CN109670973A - Social security violation detection method, device, equipment and computer storage medium - Google Patents
Social security violation detection method, device, equipment and computer storage medium Download PDFInfo
- Publication number
- CN109670973A CN109670973A CN201811530842.2A CN201811530842A CN109670973A CN 109670973 A CN109670973 A CN 109670973A CN 201811530842 A CN201811530842 A CN 201811530842A CN 109670973 A CN109670973 A CN 109670973A
- Authority
- CN
- China
- Prior art keywords
- medical data
- data
- social security
- medical
- taste
- 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.)
- Pending
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 118
- 238000003860 storage Methods 0.000 title claims abstract description 17
- 239000003814 drug Substances 0.000 claims abstract description 87
- 230000002159 abnormal effect Effects 0.000 claims abstract description 66
- 229940079593 drug Drugs 0.000 claims abstract description 56
- 244000025254 Cannabis sativa Species 0.000 claims abstract description 47
- 239000000796 flavoring agent Substances 0.000 claims abstract description 47
- 235000019634 flavors Nutrition 0.000 claims abstract description 47
- 238000007689 inspection Methods 0.000 claims abstract description 12
- 235000019640 taste Nutrition 0.000 claims description 121
- 241000411851 herbal medicine Species 0.000 claims description 29
- 230000033228 biological regulation Effects 0.000 claims description 20
- 238000011282 treatment Methods 0.000 claims description 18
- 239000011159 matrix material Substances 0.000 claims description 17
- 239000007943 implant Substances 0.000 claims description 16
- 238000013450 outlier detection Methods 0.000 claims description 14
- 230000007246 mechanism Effects 0.000 claims description 13
- 238000012360 testing method Methods 0.000 claims description 12
- 238000004140 cleaning Methods 0.000 claims description 7
- 239000000284 extract Substances 0.000 claims description 6
- 238000000605 extraction Methods 0.000 claims description 3
- 230000005856 abnormality Effects 0.000 claims 1
- 239000002699 waste material Substances 0.000 abstract description 15
- 238000011176 pooling Methods 0.000 abstract description 14
- 238000010801 machine learning Methods 0.000 abstract 1
- 238000000034 method Methods 0.000 description 20
- 238000012545 processing Methods 0.000 description 13
- 238000010586 diagram Methods 0.000 description 11
- 238000004458 analytical method Methods 0.000 description 8
- 238000004891 communication Methods 0.000 description 6
- 230000011218 segmentation Effects 0.000 description 6
- 230000006399 behavior Effects 0.000 description 5
- 208000024172 Cardiovascular disease Diseases 0.000 description 4
- 230000008901 benefit Effects 0.000 description 4
- 208000026106 cerebrovascular disease Diseases 0.000 description 4
- 230000002526 effect on cardiovascular system Effects 0.000 description 4
- 230000006870 function Effects 0.000 description 4
- 230000008520 organization Effects 0.000 description 4
- 230000008569 process Effects 0.000 description 3
- 235000009917 Crataegus X brevipes Nutrition 0.000 description 2
- 235000013204 Crataegus X haemacarpa Nutrition 0.000 description 2
- 235000009685 Crataegus X maligna Nutrition 0.000 description 2
- 235000009444 Crataegus X rubrocarnea Nutrition 0.000 description 2
- 235000009486 Crataegus bullatus Nutrition 0.000 description 2
- 235000017181 Crataegus chrysocarpa Nutrition 0.000 description 2
- 235000009682 Crataegus limnophila Nutrition 0.000 description 2
- 235000004423 Crataegus monogyna Nutrition 0.000 description 2
- 240000000171 Crataegus monogyna Species 0.000 description 2
- 235000002313 Crataegus paludosa Nutrition 0.000 description 2
- 235000009840 Crataegus x incaedua Nutrition 0.000 description 2
- 241001269238 Data Species 0.000 description 2
- 235000003143 Panax notoginseng Nutrition 0.000 description 2
- 241000180649 Panax notoginseng Species 0.000 description 2
- 244000197580 Poria cocos Species 0.000 description 2
- 235000008599 Poria cocos Nutrition 0.000 description 2
- 241000157352 Uncaria Species 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000006243 chemical reaction Methods 0.000 description 2
- 150000001875 compounds Chemical class 0.000 description 2
- 238000007405 data analysis Methods 0.000 description 2
- 206010012601 diabetes mellitus Diseases 0.000 description 2
- 239000000843 powder Substances 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000003542 behavioural effect Effects 0.000 description 1
- 229940126678 chinese medicines Drugs 0.000 description 1
- 201000010099 disease Diseases 0.000 description 1
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 1
- 235000013399 edible fruits Nutrition 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 230000014759 maintenance of location Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000000144 pharmacologic effect Effects 0.000 description 1
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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
- G06F40/289—Phrasal analysis, e.g. finite state techniques or chunking
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/30—Semantic analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Health & Medical Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Molecular Biology (AREA)
- Accounting & Taxation (AREA)
- Computing Systems (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Data Mining & Analysis (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Finance (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Economics (AREA)
- Marketing (AREA)
- Strategic Management (AREA)
- Technology Law (AREA)
- General Business, Economics & Management (AREA)
- Development Economics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The present invention provides a kind of social security violation detection method, comprising: obtains the medical data that user settles accounts medical expenditure by social security card;By the medical data input deviation detection model, through medium-height grass flavour of a drug number in data of going to a doctor described in separate-blas estimation model inspection whether within preset range;If detecting, medium-height grass flavour of a drug number is not within preset range in the medical data, it is determined that the medical data exception, and the abnormal data in the medical data is extracted, to provide the punishment certificate punished to the user.The present invention also provides a kind of social security violation detection device, equipment and computer storage mediums.The present invention, which is realized, is supervised user by the medical data that social security card is settled accounts by separate-blas estimation model based on machine learning, so as to avoid manpower waste and the waste of outpatient service risk-pooling fund caused by artificial supervision, the supervisory efficiency to outpatient service risk-pooling fund is improved.
Description
Technical field
Data monitoring technical field of the present invention more particularly to a kind of social security violation detection method, device, equipment and computer
Storage medium.
