CN109492095A - Claims Resolution data processing method, device, computer equipment and storage medium - Google Patents
Claims Resolution data processing method, device, computer equipment and storage medium Download PDFInfo
- Publication number
- CN109492095A CN109492095A CN201811201175.3A CN201811201175A CN109492095A CN 109492095 A CN109492095 A CN 109492095A CN 201811201175 A CN201811201175 A CN 201811201175A CN 109492095 A CN109492095 A CN 109492095A
- Authority
- CN
- China
- Prior art keywords
- settlement
- resolution
- classification
- information
- case
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/40—Document-oriented image-based pattern recognition
- G06V30/41—Analysis of document content
- G06V30/413—Classification of content, e.g. text, photographs or tables
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; 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 OR CALCULATING; COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/10—Character recognition
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Artificial Intelligence (AREA)
- Multimedia (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Marketing (AREA)
- Strategic Management (AREA)
- Technology Law (AREA)
- General Business, Economics & Management (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
本发明公开了一种理赔数据处理方法、装置、计算机设备及存储介质，所述方法包括：识别理赔案件文档，获取所述理赔案件文档中的理赔信息；在预设的神经网络模型中输入所述理赔信息，根据所述神经网络模型对输入的所述理赔信息进行归类，并获取所述神经网络模型输出的归类后的所述理赔信息的预设归类值；根据归类后的所述理赔信息的预设归类值评估所述理赔案件是否存在理赔风险；在理赔案件不存在理赔风险时，输出所述理赔案件对应的理赔结果。本发明通过对理赔信息进行大数据处理，从而实现全程无人工操作的自动理赔，提高了理赔效率，缩短了理赔时间验，为客户提供了极速理赔体验。
The invention discloses a claim settlement data processing method, device, computer equipment and storage medium. The method includes: identifying a claim settlement case document, acquiring the claim settlement information in the claim settlement case document; inputting all the data in a preset neural network model the claim settlement information, classify the input claim settlement information according to the neural network model, and obtain a preset classification value of the classified claim settlement information output by the neural network model; The preset classification value of the claim settlement information evaluates whether the claim settlement case has a claim settlement risk; when the claim settlement case does not have a claim settlement risk, the claim settlement result corresponding to the claim settlement case is output. The invention realizes the automatic claim settlement without manual operation in the whole process by processing the big data of the claim settlement information, improves the claim settlement efficiency, shortens the claim settlement time and provides the customers with an extremely fast claim settlement experience.
Description
Technical field
The present invention relates to big data processing fields, and in particular to a kind of Claims Resolution data processing method, device, computer equipment
And storage medium.
Background technique
Currently, increasing with insurance portfolio, the demand of settlement of insurance claim also increases therewith.In the prior art, although
Some insurance companies have had been introduced into automatic Claims Resolution system, but the automatic Claims Resolution system still needs client voluntarily to fill in Claims Resolution money
Material, the program are disadvantageous in that: since the Claims Resolution data of Claims Resolution case is filled in or provided by customer self-service, due to visitor
Family does not have the professional knowledge of insurance industry, causes the Claims Resolution data of much Claims Resolution cases to be filled in wrong, and can not be by automatic
The automatic audit of Claims Resolution system, also, still can not need to carry out manual examination and verification by audit, in this way, will lead to Claims Resolution effect
Rate is low, and Claims Resolution speed is slow.Therefore, a kind of automatic Claims Resolution scheme that can be further improved Claims Resolution treatment effeciency is currently needed.
Summary of the invention
The embodiment of the present invention provides a kind of Claims Resolution data processing method, device, computer equipment and storage medium, the present invention
, to realize the automatic Claims Resolution of whole prosthetic operation, Claims Resolution can be improved by carrying out big data processing to Claims Resolution information
Efficiency, shortening the Claims Resolution time tests, and provides very fast Claims Resolution for client and experiences.
A kind of Claims Resolution data processing method, comprising:
Identification Claims Resolution case document obtains the Claims Resolution information in the Claims Resolution case document；
The Claims Resolution information is inputted in preset neural network model, according to the neural network model to the institute of input
It states Claims Resolution information to be sorted out, and obtains the default classification of the Claims Resolution information after the classification of the neural network model output
Value；
The Claims Resolution case is assessed with the presence or absence of Claims Resolution risk according to the default classification value of the Claims Resolution information after classification；
In case of settling a claim there is no when Claims Resolution risk, the corresponding Claims Resolution result of the Claims Resolution case is exported.
A kind of Claims Resolution data processing equipment, comprising:
Identification module, case document of settling a claim for identification obtain the Claims Resolution information in the Claims Resolution case document；
Classifying module, for inputting the Claims Resolution information in preset neural network model, according to the neural network
Model sorts out the Claims Resolution information of input, and obtains the Claims Resolution after the classification of the neural network model output
The default classification value of information；
Evaluation module, for whether assessing the Claims Resolution case according to the default classification value of the Claims Resolution information after classification
In the presence of Claims Resolution risk；
Output module, for there is no when Claims Resolution risk, export the corresponding Claims Resolution knot of the Claims Resolution case in case of settling a claim
Fruit.
