CN108681811B - Decentralized data ecosystem - Google Patents
Decentralized data ecosystem Download PDFInfo
- Publication number
- CN108681811B CN108681811B CN201810435333.5A CN201810435333A CN108681811B CN 108681811 B CN108681811 B CN 108681811B CN 201810435333 A CN201810435333 A CN 201810435333A CN 108681811 B CN108681811 B CN 108681811B
- Authority
- CN
- China
- Prior art keywords
- data
- proofreading
- calibration
- party
- demander
- 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.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06395—Quality analysis or management
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q20/00—Payment architectures, schemes or protocols
- G06Q20/22—Payment schemes or models
- G06Q20/28—Pre-payment schemes, e.g. "pay before"
Abstract
The invention discloses a decentralized data ecosystem, which comprises a task allocation system, a task management system and a task management system, wherein the task allocation system is used for allocating data demand information of a data demand party to a data provider; after receiving data provided by a data provider, respectively sending a data checking task to a data checking party, and loading calibration data to a data quality detection system; the data quality detection system receives the proofreading data of the data proofreading party, proofreads the proofreading data according to the calibration data and/or AI technology, delivers the finished product data to the data demand party after the proofreading is confirmed, submits a payment instruction to the settlement system and feeds back the data production quality to the task allocation system; and the settlement system receives the prepayment cost of the data demand party and pays the settlement cost to the data provider and the data proof party from the prepayment cost according to the payment instruction. The invention can ensure the quality of manual data correction work in a decentralized mode.
Description
Technical Field
The invention relates to the technical field of supply and demand ecosystems of artificial intelligence databases, in particular to a decentralized data ecosystem.
Background
Artificial Intelligence (AI) has gained rapid growth over the last decade. With the open source of technologies such as deep learning and the rapid advancement of computing devices, artificial intelligence has begun to gradually impact various aspects of human life. Technologies such as intelligent voice, face recognition and automatic driving are no longer just hot spots of academic circles, but are about to actually walk into lives of everyone.
After having an unlimited number of possible artificial intelligence technologies and applications, it is a huge amount of quality data that provides fuel. The data is used as training and testing data of an artificial intelligence algorithm and has the same important position as a machine learning algorithm. Training data becomes especially important in the context of deep learning algorithms that are already fully open source.
In the future, the progress of the artificial intelligence technology will depend on the massive high-quality data support. On the one hand, the capacity of the existing artificial intelligence data is limited, and the future data demand cannot be supported. On the other hand, everyone has left a lot of data on the internet, but the property right of the data is unclear, and the utilization rate of the data is greatly improved. In the current database industrial chain, a plurality of levels of intermediate merchants exist between a database demand party and an actual manufacturing party, so that the data manufacturing efficiency is low, the cost is high, and the personal information data safety of an acquirer cannot be guaranteed.
The application block chaining (decentralized) technology is the best way to utilize massive fragmented internet user data and improve the data productivity and the use efficiency. The decentralized data ecosystem can enable personal data generated by mass users to be higher in safety, namely the data are only visible to the real demand side of specific data, and therefore the possibility that merchants among data at all levels obtain information contained in the data is eliminated. In addition, the decentralized data ecosystem can enable data providers to obtain full control over the types of data provided by the data providers and destinations, and therefore the possibility that data intermediaries misuse privacy and personal information of the data providers is avoided.
Although various decentralization protocols and solutions based on HASH (HASH) algorithms exist, none of these solutions can be directly applied to a decentralized data ecosystem. The most important of them are: if data (e.g., speech data, image data, etc.) is intended to contribute to an artificial intelligence algorithm (e.g., deep learning), the data itself needs to be accurately labeled. If a speech recording is to be used by a speech recognition algorithm, a text signal accompanying the speech signal (i.e. the content of the speech signal is written), which is hereinafter referred to as accompanying data, must be provided to the algorithm for training. The work of obtaining the corresponding interpretation content for a certain data, called labeling work, usually needs manual proofreading and labeling. Therefore, the existing decentralized technology can ensure the uniqueness and the safety of the data provided by the data provider and record the transaction of certain data; the uniqueness and the safety of the accompanying data generated by the proofreader of a certain pair of data and the record of the transaction of the proofread data can be ensured. However, at present, the quality of manual data correction work cannot be guaranteed in a decentralized manner. If this link cannot be guaranteed, the generated data will not be used by the artificial intelligence system.
Disclosure of Invention
The invention aims to provide a decentralized data ecosystem aiming at the technical defects in the prior art.
The technical scheme adopted for realizing the purpose of the invention is as follows:
a decentralized data ecosystem, comprising:
the task allocation system is used for allocating the data demand information of the data demand party to the data provider according to a preset allocation rule; after receiving data provided by a data provider, respectively sending a data proofreading task to a data proofreader according to a preset distribution rule, marking the data by the data proofreader, and loading calibration data into a data quality detection system;
the data quality detection system is used for receiving the proofreading data of the data proofreader, detecting the proofreading data according to the loaded calibration data and/or AI technology, delivering finished product data to the data demander after the proofreading confirmation is passed, simultaneously submitting a payment instruction to the settlement system, and simultaneously performing data production quality feedback to the task allocation system;
and the settlement system is used for receiving the prepayment cost of the data demand party, settling the cost according to the payment instruction of the data quality detection system and paying the settlement cost to the data provider and the data proofreader in prepayment.
The data demand information comprises data type, data scale, data price, data quality requirement and data quality inspection method.
After receiving the data demand information, the task allocation system firstly carries out the inspection of submitting the demand according to the rule agreed in advance for the data demand party, and if the inspection requirement is met, the task allocation system allocates a data acquisition task to the data supply party; otherwise, feeding back to the data demander, requiring the data demander to modify, and repeatedly executing until the data demander passes the verification.
The task allocation system firstly broadcasts the collection data provider and the data corrector downwards before allocating the data acquisition task to the data provider, if the minimum number of participants can not be obtained within a certain time, the result is fed back to the data demander, and the data demander is prompted to modify the demand until the minimum number of participants is achieved.
And each minimum unit data in the data submitted by the data provider and the personal identification information of the data provider are generated into a hash value with a fixed length by the task allocation system through arithmetic.
The data quality detection system receives each minimum unit data in the proofreading data submitted by the data proofreading party and generates a hash value with a fixed length by the aid of the data marking information and the personal identification information of the data proofreading party through the Hill operation.
When the data quality detection system does not pass the proofreading, a proofreading rework instruction is sent to the data proofreader, and the data proofreader needs to upload the data to the data proofreader after improving the proofreading quality.
The calibration data is finished product data provided by a data demander.
The data quality detection system detects the proofreading data through the calibration data and/or artificial intelligence.
The decentralized data ecosystem provided by the invention can ensure the quality of manual data correction work in a decentralized mode and ensure that the generated data can be used by an artificial intelligent system.
Drawings
FIG. 1 is a schematic diagram of a decentralized data ecosystem.
Detailed Description
The invention is described in further detail below with reference to the figures and specific examples. It should be understood that the specific embodiments described herein are merely illustrative of the invention and do not limit the invention.
Referring to FIG. 1, a decentralized data ecosystem, comprises:
the task allocation system is used for allocating the data demand information of the data demand party to the data provider according to a preset allocation rule; after receiving data provided by a data provider, respectively sending a data proofreading task to a data proofreader according to a preset distribution rule, marking the data by the data proofreader, and loading calibration data into a data quality detection system;
the data quality detection system is used for receiving the proofreading data of the data proofreader, detecting the proofreading data according to the loaded calibration data and/or AI technology, delivering finished product data to the data demander after the proofreading confirmation is passed, simultaneously submitting a payment instruction to the settlement system, and simultaneously performing data production quality feedback to the task allocation system;
and the settlement system is used for receiving the prepayment cost of the data demand party, settling the cost according to the payment instruction of the data quality detection system and paying the settlement cost to the data provider and the data proofreader in prepayment.
It should be noted that, in the present invention, each subsystem in the data ecosystem is constructed in a decentralized manner, that is, each system exists logically, but actually is deployed in a distributed manner in a computer providing computing power for the data ecosystem, and the transmission of the data machine instruction is performed through an encrypted internet format. Any entity with computing resources can provide computing power and benefit from the system, including data consumers, data providers, and data correctors.