Background technique
Medical insurance refers generally to basic medical insurance, due to being to compensate for labourer's economic loss caused by disease risks
The social security system established.By employing unit and personal payment, Medical Benefits Fund is established, insurant illness is just
It examines after medical expense occurs, gives certain economic compensation to it by Medical Insurance Organizations.Insured people is in designated medical organization, spy
When not being pharmacy's purchase medicine, some medium-height grass flavour of a drug numbers are no more than certain amount, if being more than certain amount, then it is assumed that insured people is in violation of rules and regulations
Operation, such case do not meet the medical insurance regulation of certain areas, great harm can be brought to medical insurance fund.
Currently, being monitored by the supervisor being equipped with to the social security behavior of insurant, the medical insurance document of settlement is carried out
It calculates, manages the expenditure of outpatient service risk-pooling fund, still, artificial supervision causes manpower to waste, and there are many China's insurant, make
It is inadequate at supervisor, so that criminal be allowed to have an opportunity to take advantage of, cause the waste of outpatient service risk-pooling fund.
Above content is only used to facilitate the understanding of the technical scheme, and is not represented and is recognized that above content is existing skill
Art.
Summary of the invention
The main purpose of the present invention is to provide the storages of a kind of social security violation detection method, device, equipment and computer to be situated between
Matter, it is intended to solve the technical issues of artificial supervision causes manpower waste and outpatient service risk-pooling fund to waste in the prior art.
To achieve the above object, the present invention provides a kind of social security violation detection method, the social security violation detection method packet
Include following steps:
Obtain the medical data that user settles accounts medical expenditure by social security card;
By the medical data input deviation detection model, pass through medium-height grass in data of going to a doctor described in separate-blas estimation model inspection
Whether flavour of a drug number is within preset range;
If detecting, medium-height grass flavour of a drug number is not within preset range in the medical data, it is determined that the medical data are different
Often, and the abnormal data in the medical data is extracted, to provide the punishment certificate punished to the user.
Optionally, described by the medical data input deviation detection model, by described in separate-blas estimation model inspection just
The step of medium-height grass flavour of a drug number is examined in data whether within preset range include:
By the medical data input deviation detection model, by separate-blas estimation model by the text in the medical data
Data are converted into standardized field, to obtain normal data;
The normal data is detected by the Outlier Detection Algorithm based on density, according to testing result described in acquisition
The taste number of Chinese herbal medicine in medical data, and determine the taste number whether within preset range;
If the taste number is not within preset range, it is determined that medium-height grass flavour of a drug number is abnormal in the medical data.
Optionally, described by the medical data input deviation detection model, it will be described medical by separate-blas estimation model
Text data in data is converted into standardized field, and to obtain normal data the step of includes:
By the medical data input deviation detection model, medical data are cleaned by separate-blas estimation model, with
The medical data of specification after being cleaned;
Medical data configuration text feature is standardized to described, to obtain term vector, and by double in separate-blas estimation model
Sentence vector matrix is converted by term vector to RNN submodel;
The operation of attention mechanism is carried out based on the sentence vector matrix, to obtain the normal data.
Optionally, described that the normal data is detected by the Outlier Detection Algorithm based on density, according to detection
As a result the taste number of Chinese herbal medicine in the medical data is obtained, and determines that the step of taste number is whether within preset range is wrapped
It includes:
The normal data is detected by the Outlier Detection Algorithm based on density, according to testing result described in acquisition
The taste number of implant treatment and Chinese herbal medicine in medical data;
It is searched in default illness library and treats the corresponding default taste number of the implant treatment, obtain the pre- of the default taste number
If range, and determine the taste number whether within the preset range.
Optionally, detection method includes the following steps in violation of rules and regulations for the social security:
The medical data of history of the user are obtained, and medium-height grass is bought based on user described in the medical data statistics of the history
The history taste number of medicine;
The variation range of the history taste number is analyzed, and determines the taste number in the medical data whether in the variation model
It encloses;
If the taste number in the medical data is not in the variation range, it is determined that medium-height grass flavour of a drug number in the medical data
It is abnormal.
Optionally, if described detect that medium-height grass flavour of a drug number is abnormal in the medical data, extracts in the medical data
Abnormal data, the step of to provide the punishment certificate punished to the user after, the social security violation detection method is also wrapped
It includes:
Default punishment rule is obtained, notice of punishment is generated based on the default punishment rule and abnormal data;
The notice of punishment and the punishment certificate are sent to customer mobile terminal, and user is punished automatically.
Optionally, the social security violation detection method further include:
The changing rule of the medical data of the history is analyzed based on variation range, and user is predicted according to the changing rule
The prediction taste number of Chinese medicine in data of going to a doctor next time;
And the prediction taste number is determined with the presence or absence of exception, exception, then it is mobile eventually to user to send abnormal prompt if it exists
End.
In addition, to achieve the above object, the present invention also provides a kind of social security violation detection device, the social security detects in violation of rules and regulations
Device includes:
Module is obtained, passes through the medical data of social security card clearing medical expenditure for obtaining user;
Detection module is used for by the medical data input deviation detection model, by described in separate-blas estimation model inspection
Whether medium-height grass flavour of a drug number is abnormal in medical data;
Extraction module, if medium-height grass flavour of a drug number is abnormal in the medical data for detecting, it is determined that the medical data
It is abnormal, and the abnormal data in the medical data is extracted, to provide the punishment certificate punished to the user.
In addition, to achieve the above object, the present invention also provides a kind of social security violation detection device, the social security detects in violation of rules and regulations
Equipment includes processor, memory and is stored on the memory and can be examined in violation of rules and regulations by the social security that the processor executes
Ranging sequence, wherein realizing when social security violation detection program is executed by the processor such as above-mentioned social security violation detection side
The step of method.
In addition, to achieve the above object, the present invention also provides a kind of computer storage medium, the computer storage medium
On be stored with social security detection program in violation of rules and regulations, wherein the social security in violation of rules and regulations realize as above-mentioned when being executed by processor by detection program
The step of social security violation detection method.
The present invention provides a kind of social security violation detection method, device, equipment and computer storage medium, and the present invention, which obtains, to be used
The medical data of medical expenditure are settled accounts at family by social security card, then by the medical data input deviation detection model, by inclined
Whether medium-height grass flavour of a drug number is within preset range in the poor detection model detection medical data, if finally detecting described medical
Medium-height grass flavour of a drug number then extracts the abnormal data in the medical data not within preset range in data, to provide to described
The punishment certificate of user's punishment;It is thus achieved that medical data user settled accounts by social security card by separate-blas estimation model into
Row supervision improves and plans as a whole to outpatient service so as to avoid manpower waste and the waste of outpatient service risk-pooling fund caused by artificial supervision
The supervisory efficiency of fund.
Detailed description of the invention
Fig. 1 is the hardware structural diagram for the social security violation detection device that various embodiments of the present invention are related to;
Fig. 2 is the flow diagram of social security violation detection method first embodiment of the present invention;
Fig. 3 is the flow diagram of social security violation detection method second embodiment of the present invention;
Fig. 4 is the flow diagram of social security violation detection method 3rd embodiment of the present invention;
Fig. 5 is the flow diagram of social security violation detection method fourth embodiment of the present invention;
Fig. 6 is the flow diagram of the 5th embodiment of social security violation detection method of the present invention;
Fig. 7 is the flow diagram of social security violation detection method sixth embodiment of the present invention;
Fig. 8 is the flow diagram of the 7th embodiment of social security violation detection method of the present invention;
Fig. 9 is the functional block diagram of social security violation detection device first embodiment of the present invention.
The embodiments will be further described with reference to the accompanying drawings for the realization, the function and the advantages of the object of the present invention.
Specific embodiment
It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention, it is not intended to limit the present invention.
The present embodiments relate to social security violation detection method be mainly used in social security violation detection device, which disobeys
Rule detection device, which can be PC (personal computer personal computer), portable computer, mobile terminal etc., has display
With the equipment of processing function.
Referring to Fig.1, Fig. 1 is the hardware configuration signal of social security violation detection device involved in the embodiment of the present invention
Figure.In the embodiment of the present invention, social security violation detection device may include (such as the central processing unit Central of processor 1001
Processing Unit, CPU), communication bus 1002, user interface 1003, network interface 1004, memory 1005.Wherein,
Communication bus 1002 is for realizing the connection communication between these components;User interface 1003 may include display screen
(Display), input unit such as keyboard (Keyboard);Network interface 1004 optionally may include that the wired of standard connects
Mouth, wireless interface (such as WI-FI interface);Memory 1005 can be high speed RAM memory, be also possible to stable memory
(non-volatile memory), such as magnetic disk storage, memory 1005 optionally can also be independently of aforementioned processor
1001 storage device.It will be understood by those skilled in the art that hardware configuration shown in Fig. 1 is not constituted to limit of the invention
It is fixed, it may include perhaps combining certain components or different component layouts than illustrating more or fewer components.
With continued reference to Fig. 1, the memory 1005 in Fig. 1 as a kind of computer storage medium may include operating system,
Network communication module and social security detect program in violation of rules and regulations.In Fig. 1, network communication module is mainly used for connecting server, with clothes
Business device carries out data communication;And processor 1001 can call the social security stored in memory 1005 detection program in violation of rules and regulations, and hold
Row social security violation detection method provided in an embodiment of the present invention.
The present invention further provides a kind of social security violation detection methods.Referring to Fig. 2, Fig. 2 is that social security of the present invention detects in violation of rules and regulations
The method schematic diagram of method first embodiment.
In the present embodiment, the executing subject of social security violation detection method of the present invention is people society core system, the people society core
Feeling concerned about system includes equipment violation detection device, and people society core system is able to detect whether medium-height grass flavour of a drug number in medical data is being preset
Within the scope of, if not existing, the medical data exception, and the abnormal data in medical data is extracted, it is not met to filter out
Abnormal data as defined in medical insurance, and using the abnormal data as the punishment certificate punished to user.
The social security violation detection method includes:
Step S10 obtains the medical data that user settles accounts medical expenditure by social security card;
In the present embodiment, insurant is settled accounts in designated medical organization (for example, hospital, pharmacy) using social security card
When, the terminal device of designated medical organization generates prescription list according to prescription information, insured people's information and cost details etc., will
The prescription list is uploaded to people society core system, and people society core system receives the prescription that the terminal device of designated medical organization is sent
It is single, the medical data in prescription list are obtained, are detected in violation of rules and regulations with carrying out social security to medical data, wherein medical data include medical
Time, medical medical institutions' title or code, insured people's information, prescription information and cost details etc., wherein insured people's letter
Breath include insured people identity information, insured people's social security card information etc., the identity information of insured people include the age, identification card number,
Weight, height etc., prescription information include medicine information, drug usage (for example, oral, once a day), Medicine prescription time etc.,
The medicine information includes nomenclature of drug, specification, unit, quantity etc..
The medical data input deviation detection model is passed through number of going to a doctor described in separate-blas estimation model inspection by step S20
According to medium-height grass flavour of a drug number whether within preset range;
In the present embodiment, which can use cluster, sequence variation, nearest-neighbors method, multidimensional data
Medical data are analyzed in analysis etc., and whether be able to detect medium-height grass flavour of a drug number in medical data by separate-blas estimation model different
Often, specifically, the medical data got are cleaned by separate-blas estimation model first, cleaning refers to in medical data
Text data carry out word segmentation processing, go to a doctor data in there are some unnecessary data or nonstandard data, in both tables
Show word itself and can be considered under the requirement of semantic distance, it is longer more multiple using the RNN submodeling analysis in separate-blas estimation model
Miscellaneous content of text carries out word segmentation processing for medical data, is all continuous between word for example, a Chinese sentence, and
The minimum unit granularity of data analysis is word, so need to carry out word segmentation processing, and for the sentence of English text, it is English
The minimum unit of sentence is word, is separated between word by space, meanwhile, it needs to be labeled part of speech, for example, noun,
From word, adjective, numeral-classifier compound etc., the purpose of part-of-speech tagging is more useful in order to allow sentence to incorporate in processing below
Language message for some text-processing tasks, can not have to part-of-speech tagging certainly.
Further, stop words can also be removed, stop words is exactly the word for not appointing contribution function to text feature, than
Such as, eh, punctuation mark etc., need to get rid of these stop words when carrying out text analyzing, certainly, different answered
With the word that parameter removal regulation part of speech can be arranged in buggy model.
It further,, will be literary to the medical data configuration text feature after cleaning after carrying out cleaning operation to medical data
This is indicated with the sequence of a vector, and word word vector is indicated, is obtained by forward direction calculating and reverse calculating splicing complete
Term vector, obtain the sequence of term vector, then using two-way RNN model by term vector coding (conversion) for a sentence vector
Matrix, distich subvector matrix carry out attention mechanism, obtain final normal data.
Further, due to the special pharmacological mechanism of Chinese herbal medicine, specific management policy is not provided in individual areas, insured
People causes the waste of medical insurance fund, so obtaining normal data for the Chinese herbal medicine of the purpose quantity purchase exception of non-treatment
Afterwards, the Outlier Detection Algorithm based on density detects normal data, and is determined in the medical data according to testing result
The taste number of Chinese herbal medicine, and the taste number is determined whether within preset range, if the taste number, not within preset range, this is medical
Medium-height grass flavour of a drug number is abnormal in data, alternatively, the taste number is greater than a certain threshold value or is less than a certain threshold value, then medium-height grass in the medical data
Flavour of a drug number is abnormal.It is of course also possible to first judge user's implant treatment, the default taste of corresponding Chinese herbal medicine is searched according to implant treatment
Number, determines whether the taste number of Chinese herbal medicine in medical data is greater than default taste number or is less than default taste number.
Step S30, if detecting, medium-height grass flavour of a drug number is not within preset range in the medical data, it is determined that it is described just
Data exception is examined, and extracts the abnormal data in the medical data, to provide the punishment certificate punished to the user.
In the present embodiment, if detecting, medium-height grass flavour of a drug number is abnormal in medical data, extracts the exception in medical data
Data, and punishment certificate is generated according to abnormal data, which includes the title for punishing object, punishment measure, punishment original
Cause, punishment mechanism, penalty minutes etc..The abnormal data can be used as the evidence that user utilizes social security card violation operation.
The social security violation detection method that the present embodiment proposes settles accounts medical expenditure just by social security card by obtaining user
Data are examined, then by the medical data input deviation detection model, by data of going to a doctor described in separate-blas estimation model inspection
Whether medium-height grass flavour of a drug number is within preset range, if finally detecting that medium-height grass flavour of a drug number is not in preset range in the medical data
Within, it is determined that the medical data exception, and the abnormal data in the medical data is extracted, to provide to the user
The punishment certificate penalized;It realizes and user is supervised by the medical data that social security card is settled accounts by separate-blas estimation model, from
And manpower waste and the waste of outpatient service risk-pooling fund caused by artificial supervision are avoided, improve the supervision to outpatient service risk-pooling fund
Efficiency.
Based on first embodiment, the second embodiment of social security violation detection method of the present invention is proposed, referring to Fig. 3, this implementation
In example, step S20 includes:
Step S21, by the medical data input deviation detection model, by separate-blas estimation model by the medical data
In text data be converted into standardized field, to obtain normal data;
In the present embodiment, which can use cluster, sequence variation, nearest-neighbors method, multidimensional data
Medical data are analyzed in analysis etc., and whether be able to detect medium-height grass flavour of a drug number in medical data by separate-blas estimation model different
Often, specifically, the medical data got are cleaned by separate-blas estimation model first, in data of going to a doctor there are it is some not
Necessary data or nonstandard data, by being cleaned to obtain normal data to these data.
Step S22 detects the normal data by the Outlier Detection Algorithm based on density, according to testing result
The taste number of Chinese herbal medicine in the medical data is obtained, and determines the taste number whether within preset range;
In the present embodiment, it is detected in medical data according to the Outlier Detection Algorithm based on density in separate-blas estimation model
Abnormal data, the Outliers Detection algorithm based on density be generally built upon distance on the basis of, in some sense it may be said that base
In the method for density be one of the method based on distance, the method based on density be the distance between will record and it is a certain to
Determine record the two parameters of number in range to combine, to obtain the concept of " density ", then determines that record is according to density
No is outlier.According to the Outlier Detection Algorithm based on density by separate-blas estimation model to the medical data of each insured people into
Row study and prediction, the medical data standardized for every identify single medium-height grass according to the distribution of medical settlement time
The abnormal data of flavour of a drug number exception obtains the taste number of Chinese herbal medicine in medical data, and determine the taste number whether preset range it
It is interior, if the taste number, not within preset range, kind herbal medicine taste number is abnormal in data of going to a doctor, and abnormal data is exported.
Step S23, if the taste number is not within preset range, it is determined that medium-height grass flavour of a drug number is different in the medical data
Often.
In the present embodiment, if detecting the taste number of Chinese herbal medicine in medical data not within preset range, it is determined that just
It is abnormal to examine medium-height grass flavour of a drug number in data, for example, detecting that single medium-height grass flavour of a drug number is lower than 5 tastes or higher than 20 tastes in medical data
When, then illustrate that medium-height grass flavour of a drug number is abnormal.
The social security violation detection method that the present embodiment proposes, by leading to the medical data input deviation detection model
It crosses separate-blas estimation model and converts standardized field for the text data in the medical data, to obtain normal data, so
The normal data is detected by the Outlier Detection Algorithm based on density afterwards, obtains the medical number according to testing result
According to the taste number of middle Chinese herbal medicine, and the taste number is determined whether within preset range, if the last taste number is not in preset range
Within, it is determined that medium-height grass flavour of a drug number is abnormal in the medical data;It realizes and medical data is carried out by separate-blas estimation model
Detection determines whether the taste number of Chinese herbal medicine in medical data is abnormal, so as to detect abnormal data automatically, avoids artificial
Manpower waste and the waste of outpatient service risk-pooling fund caused by artificial supervision are avoided, the effect of the supervision to outpatient service risk-pooling fund is improved
Rate.
Based on second embodiment, the 3rd embodiment of social security violation detection method of the present invention is proposed, referring to Fig. 4, this implementation
In example, step S21 includes:
Step S211, by the medical data input deviation detection model, by separate-blas estimation model to medical data into
Row cleaning, with the medical data of specification after clean;
In the present embodiment, which can use cluster, sequence variation, nearest-neighbors method, multidimensional data
Medical data are analyzed in analysis etc., and whether be able to detect medium-height grass flavour of a drug number in medical data by separate-blas estimation model different
Often, specifically, the medical data got are cleaned by separate-blas estimation model first, cleaning refers to in medical data
Text data carry out word segmentation processing, go to a doctor data in there are some unnecessary data or nonstandard data, in both tables
Show word itself and can be considered under the requirement of semantic distance, using the longer more complicated content of text of RNN model analysis, to medical
Data carry out word segmentation processing, are all continuous, and the minimum unit grain of data analysis for example, a Chinese sentence, between word
Degree is word, so need to carry out word segmentation processing, and for the sentence of English text, the minimum unit of English sentence is word
Language is separated by space between word, meanwhile, need to be labeled part of speech, for example, noun, from word, adjective, numeral-classifier compound
Be Deng, the purpose of part-of-speech tagging in order to allow sentence to incorporate more useful language messages in processing below, certainly, for
Some text-processing tasks can not have to part-of-speech tagging.
Further, stop words can also be removed, stop words is exactly the word for not appointing contribution function to text feature, than
Such as, eh, punctuation mark etc., need to get rid of these stop words when carrying out text analyzing, certainly, different answered
With the word that parameter removal regulation part of speech can be arranged in separate-blas estimation model.
Step S212 standardizes medical data configuration text feature to described, to obtain term vector, and passes through separate-blas estimation mould
Term vector is converted sentence vector matrix by the two-way RNN submodel of type;
In the present embodiment, the medical data of specification are obtained after carrying out cleaning operation to medical data, it is medical to specification
Data configuration text feature indicates text with the sequence of a vector, and word word vector is indicated, calculated by forward direction and
Reverse calculating splicing obtains complete term vector, obtains the sequence of term vector, is then encoded term vector using two-way RNN model
(conversion) is a sentence vector matrix, and distich subvector matrix carries out attention mechanism, obtains final normal data.
Step S213 carries out the operation of attention mechanism based on the sentence vector matrix, to obtain the normal data.
In the present embodiment, distich subvector matrix carries out attention mechanism, by the sentence vector matrix in step S212
One vector of boil down to indicates that the feedforward neural network for being sent into standard is predicted, a matrix is inputted, and exports a vector,
A context vector is extracted from input content, the context vector of the mechanism is to be taken as the parameter learning of model to obtain.
This makes attention mechanism become a pure squeeze operation, can replace any pond step.Content of text is compressed into
After one vector, final normal data is obtained, for example, a kind of class label, a real number value etc..
The social security violation detection method that the present embodiment proposes, by leading to the medical data input deviation detection model
It crosses separate-blas estimation model to clean medical data, be gone to a doctor data with the specification after being cleaned, then just to the specification
Data configuration text feature is examined, to obtain term vector, and sentence vector matrix is converted for term vector by two-way RNN model,
The operation of attention mechanism is finally carried out based on the sentence vector matrix, to obtain the normal data;It realizes and passes through deviation
Medical data are converted normal data by detection model, to be conducive to the detection of abnormal data, and then improves abnormal data inspection
The correctness of survey.
Based on second embodiment, the fourth embodiment of social security violation detection method of the present invention is proposed, referring to Fig. 5, this implementation
In example, step S22 includes:
Step S221 detects the normal data by the Outlier Detection Algorithm based on density, is tied according to detection
Fruit obtains the taste number of implant treatment and Chinese herbal medicine in the medical data;
In the present embodiment, different implant treatments may be different using the taste number of traditional Chinese medicine, according to based on the different of density
Normal detection algorithm detects normal data, is obtained in the implant treatment and medical data of medical data according to testing result
The taste number of herbal medicine, to determine whether the taste number of Chinese herbal medicine in medical data is greater than default taste number.
Step S222 is searched in default illness library and is treated the corresponding default taste number of the implant treatment, calculates described pre-
If the preset range of taste number, and determine whether the preset range is greater than preset threshold.
In the present embodiment, the corresponding default taste number of the implant treatment is searched in default illness library, counts the default taste
Several preset ranges, and the taste number of Chinese medicine in medical data is determined whether within preset range, for example, in default illness library
The Chinese medicine for searching treatment diabetes can make 5 taste medicines, be also possible to 6 taste medicines, it is determined that treat the diabetes and use taste of traditional Chinese medicine number
Preset range be 5 tastes to 6 tastes, if detect Chinese medicine in medical data taste number be 5 tastes, go to a doctor data in Chinese medicine taste number
Within the preset range, illustrate that there is no abnormal datas for medical data.
The social security violation detection method that the present embodiment proposes, by the Outlier Detection Algorithm based on density to the criterion numeral
According to being detected, the taste number of the implant treatment and Chinese herbal medicine in the medical data is obtained according to testing result, then default
It is searched in illness library and treats the corresponding default taste number of the implant treatment, obtain the preset range of the default taste number, and determine
Whether the taste number is within the preset range;It realizes according to implant treatment and determines default taste number, so as to basis
Different illnesss detects the abnormal data of Chinese medicine in medical data, and then more accurately extracts abnormal data.
Based on fourth embodiment, the 5th embodiment of social security violation detection method of the present invention is proposed, referring to Fig. 6, this implementation
In example, the social security violation detection method further include:
Step S40 obtains the medical data of history of the user, and based on user described in the medical data statistics of the history
Buy the history taste number of Chinese herbal medicine;
In the present embodiment, the Chinese medicine that certain illnesss may need doctor to open different taste numbers is taken to user, for example, can be with
It is 6 taste medicines, can be 7 taste medicines, but the variation of taste of traditional Chinese medicine number outputed of doctor can be according to the needs of illness within limits
Variation, it is possible to obtain the medical data of history of user, the history taste number of Chinese medicine in the medical data of statistical history.
Whether step S50 analyzes the variation range of the history taste number, and determine the taste number in the medical data in institute
State variation range;
In the present embodiment, the history taste number is analyzed when the history taste number of Chinese medicine in counting on the medical data of history
Variation range, and determine the taste number in this medical data whether within the variation range.
Step S60, if the taste number in the medical data is not in the variation range, it is determined that in the medical data
Herbal medicine taste number is abnormal.
In the present embodiment, if the taste number in this medical data is not within the variation range, in the medical data
Medium-height grass flavour of a drug numbers is abnormal, for example, the history that statistics obtains each prescription of cardiovascular and cerebrovascular disease user is gone to a doctor, Chinese medicine is all only in data
There is this Six-element medicine of Radix Notoginseng powder, hawthorn, uncaria, Poria cocos, Radix Salviae Miltiorrhizae, Radix Glycyrrhizae, if detecting cardiovascular and cerebrovascular disease user at this
Chinese medicine also only exists this Six-element medicine in medical data in side, then it is assumed that the medical data in this prescription are without exception, if detection
Into the medical data in this prescription of cardiovascular and cerebrovascular disease user there are other Chinese medicines in addition to this Six-element in Chinese medicine, then it is assumed that
Medical data exception in this prescription.In another example statistics obtains the medical number of history of each prescription of cardiovascular and cerebrovascular disease user
According to Chinese medicine be all Radix Notoginseng powder, hawthorn, uncaria, Poria cocos, Radix Salviae Miltiorrhizae, Radix Glycyrrhizae this Six-element medicine the five tastes or Six-element, it is determined that this is gone through
The variation of history taste number changes in the five tastes between Six-element, then the variation range five tastes of the history taste number are to Six-element, if detecting this
There are seven flavor medicines in medical data in secondary prescription, then it is assumed that the medical data exception of this prescription.
