after deleting all the main bodies which do not accord with the grouping rule, merging the current settlement grouping into the first settlement grouping set.
3. The method of claim 1, wherein said filtering all settlement packets to obtain a first set of settlement packets and a second set of settlement packets comprises:
respectively extracting expense data in each settlement group;
respectively processing the expense data in each settlement group by using a preset variation coefficient algorithm to obtain the variation coefficient of each settlement group;
respectively processing the expense data in each settlement group by using a preset overall inter-group difference coefficient algorithm to obtain an overall inter-group difference coefficient of each settlement group;
comparing the variation coefficient and the overall inter-group difference coefficient with a preset variation coefficient threshold value and an overall inter-group difference coefficient threshold value respectively;
adding settlement packets of which the coefficient of variation exceeds a preset coefficient of variation threshold and/or the overall inter-group difference coefficient exceeds a preset overall inter-group difference coefficient threshold to the first settlement packet set;
adding settlement packets of which the coefficient of variation does not exceed a preset coefficient of variation threshold and/or the overall inter-group difference coefficient does not exceed a preset overall inter-group difference coefficient threshold to the second settlement packet set.
4. The method according to claim 3, wherein the processing the fee data in each settlement group by using a preset coefficient of variation algorithm to obtain the coefficient of variation of each settlement group comprises:
calculating the standard deviation and the mean value of the cost of each settlement group according to the cost data in each settlement group;
determining the ratio of the standard deviation of the cost to the mean value of the cost as a variation coefficient of each settlement group, wherein the characteristic of the variation coefficient algorithm is described as follows:
wherein CV is a coefficient of variation of each settlement group, σ is a standard deviation of the fee of each settlement group,
the mean value of the fees for each settlement group.
5. The method according to claim 3, wherein the processing the fee data in each settlement group by using a preset overall inter-group difference coefficient algorithm to obtain the overall inter-group difference coefficient of each settlement group comprises:
respectively calculating the sum of the squares of the cost differences of each settlement group and the sum of the squares of the cost differences of all the settlement groups according to the cost data in each settlement group;
determining a ratio of the sum of squared cost deviations of each settlement group to the sum of squared cost deviations of all groups as an overall inter-group difference coefficient for each settlement group, wherein the overall inter-group difference coefficient algorithm is characterized by:
wherein RIV is the overall inter-group difference coefficient of each settlement group, SS is the sum of squared deviations of each settlement group, and SS is the sum of squared deviations of all settlement groups.
6. The method according to claim 1, wherein the analyzing a settlement manner of the fee data in the merged first settlement packet set to process the fee data according to a matching settlement manner to obtain settlement data comprises:
extracting the group identification of each settlement group in the first settlement group set and the charge data of each main body in each settlement group;
searching a corresponding settlement formula and a settlement coefficient in a pre-established settlement database according to the group identifier of each settlement group;
and processing the settlement coefficient and the expense data of each main body by using the settlement formula to obtain settlement data of each main body, wherein the settlement data is stored in a designated block of a block chain network.
7. The method of claim 1, wherein before the filtering all settlement packets to obtain the first set of settlement packets and the second set of settlement packets, the method further comprises:
and training the grouping optimization model according to the machine learning model, the grouping training sample data and the grouping demonstration label.
8. A machine learning-based settlement data processing apparatus, comprising:
an acquisition unit for acquiring all settlement packets output by the standard packetizer;
the screening unit is used for screening all settlement groups to obtain a first settlement group set and a second settlement group set, wherein the first settlement group set comprises settlement groups which do not need to be optimized, and the second settlement group set comprises settlement groups to be optimized;
an optimization unit, configured to perform packet optimization on each settlement packet in the second settlement packet set based on a packet mapping relationship of main information in the settlement packet, and merge a result of the packet optimization into the first settlement packet set;
and the settlement unit is used for analyzing the settlement mode of the expense data in the merged first settlement group set and processing the expense data according to the matched settlement mode to obtain the settlement data.
9. A storage medium having stored thereon a computer program, the storage medium having stored therein at least one executable instruction that causes a processor to perform operations corresponding to the machine learning-based settlement data processing method according to any one of claims 1 to 7.