In the invention, the data demand information or specification demand of the data demand is submitted to the task allocation system by the data demand, including but not limited to data type (such as voice, image, text, video, etc.), data scale, data price, data quality requirement, and data quality inspection method (such as adopting calibration data quality inspection provided by the data demand party or adopting artificial intelligence AI technology quality inspection).
After receiving the data demand information, the task allocation system firstly carries out the inspection of submitting the demand according to the rule agreed in advance for the data demand party, and if the inspection requirement is met, the task allocation system allocates a data acquisition task to the data supply party; otherwise, feeding back the existing problems to the data demander, and requiring the data demander to modify; and submitting the audit again after the data demander is modified. This process is repeated until the demand passes.
After the data demand of the data demand side passes system examination, the task allocation system firstly broadcasts the collection data provider and the data school side downwards before allocating a data acquisition task to the data provider, and if the minimum number of participants required by the system cannot be obtained within a certain time, the task allocation system feeds back a result to the data demand side and prompts the data demand side to modify demands (such as data price or quotation) until the minimum number of participants is achieved.
After the minimum number of the participants is achieved, the data demand side pays the fees to the settlement system in advance according to a certain proportion, then the task distribution system distributes data acquisition tasks to the data provider side, and the subsequent work such as data acquisition, production, proofreading and the like is started.
When the task allocation system allocates the data acquisition task to the data provider, the data acquisition task is allocated in consideration of the data making capacity, the making data quality and the like of the provider.
One part of the pre-paid fee is to pay the fee to the party who provides the proofreading for the data acquisition task, and the other part is used as the guarantee for the settlement of the data provider when the data production is completed. And the data is left in the data settlement system until the data finished product submitted by the data provider passes the quality detection of the data quality detection system and is submitted to the data demander.
The data provider collects data as required after receiving a collection task, and submits the data to the task distribution system after the collection is completed, wherein each minimum unit data _ capture (for example, in voice data, the minimum unit is a sentence) in the data submitted by the data provider received by the task distribution system is combined with the personal identification information userID of the data provider to generate a hash value with a fixed length through a desired operation, as shown in the following formula:
hc＝hash(data_capture+userID)
in addition, the data provider uploads the collected data to the task distribution system and simultaneously issues the hash value to the system in a broadcasting mode, and the hash value is used for declaring the right of certain collected data so as to ensure the accuracy in settlement and the accuracy of secondary sales settlement.
After certain data are uploaded to the task allocation system by the data provider, the task allocation system allocates the proofreading task to the data proofreader according to certain rules (such as capability and quality of data production according to the data proofreader).
After receiving the proofreading task, the data proofreading party performs data proofreading according to the requirement, and submits the data to the data quality detection system after the proofreading is completed, wherein the data quality detection system receives each minimum unit data _ capture (for example, in voice data, the minimum unit is a sentence) in the proofreading data submitted by the data proofreading party and generates a hash value with a fixed length together with the personal identification information userID and the data tag data _ labelling of the data proofreading party through a desired operation, as shown in the following formula.
hl＝hash(data_capture+data_labelling+userID)
The data proofreader also issues a hash value to the system in a broadcasting mode while uploading the acquisition result to the system, so as to declare the right to the data of a certain proofreading, and ensure the accuracy during settlement and the accuracy of secondary sales settlement.
Specifically, in the invention, when the data quality detection system fails in the proofreading, the data proofreading party sends a proofreading rework instruction, and the data proofreading party needs to upload the data until the data proofreading party passes the verification after improving the proofreading quality. After qualified data is provided for a data demand party, the data quality detection system feeds back the working quality of a data provider and a data school party to the task allocation system, and punishments are carried out on parties who additionally consume computing power (namely, the waste of computing resources such as additional secondary verification of the system caused by unqualified quality) in the processing process, wherein the measures include but are not limited to settlement for deducting the cost of the additional computing power, adjustment of the allocation priority of the next task and the like.
In the invention, when the quality detection is carried out, the calibration data is qualified finished product data which is provided by a data demand party and detected in a manual mode.
Since the original data is collected by the data provider, the accompanying annotation information is provided by the data corrector. Since it is desirable to verify the working quality of a data proof party, in a data quality detection system, it is possible to:
a) The data demand side provides some finished product data which are manually checked and serve as a basis for evaluating the working quality of the proofreading side;
b) Adding one or more existing artificial intelligence systems for judgment, wherein the result is used as the basis for evaluating the working quality of the data proofreading party;
c) And performing comprehensive judgment by using finished product data provided by a data demander and an artificial intelligence system, wherein the result is used as a basis for evaluating the working quality of a data corrector.