The social security violation detection method that the present embodiment proposes, the medical data of history by obtaining the user, and be based on
User described in the medical data statistics of the history buys the history taste number of Chinese herbal medicine, then analyzes the variation model of the history taste number
It encloses, and determines the taste number in the medical data whether in the variation range, if the taste number in the last medical data is not
In the variation range, it is determined that medium-height grass flavour of a drug number is abnormal in the medical data;It realizes and is gone to a doctor data according to user's history
It whether abnormal analyzes this medical data medium-height grass flavour of a drug number, further increases the supervisory efficiency to outpatient service risk-pooling fund.
Based on first embodiment, the sixth embodiment of social security violation detection method of the present invention is proposed, referring to Fig. 7, this implementation
In example, after step S60, further includes:
Step S70 obtains default punishment rule, generates notice of punishment based on the default punishment rule and abnormal data;
In the present embodiment, if detecting, there are abnormal datas in medical data, search default punishment in the database
Rule, and rule of punishing is then according to this and abnormal data generates notice of punishment, which is set by technical staff, at this
Penalizing notice includes title, punishment measure, punishment reason, punishment mechanism, penalty minutes of punishment object etc..
The notice of punishment and the punishment certificate are sent to customer mobile terminal by step S80, and automatically to user into
Row punishment.
In the present embodiment, notice of punishment is sent to mobile terminal, and user is punished automatically, wherein can also
To count the number of insured people's preset time period violation operation, the degree of punitive measures is determined, for example, warning, fine, cancellation doctor
Card etc. is protected, alternatively, can carry out related prompt when detecting that user has the pre- behavior of violation, pre- behavior refers to according to insured
Behavioural analysis prediction insured people next time of the purchase drug of people buys the behavior of drug, and reminding method can be short message prompt, electricity
Words notice or wechat message notifying etc..
The social security violation detection method that the present embodiment proposes is based on the default punishment by obtaining default punishment rule
Rule and abnormal data generate notice of punishment, and it is mobile eventually that the notice of punishment and the punishment certificate are then sent to user
End, and user is punished automatically;Realize detect user go to a doctor data deposit when abnormal, user is located automatically
It penalizes, so as to avoid manpower waste and the waste of outpatient service risk-pooling fund caused by artificial supervision is manually avoided, improves on the door
Examine the supervisory efficiency of risk-pooling fund.
Based on the 5th embodiment, the 7th embodiment of social security violation detection method of the present invention is proposed, referring to Fig. 8, this implementation
In example, the social security violation detection method further include:
Step S90 analyzes the changing rule of the medical data of the history by prediction model, and according to the changing rule
Predict the prediction taste number of Chinese medicine in medical data user's next time;
Step S100, and the prediction taste number is determined with the presence or absence of exception, exception, then send abnormal prompt and extremely use if it exists
Family mobile terminal.
In the present embodiment, the data that history can be gone to a doctor and this medical data are input to preset prediction model, obtain
The prediction result for taking prediction model to export, prediction model is capable of the changing rule of analysis of history data, according to the change of historical data
Law can predict the prediction taste number of Chinese medicine in medical data when user buys Chinese medicine next time, and whether determine the prediction taste number
Within preset range, or determine whether the prediction taste number is greater than preset threshold, the prediction taste number is not within preset range
Or it is greater than preset threshold, then it is assumed that prediction taste number has exception, then abnormal prompt is sent to customer mobile terminal, the exception
Prompt should be noted item, social security rule etc. including prediction result, purchase next time Chinese medicine.
The social security violation detection method that the present embodiment proposes, by analyzing the medical data of the history by prediction model
Changing rule, and according to the prediction taste number of Chinese medicine in changing rule prediction medical data user's next time, then and determine institute
Prediction taste number is stated with the presence or absence of exception, exception, then send abnormal prompt to customer mobile terminal if it exists;It realizes and passes through prediction
The behavior that model prediction user's next time buys Chinese medicine predicts that reminding the user that not carry out violation operation, further mentions
It is high to the supervisory efficiency to outpatient service risk-pooling fund.
In addition, the embodiment of the present invention also provides a kind of social security violation detection device.
It is the functional block diagram of social security violation detection device first embodiment of the present invention referring to Fig. 9, Fig. 9.
Social security violation detection device of the present invention is virtual bench, is stored in the storage of the detection device of social security violation shown in Fig. 1
In device 1005, the institute for realizing social security detection program in violation of rules and regulations is functional: obtaining user and passes through social security card clearing medical expenditure
Medical data;By the medical data input deviation detection model, in data of going to a doctor described in separate-blas estimation model inspection
Whether herbal medicine taste number is within preset range;If detecting, medium-height grass flavour of a drug number is not within preset range in the medical data,
The abnormal data in the medical data is extracted, then to provide the punishment certificate punished to the user.
Specifically, in the present embodiment, the social security violation detection device includes:
Module 101 is obtained, passes through the medical data of social security card clearing medical expenditure for obtaining user;
Detection module 102, for passing through separate-blas estimation model inspection institute for the medical data input deviation detection model
Whether abnormal state medium-height grass flavour of a drug number in medical data;
Extraction module 103, if medium-height grass flavour of a drug number is abnormal in the medical data for detecting, it is determined that the medical number
According to exception, and the abnormal data in the medical data is extracted, to provide the punishment certificate punished to the user.
Further, the detection module is also used to:
By the medical data input deviation detection model, by separate-blas estimation model by the text in the medical data
Data are converted into standardized field, to obtain normal data;
The normal data is detected by the Outlier Detection Algorithm based on density, according to testing result described in acquisition
The taste number of Chinese herbal medicine in medical data, and determine the taste number whether within preset range;
If the taste number is not within preset range, it is determined that medium-height grass flavour of a drug number is abnormal in the medical data.