10. A computer device comprising a processor, a memory, a communication interface, and a communication bus, wherein the processor, the memory, and the communication interface communicate with each other via the communication bus, and the memory is used for storing at least one executable instruction, which causes the processor to execute the operation corresponding to the machine learning based settlement data processing according to any one of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010578066.4A CN111724269B (en) | 2020-06-23 | 2020-06-23 | Machine learning-based settlement data processing method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010578066.4A CN111724269B (en) | 2020-06-23 | 2020-06-23 | Machine learning-based settlement data processing method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111724269A true CN111724269A (en) | 2020-09-29 |
| CN111724269B CN111724269B (en) | 2023-03-24 |
Family
ID=72568221
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010578066.4A Active CN111724269B (en) | 2020-06-23 | 2020-06-23 | Machine learning-based settlement data processing method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111724269B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111932142A (en) * | 2020-08-25 | 2020-11-13 | 望海康信(北京)科技股份公司 | Method, device, equipment and storage medium for scheme grouping and data grouping |
| CN113743936A (en) * | 2021-08-26 | 2021-12-03 | 泰康保险集团股份有限公司 | Packet settlement method, device, equipment and storage medium |
| CN114385256A (en) * | 2020-10-22 | 2022-04-22 | 华为云计算技术有限公司 | Method and device for configuring system parameters |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20170023538A1 (en) * | 2015-07-23 | 2017-01-26 | Tesoro Refining & Marketing Company LLC | Methods and Apparatuses for Spectral Qualification of Fuel Properties |
| CN109272418A (en) * | 2018-09-19 | 2019-01-25 | 中链科技有限公司 | Product manufacturing data sharing method and device based on block chain |
| CN111210356A (en) * | 2020-01-14 | 2020-05-29 | 平安医疗健康管理股份有限公司 | Medical insurance data analysis method and device, computer equipment and storage medium |
-
2020
- 2020-06-23 CN CN202010578066.4A patent/CN111724269B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20170023538A1 (en) * | 2015-07-23 | 2017-01-26 | Tesoro Refining & Marketing Company LLC | Methods and Apparatuses for Spectral Qualification of Fuel Properties |
| CN109272418A (en) * | 2018-09-19 | 2019-01-25 | 中链科技有限公司 | Product manufacturing data sharing method and device based on block chain |
| CN111210356A (en) * | 2020-01-14 | 2020-05-29 | 平安医疗健康管理股份有限公司 | Medical insurance data analysis method and device, computer equipment and storage medium |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111932142A (en) * | 2020-08-25 | 2020-11-13 | 望海康信(北京)科技股份公司 | Method, device, equipment and storage medium for scheme grouping and data grouping |
| CN114385256A (en) * | 2020-10-22 | 2022-04-22 | 华为云计算技术有限公司 | Method and device for configuring system parameters |
| CN113743936A (en) * | 2021-08-26 | 2021-12-03 | 泰康保险集团股份有限公司 | Packet settlement method, device, equipment and storage medium |
| CN113743936B (en) * | 2021-08-26 | 2023-12-15 | 泰康保险集团股份有限公司 | Packet settlement method, apparatus, device and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111724269B (en) | 2023-03-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111724269B (en) | Machine learning-based settlement data processing method and device | |
| CN110443798B (en) | Autism detection method, device and system based on magnetic resonance image | |
| CN109785976B (en) | Gout disease stage prediction system based on Soft-Voting | |
| WO2018112783A1 (en) | Image recognition method and device | |
| CN111739076A (en) | Unsupervised content protection domain adaptation method for multiple CT lung texture recognition | |
| CN113011973B (en) | Method and equipment for financial transaction supervision model based on intelligent contract data lake | |
| WO2020119114A1 (en) | Method, device, and equipment for test data screening, and storage medium | |
| CN113808738A (en) | Disease identification system based on self-identification image | |
| CN111539657A (en) | Typical electricity consumption industry load characteristic classification and synthesis method combined with user daily electricity consumption curve | |
| CN107480426B (en) | Self-iteration medical record file clustering analysis system | |
| CN113674862A (en) | Acute renal function injury onset prediction method based on machine learning | |
| CN112989990A (en) | Medical bill identification method, device, equipment and storage medium | |
| CN116681923A (en) | Automatic ophthalmic disease classification method and system based on artificial intelligence | |
| CN112690802B (en) | Method, device, terminal and storage medium for detecting electrocardiosignals | |
| CN113052236A (en) | Pneumonia image classification method based on NASN | |
| WO2022205780A1 (en) | Method and apparatus for classifying eye examination data on basis of cross-modal relationship inference | |
| Bello et al. | Features-Based Individual Cattle Instance identification Method Using Hybrid Deep Learning Models for Sustainable Livestock Management | |
| Sharma et al. | Detection of white blood cells using optimized qgwo | |
| CN114121243A (en) | Method for quantifying case complexity and difficulty degree | |
| CN112967789A (en) | Resource consumption calculation and determination method for special disease species of oral cavity | |
| CN107678841B (en) | Method and device for controlling underwriting processing, computer equipment and storage medium | |
| WO2021189982A1 (en) | Health information distribution determination method and apparatus based on machine learning | |
| CN110377592B (en) | Data preprocessing method and device for quantifying variable to virtual variable and terminal equipment | |
| Chetan et al. | Plant Disease Detection using Deep Learning in Banana and Sunflower. | |
| CN114463330B (en) | CT data collection system, method and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |