CN108073699A

CN108073699A - Big data polymerization analysis method and device

Info

Publication number: CN108073699A
Application number: CN201711319640.9A
Authority: CN
Inventors: 姚韬; 范勇杰; 蒋小燕
Original assignee: CHINA SECTRUST Corp Ltd; China United Network Communications Group Co Ltd
Current assignee: CHINA SECTRUST Corp Ltd; China United Network Communications Group Co Ltd
Priority date: 2017-12-12
Filing date: 2017-12-12
Publication date: 2018-05-25
Anticipated expiration: 2037-12-12
Also published as: CN108073699B

Abstract

The present invention provides a kind of big data polymerization analysis method and device,The embodiment of the present invention by polymerization analysis model by being divided into two levels of basic model and polymerization model,The sequence information that polymerization analysis center receives includes at least the mark of object module and the corresponding at least one data structures mark of object module,Sequence information is sent to each data structures and identifies corresponding data structures,Intermediate data is generated according to initial data and basic model by each data structures,Summarize the intermediate data of each data structures in polymerization analysis center,Polymerization analysis is carried out to each intermediate data by the polymerization model of object module and obtains polymerization analysis result,It polymerization analysis center need not be from each data structures bulk purchase or exchange initial data,Polymerization analysis processing procedure can be completed with each data structures cooperation,Obtain polymerization analysis result,Improve the enthusiasm of each data structures cooperation,Simplify the flow of big data polymerization analysis,Substantially increase the timeliness of polymerization analysis result.

Description

Big data polymerization analysis method and device

Technical field

The present invention relates to big data technical field more particularly to a kind of big data polymerization analysis method and devices.

Background technology

With the development of big data and big data technology, the data structures for possessing big data emerge in an endless stream, big data conduct Commodity become the object of each customer transaction.To avoid that the competitive advantage of itself, mesh may be damaged after initial data is shared Preceding each data structures are mostly based on own data and sell big data commodity using the pattern marketed one's own products.

The self-built polymerization analysis center of data structures, polymerization analysis center is according to user demand, self-developing data analysis mould Type carries out big data analysis, so as to obtain big data analysis based on own initial data using self-built Data Analysis Model As a result, using big data analysis result as commodity selling to user.If own initial data can not meet user demand, By way of buying or exchanging required initial data need to be obtained from other data structures batch.Polymerization analysis center is artificial After confirmation receives whole initial data, data summarization is cleaned, operation analysis model is obtained a result, and delivers user.

But the cost of polymerization analysis center from other data structures bulk purchase initial data is very big, with other data Mechanism exchange initial data may because the competitive advantage that itself is lost when sharing of proprietary data, and with other data structures It is complicated to consult the flows such as cooperation, it is time-consuming longer, it can not ensure the timeliness of analysis result.

The content of the invention

The present invention provides a kind of big data polymerization analysis method and device, to solve polymerization analysis center in the prior art Very big from the cost of other data structures bulk purchase initial data, exchanging initial data with other data structures may be because only Have sharing for data and lose the competitive advantage of itself, and it is complicated with other data structures to consult the flows such as cooperation, take compared with The problem of growing, can not ensureing the timeliness of analysis result.

It is an aspect of the invention to provide a kind of big data polymerization analysis method, including：

Polymerization analysis center receives the sequence information that user submits, and the sequence information includes at least the mark of object module At least one data structures mark corresponding with the object module, the object module is any announced polymerization analysis mould Type, the object module include polymerization model and at least one basic model, and each basic model corresponds to unique one A data structures；

The sequence information is sent to each data structures and identifies corresponding data structures by the polymerization analysis center, So that the mark of object module of each data structures in the sequence information and the object module of storage, really The fixed basic model for belonging to the object module corresponding with itself, and corresponding with itself belong to the target mould according to described The basic model generation intermediate data of type, and the intermediate data is sent to the polymerization analysis center；

The polymerization analysis center receives the intermediate data that each data structures are sent；

The polymerization analysis carries out each intermediate data polymerization point centrally through the polymerization model of the object module Analysis, obtains polymerization analysis result.

Another aspect of the present invention is to provide a kind of big data polymerization analysis method, including：

Data structures receive the sequence information that polymerization analysis center is sent, and the sequence information includes at least object module Mark at least one data structures mark corresponding with the object module, the object module are any announced polymerization point Model is analysed, the object module includes polymerization model and at least one basic model, and each basic model corresponds to unique The data structures；

The mark of object module of the data structures in the sequence information and itself storage it is each described poly- Analysis model is closed, determines the basic model for belonging to the object module corresponding with itself；

The data structures generate mediant according to the basic model for belonging to the object module corresponding with itself According to；

The intermediate data is sent to the polymerization analysis center by the data structures, so that the polymerization analysis center By the polymerization model of the object module, polymerization analysis is carried out to each intermediate data received, obtains polymerization analysis As a result.

Another aspect of the present invention is to provide a kind of big data polymerization analysis device, including：

First communication module receives the sequence information of user's submission for polymerization analysis center, and the sequence information is at least The corresponding at least one data structures mark of mark and the object module including object module, the object module is any Announced polymerization analysis model, the object module include polymerization model and at least one basic model, each basis Model corresponds to unique data structures；

The sequence information is sent to each data structures mark by second communication module for the polymerization analysis center Corresponding data structures are known, so that the mark of object module of each data structures in the sequence information and storage The object module, the basic model for belonging to the object module corresponding with itself is determined, and according to described right with itself The basic model for belonging to the object module generation intermediate data answered, and the intermediate data is sent to the polymerization analysis Center；

The second communication module be additionally operable to that the polymerization analysis center receives that each data structures send it is described in Between data；

Polymerization analysis module, for the polymerization analysis centrally through the polymerization model of the object module, to each described Intermediate data carries out polymerization analysis, obtains polymerization analysis result.

Receiving module receives the sequence information of polymerization analysis center transmission for data structures, and the sequence information is at least The corresponding at least one data structures mark of mark and the object module including object module, the object module is any Announced polymerization analysis model, the object module include polymerization model and at least one basic model, each basis Model corresponds to unique data structures；

Determining module, for the mark of object module of the data structures in the sequence information and itself Each polymerization analysis model of storage determines the basic model for belonging to the object module corresponding with itself；

Generation module, for the data structures according to the basic mould for belonging to the object module corresponding with itself Type generates intermediate data；

The intermediate data is sent to the polymerization analysis center, so that institute by sending module for the data structures Polymerization model of the polymerization analysis centrally through the object module is stated, polymerization point is carried out to each intermediate data received Analysis, obtains polymerization analysis result.

Big data polymerization analysis method and device provided by the invention, the embodiment of the present invention is by the way that polymerization analysis model is divided For two levels of basic model and polymerization model, the sequence information that polymerization analysis center receives includes at least the mark of object module Know at least one data structures mark corresponding with object module, sequence information is sent to each data structures by polymerization analysis center Corresponding data structures are identified, by each data structures according to itself initial data and basic model generation intermediate data, then by The intermediate data of each data structures generation is summarized at polymerization analysis center, by the polymerization model of object module, to each intermediate data Carry out polymerization analysis and obtain polymerization analysis as a result, so as to directly provide intermediate data by each data structures, polymerization analysis center without Need to from each data structures bulk purchase or exchange initial data, so as to each data structures without with polymerization analysis center or its On the premise of his data structures share initial data, polymerization analysis processing procedure is completed in cooperation, obtains each polymerization analysis model pair The polymerization analysis answered simplifies the stream of big data polymerization analysis as a result, so as to improve the enthusiasm of each data structures cooperation Journey substantially increases the timeliness of polymerization analysis result.

Description of the drawings

Attached drawing herein is merged in specification and forms the part of this specification, shows the implementation for meeting the present invention Example, and the principle for explaining the present invention together with specification.

Fig. 1 is the big data polymerization analysis method flow diagram that the embodiment of the present invention one provides；

Fig. 2 is the big data polymerization analysis method flow diagram that the embodiment of the present invention three provides；

Fig. 3 is the structure diagram for the big data polymerization analysis device that the embodiment of the present invention five provides；

Fig. 4 is the structure diagram for the big data polymerization analysis device that the embodiment of the present invention seven provides.

Pass through above-mentioned attached drawing, it has been shown that the specific embodiment of the present invention will be hereinafter described in more detail.These attached drawings It is not intended to limit the scope of present inventive concept by any mode with word description, but is by reference to specific embodiment Those skilled in the art illustrate idea of the invention.

Specific embodiment

Here exemplary embodiment will be illustrated in detail, example is illustrated in the accompanying drawings.Following description is related to During attached drawing, unless otherwise indicated, the same numbers in different attached drawings represent the same or similar element.Following exemplary embodiment Described in embodiment do not represent and the consistent all embodiments of the present invention.On the contrary, they be only with it is such as appended The example of the consistent apparatus and method of some aspects being described in detail in claims, of the invention.

Noun according to the present invention is explained first：

Intelligent contract：It can realize the agreement terms that basis is made an appointment, when the condition that one is finished in advance is triggered, It can the corresponding agreement terms of automated execution.Intelligent contract is not disrupted during can guarantee automated execution agreement terms.Intelligence This term of contract at least can trace back to nineteen ninety-five, be by the cross-cutting jurisprudent Ni Kesabo (Nick of fecund Szabo) put forward.He refers to the theory of intelligent contract in several articles for being published in the website of oneself.His definition It is as follows：" an intelligent contract is a set of promise (promises) defined in digital form, can be upper including contract participant Face performs the agreement of these promises." digital form means that contract must not be not written into computer-readable code.

Cluster：It is the parallel or distributed system that the computer being mutually connected to each other by some is formed.

Timeliness：Refer to that information has decision-making only in certain period of time the attribute of value.The timeliness of decision-making is very big Restrict the objective effect of decision-making in degree.

It is multiple：Refer to two and more than.

These specific embodiments can be combined with each other below, may be at certain for the same or similar concept or process It is repeated no more in a little embodiments.Below in conjunction with attached drawing, the embodiment of the present invention is described.

Embodiment one

Fig. 1 is the big data polymerization analysis method flow diagram that the embodiment of the present invention one provides.The embodiment of the present invention is for existing The cost for having polymerization analysis center from other data structures bulk purchase initial data in technology is very big, is handed over other data structures Changing initial data may consult to close because the competitive advantage for losing itself when sharing of proprietary data, and with other data structures It is complicated the flows such as to make, takes longer, the problem of can not ensureing the timeliness of analysis result, provides big data polymerization analysis side Method.As shown in Figure 1, this method is as follows：

Step S101, polymerization analysis center receives the sequence information that user submits, and sequence information includes at least object module Mark and the corresponding at least one data structures mark of object module, object module is any announced polymerization analysis mould Type, object module include polymerization model and at least one basic model, and each basic model corresponds to a unique modem Structure.

In the present embodiment, polymerization analysis center is independently of computer equipments such as server, the clusters of each data structures.It is poly- It closes analysis center and is not belonging to any one data structures, but the administrative center of multiple data structures, it is total to multiple data structures With the transaction platform for forming a big data, big data commodity are externally provided.

Wherein, polymerization analysis model includes polymerization model and at least one basic model, and each basic model corresponds to only One data structures.For any one polymerization analysis model, carry out polymerization analysis and obtain the process of polymerization analysis result It is as follows：

For any basic model of the polymerization analysis model, by the corresponding data structures of the basic model according to itself Initial data generates intermediate data, the intermediate data as the basic model by the basic model；Summarize at polymerization analysis center The intermediate data of each basic model of the polymerization analysis model, by polymerization model to each basic model of the polymerization analysis model Intermediate data carry out polymerization analysis, obtain the polymerization analysis result of the polymerization analysis model.

In this implementation, the big data commodity that user is ordered by sequence information refer to the mark of the object module in sequence information Know the identified corresponding polymerization analysis result of object module.User believes according to the issue of issued polymerization analysis model Breath can need to select one of them issued polymerization analysis model as object module according to oneself, submission order, To order the corresponding polymerization analysis result of object module.

Step S102, sequence information is sent to each data structures and identifies corresponding data structures by polymerization analysis center, with Make the mark of object module of each data structures in sequence information and the object module of storage, determine corresponding with itself Belong to the basic model of object module, and intermediate data generated according to the basic model for belonging to object module corresponding with itself, And intermediate data is sent to polymerization analysis center.

In the step, sequence information is transmitted to each data structures and identifies corresponding data structures, tool by polymerization analysis center Following manner realization may be employed in body：

Polymerization analysis center determines that each data structures in sequence information identify corresponding modem according to sequence information Structure identifies corresponding data structures to each data structures and sends sequence information, so that each data structures are according in sequence information The mark of object module and pre-stored each polymerization analysis model determine object module corresponding with itself, further really The fixed basic model for belonging to object module corresponding with itself, and according to the basic model for belonging to object module corresponding with itself Intermediate data is generated, and intermediate data is sent to polymerization analysis center.Alternatively, sequence information is transmitted to by polymerization analysis center All data structures to communicate are gathered by the mark of object module of each data structures in sequence information and each of storage Close analysis model, it is determined whether there is the basic model for belonging to object module corresponding with itself, if needed in the presence of if participate in this Secondary polymerization analysis process, the basic model for belonging to object module according to corresponding with itself generate intermediate data, and by mediant According to being sent to polymerization analysis center.

Step S103, polymerization analysis center receives the intermediate data that each data structures are sent.

Wherein, intermediate data includes the mark of its corresponding basic model, so that polymerization analysis center can be by basic mould Type carries out corresponding with intermediate data.Optionally, intermediate data can also include the mark of object module, that is to say intermediate data pair The mark of polymerization analysis model belonging to the basic model answered, in order to which polymerization analysis center handles multiple sequence informations at the same time When, the intermediate data of different object modules can be distinguished.

Step S104, polymerization analysis centrally through object module polymerization model, to each intermediate data carry out polymerization analysis, Obtain polymerization analysis result.

Optionally, polymerization analysis center can be a server or can also include multiple aggregations, different Aggregation is each responsible for the polymerization analysis processing procedure of different polymerization analysis models, for example, it may be by multiple aggregations The cluster of composition, to accelerate the computational efficiency at polymerization analysis center, so as to improve the efficiency of polymerization analysis processing.

In addition, in the present embodiment, sequence information can include the mark of one or more object module, respectively for every One object module carries out polymerization analysis by step S102-S104 and handles to obtain polymerization analysis knot corresponding with the object module Fruit, identical to the process of each object module progress polymerization analysis processing, details are not described herein again for the present embodiment.

The embodiment of the present invention by polymerization analysis model by being divided into two levels of basic model and polymerization model, polymerization analysis The sequence information that center receives includes at least the mark of object module and the corresponding at least one data structures mark of object module Know, sequence information is sent to each data structures and identifies corresponding data structures by polymerization analysis center, by each data structures according to The initial data and basic model of itself generate intermediate data, then summarize the centre of each data structures generation by polymerization analysis center Data by the polymerization model of object module, carry out each intermediate data polymerization analysis and obtain polymerization analysis as a result, so as to by each Data structures directly provide intermediate data, and polymerization analysis center need not be from each data structures bulk purchase or exchange original number According to, so as to it need not share initial data with polymerization analysis center or other data structures in each data structures on the premise of, cooperation Polymerization analysis processing procedure is completed, obtains the corresponding polymerization analysis of each polymerization analysis model as a result, so as to improve each data The enthusiasm of institution cooperation simplifies the flow of big data polymerization analysis, substantially increases the timeliness of polymerization analysis result.

Embodiment two

On the basis of above-described embodiment one, in the present embodiment, polymerization analysis center receives the sequence information that user submits Before, further include：Polymerization analysis center obtains at least one polymerization analysis model；Polymerization analysis center is by each polymerization analysis model It is sent to each data structures；Each polymerization analysis model is issued at polymerization analysis center, so that user is according to each polymerization analysis Sequence information is submitted in releasing news for model, to order the polymerization analysis result of object module.It is provided by Embodiment 2 of the present invention Big data polymerization analysis method is as follows：

Step S201, polymerization analysis center obtains at least one polymerization analysis model.

In the present embodiment, polymerization analysis center obtains at least one polymerization analysis model, and following manner specifically may be employed It realizes：

What the sample data and metadata that the polymerization analysis center acquisition third-party institution provides according to each data structures generated First model, the first model are the polymerization analysis model without test.First model is sent to each number by polymerization analysis center According to mechanism, so that each data structures test the basic model of the first model, and each first is fed back to polymerization analysis center The intermediate data of the basic model for the first model that the test result of the basic model of model and test pass through.Polymerization analysis The intermediate data of the basic model for the first model that center passes through according to test result and test, the polymerization to the first model Model is tested；Polymerization analysis center by test by the first model be used as polymerization analysis model.

Wherein, sample data refer to that each data structures provide with the corresponding sample data of initial data；Metadata is Refer to the format information of initial data and sample data.

Optionally, polymerization analysis center by test by the first model be used as polymerization analysis model after, polymerization divide Analysis center can jointly decide through consultation with data structures and the third-party institution and fix a price for polymerization analysis model, can also arrange to correspond to In the deserved expense of each polymerization analysis model data mechanism and the third-party institution.

Optionally, polymerization analysis center by test by the first model be used as polymerization analysis model after, inform test By the corresponding third-party institution of the first model, and according to prior pact orientation third-party institution payment expense.

Optionally, the sample data and metadata that the third-party institution can provide according to each data structures, are therefrom freely selected Sample number and metadata are taken, polymerization analysis model is developed according to the sample data of selection and metadata, as the first model, by the One model and its corresponding sample data are submitted to polymerization analysis center, and each data structures of polymerization analysis center complex are to the first mould Type is tested, test by the first model can be used as polymerization analysis model.

In the present embodiment, the polymerization model of the first model includes at least following information：Model identification, intermediate data source letter Breath, basic model mark, analysis result, die body, third-party institution's mark and eap-message digest.Wherein, model identification is used for The unique mark polymerization model；Intermediate data source-information is for intermediate data of the explanation as the input data of the polymerization model And its source, that is, intermediate data and its corresponding basic model mark；Basic model mark can be multiple basic moulds The mark of type, for illustrating that the polymerization model needs the corresponding intermediate data of which basic model；Analysis result is used to illustrate to gather The information for the polymerization analysis result that molding type should obtain；The third-party institution is identified as the third party's machine for completing polymerization model exploitation The mark of structure；Eap-message digest is the information according to included by polymerization model in addition to eap-message digest and Message Digest 5 generation Summary info, wherein Message Digest 5 can be any one existing Message Digest 5.In addition, polymerization model can be with The information such as including function declaration, in order to polymerization analysis center and the function of each data structures understanding polymerization model.

The basic model of first model includes at least following information：Model identification, sample data source, intermediate data, mould Type main body, third-party institution's mark and eap-message digest.Wherein, model identification is for the unique mark basic model；Sample data For illustrating the sample data of the input data as the basic model and its source, intermediate data is used to illustrate the basis in source The information for the intermediate data that model should obtain；The third-party institution is identified as the mark for the third-party institution for completing basic model exploitation Know；Information in addition to eap-message digest and Message Digest 5 generation according to eap-message digest included by the basic model are plucked Information is wanted, wherein Message Digest 5 can be any one existing Message Digest 5.In addition, the basic model can also wrap The information such as function declaration are included, in order to which polymerization analysis center and each data structures understand the function of the basic model.

In the present embodiment, data structures test basic model, can verify basis by verifying eap-message digest The validity and security of model.

Further, the basic model pair for the first model that polymerization analysis center passes through according to test result and test The intermediate data answered, tests the polymerization model of the first model, and following manner realization specifically may be employed：

Polymerization analysis center is using each first model as model to be measured, according to test result, if any of model to be measured Basic model test does not pass through, it is determined that model measurement to be measured does not pass through；

Pass through if each basic model of model to be measured is tested, according to the corresponding mediant of the basic model of model to be measured According to testing the polymerization model of model to be measured；If the polymerization model test of model to be measured is passed through, it is determined that model to be measured Test passes through.

In the present embodiment, according to the corresponding intermediate data of the basic model of model to be measured, to the polymerization model of model to be measured It is tested, can verify the validity of polymerization model and security by verifying the eap-message digest of polymerization model.

In the present embodiment, initially set up a block chain, the block chain can by polymerization analysis center, each data structures and The third-party institution shares, and the key message in step S201-S209 can be recorded in the block chain, to realize entire processing The record of crucial processing procedure in the process, so that polymerization analysis center, each data structures and the third-party institution can be to whole A processing procedure exercises supervision, convenient for finding malicious act.

Optionally, after each data structures issue new sample data and metadata, according to the sample data newly issued and Metadata generates new block, which can include the summary info of sample data and metadata, data structures mark, issue The information such as time, and the block is added to the end of block chain.

Optionally, after each data structures are to the test of the basic data of the first model, generation is new comprising test result The block is added to the end of block chain by block.

Optionally, after the first model is submitted to polymerization analysis center by the third-party institution, given birth to according to the information of the first model The block of Cheng Xin, the block can include：Third-party institution's mark, the first model identification, the eap-message digest of polymerization model and The information such as the eap-message digest of basic model, and the block is connected in block chain.

Step S202, each polymerization analysis model is sent to each data structures by polymerization analysis center, so that each data structures Store each polymerization analysis model.

After each polymerization analysis model is got, in the step, each polymerization analysis model is sent to by polymerization analysis center Each data structures, so that each data structures store each polymerization analysis model, in order to the target subsequently specified according to sequence information The mark of model is assured that the basic model of object module, and may further determine that and corresponding with itself belong to target mould The basic model of type, so as to according to the basic model for belonging to object module corresponding with itself and pre-stored original Beginning data generate intermediate data.

Optionally, in the step, after each polymerization analysis model is sent to each data structures by polymerization analysis center, polymerization Analysis center generates new block according to polymerization analysis model, which can include polymerization analysis model identification, data structures The information such as mark and third-party institution's mark, and the block is added to the end of block chain.

Step S203, each polymerization analysis model is issued at polymerization analysis center, so that user is according to each polymerization analysis Sequence information is submitted in releasing news for model, to order the polymerization analysis result of object module.

In the step, each polymerization analysis model is issued at polymerization analysis center, by the issue of each polymerization analysis model Information is externally issued so that user can consult releasing news for polymerization analysis model, wherein polymerization can be included by releasing news Data content summary info, purposes, application method and price of analysis model etc., so as to be believed according to the description of polymerization analysis model The information such as the breath data content that briefly understanding polymerization analysis model includes or purposes, so that user is according to each polymerization analysis Sequence information is submitted in releasing news for model, to order the polymerization analysis result of object module.