A kind of computer equipment, including memory, processor and storage are in the memory and can be in the processing
The computer program run on device, the processor realize above-mentioned Claims Resolution data processing method when executing the computer program
Step.
A kind of computer readable storage medium, the computer-readable recording medium storage have computer program, the meter
The step of calculation machine program realizes above-mentioned Claims Resolution data processing method when being executed by processor.
Claims Resolution data processing method, device, computer equipment and storage medium provided by the invention, are known by optical character
Other model carries out automatic identification to the Claims Resolution case document, obtains Claims Resolution information；And according to preset neural network model pair
The Claims Resolution information is sorted out, while there is no reasons assessing the Claims Resolution case automatically according to the Claims Resolution information after classification
After paying for risk, settled a claim automatically.This motion can be by carrying out big data processing to Claims Resolution information, to realize whole nothing
Manually-operated automatic Claims Resolution, improves Claims Resolution efficiency, shortens the Claims Resolution time and tests, provides very fast Claims Resolution for client and experience.
Detailed description of the invention
In order to illustrate the technical solution of the embodiments of the present invention more clearly, below by institute in the description to the embodiment of the present invention
Attached drawing to be used is needed to be briefly described, it should be apparent that, the accompanying drawings in the following description is only some implementations of the invention
Example, for those of ordinary skill in the art, without creative efforts, can also obtain according to these attached drawings
Obtain other attached drawings.
Fig. 1 is the application environment schematic diagram of Claims Resolution data processing method in one embodiment of the invention；
Fig. 2 is the flow chart of Claims Resolution data processing method in one embodiment of the invention；
Fig. 3 is the flow chart of the step S10 of Claims Resolution data processing method in one embodiment of the invention；
Fig. 4 is the flow chart of the step S10 of Claims Resolution data processing method in another embodiment of the present invention；
Fig. 5 is the flow chart of the step S20 of Claims Resolution data processing method in one embodiment of the invention；
Fig. 6 is the flow chart of the step S202 of Claims Resolution data processing method in one embodiment of the invention；
Fig. 7 is the functional block diagram of Claims Resolution data processing equipment in one embodiment of the invention；
Fig. 8 is the functional block diagram of the identification module of Claims Resolution data processing equipment in one embodiment of the invention；
Fig. 9 is the functional block diagram of Claims Resolution data processing equipment in another embodiment of the present invention；
Figure 10 is the schematic diagram of computer equipment in one embodiment of the invention.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiments are some of the embodiments of the present invention, instead of all the embodiments.Based on this hair
Embodiment in bright, every other implementation obtained by those of ordinary skill in the art without making creative efforts
Example, shall fall within the protection scope of the present invention.
Claims Resolution data processing method provided by the invention, can be applicable in the application environment such as Fig. 1, wherein client (meter
Calculate machine equipment) it is communicated by network with server.Wherein, client (computer equipment) is including but not limited to various
People's computer, laptop, smart phone, tablet computer, camera and portable wearable device.Server can be with solely
The server clusters of the either multiple servers compositions of vertical server is realized.
In one embodiment, it as shown in Fig. 2, providing a kind of Claims Resolution data processing method, applies in Fig. 1 in this way
It is illustrated for server, comprising the following steps:
S10, identification Claims Resolution case document, obtain the Claims Resolution information in the Claims Resolution case document.
That is, in the present embodiment, it is necessary first to obtain the document for the Claims Resolution case settled a claim, and then according to institute
It states Claims Resolution case document and obtains Claims Resolution information therein, to calculate Claims Resolution automatically according to the Claims Resolution information as a result, real in turn
Now automatic Claims Resolution.
In one embodiment, as shown in figure 3, the step S10 the following steps are included:
S101 receives identification instruction, transfers Claims Resolution case document according to the case information for including in the identification instruction.
Wherein, the identification instruction refers to that user triggers pre-set button by the triggering modes such as click or sliding in client
It is sent to the identification instruction of server later.Since client is before click pre-set button sends identification instruction, just necessarily
Claims Resolution case that confirmed needs are settled a claim automatically simultaneously selects it, therefore, contains in the identification instruction described
The case information for case of settling a claim necessarily includes that title of a cause, case unique encodings, case type etc. are used in the case information
Distinguish the information of the Claims Resolution case and other Claims Resolution cases.
And in the database of server, when establishing the Claims Resolution case, it has been stored with the correlation text of the Claims Resolution case
Shelves (namely Claims Resolution case document), the Claims Resolution case document includes the Claims Resolution for having to submit when the Claims Resolution case is settled a claim
The scanned copy etc. of material, for example, in accident/injury insurance Claims Resolution case, it is desirable to provide Claims Resolution case document include medicine
Diagnosis proves, the sudden hurt accident provided of relevant department proves, payment for medical care original invoice and prescription, personal identification papers or household register
Prove etc..
In one embodiment, before the step S101, Claims Resolution litigious party (user for receiving Claims Resolution) or Claims Resolution
Service administrators (administrator for managing the Claims Resolution case), which upload the Claims Resolution case document, (can scan the Claims Resolution
Scanned copy is uploaded after case document, or scans the paper while submitting Claims Resolution case document paper on self-service equipment
Part simultaneously uploads scanned copy) it is closed into the database of the server, and by the Claims Resolution case document and the case information
Connection storage.
Server after receiving identification instruction, according to include in the identification instruction case information (such as
Title of a cause or case unique encodings) Claims Resolution case document corresponding with the case information in called data library, and in step
It is identified in S102.
S102 identifies the Claims Resolution case document by optical character identification model, obtains the Claims Resolution case
Claims Resolution information in document.
That is, in the present embodiment, can by optical character identification (Optical Character Recognition,
Referred to as OCR) model identifies the Claims Resolution case document, to extract the specific Claims Resolution in the Claims Resolution case document
Information, for example, the medical item in payment for medical care original invoice can be read for the payment for medical care original invoice in Claims Resolution case document
Mesh and the project are corresponding digital (representing expense), and the medical item and corresponding digital auto-associating are stored.For another example,
The medical history of patient can also be read, and obtains the medical history corresponding time, and the medical history and association in time are stored.In this reality
It applies in example, timeliness is faster, error rate is lower, data is identified to Claims Resolution case document by optical character identification model
Granularity is more preferable；And the identification process will not generate the time break that manual operation needs to rest, therefore without manually being operated
Recognition speed can be consistent, and can identify that recognition efficiency is higher to multiple Claims Resolution case documents simultaneously.
Understandably, in one embodiment, before the step S102 further include: obtain document to be identified, and according to institute
It states document training to be identified and generates optical character identification model.
The document to be identified is the scanned copy settled a claim and have to the Claims Resolution material submitted when case is settled a claim；For example,
In the training optical character identification model identification payment for medical care original invoice, the payment for medical care that 2000 same types can be used is former
Ticket is originated as being required to after being learnt each time according to a payment for medical care original invoice according to Training document
Content correction is practised as a result, generating the optical character identification mould that can identify payment for medical care original invoice after by repetition learning
Type.Further, in the learning process, it can also increase what the optical character identification model distinguished the true and false of invoice etc.
Training, the training process can by learn document format, provide document unit official seal concrete shape and construction come into
Row.
In one embodiment, as shown in figure 4, before the step S102, namely by optical character identification model to institute
Claims Resolution case document is stated to be identified, further comprising the steps of before obtaining the Claims Resolution information in the Claims Resolution case document:
S103 extracts the sensitive keys word in the Claims Resolution case document, obtains the sensitive keys word in the Claims Resolution
It include the sensitive content of the sensitive keys word on locating sensitive position and the sensitive position in case document, to described
The sensitive content on sensitive position carries out delete processing or mosaic processing.
In this embodiment, since in the identification process of Claims Resolution case document, may exist will settle a claim outside case document
The case where hair is identified is (for example, the identification process of step S102 is that the identification module of external interface connection is called to be known
When other), therefore, when the Claims Resolution case document is carried out outgoing, need first to carry out desensitization process to it, at this time first by institute
The sensitive content removal (such as claims adjuster's identity information etc.) in Claims Resolution case document is stated, then the Claims Resolution case after desensitization is literary
Shelves are identified.
Therefore, in the present embodiment, it is necessary first to extract the sensitive keys word in the Claims Resolution case document, the sensitivity
Keyword can be preset, for example, the sensitive keys word is set as name, identification card number, at this point, the Claims Resolution case
The sensitive content on the sensitive position in document may be to be following comprising the sensitive keys word " name " and " identity card
Number " following the description: " name: king is small by two；Identification card number: XX ... XX ".And the sensitive position is the sensitive content pair
The position answered.
The delete processing and mosaic processing are one of desensitization process mode, the desensitization process side in the present invention
Formula can also be not limited to it is above-mentioned, as long as can achieve the effect that can by it is described Claims Resolution case document in sensitive content removal i.e.
It can.
S20 inputs the Claims Resolution information, according to the neural network model to input in preset neural network model
The Claims Resolution information sorted out, and obtain the default of the Claims Resolution information after the classification of neural network model output
Classification value.
Wherein, the Claims Resolution information is the Claims Resolution information in the Claims Resolution case document obtained in above-mentioned steps S10, described pre-
If classification value refers to the designated value of the Claims Resolution information after being sorted out；For example, can be by " the inspection in payment for medical care original invoice
Look into expense " it is set as designated value, but " Laboratory Fee " can be written as to " procuratorial work expense " in having some payment for medical care original invoices or " examined
Take " etc., at this point, the Claims Resolution information (" procuratorial work expense " or " survey fees ") identified is different from default classification value, institute can be passed through
It states neural network model and it is uniformly classified as to " Laboratory Fee " this designated value (namely above-mentioned default classification value).
It can be directly that default classification value or correspondence are revised as presetting by the Claims Resolution information labeling after being sorted out
Classification value can be convenient quickly corresponding storage reason after extracting the Claims Resolution information in Claims Resolution case document
Pay for information.
In one embodiment, as shown in figure 5, it is before the step S20 namely described according to preset neural network model
Before sorting out to the Claims Resolution information, further includes:
S201 is obtained and is sorted out training sample；The training sample of sorting out is the history Claims Resolution in history Claims Resolution case document
Information；
That is, using the Claims Resolution information of the above-mentioned Claims Resolution case document after desensitization as classification training sample.