The method of detecting the quality of data will be described below.
a) Soft checking method for providing small amount of calibration data by data demander
If the data required by the data demander is N units (for example, a speech signal in a speech recognition database can be one unit of data), M pieces of calibration data (M < < N) can be prepared simultaneously according to the requirement of the data demander on the quality of the database.
Specifically, the data may be randomly inserted into the data proofreading process, and the data proofreader is required to perform proofreading on the corresponding data as required (but does not inform which of the required proofreading data are calibration data). For example, in a voice database, the audio signals in the calibration data and the recorded audio data collected by the data provider to be calibrated are randomly sent to the data calibrator in a certain proportion for calibration.
Minimum of M pieces of calibration data is
M min ＝I max *m
Wherein, I max Maximum number of proofreading allowed to be accepted by a single data proof party (each data proof party accepts I at most according to initial submission of a proof application) max A secondary proofreading task, wherein data is distributed to a data proofreading party in batches according to a certain unit because data acquisition needs a certain period, such as data of 1000 units for proofreading each time; m is the amount of calibration data added for each calibration (e.g., each time)The data of 1000 units of calibration may contain 20 calibration data, i.e. 980 units of actual annotation data). This ensures that the calibration data (i.e., the verification data) used by each of the verification partners is not duplicated.
In the data quality detection system, part of calibration data in a data set is extracted from unit data submitted by a certain data calibration party for calibration. The verification method is to compare the deviation between the result of the data verification party on the verification data and the verification result provided by the supplier. For example, for image data, verification of the pixel level deviation of the annotation object can be performed.
Taking the most commonly used image box notation as an example, if the graph Pi indicates that the upper left vertex coordinates of the box are (a, B), the lower right vertex coordinates are (C, D), the upper left vertex coordinates of the verification data are (a, B), and the lower right vertex coordinates are (C, D), then the deviation diff is:
wherein, diff LT Error for labeling the Top Left (Top) vertex of the box and the Top Left vertex of the verification box; diff is a unit of a product RB Error of a Right Bottom (Bottom) vertex of the label box and a Right Bottom vertex of the verification box;
for any picture Pi, if the deviation is less than the preset value (diff) target ) If the picture label is valid, that is:
if the labeled box deviates 1 pixel from the upper left and 1 pixel from the lower right, then diff LT And diff RB Are all 2. If diff target If greater than 2, the label may pass.
And if the accuracy of the m units of calibration data is greater than the preset value, judging that the corresponding actual data is valid.
b) Verification method for providing extra computing power by using artificial intelligence system
If the data demander cannot provide the calibration data, the system can provide extra calculation power to check the calibration quality by using an artificial intelligence system.
Often for a data collection task, the goal is to improve the performance of existing artificial intelligence systems. It can be verified using existing artificial intelligence systems. If the required data has the correct rate P d Lower than the recognition rate P of the existing artificial intelligence system s Comparing the result of the verified data generated by the existing artificial intelligence system with the verification result, and if the consistent ratio is more than or equal to P d The data may be validated.
If the required data has the correct rate P d Higher than the recognition rate P of the existing artificial intelligence system s Comparing the result of the verified data generated by the existing artificial intelligence system with the verification result, and if the consistent proportion is less than P s The data direct verification fails; if the consistent ratio is greater than P s Part of the data is proposed as redundant calibration data and is simultaneously distributed to a plurality of data correctors for recalibration. Taking the proofreading answer with more people as the correct value, comparing with the part of original proofreading data, and if the consistency rate is more than P d The proof of check passes, otherwise fails. The more verified proofreading answer in the redundant calibration data can be used as the redundant calibration data of the system to be distributed to other proofreaders who do not use the batch of calibration data to perform the operation in a).
The data verification method of the data quality detection system can verify whether a certain data set meets the specified verification requirements according to the preset verification method and the ratio, namely, the verified data set is allowed to have certain errors, errors and the like, and the data set can be regarded as the whole to meet the requirements as long as the preset accuracy, precision and the like can be met.