Further, the detection module is also used to:
By the medical data input deviation detection model, medical data are cleaned by separate-blas estimation model, with
The medical data of specification after being cleaned;
Medical data configuration text feature is standardized to described, to obtain term vector, and by double in separate-blas estimation model
Sentence vector matrix is converted by term vector to RNN submodel;
The operation of attention mechanism is carried out based on the sentence vector matrix, to obtain the normal data.
Further, the detection module is also used to: by the Outlier Detection Algorithm based on density to the normal data
It is detected, obtains the taste number of the implant treatment and Chinese herbal medicine in the medical data according to testing result;
It is searched in default illness library and treats the corresponding default taste number of the implant treatment, obtain the pre- of the default taste number
If range, and determine the taste number whether within the preset range.
Further, the social security violation detection device further include:
Module is obtained, the medical data of the history for obtaining the user, and based on the medical data statistics institute of the history
State the history taste number that user buys Chinese herbal medicine;
Analysis module for analyzing the variation range of the history taste number, and determines that the taste number in the medical data is
It is no in the variation range;
If the taste number in the medical data is not in the variation range, it is determined that medium-height grass flavour of a drug number in the medical data
It is abnormal.
Further, the social security violation detection device further include:
Generation module generates punishment based on the default punishment rule and abnormal data for obtaining default punishment rule
Notice;
Module is punished, for the notice of punishment and the punishment certificate to be sent to customer mobile terminal, and it is automatic right
User punishes.
Further, the social security violation detection device further include:
Prediction module, for analyzing the changing rule of the medical data of the history by prediction model, and according to the change
Law predicts the prediction taste number of Chinese medicine in medical data user's next time;
Cue module is used for and determines the prediction taste number with the presence or absence of exception, and exception, then send abnormal prompt if it exists
To customer mobile terminal.
In addition, the embodiment of the present invention also provides a kind of computer storage medium.
Social security detection program in violation of rules and regulations is stored in computer storage medium of the present invention, wherein the social security detects program in violation of rules and regulations
When being executed by processor, realize such as the step of above-mentioned social security violation detection method.
Wherein, social security violation detection program, which is performed realized method, can refer to social security violation detection method of the present invention
Each embodiment, details are not described herein again.
It should be noted that, in this document, the terms "include", "comprise" or its any other variant are intended to non-row
His property includes, so that the process, method, article or the system that include a series of elements not only include those elements, and
And further include other elements that are not explicitly listed, or further include for this process, method, article or system institute it is intrinsic
Element.In the absence of more restrictions, the element limited by sentence "including a ...", it is not excluded that including being somebody's turn to do
There is also other identical elements in the process, method of element, article or system.
The serial number of the above embodiments of the invention is only for description, does not represent the advantages or disadvantages of the embodiments.
Through the above description of the embodiments, those skilled in the art can be understood that above-described embodiment side
Method can be realized by means of software and necessary general hardware platform, naturally it is also possible to by hardware, but in many cases
The former is more preferably embodiment.Based on this understanding, technical solution of the present invention substantially in other words does the prior art
The part contributed out can be embodied in the form of software products, which is stored in one as described above
In storage medium (such as ROM/RAM, magnetic disk, CD), including some instructions are used so that terminal device (it can be mobile phone,
Computer, server, air conditioner or network equipment etc.) execute method described in each embodiment of the present invention.
The above is only a preferred embodiment of the present invention, is not intended to limit the scope of the invention, all to utilize this hair
Equivalent structure or equivalent flow shift made by bright specification and accompanying drawing content is applied directly or indirectly in other relevant skills
Art field, is included within the scope of the present invention.
Claims (10)
1. a kind of social security violation detection method, which is characterized in that detection method includes the following steps in violation of rules and regulations for the social security:
Obtain the medical data that user settles accounts medical expenditure by social security card;
By the medical data input deviation detection model, pass through medium-height grass flavour of a drug in data of going to a doctor described in separate-blas estimation model inspection
Whether number is within preset range;
If detecting, medium-height grass flavour of a drug number is not within preset range in the medical data, it is determined that the medical data exception,
And the abnormal data in the medical data is extracted, to provide the punishment certificate punished to the user.
2. social security violation detection method as described in claim 1, which is characterized in that described by the medical data input deviation
Detection model, by described in separate-blas estimation model inspection go to a doctor data in medium-height grass flavour of a drug number whether within preset range the step of
Include:
By the medical data input deviation detection model, by separate-blas estimation model by the text data in the medical data
It is converted into standardized field, to obtain normal data;
The normal data is detected by the Outlier Detection Algorithm based on density, is obtained according to testing result described medical
The taste number of Chinese herbal medicine in data, and determine the taste number whether within preset range;
If the taste number is not within preset range, it is determined that medium-height grass flavour of a drug number is abnormal in the medical data.
3. social security violation detection method as claimed in claim 2, which is characterized in that described by the medical data input deviation
Detection model converts standardized field for the text data in the medical data by separate-blas estimation model, to obtain
The step of normal data includes:
By separate-blas estimation model medical data are cleaned, the medical data input deviation detection model with to obtain
The medical data of specification after cleaning;
Medical data configuration text feature is standardized to described, to obtain term vector, and passes through the two-way RNN in separate-blas estimation model
Term vector is converted sentence vector matrix by submodel;
The operation of attention mechanism is carried out based on the sentence vector matrix, to obtain the normal data.
4. social security violation detection method as claimed in claim 2, which is characterized in that described to pass through the abnormality detection based on density
Algorithm detects the normal data, obtains the taste number of Chinese herbal medicine in the medical data according to testing result, and determine
The step of whether the taste number is within preset range include:
The normal data is detected by the Outlier Detection Algorithm based on density, is obtained according to testing result described medical
The taste number of implant treatment and Chinese herbal medicine in data;
It is searched in default illness library and treats the corresponding default taste number of the implant treatment, obtain the default model of the default taste number
It encloses, and determines the taste number whether within the preset range.
5. social security violation detection method as claimed in claim 4, which is characterized in that the social security violation detection method include with
Lower step:
The medical data of history of the user are obtained, and Chinese herbal medicine is bought based on user described in the medical data statistics of the history
History taste number;
The variation range of the history taste number is analyzed, and determines the taste number in the medical data whether in the variation range;
If the taste number in the medical data is not in the variation range, it is determined that medium-height grass flavour of a drug number is different in the medical data
Often.