Optionally, releasing news for polymerization analysis model includes at least：Model identification, initial data source, basic model Mark, polymerization model mark, third-party institution's mark, pricing information and eap-message digest.Wherein, model identification is used for unique mark One polymerization analysis model；Initial data source is used for the information for illustrating the initial data of the polymerization analysis model required input； Basic model is identified as the mark for each basic model that the polymerization analysis model includes；Polymerization model is identified as the polymerization analysis mould The mark for the polymerization model that type includes；The third-party institution is identified as the mark for the third-party institution for completing the polymerization analysis model development Know；Pricing information refers to the pricing information of the polymerization analysis model；The polymerization analysis model includes according to eap-message digest The summary info of information and Message Digest 5 generation in addition to eap-message digest, wherein Message Digest 5 can be existing It anticipates a kind of Message Digest 5.In addition, the basic model can also including function declaration etc. information, in order to user, to understand this poly- Close the function of analysis model.

Optionally, in the step, after each polymerization analysis model is issued at polymerization analysis center, polymerization analysis center New block is generated according to the polymerization analysis model of issue, which can include the letters such as the mark of polymerization analysis model of issue It ceases, and the block is added to the end of block chain.

It should be noted that the process of above-mentioned steps S201-S203 is by polymerization analysis center and each data structures, the 3rd Square mechanism combines the process for getting polymerization analysis model, which can periodically carry out according to actual needs, with the cycle Update polymerization analysis model to property；Alternatively, each data structures can aperiodically issue new sample data and metadata, whenever When data structures issue new sample data and metadata, sample data that the third-party institution is newly issued according to each data structures and Metadata and original sample data and metadata generate and the first new model are submitted to polymerization analysis center, polymerizeing After analysis center and corresponding data structures are verified, as new polymerization analysis model.It, can be by technology in the present embodiment Personnel set the operation opportunity of the process of above-mentioned steps S201-S203 according to actual needs, and the present embodiment does not do this specific limit It is fixed.

Step S204, polymerization analysis center receives the sequence information that user submits, and sequence information includes at least object module Mark and the corresponding at least one data structures mark of object module, object module is any announced polymerization analysis mould Type, object module include polymerization model and at least one basic model, and each basic model corresponds to a unique modem Structure.

Optionally, sequence information can include order note identification, object module mark, user identifier, data structures identify and The information such as eap-message digest.Polymerization analysis centrally through the object module mark in sequence information can determine default market settlement, The information such as the corresponding third-party institution of object module and the corresponding public key of user identifier.In addition, sequence information can also be included in advance If the information such as the corresponding third-party institution's mark of market settlement, object module and client public key.It wherein, should according to eap-message digest The summary info for information in addition to eap-message digest and the Message Digest 5 generation that sequence information includes, wherein eap-message digest are calculated Method can be any one existing Message Digest 5.

Optionally, in the step, after polymerization analysis center receives the sequence information that user submits, polymerization analysis Central Radical Include the new block of the sequence information according to sequence information generation, and the block is added to the end of block chain.

Step S205, sequence information is sent to each data structures and identifies corresponding data structures by polymerization analysis center, with Make the mark of object module of each data structures in sequence information, determine the basis for belonging to object module corresponding with itself Model, and intermediate data is generated according to the basic model for belonging to object module corresponding with itself, and intermediate data is sent to Polymerization analysis center.

The step is identical with above-mentioned steps S102, and details are not described herein again for the present embodiment.

In the present embodiment, intermediate data can include following information：Polymerization analysis model identification, user identifier, modem Structure mark, polymerization model mark, intermediate data main body and eap-message digest.Wherein, polymerization analysis model identification is in sequence information Object module mark；User identifier corresponds to the user identifier in sequence information；Data structures, which are identified as, generates the mediant According to data structures itself mark；Polymerization model is identified as the corresponding mesh of polymerization analysis model identification that the intermediate data includes Mark the mark of the polymerization model of model.The information in addition to eap-message digest and disappear that the intermediate data includes according to eap-message digest The summary info of digest algorithm generation is ceased, wherein Message Digest 5 can be any one existing Message Digest 5.

Step S206, polymerization analysis center receives the intermediate data that each data structures are sent.

The step is identical with above-mentioned steps S103, and details are not described herein again for the present embodiment.

Optionally, in the step, after polymerization analysis center receives the intermediate data that each data structures are sent, polymerization analysis Center generates new block according to the intermediate data received, which includes the letters such as intermediate data, order note identification, receiving time It ceases, and the block is added to the end of block chain.

Step S207, polymerization analysis centrally through object module polymerization model, to each intermediate data carry out polymerization analysis, Obtain polymerization analysis result.

Optionally, polymerization analysis result can include following information：Polymerization analysis model identification, user identifier, polymerization mould Type mark, polymerization analysis result main body and eap-message digest etc..Wherein, polymerization analysis model identification is the target mould in sequence information Type identifies；User identifier corresponds to the user identifier in sequence information；Polymerization model is identified as what the polymerization analysis result included The mark of the polymerization model of the corresponding object module of polymerization analysis model identification；The intermediate data includes according to eap-message digest The summary info of information and Message Digest 5 generation in addition to eap-message digest, wherein Message Digest 5 can be existing It anticipates a kind of Message Digest 5.

In addition, in the present embodiment, sequence information can include the mark of one or more object module, respectively for every One object module carries out polymerization analysis by step S204-S207 and handles to obtain polymerization analysis knot corresponding with the object module Fruit, identical to the process of each object module progress polymerization analysis processing, details are not described herein again for the present embodiment.

Optionally, in the step, after polymerization analysis center obtains polymerization analysis result, polymerization analysis center is according to obtaining Polymerization analysis result generate new block, which includes the letters such as object module mark, order note identification, polymerization analysis result It ceases, and the block is added to the end of block chain.

Step S208, polymerization analysis result is fed back to user by polymerization analysis center.

In the present embodiment, at polymerization analysis center according to each intermediate data and the polymerization model of object module, it is polymerize Polymerization analysis result, after obtaining polymerization analysis result, can also be fed back to user, completed to user's sequence information by analysis Response, so as to complete the transaction with the big data commodity of user.

Optionally, in order to ensure the security of data, polymerization analysis center is using the public key of user to polymerization analysis result It is encrypted, so that user after the encryption data of polymerization analysis mechanism is received, is decoded using the private key of oneself, so as to It can obtain polymerization analysis result.

Step S209, polymerization analysis carries out disbursement and sattlement centrally through the mode of intelligent contract according to default market settlement.

Wherein, preset market settlement including object module price, should be by corresponding to each data structures of the object module The expense that the ratio of the expense occupancy family payment expense of payment and the third-party institution should be paid for occupies family payment expense Ratio etc..Default market settlement can be consulted to set in advance by polymerization analysis center and each data structures and the third-party institution, this Embodiment is not specifically limited for presetting the particular content of market settlement.

In the step, when polymerization analysis result is fed back to user, intelligent contract is triggered, it is automatically real by intelligent contract Now according to default market settlement, each data structures are calculated and/or expense that the third-party institution should be paid for, complete expense Automatic-settlement.

Optionally, after disbursement and sattlement is carried out according to default market settlement by way of intelligent contract, polymerization analysis Center can also generate the information that reckons up the bill, to record this clearing.

The information that reckons up the bill can include：Order note identification, polymerization analysis model identification, user identifier, data structures mark, The information such as third-party institution's mark, settlement information and eap-message digest.Wherein, order note identification corresponds to ordering in sequence information Single mark；Polymerization analysis model identification corresponds to the mark of the object module in sequence information；User identifier is believed corresponding to order User identifier in breath；Data structures mark includes the corresponding target mould of polymerization analysis model identification in the information that reckons up the bill The mark of the corresponding data structures of each basic model of type；The third-party institution is identified as the polymerization completed in the information that reckons up the bill Analysis model identifies the mark of the third-party institution of corresponding object module exploitation；Settlement information can include this disbursement and sattlement In the cost information that obtains respectively of each data structures, the third-party institution and polymerization analysis center；The clearing according to eap-message digest The summary info for information in addition to eap-message digest and the Message Digest 5 generation that bill information includes, wherein eap-message digest are calculated Method can be any one existing Message Digest 5.

Optionally, in the step, polymerization analysis carries out expense centrally through the mode of intelligent contract according to default market settlement After clearing, polymerization analysis center generates new block according to this disbursement and sattlement process, which includes the letter that reckons up the bill It ceases, and the block is added to the end of block chain.In the present embodiment, in order to ensure the security of data, in polymerization analysis Secret key both arbitrary during interacting in the heart, data structures and the third-party institution, being made an appointment using the two Data are encrypted with transmission, to ensure the security of data transmission.

In addition, in an alternative embodiment of the invention, polymerization analysis center, data structures can pass through the side of intelligent contract Formula realizes the automated execution of above-mentioned function, and details are not described herein again for the present embodiment.

Preferably, the polymerization analysis centrally through the object module polymerization model, to each intermediate data into Row polymerization analysis obtains polymerization analysis as a result, in the following manner realization may be employed：The polymerization analysis center is using intelligent contract Mode, polymerization analysis is carried out to each intermediate data by the polymerization model of the object module, obtains polymerization analysis knot Fruit.It is not disrupted during can guarantee automated execution appointment function due to intelligent contract, is realized by way of intelligent contract The process of polymerization analysis can improve the security of polymerization analysis process so that processing of each data structures to polymerization analysis center Process is more trusted.

Optionally, the polymerization analysis center carries out data biography by the way of intelligent contract with each data structures Defeated, the data transmission carried out between each data structures, between polymerization analysis center and the third-party institution is also using intelligent contract Mode is realized, it can be ensured that the security of data transmission procedure between each mechanism.

In another real-time mode of the present embodiment, the polymerization model includes at least two layers of polymer submodels.It is described poly- Polymerization model of the analysis center by the object module is closed, polymerization analysis is carried out to each intermediate data, obtains polymerization point Analysis as a result, including：The polymerization analysis centrally through the object module the first layers of polymer submodel, to each mediant According to polymerization analysis is carried out, first layer intermediate analysis result is obtained；The polymerization analysis is centrally through except first layers of polymer Each layers of polymer submodel outside model carries out polymerization analysis to the intermediate analysis result of last layer polymerization submodel, obtains this The intermediate analysis result of layer；The intermediate analysis knot of last layer that polymerization analysis obtains is carried out by last layers of polymer submodel Fruit is as the polymerization analysis result.

The embodiment of the present invention is by being responsible for specially sample data and member according to the offer of each data structures by the third-party institution First model of data generation, jointly tests the first model by polymerization analysis center and each data structures, and test is logical The first model crossed can ensure the correctness of polymerization analysis model, so as to realize by polymerizeing as polymerization analysis model Polymerization analysis processing procedure is completed in analysis center and each data structures cooperation, and polymerization analysis center need not be from each data structures batch Purchase exchanges initial data, so as to improve the enthusiasm of each data structures cooperation, simplifies big data polymerization analysis Flow, substantially increase the timeliness of polymerization analysis result.

Embodiment three

Fig. 2 is the big data polymerization analysis method flow diagram that the embodiment of the present invention three provides.The embodiment of the present invention is for existing The cost for having polymerization analysis center from other data structures bulk purchase initial data in technology is very big, is handed over other data structures Changing initial data may consult to close because the competitive advantage for losing itself when sharing of proprietary data, and with other data structures It is complicated the flows such as to make, takes longer, the problem of can not ensureing the timeliness of analysis result, provides big data polymerization analysis side Method.As shown in Fig. 2, this method is as follows：

Step S301, data structures receive the sequence information that polymerization analysis center is sent, and sequence information includes at least target The corresponding at least one data structures mark of mark and object module of model, object module is any announced polymerization analysis Model, object module include polymerization model and at least one basic model, and each basic model corresponds to unique data Mechanism.

In this implementation, the big data commodity that user is ordered by sequence information refer to the mark of the object module in sequence information Know the identified corresponding polymerization analysis result of object module.User believes according to the issue of issued polymerization analysis model Breath can need to select one of them issued polymerization analysis model as object module according to oneself, submission order, To order the corresponding polymerization analysis result of object module.Polymerization site can turn sequence information after sequence information is received Issue data structures.

Step S302, each polymerization of the mark of object module of the data structures in sequence information and itself storage Analysis model determines the basic model for belonging to object module corresponding with itself.

In the present embodiment, each data structures are previously stored with each polymerization analysis model.In the step, data structures according to The mark of object module in sequence information and pre-stored each polymerization analysis model determine target corresponding with itself Model further determines that the basic model for belonging to object module corresponding with itself, and belongs to target according to corresponding with itself The basic model of model and the Raw Data Generation intermediate data of itself, and intermediate data is sent to polymerization analysis center.

Step S303, data structures generate intermediate data according to the basic model for belonging to object module corresponding with itself.

Wherein, intermediate data includes the mark of its corresponding basic model, so that polymerization analysis center can be by basic mould Type carries out corresponding with intermediate data.

Optionally, intermediate data can also include the mark of object module, that is to say the corresponding basic model of intermediate data The mark of affiliated polymerization analysis model when handling multiple sequence informations at the same time in order to polymerization analysis center, can be distinguished The intermediate data of different object modules.

Step S304, intermediate data is sent to polymerization analysis center by data structures, so that polymerization analysis is centrally through mesh The polymerization model of model is marked, polymerization analysis is carried out to each intermediate data received, obtains polymerization analysis result.

Example IV

On the basis of above-described embodiment three, in the present embodiment, data structures receive the order that polymerization analysis center is sent Before information, further include：Data structures receive each polymerization analysis model that polymerization analysis center is sent, and store.It is of the invention real The big data polymerization analysis method for applying the offer of example four is as follows：

Step S401, data structures receive the first model that polymerization analysis center is sent, and the first model is that the first model is The polymerization analysis model without test that the sample data and metadata that the third-party institution provides according to data structures generate.

In the present embodiment, data structures can aperiodically issue new sample data and metadata, so that third party's machine The sample data and metadata that structure can be provided according to each data structures, therefrom freely choose sample number and metadata, according to choosing Sample data and metadata the exploitation polymerization analysis model taken, as the first model, by the first model and its corresponding sample number According to polymerization analysis center is submitted to, each data structures of polymerization analysis center complex test the first model, test what is passed through First model can be used as polymerization analysis model.

Step S402, data structures test the basic model of the first model.

In the present embodiment, the basic model of the first model includes at least following information：Model identification, sample data source, Intermediate data, die body, third-party institution's mark and eap-message digest.Wherein, model identification is for the unique mark basis mould Type；For illustrating the sample data of the input data as the basic model and its source, intermediate data is used in sample data source In the information of intermediate data that illustrates the basic model and should obtain；The third-party institution, which is identified as, completes the of basic model exploitation The mark of tripartite mechanism；The information and eap-message digest in addition to eap-message digest according to eap-message digest included by the basic model The summary info of algorithm generation, wherein Message Digest 5 can be any one existing Message Digest 5.In addition, the basis Model can also including function declaration etc. information, in order to which polymerization analysis center and each data structures understand the work(of the basic model Energy.

In the present embodiment, information that data structures include according to the initial data of itself and the basic model of the first model, Basic model is tested, is verified whether the basic model can obtain expected intermediate data in the basic model. In addition, data structures can also verify the validity of basic model and security by verifying basic model eap-message digest.

In the present embodiment, a block chain is initially set up, this can be recorded in the key message in step S201-S209 In block chain, to realize the record of crucial processing procedure in entire processing procedure, convenient for finding malicious act.

Optionally, in the step, after each data structures issue new sample data and metadata, according to the sample newly issued Notebook data and metadata generate new block, which can include summary info, the data structures of sample data and metadata The information such as mark, issuing time, and the block is added to the end of block chain.

If step S403, passing through to the basic model test of the first model, data structures generate the basis of the first model The intermediate data of model.

In the step, if passing through to the basic model test of the first model, data structures are according to initial data and the verification Intermediate data is generated by basic model.

Step S404, data structures fed back to polymerization analysis center the basic model of each first model test result, with And test the intermediate data of the basic model of the first model passed through.

In the step, data structures according to initial data and this be verified basic model generation intermediate data after, to It feeds back the test result of the basic model of each first model and tests the basic mould of the first model passed through in polymerization analysis center The intermediate data of type, so that polymerization analysis center can be logical according to the test result of the basic model of the first model and test The intermediate data of the basic model for the first model crossed, tests the polymerization model of the first model.

Step S405, data structures receive each polymerization analysis model that polymerization analysis center is sent, and store.

After each polymerization analysis model is got, each polymerization analysis model can be sent to each data by polymerization analysis center Mechanism, each data structures receive and store each polymerization analysis model of polymerization analysis center transmission, in order to subsequently according to order The mark for the object module that information is specified is assured that the basic model of object module, and may further determine that it is right with itself That answers belongs to the basic model of object module, so as to according to the basic model for belonging to object module corresponding with itself, with And pre-stored initial data, generate intermediate data.

It should be noted that the process of above-mentioned steps S401-S405 is by polymerization analysis center and each data structures, the 3rd Square mechanism combines the process for getting polymerization analysis model, which can periodically carry out according to actual needs, with the cycle Update polymerization analysis model to property；Alternatively, each data structures can aperiodically issue new sample data and metadata, whenever When data structures issue new sample data and metadata, sample data that the third-party institution is newly issued according to each data structures and Metadata and original sample data and metadata generate and the first new model are submitted to polymerization analysis center, polymerizeing After analysis center and corresponding data structures are verified, as new polymerization analysis model.It, can be by technology in the present embodiment Personnel set the operation opportunity of the process of above-mentioned steps S401-S405 according to actual needs, and the present embodiment does not do this specific limit It is fixed.

Step S406, data structures receive the sequence information that polymerization analysis center is sent, and sequence information includes at least target The corresponding at least one data structures mark of mark and object module of model, object module is any announced polymerization analysis Model, object module include polymerization model and at least one basic model, and each basic model corresponds to unique data Mechanism.

The step is identical with above-mentioned steps S301, and details are not described herein again for the present embodiment.

Step S407, each polymerization of the mark of object module of the data structures in sequence information and itself storage Analysis model determines the basic model for belonging to object module corresponding with itself.

The step is identical with above-mentioned steps S302, and details are not described herein again for the present embodiment.

Step S408, data structures generate intermediate data according to the basic model for belonging to object module corresponding with itself.

The step is identical with above-mentioned steps S303, and details are not described herein again for the present embodiment.

Optionally, in the step, data structures are according to the basic model for belonging to object module corresponding with itself, in generation Between after data, data structures generate new block according to the intermediate data of generation, which includes intermediate data, order mark Information such as knowledge, generated time, and the block is added to the end of block chain, so that polymerization analysis center is to the centre that receives The authenticity and integrity of data is verified.

Step S409, intermediate data is sent to polymerization analysis center by data structures, so that polymerization analysis is centrally through mesh The polymerization model of model is marked, polymerization analysis is carried out to each intermediate data received, obtains polymerization analysis result.

In addition, in the present embodiment, sequence information can include the mark of one or more object module, respectively for every One object module carries out polymerization analysis and handles to obtain polymerization analysis corresponding with the object module as a result, to each object module The process of progress polymerization analysis processing is identical, and details are not described herein again for the present embodiment.

In the present embodiment, in order to ensure the security of data, in polymerization analysis center, data structures and the third-party institution Both arbitrary transmission is encrypted in the secret key pair data made an appointment using the two during interacting, to ensure The security of data transmission.

Embodiment five

Fig. 3 is the structure diagram for the big data polymerization analysis device that the embodiment of the present invention five provides.The embodiment of the present invention The big data polymerization analysis device of offer can perform the process flow of big data polymerization analysis embodiment of the method offer.Such as Fig. 3 Shown, which includes：First communication module 501, second communication module 502 and polymerization analysis module 503.

Specifically, first communication module 501 receives the sequence information of user's submission, sequence information for polymerization analysis center The corresponding at least one data structures mark of mark and object module including at least object module, object module have been sent out to be any The polymerization analysis model of cloth, object module include polymerization model and at least one basic model, and each basic model corresponds to only One data structures.

Sequence information is sent to each data structures for polymerization analysis center and identifies corresponding number by second communication module 502 According to mechanism so that the mark of object module of each data structures in sequence information and the object module of storage, determine with Itself corresponding basic model for belonging to object module, and generated according to the basic model for belonging to object module corresponding with itself Intermediate data, and intermediate data is sent to polymerization analysis center.

Second communication module 502 is additionally operable to polymerization analysis center and receives the intermediate data that each data structures are sent.

Polymerization analysis module 503 for polymerization analysis centrally through object module polymerization model, to each intermediate data into Row polymerization analysis obtains polymerization analysis result.

Optionally, first communication module 501 and second communication module 502 can be gateway.

Device provided in an embodiment of the present invention can be specifically used for performing the embodiment of the method that above-described embodiment one is provided, Details are not described herein again for concrete function.

Embodiment six

On the basis of above-described embodiment five, in the present embodiment, device 50 further includes：Model generation module and issue mould Block.

Specifically, model generation module obtains at least one polymerization analysis model for polymerization analysis center.

Second communication module is additionally operable to polymerization analysis center and each polymerization analysis model is sent to each data structures, so that respectively Data structures store each polymerization analysis model.

Release module issues each polymerization analysis model for polymerization analysis center, so that user is according to each polymerization point Sequence information is submitted in releasing news for analysis model, to order the polymerization analysis result of object module.

Optionally, model generation module includes：First acquisition submodule and test submodule.

Wherein, the first acquisition submodule obtains what the third-party institution provided according to each data structures for polymerization analysis center Sample data and the first model of metadata generation, the first model are the polymerization analysis model without test.

Second communication submodule is additionally operable to polymerization analysis center and the first model is sent to each data structures, so that each data Mechanism tests the basic model of the first model, and the survey of the basic model to each first model of polymerization analysis center feedback The intermediate data of the basic model for the first model that test result and test pass through.

The basic mould for the first model that test submodule passes through for polymerization analysis center according to test result and test The intermediate data of type tests the polymerization model of the first model；Polymerization analysis center will test the first model work passed through For polymerization analysis model.

In the present embodiment, test submodule is additionally operable to：Polymerization analysis center is using each first model as model to be measured, root According to test result, if any basic model test of model to be measured does not pass through, it is determined that model measurement to be measured does not pass through；It is if to be measured Each basic model of model, which is tested, to be passed through, then according to the corresponding intermediate data of basic model of model to be measured, to model to be measured Polymerization model tested；If the polymerization model test of model to be measured is passed through, it is determined that model measurement to be measured passes through.

Optionally, which can also include settlement module.

First communication module is additionally operable to polymerization analysis center and polymerization analysis result is fed back to user.

Settlement module carries out expense knot centrally through the mode of intelligent contract for polymerization analysis according to default market settlement It calculates.

Device provided in an embodiment of the present invention can be specifically used for performing the embodiment of the method that above-described embodiment two is provided, Details are not described herein again for concrete function.

Embodiment seven

Fig. 4 is the structure diagram for the big data polymerization analysis device that the embodiment of the present invention seven provides.The embodiment of the present invention The big data polymerization analysis device of offer can perform the process flow of big data polymerization analysis embodiment of the method offer.Such as Fig. 4 Shown, which includes：Receiving module 701, determining module 702, generation module 703 and sending module 704.

Specifically, receiving module 701 receives the sequence information of polymerization analysis center transmission, sequence information for data structures The corresponding at least one data structures mark of mark and object module including at least object module, object module have been sent out to be any The polymerization analysis model of cloth, object module include polymerization model and at least one basic model, and each basic model corresponds to only One data structures.

Determining module 702 for object module of the data structures in sequence information mark and itself storage Each polymerization analysis model determines the basic model for belonging to object module corresponding with itself.

Generation module 703 for data structures according to the basic model for belonging to object module corresponding with itself, in generation Between data.

Intermediate data is sent to polymerization analysis center by sending module 704 for data structures, so that polymerization analysis center By the polymerization model of object module, polymerization analysis is carried out to each intermediate data received, obtains polymerization analysis result.

Device provided in an embodiment of the present invention can be specifically used for performing the embodiment of the method that above-described embodiment three is provided, Details are not described herein again for concrete function.

Embodiment eight

On the basis of above-described embodiment seven, in the present embodiment, device 70 further includes：Test module.

In the present embodiment, receiving module is additionally operable to：

Data structures receive each polymerization analysis model that polymerization analysis center is sent, and store.

Receiving module is additionally operable to data structures and receives the first model that polymerization analysis center is sent, and the first model is the first mould The polymerization analysis mould without test that the sample data and metadata that type provides for the third-party institution according to data structures generate Type.

Test module tests the basic model of the first model for data structures.

If the basic model test that generation module is additionally operable to the first model passes through, the basic model of the first model is generated Intermediate data.

Sending module is additionally operable to the test knot that data structures feed back the basic model of each first model to polymerization analysis center The intermediate data of the basic model for the first model that fruit and test pass through.

Optionally, which can also include release module.

Release module can aperiodically issue new sample data and metadata for data structures, so that third party's machine The sample data and metadata that structure can be provided according to each data structures, therefrom freely choose sample number and metadata, according to choosing Sample data and metadata the exploitation polymerization analysis model taken, as the first model, by the first model and its corresponding sample number According to being submitted to polymerization analysis center, and then each data structures of polymerization analysis center complex is made to test the first model, tested By the first model can be used as polymerization analysis model.

Device provided in an embodiment of the present invention can be specifically used for performing the embodiment of the method that above-described embodiment four is provided, Details are not described herein again for concrete function.

In several embodiments provided by the present invention, it should be understood that disclosed apparatus and method can pass through it Its mode is realized.For example, the apparatus embodiments described above are merely exemplary, for example, the division of unit, is only A kind of division of logic function, can there is an other dividing mode in actual implementation, for example, multiple units or component can combine or Person is desirably integrated into another system or some features can be ignored or does not perform.Another, shown or discussed is mutual Between coupling, direct-coupling or communication connection can be INDIRECT COUPLING or communication link by some interfaces, device or unit It connects, can be electrical, machinery or other forms.

The unit illustrated as separating component may or may not be physically separate, be shown as unit Component may or may not be physical location, you can be located at a place or can also be distributed to multiple networks On unit.Some or all of unit therein can be selected to realize the purpose of this embodiment scheme according to the actual needs.

Those skilled in the art will readily occur to the present invention its after considering specification and putting into practice invention disclosed herein Its embodiment.It is contemplated that cover the present invention any variations, uses, or adaptations, these modifications, purposes or Person's adaptive change follows the general principle of the present invention and including undocumented common knowledge in the art of the invention Or conventional techniques.Description and embodiments are considered only as illustratively, and true scope and spirit of the invention are by following Claims are pointed out.

It should be appreciated that the invention is not limited in the precision architecture for being described above and being shown in the drawings, and And various modifications and changes may be made without departing from the scope thereof.The scope of the present invention is only limited by appended claims System.

Claims

A kind of 1. big data polymerization analysis method, which is characterized in that including：

Polymerization analysis center receives the sequence information that user submits, and the sequence information includes at least mark and the institute of object module The corresponding at least one data structures mark of object module is stated, the object module is any announced polymerization analysis model, The object module includes polymerization model and at least one basic model, and each basic model corresponds to a unique institute State data structures；

The sequence information is sent to each data structures and identifies corresponding data structures by the polymerization analysis center, so that The mark of object module of each data structures in the sequence information and the object module of storage, determine with Itself corresponding basic model for belonging to the object module, and corresponding with itself belong to the object module according to described Basic model generates intermediate data, and the intermediate data is sent to the polymerization analysis center；

The polymerization analysis center receives the intermediate data that each data structures are sent；

The polymerization analysis carries out polymerization analysis centrally through the polymerization model of the object module to each intermediate data, Obtain polymerization analysis result.
2. according to the method described in claim 1, it is characterized in that, the polymerization analysis center receives the order letter that user submits Before breath, further include：

The polymerization analysis center obtains at least one polymerization analysis model；

Each polymerization analysis model is sent to each data structures by the polymerization analysis center, so that the storage of each data structures is each The polymerization analysis model；

Each polymerization analysis model is issued at the polymerization analysis center, so that the user is according to each polymerization point The sequence information is submitted in releasing news for analysis model, to order the polymerization analysis result of the object module.
3. according to the method described in claim 2, it is characterized in that, the polymerization analysis center obtains at least one polymerization Analysis model, including：

What the sample data and metadata that the polymerization analysis center acquisition third-party institution provides according to each data structures generated First model, first model are the polymerization analysis model without test；

First model is sent to each data structures by the polymerization analysis center, so that each data structures are to institute The basic model for stating the first model is tested, and the basic model of each first model is fed back to the polymerization analysis center Test result and the intermediate data of the basic model of first model that passes through of test；

The polymerization analysis center is according to the test result and the basic model for testing first model passed through Intermediate data, the polymerization model of first model is tested；

The polymerization analysis center by test by the first model be used as the polymerization analysis model.
4. according to the method described in claim 3, it is characterized in that, the polymerization analysis center according to the test result, with And the corresponding intermediate data of basic model for testing first model passed through, to the polymerization model of first model It is tested, including：

The polymerization analysis center will each first model be as model to be measured, according to the test result, if described treat Any basic model test for surveying model does not pass through, it is determined that the model measurement to be measured does not pass through；

Pass through if each basic model of the model to be measured is tested, according to the basic model of the model to be measured it is corresponding in Between data, the polymerization model of the model to be measured is tested；

If the polymerization model test to the model to be measured passes through, it is determined that the model measurement to be measured passes through.
5. according to the method described in claim 1, it is characterized in that, the polymerization model include at least two layers of polymer submodels,

The polymerization analysis carries out polymerization analysis centrally through the polymerization model of the object module to each intermediate data, Obtain polymerization analysis as a result, including：

The polymerization analysis gathers each intermediate data centrally through the first layers of polymer submodel of the object module Analysis is closed, obtains first layer intermediate analysis result；

The polymerization analysis gathers last layer centrally through each layers of polymer submodel in addition to the first layers of polymer submodel The intermediate analysis result of zygote model carries out polymerization analysis, obtains the intermediate analysis result of this layer；

The intermediate analysis result that last layer that polymerization analysis obtains is carried out by last layers of polymer submodel is used as described gather Close analysis result.
6. according to claim 1-5 any one of them methods, which is characterized in that

The polymerization analysis center is carried out data transmission by the way of intelligent contract with each data structures；

The polymerization analysis carries out polymerization analysis centrally through the polymerization model of the object module to each intermediate data, Obtain polymerization analysis as a result, including：

The polymerization analysis center is by the way of intelligent contract, by the polymerization model of the object module to each centre Data carry out polymerization analysis, obtain polymerization analysis result.
A kind of 7. big data polymerization analysis method, which is characterized in that including：

Data structures receive the sequence information that polymerization analysis center is sent, and the sequence information includes at least the mark of object module At least one data structures mark corresponding with the object module, the object module is any announced polymerization analysis mould Type, the object module include polymerization model and at least one basic model, and each basic model corresponds to unique one A data structures；

The mark of object module of the data structures in the sequence information and each polymerization point of itself storage Model is analysed, determines the basic model for belonging to the object module corresponding with itself；

The data structures generate intermediate data according to the basic model for belonging to the object module corresponding with itself；

The intermediate data is sent to the polymerization analysis center by the data structures so that the polymerization analysis centrally through The polymerization model of the object module carries out polymerization analysis to each intermediate data received, obtains polymerization analysis result.
8. the method according to the description of claim 7 is characterized in that the data structures receive ordering for polymerization analysis center transmission Before single information, further include：

The data structures receive each polymerization analysis model that the polymerization analysis center is sent, and store.
9. the method according to claim 7 or 8, which is characterized in that the data structures receive the polymerization analysis center Before each polymerization analysis model sent, further include：

The data structures receive the first model that the polymerization analysis center is sent, and first model is first model The polymerization analysis without test that the sample data and metadata provided for the third-party institution according to the data structures generates Model；

The data structures test the basic model of first model；

If passing through to the basic model test of first model, the mediant of the basic model of first model is generated According to；

The data structures fed back to the polymerization analysis center basic model of each first model test result and Test the intermediate data of the basic model of first model passed through.
10. a kind of big data polymerization analysis device, which is characterized in that including：

First communication module, the sequence information of user's submission is received for polymerization analysis center, and the sequence information includes at least The mark of object module and the corresponding at least one data structures mark of the object module, the object module have been sent out to be any The polymerization analysis model of cloth, the object module include polymerization model and at least one basic model, each basic model Corresponding to unique data structures；

The sequence information is sent to each data structures mark pair by second communication module for the polymerization analysis center The data structures answered, so that the institute of the mark of object module of each data structures in the sequence information and storage Object module is stated, determines the basic model for belonging to the object module corresponding with itself, and according to described corresponding with itself Belong to the basic model generation intermediate data of the object module, and the intermediate data is sent in the polymerization analysis The heart；

The second communication module is additionally operable to the polymerization analysis center and receives the mediant that each data structures are sent According to；

Polymerization analysis module, for the polymerization analysis centrally through the polymerization model of the object module, to each centre Data carry out polymerization analysis, obtain polymerization analysis result.
11. device according to claim 10, which is characterized in that further include：

Model generation module obtains at least one polymerization analysis model for the polymerization analysis center；

The second communication module is additionally operable to the polymerization analysis center and each polymerization analysis model is sent to each modem Structure, so that each data structures store each polymerization analysis model；

Release module issues each polymerization analysis model for the polymerization analysis center, so that user's root The sequence information is submitted according to releasing news for each polymerization analysis model, to order the polymerization analysis knot of the object module Fruit.
12. according to the devices described in claim 11, which is characterized in that the model generation module includes：First obtains submodule Block and test submodule；

First acquisition submodule obtains what the third-party institution provided according to each data structures for the polymerization analysis center Sample data and the first model of metadata generation, first model are the polymerization analysis model without test；

The second communication submodule is additionally operable to the polymerization analysis center and first model is sent to each modem Structure, so that each data structures test the basic model of first model, and it is anti-to the polymerization analysis center It presents the test result of the basic model of each first model and tests in the basic model of first model passed through Between data；

The test submodule is for the polymerization analysis center according to passing through the test result and the test The intermediate data of the basic model of first model tests the polymerization model of first model；In the polymerization analysis The heart by test by the first model be used as the polymerization analysis model.
13. device according to claim 12, which is characterized in that the test submodule is additionally operable to：

The polymerization analysis center will each first model be as model to be measured, according to the test result, if described treat Any basic model test for surveying model does not pass through, it is determined that the model measurement to be measured does not pass through；

Pass through if each basic model of the model to be measured is tested, according to the basic model of the model to be measured it is corresponding in Between data, the polymerization model of the model to be measured is tested；

If the polymerization model test to the model to be measured passes through, it is determined that the model measurement to be measured passes through.
14. device according to claim 10, which is characterized in that the polymerization model includes at least two layers of polymer submodules Type, the polymerization analysis module are additionally operable to：

The polymerization analysis gathers each intermediate data centrally through the first layers of polymer submodel of the object module Analysis is closed, obtains first layer intermediate analysis result；

The polymerization analysis gathers last layer centrally through each layers of polymer submodel in addition to the first layers of polymer submodel The intermediate analysis result of zygote model carries out polymerization analysis, obtains the intermediate analysis result of this layer；

The intermediate analysis result that last layer that polymerization analysis obtains is carried out by last layers of polymer submodel is used as described gather Close analysis result.
15. a kind of big data polymerization analysis device, which is characterized in that including：

Receiving module, the sequence information of polymerization analysis center transmission is received for data structures, and the sequence information includes at least The mark of object module and the corresponding at least one data structures mark of the object module, the object module have been sent out to be any The polymerization analysis model of cloth, the object module include polymerization model and at least one basic model, each basic model Corresponding to unique data structures；

Determining module, for the mark of object module of the data structures in the sequence information and itself storage Each polymerization analysis model, determine the basic model for belonging to the object module corresponding with itself；

Generation module, for the data structures according to the basic model for belonging to the object module corresponding with itself, Generate intermediate data；

The intermediate data is sent to the polymerization analysis center by sending module for the data structures, so that described poly- Polymerization model of the analysis center by the object module is closed, polymerization analysis is carried out to each intermediate data received, is obtained To polymerization analysis result.
16. device according to claim 15, which is characterized in that the receiving module is additionally operable to：

The data structures receive each polymerization analysis model that the polymerization analysis center is sent, and store.
17. the device according to claim 15 or 16, which is characterized in that further include：Test module,

The receiving module is additionally operable to the data structures and receives the first model that the polymerization analysis center is sent, and described first Model is that first model is that the sample data that the third-party institution provides according to the data structures and metadata generate not Polymerization analysis model by test；

The test module tests the basic model of first model for the data structures；

The generation module if the basic model test being additionally operable to first model passes through, generates first model Basic model intermediate data；

The sending module is additionally operable to the basis that the data structures feed back each first model to the polymerization analysis center The intermediate data of the basic model for first model that the test result of model and test pass through.