CN112835893A - Method and system for detecting medical insurance fraud behavior based on clustering - Google Patents
Method and system for detecting medical insurance fraud behavior based on clustering Download PDFInfo
- Publication number
- CN112835893A CN112835893A CN202110063078.8A CN202110063078A CN112835893A CN 112835893 A CN112835893 A CN 112835893A CN 202110063078 A CN202110063078 A CN 202110063078A CN 112835893 A CN112835893 A CN 112835893A
- Authority
- CN
- China
- Prior art keywords
- clustering
- institution
- data
- information
- abnormal
- 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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/22—Indexing; Data structures therefor; Storage structures
- G06F16/2291—User-Defined Types; Storage management thereof
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
- G06F16/2465—Query processing support for facilitating data mining operations in structured databases
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
- G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Databases & Information Systems (AREA)
- Finance (AREA)
- Software Systems (AREA)
- Probability & Statistics with Applications (AREA)
- Accounting & Taxation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Fuzzy Systems (AREA)
- Mathematical Physics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computational Linguistics (AREA)
- Evolutionary Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Marketing (AREA)
- Strategic Management (AREA)
- Technology Law (AREA)
- General Business, Economics & Management (AREA)
- Medical Treatment And Welfare Office Work (AREA)
Abstract
The invention belongs to the technical field of medical insurance data processing, and particularly relates to a cluster-based medical insurance fraud detection method and system. A method for detecting medical insurance fraud based on clustering comprises the following steps: s1, data extraction; s2, normalization processing; s3, clustering data; and S4, obtaining a result, and repeating the step S3 until the clustering is converged to obtain the mechanism information of the suspicious abnormal behavior. The invention provides a clustering-based medical insurance fraud detection method and system for medical institutions which solve the problem of massage item abnormity in actual scenes and detect suspected medical insurance fraud abnormity behaviors.
Description
Technical Field
The invention belongs to the technical field of medical insurance data processing, and particularly relates to a cluster-based medical insurance fraud detection method and system.
Background
The medical insurance plays an important role in solving the problems of the people. However, with the advance of medical insurance, the basic medical insurance system of China is continuously sound, and a basic medical insurance system covering a wide range of people is established. The income of the medical insurance fund is increased and simultaneously influenced by factors such as aging and protection of population, improvement of guarantee level and the like, and great pressure is brought to the long-term stable operation of the medical insurance fund. The medical insurance fund is limited by factors such as an imperfect supervision system, an imperfect incentive constraint mechanism and the like, the medical insurance fund is low in use efficiency, fraud and protection problems frequently occur commonly, and the fund supervision situation is severe.
In order to improve the use efficiency of medical insurance funds and inhibit the phenomenon that fraud cheating insurance problems are frequently and commonly sent, big data and related machine learning technology are introduced. The method can solve the problems of complex data, high analysis difficulty, time consumption and the like when medical insurance personnel face mass data, and can directly screen suspicious cases of cheating and insurance by using a model and a large data analysis means. Abnormal behaviors in medical insurance consumption are various, abnormal consumption conditions exist in the physical therapy items of massage acupuncture and moxibustion, the behaviors can be caused by one side of a medical institution or combination of the medical institution and a doctor, the behaviors have certain time sequence and regionality, the abnormality cannot be found from a single record, the possible abnormal behaviors can be detected from the aspects of expense, item type, time sequence and the like by using a method of using big data, the condition that related personnel of the medical insurance department search in a large amount of data is avoided, the related personnel of the medical insurance department only need to suspect related data of the auditing institution, the workload of the medical insurance department is greatly reduced, and the working efficiency is improved.
Disclosure of Invention
The invention aims to solve the technical problem of providing a clustering-based medical insurance fraud detection method and system for medical institutions with suspected medical insurance fraud abnormal behaviors, which solve the problem of massage item abnormality in an actual scene. Therefore, the invention adopts the following technical scheme:
a method for detecting medical insurance fraud based on clustering comprises the following steps:
s1, data extraction, namely extracting multi-dimensional information of medical institutions and patients from the desensitized database;
s2, normalization processing, wherein the normalization processing is respectively carried out on the medical institution and the data of the patient obtained in the step S1;
s3, clustering data, namely clustering the medical institution data and the patient data obtained in the step S2 respectively, and then performing information simultaneous training;
and S4, obtaining a result, and repeating the step S3 until the clustering is converged to obtain the mechanism information of the suspicious abnormal behavior.
Because the original data with personal sensitive information is stored in the government medical insurance system, the data needs to be exported to the working system on the premise of sufficient desensitization of the data and then transferred to the safe working system for data storage.
On the basis of the technical scheme, the invention can also adopt the following further technical scheme:
the multidimensional information of the medical institution in the step S1 at least includes institution code, institution name, institution address, daily number of treatment items, daily number of patients, and daily working duration.
The multidimensional information of the patient in step S1 includes at least a patient code, a patient name, an age, a disease type, a disease number, an item number, a patient institution code, a patient institution name, a patient institution number, and a median distance between institutions.
In step S2, in addition to the institution code, institution name, patient code, and patient name, appropriate normalization operation of the data is required in consideration of differences in the respective attributes.
The step S3 further includes the steps of:
s31, clustering the data of the patients, and extracting a related organization information list of suspected abnormal people;
s32, carrying out rough clustering on the data of the medical institution to obtain a suspected abnormal institution;
s33, a suspected abnormal person correlation mechanism information list and suspected abnormal mechanism information obtained in the step S31 and the step S32 are combined;
and S34, ranking the suspected abnormal mechanisms obtained in the step S32, taking the mechanisms with the highest ranking as new abnormal clusters and taking the mechanisms with the lowest ranking as new normal clusters, and continuing training.
The step S31 further includes clustering the data of the people in need of treatment to obtain information of suspected abnormal people, and further obtain an organization information list of the people who have passed through the information.
The step S33 further includes obtaining an intersection of the suspected abnormal mechanism information and the mechanism information list, and obtaining the intersection mechanism information.
The step S34 further includes re-dividing the class clusters of the mechanism clusters by using the intersection mechanism information obtained in the step S33, dividing the mechanism belonging to the mechanism intersection into new abnormal clusters, and dividing the other mechanisms into new normal clusters, and continuing to train the model.
Specifically, a k-means method is used for clustering, the distance between two types of vectors is calculated by using cosine distance, the distance between two types of vectors is divided into a class cluster in a short distance, and finally the maximum effect is achieved by using the distance between the class clusters, wherein the k value is 2. Firstly, clustering the data of the patients to obtain suspected abnormal personnel information, and further obtaining an organization information list of the suspected abnormal personnel. And clustering the mechanism data to obtain suspected mechanism information, and taking the intersection of the suspected abnormal mechanism information and the mechanism information list to obtain the intersected mechanism information. And re-dividing the class clusters of the mechanism clusters by using the intersected mechanism information, dividing the mechanism belonging to the mechanism intersection into new abnormal clusters, dividing the mechanism belonging to the mechanism intersection into new normal clusters, and continuing to train the model in the mode until convergence.
The cosine distance is calculated as follows:
where a, b are two different feature vectors.
The invention also provides the following technical scheme:
a system for cluster-based detection of medical insurance fraud, comprising:
a memory storing computer-executable instructions and data for use or production in executing the computer-executable instructions;
a processor communicatively coupled to the memory and configured to execute computer-executable instructions stored by the memory;
the computer executable instructions, when executed, implement the detection method described above.
Compared with the prior art, the invention has the following beneficial effects:
(1) the mechanism is rechecked under the condition of the concurrent patients, and the condition of less mechanism quantity is compensated by using a large amount of data of the patients;
(2) the method for assisting clustering is provided, and the clustering of mechanism data is assisted by using the distinguishing effect of suspicious patients so as to improve the reliability of mechanism clustering;
(3) to a certain extent, suspicious behaviors can be quickly positioned, the workload of medical insurance inspectors is greatly reduced, and the working efficiency is improved.
Drawings
Fig. 1 is a clustering scheme structure diagram of a medical insurance fraud detection method and system based on clustering.
Detailed Description
For further understanding of the present invention, the following describes a method and a system for detecting fraud in medical insurance based on clustering according to the present invention with reference to specific embodiments, but the present invention is not limited thereto, and those skilled in the art can make insubstantial improvements and adjustments under the core guidance of the present invention, and still fall within the scope of the present invention.
The embodiment I provides a method for detecting medical insurance fraud behaviors based on clustering, which comprises the following steps:
and S1, data extraction, namely extracting the multidimensional information of the medical institution and the patient from the desensitized database, and respectively arranging the data of the patient and the medical institution into data in human units and data in institution units.
Specifically, the method comprises the following two parts:
and extracting multi-dimensional information of the medical institution in the specified time period from the desensitization database, wherein the specific information comprises institution codes, institution names, institution addresses, daily number of treatment projects, daily number of patients and daily working time of the institution.
And extracting multidimensional information of the patient in the specified time period from the desensitization database, wherein the specific information comprises the patient code, the patient name, the age, the disease type, the disease number, the project frequency, the treated institution code, the treated institution name, the institution number and the median distance between institutions.
Wherein, the meaning of some variables needs to be explained. Assuming that the patient has totally 4 medical institutions within a certain time period, the Euclidean distance between every two institutions is calculated, and then the median of the distance value is taken.
And S2, normalization processing, namely, normalization processing is respectively carried out on the medical institution and the patient data obtained in the step S1.
And S3, clustering the medical institution data and the patient data obtained in the step S2 respectively, and then performing information simultaneous training.
Specifically, step S3 includes the steps of:
and S31, clustering the data of the patient to obtain suspected abnormal personnel information, and further obtaining a mechanism information list of the personnel who go to the suspected abnormal personnel information.
S32, carrying out rough clustering on the data of the medical institution to obtain a suspected abnormal institution;
and S33, combining the suspected abnormal person related mechanism information list and the suspected abnormal mechanism information obtained in the steps S31 and S32, and taking the intersection of the suspected abnormal mechanism information and the mechanism information list to obtain the intersected mechanism information.
S34, ranking the suspected abnormal mechanisms obtained in the step S32, taking the mechanisms with the highest ranking as new abnormal clusters and the mechanisms with the lowest ranking as new normal clusters, namely, re-dividing the clusters of the mechanism clusters by using the intersection mechanism information obtained in the step S33, dividing the mechanisms belonging to the intersection of the mechanisms into new abnormal clusters, and dividing the mechanisms except the intersection of the mechanisms into new normal clusters, and continuing to train the model.
Specifically, a k-means method is used for clustering, the distance between two types of vectors is calculated by using cosine distance, the distance between two types of vectors is divided into a class cluster in a short distance, and finally the maximum effect is achieved by using the distance between the class clusters, wherein the k value is 2. Firstly, clustering data of a patient, and after convergence, extracting medical institution information of a suspicious person to obtain a medical institution information list; and then carrying out rough clustering on the data of the medical institutions to obtain preliminary normal and abnormal institution information, and ranking the normal and abnormal institutions, wherein the abnormal institutions are ranked in front of the ranking of the normal institutions and the abnormal institutions are ranked behind the ranking of the normal institutions. A simultaneous medical institution information list, wherein if an institution appears in the medical institution information list and the institution is originally in an abnormal type, the ranking is unchanged; if the mechanism is not in the medical information list and the mechanism is originally in the abnormal type, moving the mechanism to the end of the ranking of the abnormal mechanism; if the institution is in the medical institution information list and the institution is originally in the normal type, moving the institution to the end of the abnormal institution ranking; and then respectively taking 5 mechanisms at the front end of the abnormal ranking and 5 mechanisms at the front end of the normal ranking to respectively form initial central points, and continuing clustering training until convergence.
The K-Means clustering algorithm specifically comprises the following steps:
step 1, determining the clustering category as 2, and randomly initializing 2 central points;
step 2: calculating the distance from each data point to 2 central points, and distributing each data point to a central class which is closer to the data point;
step 3: recalculating the respective center points of class 2;
step 4, repeating the Step 2 → Step 3 until the central point of each type does not change greatly after each iteration;
the cosine distance is calculated as follows:
where a, b are two different feature vectors.
And S4, obtaining a result, and repeating the step S3 until the clustering is converged to obtain the mechanism information of the suspicious abnormal behavior.
In a second embodiment, a system for detecting medical insurance fraud abnormal behavior based on clustering includes:
a memory storing computer-executable instructions and data for use or production in executing the computer-executable instructions;
a processor communicatively coupled to the memory and configured to execute computer-executable instructions stored by the memory,
when being executed, the computer executable instructions implement the method for detecting the medical insurance fraud abnormal behavior based on clustering of the first embodiment.
While the invention has been shown and described with reference to a preferred embodiment, it will be understood by those skilled in the art that various changes in form and detail may be made therein without departing from the scope of the appended claims.
Claims (8)
1. A method for detecting medical insurance fraud based on clustering is characterized by comprising the following steps:
s1, data extraction, namely extracting multi-dimensional information of medical institutions and patients from the desensitized database;
s2, normalization processing, wherein the normalization processing is respectively carried out on the medical institution and the data of the patient obtained in the step S1;
s3, clustering data, namely clustering the medical institution data and the patient data obtained in the step S2 respectively, and then performing information simultaneous training;
and S4, obtaining a result, and repeating the step S3 until the clustering is converged to obtain the mechanism information of the suspicious abnormal behavior.
2. The method of claim 1, wherein the multidimensional information of the medical institution in the step S1 at least includes institution code, institution name, institution address, number of treatment items per day, number of patients per day, and working hours per day.
3. The method of claim 1, wherein the multidimensional information of the doctor in step S1 at least includes doctor code, doctor name, age, disease category, disease amount, item number, treated institution code, treated institution name, institution number, and median distance between institutions.
4. The method of claim 1, wherein the step S3 further comprises the steps of:
s31, clustering the data of the patients, and extracting a related organization information list of suspected abnormal people;
s32, carrying out rough clustering on the data of the medical institution to obtain a suspected abnormal institution;
s33, a suspected abnormal person correlation mechanism information list and suspected abnormal mechanism information obtained in the step S31 and the step S32 are combined;
and S34, ranking the suspected abnormal mechanisms obtained in the step S32, taking the mechanisms with the highest ranking as new abnormal clusters and taking the mechanisms with the lowest ranking as new normal clusters, and continuing training.
5. The method according to claim 4, wherein the step S31 further comprises clustering the data of the number of people in visit to obtain information about suspected abnormal people, and further obtaining a list of past institution information.
6. The method of claim 4, wherein the step S33 further comprises obtaining an intersection of the suspected abnormal mechanism information and the mechanism information list to obtain the intersection mechanism information.
7. The method of claim 6, wherein the step S34 further comprises re-classifying the clusters of the mechanism clusters according to the intersection mechanism information obtained in step S33, classifying the mechanisms belonging to the mechanism intersection as new abnormal clusters, and classifying the other mechanisms as new normal clusters, and continuing to train the model.
8. A system for detecting medical insurance fraud based on clustering is characterized by comprising:
a memory storing computer-executable instructions and data for use or production in executing the computer-executable instructions;
a processor communicatively coupled to the memory and configured to execute computer-executable instructions stored by the memory;
the computer executable instructions, when executed, implement a detection method as claimed in any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110063078.8A CN112835893B (en) | 2021-01-18 | 2021-01-18 | Method and system for detecting medical insurance fraud behavior based on clustering |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110063078.8A CN112835893B (en) | 2021-01-18 | 2021-01-18 | Method and system for detecting medical insurance fraud behavior based on clustering |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112835893A true CN112835893A (en) | 2021-05-25 |
| CN112835893B CN112835893B (en) | 2023-03-21 |
Family
ID=75928459
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110063078.8A Active CN112835893B (en) | 2021-01-18 | 2021-01-18 | Method and system for detecting medical insurance fraud behavior based on clustering |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112835893B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113657549A (en) * | 2021-08-31 | 2021-11-16 | 平安医疗健康管理股份有限公司 | Medical data auditing method, device, equipment and storage medium |
| CN115080997A (en) * | 2022-06-02 | 2022-09-20 | 武汉金豆医疗数据科技有限公司 | Mobile checking method and device for medical insurance fund, computer equipment and storage medium |
| CN116052887A (en) * | 2023-03-01 | 2023-05-02 | 联仁健康医疗大数据科技股份有限公司 | Method and device for detecting excessive inspection, electronic equipment and storage medium |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109636650A (en) * | 2018-12-13 | 2019-04-16 | 平安医疗健康管理股份有限公司 | Recognition methods, device, terminal and the readable storage medium storing program for executing of therapeutic regimen exception |
| CN109636653A (en) * | 2018-12-13 | 2019-04-16 | 平安医疗健康管理股份有限公司 | Monitoring method, monitoring service end and the storage medium of medical expense exception |
| CN109636644A (en) * | 2018-12-13 | 2019-04-16 | 平安医疗健康管理股份有限公司 | Monitoring method, monitoring service end and the storage medium of medical reimbursement project exception |
| CN109636192A (en) * | 2018-12-13 | 2019-04-16 | 平安医疗健康管理股份有限公司 | Recognition methods, device, terminal and the computer readable storage medium of abnormal purchase medicine |
| CN111582879A (en) * | 2019-01-30 | 2020-08-25 | 浙江远图互联科技股份有限公司 | Anti-fraud medical insurance identification method based on genetic algorithm |
| CN111612636A (en) * | 2020-04-29 | 2020-09-01 | 山东大学 | Abnormal medical insurance data detection system and method based on dual clustering algorithm |
| CN111785384A (en) * | 2020-06-29 | 2020-10-16 | 平安医疗健康管理股份有限公司 | Abnormal data identification method based on artificial intelligence and related equipment |
| CN111899114A (en) * | 2020-06-23 | 2020-11-06 | 山东大学 | Doctor-seeking fraud detection method and system based on multi-view double clustering |
-
2021
- 2021-01-18 CN CN202110063078.8A patent/CN112835893B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109636650A (en) * | 2018-12-13 | 2019-04-16 | 平安医疗健康管理股份有限公司 | Recognition methods, device, terminal and the readable storage medium storing program for executing of therapeutic regimen exception |
| CN109636653A (en) * | 2018-12-13 | 2019-04-16 | 平安医疗健康管理股份有限公司 | Monitoring method, monitoring service end and the storage medium of medical expense exception |
| CN109636644A (en) * | 2018-12-13 | 2019-04-16 | 平安医疗健康管理股份有限公司 | Monitoring method, monitoring service end and the storage medium of medical reimbursement project exception |
| CN109636192A (en) * | 2018-12-13 | 2019-04-16 | 平安医疗健康管理股份有限公司 | Recognition methods, device, terminal and the computer readable storage medium of abnormal purchase medicine |
| CN111582879A (en) * | 2019-01-30 | 2020-08-25 | 浙江远图互联科技股份有限公司 | Anti-fraud medical insurance identification method based on genetic algorithm |
| CN111612636A (en) * | 2020-04-29 | 2020-09-01 | 山东大学 | Abnormal medical insurance data detection system and method based on dual clustering algorithm |
| CN111899114A (en) * | 2020-06-23 | 2020-11-06 | 山东大学 | Doctor-seeking fraud detection method and system based on multi-view double clustering |
| CN111785384A (en) * | 2020-06-29 | 2020-10-16 | 平安医疗健康管理股份有限公司 | Abnormal data identification method based on artificial intelligence and related equipment |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113657549A (en) * | 2021-08-31 | 2021-11-16 | 平安医疗健康管理股份有限公司 | Medical data auditing method, device, equipment and storage medium |
| CN115080997A (en) * | 2022-06-02 | 2022-09-20 | 武汉金豆医疗数据科技有限公司 | Mobile checking method and device for medical insurance fund, computer equipment and storage medium |
| CN115080997B (en) * | 2022-06-02 | 2024-01-09 | 武汉金豆医疗数据科技有限公司 | Mobile checking method and device for medical insurance fund, computer equipment and storage medium |
| CN116052887A (en) * | 2023-03-01 | 2023-05-02 | 联仁健康医疗大数据科技股份有限公司 | Method and device for detecting excessive inspection, electronic equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112835893B (en) | 2023-03-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112835893B (en) | Method and system for detecting medical insurance fraud behavior based on clustering | |
| Jain et al. | An efficient hybrid feature selection model for dimensionality reduction | |
| Phillips et al. | Meta-analysis of face recognition algorithms | |
| CA2791597A1 (en) | Biometric training and matching engine | |
| CN111785384A (en) | Abnormal data identification method based on artificial intelligence and related equipment | |
| Liu et al. | Finger vein recognition with superpixel-based features | |
| Ni et al. | Multiscale Fine‐Grained Heart Rate Variability Analysis for Recognizing the Severity of Hypertension | |
| CN114190897B (en) | Training method of sleep stage model, sleep stage method and device | |
| CN111612636A (en) | Abnormal medical insurance data detection system and method based on dual clustering algorithm | |
| CN111899114B (en) | Doctor-seeking fraud detection method and system based on multi-view double clustering | |
| CN111383732B (en) | Medicine auditing method, device, computer system and readable storage medium based on mutual exclusion identification | |
| CN114005509A (en) | Treatment scheme recommendation system, method, device and storage medium | |
| CN111128233B (en) | Recording detection method and device, electronic equipment and storage medium | |
| Li et al. | Fusing deep metric learning with KNN for 12-lead multi-labelled ECG classification | |
| CN115240873A (en) | Medicine recommendation method based on machine learning, electronic equipment and computer-readable storage medium | |
| CN117690597B (en) | Biological medicine information clustering analysis method based on Internet | |
| Lehman et al. | Similarity-based searching in multi-parameter time series databases | |
| Ding et al. | Automatic classification of snoring sounds from excitation locations based on prototypical network | |
| CN114649071A (en) | Real world data-based peptic ulcer treatment scheme prediction system | |
| Ghassemi et al. | Topic models for mortality modeling in intensive care units | |
| Abdullah et al. | MERS-CoV disease estimation (MDE) A study to estimate a MERS-CoV by classification algorithms | |
| CN112289444B (en) | Method and device for determining potential important information of patient | |
| CN113707255B (en) | Health guidance method, device, computer equipment and medium based on similar patients | |
| Hayashi et al. | Patient deterioration detection using one-class classification via cluster period estimation subtask | |
| KR20200094815A (en) | Ensemble network-based data learning method and person identification method |
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 |