S202 is obtained when sorting out by the inclusion of the neural network model of initial parameter to the classification training sample
The whole degree of deviation, the entirety degree of deviation are the whole deviation between the Claims Resolution value of information and default classification value obtained after sorting out
Degree.
S203, judges whether the whole degree of deviation is greater than preset first threshold；The first threshold can be according to need
It asks and is set.
S204, if the entirety degree of deviation is greater than the first threshold, to the initial parameter of the neural network model
It is adjusted, and returns and execute the entirety calculated when sorting out using neural network model to the classification training sample
The degree of deviation, until the whole degree of deviation is less than or equal to the first threshold；
S205 prompts the neural network model to instruct if the entirety degree of deviation is less than or equal to the first threshold
Practice and completes.At this point, the neural network model training is completed.The neural network model determined have passed through a large amount of sample instruction
Practice, and its whole degree of deviation is maintained in a lesser range (being less than or equal to first threshold), uses the neural network mould
Type pair is worth the Claims Resolution information that the substantially identical but form of expression is not inconsistent with default classification and handles, and default classification value can be obtained.
In one embodiment, as shown in fig. 6, in the step S202, the nerve obtained by the inclusion of initial parameter
Whole degree of deviation when network model sorts out the classification training sample, comprising the following steps:
S2021 chooses the classification training sample that one is not yet selected for sorting out from the classification training sample and makees
For current sample.Sample selection sequence can be it is random, be also possible to according to preset sequence carry out, for example, in advance
Label can be carried out to the classification training sample, then successively be chosen according to the sequence of label from small to large.
S2022 is handled the Claims Resolution information in the current sample using the neural network model, is obtained described
The Claims Resolution value of information after current sample classification.That is, sample current for first, is using the nerve comprising initial parameter
Network model handles Claims Resolution information therein, obtains the Claims Resolution value of information after first current sample is sorted out.But it is right
In other subsequent current samples, exactly with the neural network model adjusted in step S204 after the initial parameter
It is handled, Claims Resolution information therein is handled, obtain the Claims Resolution value of information after the sample is sorted out.
S2023 determines the Claims Resolution value of information and institute after the current sample classification according to preset deviation decision rule
State the sample bias degree between default classification value.According to the semantic association relationship etc. between word and word in the deviation decision rule
Set the deviation ratio between different the Claims Resolution value of information and the default classification value.In the present embodiment, according to preset inclined
Poor decision rule obtains the deviation between the Claims Resolution value of information and the default classification value after the current sample is sorted out
The deviation ratio is recorded as sample bias degree by ratio.
S2024 judges in the classification training sample with the presence or absence of the classification training sample for being not yet selected for sorting out.
S2025 is not yet selected for the classification training sample sorted out if it exists, then continues from the classification training sample
The classification training sample that middle selection one is not yet selected for sorting out is as current sample；That is, receipt row step S2021 with
And subsequent step.
S2026 is not yet selected for the classification training sample sorted out if it does not exist, will be selected for sorting out all
The sum of the sample bias degree for sorting out training sample is determined as the whole degree of deviation.
In one embodiment, after the step S20 further include:
Attribute value corresponding with the Claims Resolution information after classification in Claims Resolution data list is obtained, and by the institute after classification
It states Claims Resolution information and stores the position corresponding with the attribute value into Claims Resolution data list.
In the present embodiment, after sorting out to all Claims Resolution information, namely the statement of each Claims Resolution information is become
After standard is unified, it can be corresponded in insertion Claims Resolution data list, include each Claims Resolution information in the Claims Resolution data list
And its attribute；The attribute includes the contents such as hospital name, the amount of money, inspection or pharmaceutical items；Obtain the Claims Resolution information it
Afterwards, system can be according to all Claims Resolution information in the corresponding Claims Resolution case document of the Claims Resolution information got (if the ratio Claims Resolution
Information is corresponding with the amount of money, it should which it is inspection or pharmaceutical items that anticipation, which corresponds to the amount of money,；If detecting number and having below
Printed words such as " members ", can be judged to check in advance or the expense of pharmaceutical items) and document format etc. prejudge the attribute of the Claims Resolution information, and
In this step, the attribute value with the attributes match of anticipation is found in Claims Resolution data list；Hereafter, the Claims Resolution is believed
Breath correspondence is inserted into position corresponding with the attribute value in the Claims Resolution data list.
S30 assesses the Claims Resolution case with the presence or absence of Claims Resolution according to the default classification value of the Claims Resolution information after classification
Risk.
Wherein, Claims Resolution risk can be assessed by the default classification value of the Claims Resolution information after sorting out, for example,
By the case type with the Claims Resolution case, (case type is one in the default classification value of the Claims Resolution information, in step
The case type of Claims Resolution case is obtained in S10, for example is the case types such as certain state of an illness in accident insurance, serious illness insurance) it is corresponding
Amount for which loss settled highest amount assessed, the amount for which loss settled being calculated according to above-mentioned Claims Resolution information be more than preset Claims Resolution
When amount of money highest amount, assessment show that the Claims Resolution case has Claims Resolution risk.
Risk of settling a claim can also pass through Claims Resolution project corresponding with the case type of the Claims Resolution case and list of charges (Claims Resolution
Project and list of charges are also the default classification value of the Claims Resolution information, and specifically, the Claims Resolution project and list of charges include
The ultimate cost of the corresponding Claims Resolution project of the case type and the Claims Resolution project) it is assessed, according in above-mentioned Claims Resolution information
Claims Resolution project it is more than the Claims Resolution project in the Claims Resolution project and list of charges or the Claims Resolution project that has more is more than default threshold
The corresponding total cost of Claims Resolution project be worth, having more is more than preset exceeded cost value, the expense of some or multiple Claims Resolution projects
When occurring with one or more when being more than its regular fee range, assessment show that the Claims Resolution case has wind of settling a claim
Danger.
When assessment show that the Claims Resolution case is commented in the presence of Claims Resolution risk, prompts Claims Resolution risk and there is the original of Claims Resolution risk
Cause；The case can be transferred to preset business personnel and carry out manual examination and verification.
S40 exports the corresponding Claims Resolution result of the Claims Resolution case in case of settling a claim there is no when Claims Resolution risk.
Understandably, assessment settle a claim risk during, if all Claims Resolution information with the case type of anticipation
The amount for which loss settled of similar Claims Resolution case is consistent, at this point, assessment result is that the Claims Resolution case is normal, can enter automatic Claims Resolution stream
Journey exports the corresponding Claims Resolution result of the Claims Resolution case.The Claims Resolution result includes the Claims Resolution gold calculated according to the Claims Resolution information
Volume simultaneously pays the amount for which loss settled to Claims Resolution litigious party automatically by preset clearing side；Realize whole prosthetic behaviour
The automatic Claims Resolution made, greatly optimizes Claims Resolution efficiency.
Claims Resolution data processing method provided by the invention, by optical character identification model to the Claims Resolution case document into
Row automatic identification obtains Claims Resolution information；And the Claims Resolution information is sorted out according to preset neural network model, while
The Claims Resolution case is assessed automatically there is no after Claims Resolution risk according to the Claims Resolution information after classification, is settled a claim automatically.This
Motion, to realize the automatic Claims Resolution of whole prosthetic operation, can be improved by carrying out big data processing to Claims Resolution information
It settles a claim efficiency, shortening the Claims Resolution time tests, and provides very fast Claims Resolution for client and experiences.
In one embodiment, as shown in fig. 7, providing a kind of Claims Resolution data processing equipment, the Claims Resolution data processing equipment with
Data processing method of settling a claim in above-described embodiment corresponds.The Claims Resolution data processing equipment includes:
Identification module 11, case document of settling a claim for identification obtain the Claims Resolution information in the Claims Resolution case document；
Classifying module 12, for inputting the Claims Resolution information in preset neural network model, according to the nerve net
Network model sorts out the Claims Resolution information of input, and obtains the reason after the classification of the neural network model output
Pay for the default classification value of information；
Evaluation module 13, for being according to the default classification value of the Claims Resolution information after the classification assessment Claims Resolution case
It is no to there is Claims Resolution risk；
Output module 14, for, there is no when Claims Resolution risk, exporting the corresponding Claims Resolution of the Claims Resolution case in case of settling a claim
As a result.
Claims Resolution data processing equipment provided by the invention, by optical character identification model to the Claims Resolution case document into
Row automatic identification obtains Claims Resolution information；And the Claims Resolution information is sorted out according to preset neural network model, while
The Claims Resolution case is assessed automatically there is no after Claims Resolution risk according to the Claims Resolution information after classification, is settled a claim automatically.This
Motion, to realize the automatic Claims Resolution of whole prosthetic operation, can be improved by carrying out big data processing to Claims Resolution information
It settles a claim efficiency, shortening the Claims Resolution time tests, and provides very fast Claims Resolution for client and experiences.
In one embodiment, as shown in figure 8, the identification module 11 includes:
Submodule 111 is transferred, for receiving identification instruction, reason is transferred according to the case information for including in the identification instruction
Pay for case document；
Acquisition submodule 112 is obtained for being identified by optical character identification model to the Claims Resolution case document
Claims Resolution information in the Claims Resolution case document.
In one embodiment, the identification module 11 further include:
Desensitize module, for extracting the sensitive keys word in the Claims Resolution case document, obtains the sensitive keys word and exists
In the Claims Resolution case document on locating sensitive position and the sensitive position in the sensitivity comprising the sensitive keys word
Hold, delete processing or mosaic processing are carried out to the sensitive content on the sensitive position.
In one embodiment, as shown in figure 9, described device further include:
Sample acquisition module 15 sorts out training sample for obtaining；The classification training sample is history Claims Resolution case text
History Claims Resolution information in shelves；
Deviation obtains module 16, for obtaining the neural network model by the inclusion of initial parameter to classification training sample
Whole degree of deviation when this is sorted out, the entirety degree of deviation are the Claims Resolution value of information and default classification value obtained after sorting out
Between whole extent of deviation；
Judgment module 17, for judging whether the whole degree of deviation is greater than preset first threshold；
Module 18 is adjusted, is used for when the whole degree of deviation is greater than the first threshold, to the neural network model
Initial parameter be adjusted, and return execute it is described calculating the classification training sample is returned using neural network model
Whole degree of deviation when class, until the whole degree of deviation is less than or equal to the first threshold；
Cue module 19, for prompting the nerve when the whole degree of deviation is less than or equal to the first threshold
Network model training is completed.
In one embodiment, the deviation obtains module 16 further include:
Submodule is chosen, for choosing the classification instruction for being not yet selected for sorting out from the classification training sample
Practice sample as current sample；
Handle submodule, for using the neural network model to the Claims Resolution information in the current sample at
Reason obtains the Claims Resolution value of information after the current sample is sorted out；
Submodule is set, for determining that the Claims Resolution after the current sample classification is believed according to preset deviation decision rule
Sample bias degree between breath value and the default classification value；
Judging submodule is returned in the settlement of insurance claim sample database with the presence or absence of be not yet selected for sorting out for judging
Class training sample；
Continue to choose submodule, for continuing from institute when there is the classification training sample for being not yet selected for classification
It states selection one in classification training sample and is not yet selected for the classification training sample of classification as current sample；
Determine submodule, for when there is no the classification training sample for being not yet selected for sorting out, will be selected into
The sum of the sample bias degree for all classification training samples that row is sorted out is determined as the whole degree of deviation.
Specific about Claims Resolution data processing equipment limits the limit that may refer to above for Claims Resolution data processing method
Fixed, details are not described herein.Modules in above-mentioned Claims Resolution data processing equipment can fully or partially through software, hardware and its
Combination is to realize.Above-mentioned each module can be embedded in the form of hardware or independently of in the processor in computer equipment, can also be with
It is stored in the memory in computer equipment in a software form, in order to which processor calls the above modules of execution corresponding
Operation.
In one embodiment, a kind of computer equipment is provided, which can be server, internal junction
Composition can be as shown in Figure 10.The computer equipment include by system bus connect processor, memory, network interface and
Database.Wherein, the processor of the computer equipment is for providing calculating and control ability.The memory packet of the computer equipment
Include non-volatile memory medium, built-in storage.The non-volatile memory medium is stored with operating system, computer program and data
Library.The built-in storage provides environment for the operation of operating system and computer program in non-volatile memory medium.The calculating
To realize a kind of Claims Resolution data processing method when machine program is executed by processor.
In one embodiment, a kind of computer equipment is provided, including memory, processor and storage are on a memory
And the computer program that can be run on a processor, processor perform the steps of when executing computer program
Identification Claims Resolution case document obtains the Claims Resolution information in the Claims Resolution case document；In preset neural network mould
The Claims Resolution information is inputted in type, is sorted out according to the Claims Resolution information of the neural network model to input, and obtain
The default classification value of the Claims Resolution information after the classification of the neural network model output；According to the Claims Resolution letter after classification
The default classification value of breath assesses the Claims Resolution case with the presence or absence of Claims Resolution risk；It is defeated when Claims Resolution risk is not present in case of settling a claim
The corresponding Claims Resolution result of the Claims Resolution case out.
Computer equipment provided by the invention carries out the Claims Resolution case document by optical character identification model automatic
Identification obtains Claims Resolution information；And the Claims Resolution information is sorted out according to preset neural network model, while returning in basis
Claims Resolution information after class assesses the Claims Resolution case there is no after Claims Resolution risk automatically, is settled a claim automatically.This motion can
To realize the automatic Claims Resolution of whole prosthetic operation, to improve Claims Resolution effect by carrying out big data processing to Claims Resolution information
Rate, shortening the Claims Resolution time tests, and provides very fast Claims Resolution for client and experiences.
In one embodiment, a kind of computer readable storage medium is provided, computer program is stored thereon with, is calculated
Machine program performs the steps of when being executed by processor
Identification Claims Resolution case document obtains the Claims Resolution information in the Claims Resolution case document；In preset neural network mould
The Claims Resolution information is inputted in type, is sorted out according to the Claims Resolution information of the neural network model to input, and obtain
The default classification value of the Claims Resolution information after the classification of the neural network model output；According to the Claims Resolution letter after classification
The default classification value of breath assesses the Claims Resolution case with the presence or absence of Claims Resolution risk；It is defeated when Claims Resolution risk is not present in case of settling a claim
The corresponding Claims Resolution result of the Claims Resolution case out.
Storage medium provided by the invention knows the Claims Resolution case document by optical character identification model automatically
Not, Claims Resolution information is obtained；And the Claims Resolution information is sorted out according to preset neural network model, while according to classification
Claims Resolution information later assesses the Claims Resolution case there is no after Claims Resolution risk automatically, is settled a claim automatically.This motion can be with
By carrying out big data processing to Claims Resolution information, to realize the automatic Claims Resolution of whole prosthetic operation, Claims Resolution efficiency is improved,
Shortening the Claims Resolution time tests, and provides very fast Claims Resolution for client and experiences.
Those of ordinary skill in the art will appreciate that realizing all or part of the process in above-described embodiment method, being can be with
Relevant hardware is instructed to complete by computer program, the computer program can be stored in a non-volatile computer
In read/write memory medium, the computer program is when being executed, it may include such as the process of the embodiment of above-mentioned each method.Wherein,
To any reference of memory, storage, database or other media used in each embodiment provided by the present invention,
Including non-volatile and/or volatile memory.Nonvolatile memory may include read-only memory (ROM), programming ROM
(PROM), electrically programmable ROM (EPROM), electrically erasable ROM (EEPROM) or flash memory.Volatile memory may include
Random access memory (RAM) or external cache.By way of illustration and not limitation, RAM is available in many forms,
Such as static state RAM (SRAM), dynamic ram (DRAM), synchronous dram (SDRAM), double data rate sdram (DDRSDRAM), enhancing
Type SDRAM (ESDRAM), synchronization link DRAM (SLDRAM), the direct RAM of memory bus (RDRAM), direct memory bus
Dynamic ram (DRDRAM) and memory bus dynamic ram (RDRAM) etc..
It is apparent to those skilled in the art that for convenience of description and succinctly, only with above-mentioned each function
Can unit or module division progress for example, in practical application, can according to need and by above-mentioned function distribution by difference
Functional unit or module complete, i.e., the internal structure of described device is divided into different functional unit or module, with complete
All or part of function described above.
Embodiment described above is merely illustrative of the technical solution of the present invention, rather than its limitations；Although referring to aforementioned reality
Applying example, invention is explained in detail, those skilled in the art should understand that: it still can be to aforementioned each
Technical solution documented by embodiment is modified or equivalent replacement of some of the technical features；And these are modified
Or replacement, the spirit and scope for technical solution of various embodiments of the present invention that it does not separate the essence of the corresponding technical solution should all
It is included within protection scope of the present invention.
Claims (10)
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811201175.3A CN109492095A (en) | 2018-10-16 | 2018-10-16 | Claims Resolution data processing method, device, computer equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811201175.3A CN109492095A (en) | 2018-10-16 | 2018-10-16 | Claims Resolution data processing method, device, computer equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109492095A true CN109492095A (en) | 2019-03-19 |
Family
ID=65690259
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811201175.3A Pending CN109492095A (en) | 2018-10-16 | 2018-10-16 | Claims Resolution data processing method, device, computer equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109492095A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110136010A (en) * | 2019-04-18 | 2019-08-16 | 中国平安财产保险股份有限公司 | Method, device and computer equipment for judging risk events based on neural network |
| CN110610431A (en) * | 2019-08-15 | 2019-12-24 | 中国平安人寿保险股份有限公司 | Intelligent claim settlement method and intelligent claim settlement system based on big data |
| CN112528887A (en) * | 2020-12-16 | 2021-03-19 | 支付宝(杭州)信息技术有限公司 | Auditing method and device |
| CN113792570A (en) * | 2020-11-30 | 2021-12-14 | 京东安联财产保险有限公司 | Claims data processing method and device, electronic equipment and readable medium |
| CN114170029A (en) * | 2021-11-26 | 2022-03-11 | 深圳麦亚信科技股份有限公司 | Data processing method and device, computer equipment and storage medium |
| CN116228441A (en) * | 2023-03-16 | 2023-06-06 | 平安健康保险股份有限公司 | Method, device, electronic equipment and storage medium for recommending self-checking rules of claim settlement |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6049773A (en) * | 1997-10-14 | 2000-04-11 | Reclaim Technology And Services Limited | Automated method for identification of reinsurance claims |
| CN1989512A (en) * | 2004-06-08 | 2007-06-27 | 格林莱恩系统公司 | System and method for risk assessment and management in a variety of systems and subsystems |
| CN107316263A (en) * | 2016-04-27 | 2017-11-03 | 深圳关心万家健康管理有限公司 | A kind of method of case history processing |
| US20170352105A1 (en) * | 2014-08-20 | 2017-12-07 | Midwest Employers Casualty Company | Machine learning risk factor identification and mitigation system |
| CN107871284A (en) * | 2017-11-22 | 2018-04-03 | 平安科技（深圳）有限公司 | The appraisal procedure and device of risk Claims Resolution |
| CN107895168A (en) * | 2017-10-13 | 2018-04-10 | 平安科技（深圳）有限公司 | The method of data processing, the device of data processing and computer-readable recording medium |
| CN107909299A (en) * | 2017-12-11 | 2018-04-13 | 凯泰铭科技（北京）有限公司 | People hinders Claims Resolution data risk checking method and system |
| US20180121756A1 (en) * | 2016-10-28 | 2018-05-03 | Intuit Inc. | Image quality assessment and improvement for performing optical character recognition |
| CN108256720A (en) * | 2017-11-07 | 2018-07-06 | 中国平安财产保险股份有限公司 | A kind of settlement of insurance claim methods of risk assessment and terminal device |
| CN108346036A (en) * | 2018-03-06 | 2018-07-31 | 平安科技（深圳）有限公司 | Insurance policy concentrates vouching method, electronic device and readable storage medium storing program for executing |
| CN108388575A (en) * | 2018-01-12 | 2018-08-10 | 中国平安人寿保险股份有限公司 | A kind of appraisal procedure, device, terminal device and the storage medium of settlement of insurance claim expense |
-
2018
- 2018-10-16 CN CN201811201175.3A patent/CN109492095A/en active Pending
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6049773A (en) * | 1997-10-14 | 2000-04-11 | Reclaim Technology And Services Limited | Automated method for identification of reinsurance claims |
| CN1989512A (en) * | 2004-06-08 | 2007-06-27 | 格林莱恩系统公司 | System and method for risk assessment and management in a variety of systems and subsystems |
| US20170352105A1 (en) * | 2014-08-20 | 2017-12-07 | Midwest Employers Casualty Company | Machine learning risk factor identification and mitigation system |
| CN107316263A (en) * | 2016-04-27 | 2017-11-03 | 深圳关心万家健康管理有限公司 | A kind of method of case history processing |
| US20180121756A1 (en) * | 2016-10-28 | 2018-05-03 | Intuit Inc. | Image quality assessment and improvement for performing optical character recognition |
| CN107895168A (en) * | 2017-10-13 | 2018-04-10 | 平安科技（深圳）有限公司 | The method of data processing, the device of data processing and computer-readable recording medium |
| CN108256720A (en) * | 2017-11-07 | 2018-07-06 | 中国平安财产保险股份有限公司 | A kind of settlement of insurance claim methods of risk assessment and terminal device |
| CN107871284A (en) * | 2017-11-22 | 2018-04-03 | 平安科技（深圳）有限公司 | The appraisal procedure and device of risk Claims Resolution |
| CN107909299A (en) * | 2017-12-11 | 2018-04-13 | 凯泰铭科技（北京）有限公司 | People hinders Claims Resolution data risk checking method and system |
| CN108388575A (en) * | 2018-01-12 | 2018-08-10 | 中国平安人寿保险股份有限公司 | A kind of appraisal procedure, device, terminal device and the storage medium of settlement of insurance claim expense |
| CN108346036A (en) * | 2018-03-06 | 2018-07-31 | 平安科技（深圳）有限公司 | Insurance policy concentrates vouching method, electronic device and readable storage medium storing program for executing |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110136010A (en) * | 2019-04-18 | 2019-08-16 | 中国平安财产保险股份有限公司 | Method, device and computer equipment for judging risk events based on neural network |
| CN110610431A (en) * | 2019-08-15 | 2019-12-24 | 中国平安人寿保险股份有限公司 | Intelligent claim settlement method and intelligent claim settlement system based on big data |
| CN110610431B (en) * | 2019-08-15 | 2023-10-27 | 中国平安人寿保险股份有限公司 | Intelligent claim settlement method and intelligent claim settlement system based on big data |
| CN113792570A (en) * | 2020-11-30 | 2021-12-14 | 京东安联财产保险有限公司 | Claims data processing method and device, electronic equipment and readable medium |
| CN112528887A (en) * | 2020-12-16 | 2021-03-19 | 支付宝(杭州)信息技术有限公司 | Auditing method and device |
| CN114170029A (en) * | 2021-11-26 | 2022-03-11 | 深圳麦亚信科技股份有限公司 | Data processing method and device, computer equipment and storage medium |
| CN116228441A (en) * | 2023-03-16 | 2023-06-06 | 平安健康保险股份有限公司 | Method, device, electronic equipment and storage medium for recommending self-checking rules of claim settlement |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109492095A (en) | Claims Resolution data processing method, device, computer equipment and storage medium | |
| US11526579B1 (en) | System and methods for performing automatic data aggregation | |
| Intarot et al. | Influencing factor in e-wallet acceptant and use | |
| CN108509485B (en) | Data preprocessing method and device, computer equipment and storage medium | |
| US20220215243A1 (en) | Risk-Reliability Framework for Evaluating Synthetic Data Models | |
| US20210342743A1 (en) | Model aggregation using model encapsulation of user-directed iterative machine learning | |
| US20140258169A1 (en) | Method and system for automated verification of customer reviews | |
| US11847390B2 (en) | Generation of synthetic data using agent-based simulations | |
| CN112035595B (en) | Method and device for constructing auditing rule engine in medical field and computer equipment | |
| CN109299094A (en) | Tables of data processing method, device, computer equipment and storage medium | |
| US20220215242A1 (en) | Generation of Secure Synthetic Data Based On True-Source Datasets | |
| US12106026B2 (en) | Extensible agents in agent-based generative models | |
| CN112288279A (en) | Business risk assessment method and device based on natural language processing and linear regression | |
| CN109816200A (en) | Task method for pushing, device, computer equipment and storage medium | |
| US20220215262A1 (en) | Augmenting Datasets with Synthetic Data | |
| KR20220071875A (en) | Device and method for underwriting a person who will subscribe to insurance based on artificial neural network | |
| CN113901817A (en) | Document classification method and device, computer equipment and storage medium | |
| Lu et al. | Embedding for anomaly detection on health insurance claims | |
| Viviani et al. | What design topics do developers discuss? | |
| CN117251799A (en) | Financial certificate processing method and device, computer equipment and storage medium | |
| Klosterman | Data Science Projects with Python: A case study approach to successful data science projects using Python, pandas, and scikit-learn | |
| CN119378494A (en) | An entity relationship extraction method and system for building knowledge graphs in the financial field | |
| CN120105736A (en) | Scenario simulation generation method, device, equipment and medium | |
| CN111967999A (en) | Investigation processing method, device, computer equipment and storage medium | |
| WO2022150343A1 (en) | Generation and evaluation of secure synthetic data |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190319 |