The foregoing is only a preferred embodiment of the present invention, and it should be noted that, for those skilled in the art, many modifications and adaptations can be made without departing from the principle of the present invention, and such modifications and adaptations should also be considered as the scope of the present invention.
Claims (4)
1. A decentralized data ecosystem, comprising:
the task distribution system is used for distributing the data demand information of the data demand party to the data provider according to a preset distribution rule; after receiving data provided by a data provider, respectively sending a data proofreading task to a data proofreader according to a preset distribution rule, marking the data by the data proofreader, and loading calibration data into a data quality detection system;
the data quality detection system is used for receiving the proofreading data of the data proofreader, detecting the proofreading data according to the loaded calibration data and/or AI technology, delivering finished product data to the data demander after the proofreading confirmation is passed, simultaneously submitting a payment instruction to the settlement system, and simultaneously performing data production quality feedback to the task allocation system;
when the data quality detection system detects, if a data demander provides calibration data, the calibration data provided by the data demander is randomly inserted into data to be calibrated and is sent to a data calibrator for calibration, and for unit data submitted by a certain data calibrator, the calibration data in a unit data set is extracted for calibration; the verification method comprises the steps of comparing the result of a data corrector on the verification data with the deviation of the verification result provided by a data demander, wherein the deviation comprises the pixel level deviation of a marked object of the image data; if the deviation is smaller than the preset value, the proofreading is passed, otherwise, the proofreading is not passed;
if the data demander can not provide the calibration data, an artificial intelligence system is used for providing extra calculation power to check the calibration quality: if the accuracy P of the required data is correct d Lower than the recognition rate P of the artificial intelligence system s Comparing the verified data with the verification result generated by the artificial intelligence system, and if the consistent ratio is more than or equal to P d The data is verified; if the required data areAccuracy P d Recognition rate P higher than artificial intelligence system s Comparing the verified data with the verification result generated by the artificial intelligence system, and if the consistent proportion is less than P s The data verification fails; if the consistent ratio is greater than P s Distributing partial data as redundant calibration data to multiple data proofreading parties for proofreading again, taking the proofreading answer with more people as correct value, comparing with the original proofreading data, and if the consistency rate is more than P d If the verification is passed, otherwise, the verification is not passed;
the settlement system is used for receiving the prepayment cost of the data demand party, settling the cost according to the payment instruction of the data quality detection system and paying the settlement cost to the data provider and the data proof party from the prepayment;
each minimum unit data in the data submitted by the data provider and received by the task allocation system is combined with the personal identification information of the data provider to generate a hash value with a fixed length through arithmetic;
the data quality detection system receives each minimum unit data in the proofreading data submitted by the data proofreading party and generates a hash value with a fixed length together with the data marking information and the personal identification information of the data proofreading party through the Hi-Ching operation;
the calibration data is finished product data provided by a data demander;
the data demand information comprises data type, data scale, data price, data quality requirement and data quality inspection method.
2. The decentralized data ecosystem according to claim 1, wherein the task allocation system, after receiving the data demand information, performs a demand submission check according to a rule predetermined by a data demand party, and if the check requirement is met, allocates a data acquisition task to a data provider; otherwise, feeding back to the data demander, requiring the data demander to modify, and repeatedly executing until the data demander passes the check.
3. The decentralized data ecosystem of claim 1, wherein the task allocation system broadcasts the collection data provider and the data corrector downwards before allocating the data collection task to the data provider, and if the minimum number of participants cannot be obtained within a certain time, the task allocation system feeds back a result to the data demander and prompts the data demander to modify the demand until the minimum number of participants is reached.