6. social security violation detection method as described in claim 1, which is characterized in that if described detect in the medical data
Medium-height grass flavour of a drug number is abnormal, then extracts the abnormal data in the medical data, to provide the punishment certificate punished to the user
The step of after, the social security violation detection method further include:
Default punishment rule is obtained, notice of punishment is generated based on the default punishment rule and abnormal data;
The notice of punishment and the punishment certificate are sent to customer mobile terminal, and user is punished automatically.
7. social security violation detection method as claimed in claim 5, which is characterized in that the social security violation detection method is also wrapped
It includes:
The changing rule of the medical data of the history is analyzed by prediction model, and user's next time is predicted according to the changing rule
The prediction taste number of Chinese medicine in medical data;
And the prediction taste number is determined with the presence or absence of exception, exception, then send abnormal prompt to customer mobile terminal if it exists.
8. a kind of social security violation detection device, which is characterized in that the social security violation detection device includes:
Module is obtained, passes through the medical data of social security card clearing medical expenditure for obtaining user;
Detection module is used for the medical data input deviation detection model, by going to a doctor described in separate-blas estimation model inspection
Whether medium-height grass flavour of a drug number is abnormal in data;
Extraction module, if medium-height grass flavour of a drug number is abnormal in the medical data for detecting, it is determined that the medical data exception,
And the abnormal data in the medical data is extracted, to provide the punishment certificate punished to the user.
9. a kind of social security violation detection device, which is characterized in that the social security violation detection device include processor, memory,
And be stored on the memory and program can be detected by the social security that the processor executes in violation of rules and regulations, wherein the social security is in violation of rules and regulations
When detection program is executed by the processor, realizing the social security violation detection method as described in any one of claims 1 to 7
Step.
10. a kind of computer storage medium, which is characterized in that credit evaluation program is stored in the computer storage medium,
When wherein the social security violation detection program is executed by processor, realize that the social security as described in any one of claims 1 to 7 is disobeyed
The step of advising detection method.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811530842.2A CN109670973A (en) | 2018-12-13 | 2018-12-13 | Social security violation detection method, device, equipment and computer storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811530842.2A CN109670973A (en) | 2018-12-13 | 2018-12-13 | Social security violation detection method, device, equipment and computer storage medium |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109670973A true CN109670973A (en) | 2019-04-23 |
Family
ID=66143915
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811530842.2A Pending CN109670973A (en) | 2018-12-13 | 2018-12-13 | Social security violation detection method, device, equipment and computer storage medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109670973A (en) |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103824130A (en) * | 2014-02-27 | 2014-05-28 | 武汉理工大学 | Grain condition forecasting and early warning method and system based on SVM |
CN104050787A (en) * | 2013-03-12 | 2014-09-17 | 霍尼韦尔国际公司 | System and Method of Anomaly Detection with Categorical Attributes |
CN104196506A (en) * | 2014-08-01 | 2014-12-10 | 中国石油大学(北京) | Injection and production parameter joint debugging method, device and system for SAGD single well set |
CN107066613A (en) * | 2017-05-02 | 2017-08-18 | 南京航空航天大学 | A kind of Outlier Detection Algorithm based on relative density |
-
2018
- 2018-12-13 CN CN201811530842.2A patent/CN109670973A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104050787A (en) * | 2013-03-12 | 2014-09-17 | 霍尼韦尔国际公司 | System and Method of Anomaly Detection with Categorical Attributes |
CN103824130A (en) * | 2014-02-27 | 2014-05-28 | 武汉理工大学 | Grain condition forecasting and early warning method and system based on SVM |
CN104196506A (en) * | 2014-08-01 | 2014-12-10 | 中国石油大学(北京) | Injection and production parameter joint debugging method, device and system for SAGD single well set |
CN107066613A (en) * | 2017-05-02 | 2017-08-18 | 南京航空航天大学 | A kind of Outlier Detection Algorithm based on relative density |
Non-Patent Citations (1)
Title |
---|
周如意: ""基于BP神经网络和关联规则的智能医疗保险稽核系统研究"", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Masry et al. | Chartqa: A benchmark for question answering about charts with visual and logical reasoning | |
CN108921710A (en) | The method and system of medical insurance abnormality detection | |
CN109635044A (en) | Hospitalization data method for detecting abnormality, device, equipment and readable storage medium storing program for executing | |
CN109685670A (en) | Social security violation detection method, device, equipment and computer readable storage medium | |
Yang et al. | Using health-consumer-contributed data to detect adverse drug reactions by association mining with temporal analysis | |
Lee et al. | Using Mahalanobis–Taguchi system, logistic regression, and neural network method to evaluate purchasing audit quality | |
CN109636645A (en) | Medical insurance monitoring and managing method, unit and computer readable storage medium | |
CN109671476A (en) | Recognition methods, device, terminal and the computer readable storage medium of unrelated medication | |
CN109102386A (en) | Risk control method and device based on big data | |
CN109409619A (en) | Prediction technique, device, medium and the electronic equipment of public sentiment trend | |
Naidoo et al. | Unsupervised anomaly detection of healthcare providers using generative adversarial networks | |
CN109636648A (en) | Social security violation detection method, device, equipment and computer storage medium | |
CN109598633A (en) | Social security violation detection method, device, equipment and computer storage medium | |
CN109559242A (en) | Processing method, device, equipment and the computer readable storage medium of abnormal data | |
CN109635113A (en) | Abnormal insured people purchases medicine data detection method, device, equipment and storage medium | |
Gao et al. | Predicting opioid use disorder and associated risk factors in a Medicaid managed care population. | |
CN109615538A (en) | Social security violation detection method, device, equipment and computer storage medium | |
CN109616185A (en) | The method and relevant device of inspection item behavior are issued in detection in violation of rules and regulations | |
CN109658267A (en) | Social security violation detection method, device, equipment and computer storage medium | |
CN112382388A (en) | Early warning method for adverse pressure sore event | |
CN109670973A (en) | Social security violation detection method, device, equipment and computer storage medium | |
JP5933863B1 (en) | Data analysis system, control method, control program, and recording medium | |
CN109993365A (en) | Probability of illness prediction technique, device, computer equipment and storage medium | |
CN109657059A (en) | Abnormal insured people purchases medicine data detection method, device, equipment and storage medium | |
Rajapaksha et al. | Identifying adverse drug reactions by analyzing Twitter messages |
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 | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190423 |
|
RJ01 | Rejection of invention patent application after publication |