4. The decentralized data ecosystem according to claim 1, wherein the data quality detection system sends a proofreading rework instruction to the data proofreader when the proofreading fails, and the data proofreader needs to upload the data until the data proofreading fails after improving the proofreading quality.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810435333.5A CN108681811B (en) | 2018-05-09 | 2018-05-09 | Decentralized data ecosystem |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810435333.5A CN108681811B (en) | 2018-05-09 | 2018-05-09 | Decentralized data ecosystem |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108681811A CN108681811A (en) | 2018-10-19 |
| CN108681811B true CN108681811B (en) | 2022-10-18 |
Family
ID=63805431
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810435333.5A Active CN108681811B (en) | 2018-05-09 | 2018-05-09 | Decentralized data ecosystem |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108681811B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110534168A (en) * | 2019-08-30 | 2019-12-03 | 北京百度网讯科技有限公司 | Medicine advises indicating risk method, apparatus, electronic equipment and storage medium |
| CN111311832A (en) * | 2020-03-04 | 2020-06-19 | 长春市万易科技有限公司 | Gas payment method based on block chain |
| CN114119216A (en) * | 2020-09-30 | 2022-03-01 | 京东方科技集团股份有限公司 | Data communication method and electronic equipment |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104517212A (en) * | 2013-09-29 | 2015-04-15 | 林国华 | Specific field information service transaction realization method and system |
| CN106209874A (en) * | 2016-07-18 | 2016-12-07 | 沈阳师范大学 | A kind of intelligent perception distribution system and method for allocating tasks thereof |
| CN106779488A (en) * | 2017-01-20 | 2017-05-31 | 深圳市天朗时代科技有限公司 | The proofreading method and system of digital version e-book |
| CN107169764A (en) * | 2017-05-10 | 2017-09-15 | 山东大学 | Fair data trade method based on block chain |
| CN107729378A (en) * | 2017-07-13 | 2018-02-23 | 华中科技大学 | A kind of data mask method |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8229809B2 (en) * | 2009-11-03 | 2012-07-24 | Hartford Fire Insurance Company | System and method for use in management of financial assets |
-
2018
- 2018-05-09 CN CN201810435333.5A patent/CN108681811B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104517212A (en) * | 2013-09-29 | 2015-04-15 | 林国华 | Specific field information service transaction realization method and system |
| CN106209874A (en) * | 2016-07-18 | 2016-12-07 | 沈阳师范大学 | A kind of intelligent perception distribution system and method for allocating tasks thereof |
| CN106779488A (en) * | 2017-01-20 | 2017-05-31 | 深圳市天朗时代科技有限公司 | The proofreading method and system of digital version e-book |
| CN107169764A (en) * | 2017-05-10 | 2017-09-15 | 山东大学 | Fair data trade method based on block chain |
| CN107729378A (en) * | 2017-07-13 | 2018-02-23 | 华中科技大学 | A kind of data mask method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108681811A (en) | 2018-10-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110263024B (en) | Data processing method, terminal device and computer storage medium | |
| US11853935B2 (en) | Automated recommendations for task automation | |
| KR101887415B1 (en) | Program and method for checking data labeling product | |
| US20190272488A1 (en) | System and method for outsourcing computer-based tasks | |
| CN108681811B (en) | Decentralized data ecosystem | |
| US9892106B1 (en) | Methods, systems and articles for correcting errors in electronic government forms | |
| US20040049445A1 (en) | Financial services automation | |
| US20210065304A1 (en) | Contract automation with blockchain based interaction and recording | |
| WO2020233402A1 (en) | Accounts payable order validation method, apparatus and device, and storage medium | |
| US20200042624A1 (en) | Electronic Document Workflow | |
| CN110222323A (en) | A kind of voucher generation method, device, readable storage medium storing program for executing and terminal device | |
| CN113256299A (en) | System and method for automatically generating risk examination order and auditing mechanism | |
| CN113781048B (en) | Transaction information verification and settlement method based on block chain | |
| US20230245237A1 (en) | Systems and methods for allocating assets to directed and interest-based participants | |
| CN115391758B (en) | Self-service business platform system | |
| US20200051167A1 (en) | System and method for financial matching | |
| KR102476181B1 (en) | Method for generating automatically of certificated emissions reduction document and the system thereof | |
| KR102349164B1 (en) | System for appointment transaction of insurance planner | |
| CN114663201A (en) | Electronic bidding management system and method | |
| CN113705180A (en) | Document editing and reviewing method and device, electronic equipment and storage medium | |
| CN113255836A (en) | Job data processing method and device, computer equipment and storage medium | |
| CN110852679A (en) | Document adjusting method, device and equipment | |
| RU2680354C1 (en) | Trades in real time mode on a multi-variant scheme carrying out method and system | |
| CN112541732A (en) | Intelligent bidding contract generation method and device and readable storage medium | |
| CN110930206A (en) | Invoice matching method and device and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |