CN108510281A - Data detection method and device, electronic equipment and storage medium - Google Patents
Data detection method and device, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN108510281A CN108510281A CN201810276790.4A CN201810276790A CN108510281A CN 108510281 A CN108510281 A CN 108510281A CN 201810276790 A CN201810276790 A CN 201810276790A CN 108510281 A CN108510281 A CN 108510281A
- Authority
- CN
- China
- Prior art keywords
- address
- electronic
- target
- fraud
- neural network
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
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
- G06Q20/00—Payment architectures, schemes or protocols
- G06Q20/38—Payment protocols; Details thereof
- G06Q20/40—Authorisation, e.g. identification of payer or payee, verification of customer or shop credentials; Review and approval of payers, e.g. check credit lines or negative lists
- G06Q20/401—Transaction verification
- G06Q20/4016—Transaction verification involving fraud or risk level assessment in transaction processing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
-
- 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
- G06Q20/04—Payment circuits
- G06Q20/06—Private payment circuits, e.g. involving electronic currency used among participants of a common payment scheme
- G06Q20/065—Private payment circuits, e.g. involving electronic currency used among participants of a common payment scheme using e-cash
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Accounting & Taxation (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Finance (AREA)
- Strategic Management (AREA)
- General Business, Economics & Management (AREA)
- Artificial Intelligence (AREA)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computer Security & Cryptography (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Health & Medical Sciences (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)
Abstract
The embodiment of the invention provides a data detection method and device, electronic equipment and a storage medium. The method comprises the following steps: determining a target electronic currency transaction address to be transacted; obtaining a transaction record of a target electronic currency transaction address in a first preset time period as a first target transaction record; coding the first target transaction record by adopting a preset coding mode to obtain a target coding result; inputting a target coding result into a pre-constructed neural network model; the neural network model is used for: determining a fraud detection result of the electronic money transaction address corresponding to the target coding result; and obtaining a fraud detection result output by the neural network model as a fraud detection result of the target electronic currency transaction address. By applying the embodiment of the invention, whether the transaction to be carried out is a fraud transaction can be determined, thereby reducing the fraud transaction and protecting the asset security of the user.
Description
Technical field
The present invention relates to field of computer technology, are situated between more particularly to data detection method, device, electronic equipment and storage
Matter.
Background technology
With the development of technology, more and more users are traded using electronic money to substitute conventional currency.Wherein,
Part electronic money is the encryption currency based on block chain technology, such as bit coin and ether coin etc..
But since the transaction in block chain technology is anonymous deal, thus when user wants to carry out electronic money trading
When, user can not know the identity information of counterpart.So that user is difficult to distinguish by whether the transaction of progress is fraud
Transaction, causes usually to generate fraudulent trading, user's assets is made to suffer a loss.
Invention content
The embodiment of the present invention is designed to provide a kind of data detection method, device, electronic equipment and storage medium, with
It can determine whether pending transaction is fraudulent trading, to reduce fraudulent trading, protect the assets security of user.Specific skill
Art scheme is as follows:
In a first aspect, an embodiment of the present invention provides a kind of data detection method, this method may include:
Determine the target electronic moneytary operations address of pending transaction;
Transaction record of the target electronic moneytary operations address in the first preset time period is obtained, is merchandised as first object
Record;
First object transaction record is encoded using pre-arranged code mode, obtains target code result;
Target code result is input to the neural network model built in advance;Wherein, neural network model is used for:It determines
The fraud detection result of target code result corresponding target electronic moneytary operations address;
The fraud detection of neural network model output is obtained as a result, fraud detection as target electronic moneytary operations address
As a result.
Optionally, before target code result is input to the neural network model built in advance, method can also wrap
It includes:
Build neural network model;
Correspondingly, neural network model is built, may include:
Determine N number of preset electronic moneytary operations address, wherein N number of preset electronic moneytary operations address includes:With taking advantage of
The electronic money trading address of swindleness behavior, and/or, do not have the electronic money trading address of fraud;Wherein, N >=2;
Transaction record of each preset electronic moneytary operations address in the second preset time period is obtained respectively, as each
The corresponding second target transaction record in preset electronic moneytary operations address;
Each second target transaction record is encoded respectively using pre-arranged code mode, each second target is obtained and hands over
Easily record corresponding coding result;
It trains to obtain neural network model using default neural network algorithm and N number of training sample, wherein a trained sample
This includes:One the second target transaction records corresponding coding result and fraud identification information;Fraud identification information is used for:Mark
It is fraud address or non-fraud address to know second target transaction with recording corresponding preset electronic moneytary operations.
Optionally, transaction record of each preset electronic moneytary operations address in the second preset time period is obtained respectively,
As the corresponding second target transaction record in each preset electronic moneytary operations address, may include:
First of the preset electronic moneytary operations address with fraud is obtained respectively extends transaction record and second
Transaction record in preset time period, and/or, the preset electronic moneytary operations address without fraud is obtained respectively the
Transaction record in two preset time periods, as the corresponding second target transaction record in corresponding preset electronic moneytary operations address;
First, which extends transaction record, includes:Receive the transaction note of the electronic money trading address of preset electronic moneytary operations address remittance
Record.
Optionally, transaction record of the target electronic moneytary operations address in the first preset time period is obtained, as first
The step of target transaction record, may include:
Obtain target electronic moneytary operations address second extends transaction record and the transaction in the first preset time period
Record, as first object transaction record;Second, which extends transaction record, includes:Receive target electronic moneytary operations address remittance
The transaction record of electronic money trading address.
Optionally, when N number of preset electronic moneytary operations address includes:Electronic money trading address with fraud
When with electronic money trading address without fraud, the total number of the electronic money trading address with fraud with
The absolute difference of the total number of electronic money trading address without fraud is less than preset number.
Optionally, after training obtains neural network model, this method can also include:
Tuning is carried out to neural network model using M optimization sample, obtains tuning neural network model;Wherein, one
Optimizing sample includes:The corresponding coding result in electronic money trading address and fraud identification information for Optimized model;Its
In, M >=2;
Correspondingly, target code result is input to the neural network model built in advance, may include:
Target code result is input to tuning neural network model.
Optionally, default neural network algorithm may include:Recognition with Recurrent Neural Network algorithm, deep neural network algorithm and volume
Any one in product neural network algorithm.
Optionally, pre-arranged code mode may include:Term vector encodes item2vec modes or one-hot coding one-hot
Encoding modes.
Optionally, first object transaction record includes at least one sub- transaction record, and the first sub- transaction record includes:Target
Electronic money trading address, electronic money trading partner address, the revenue and expenditure type of target electronic moneytary operations address, trade gold
Volume, target electronic moneytary operations address transactions balances and exchange hour;First sub- transaction record is first object transaction record
In any one sub- transaction record;
Wherein, electronic money trading partner address includes:It receives target electronic moneytary operations address remittance or remits money to mesh
Mark the electronic money trading address of electronic money trading address.
Second aspect, the embodiment of the present invention additionally provide a kind of data detection device, which may include:
Determination unit, the target electronic moneytary operations address for determining pending transaction;
First obtains unit, for obtaining transaction note of the target electronic moneytary operations address in the first preset time period
Record, as first object transaction record;
Coding unit obtains target code for being encoded to first object transaction record using pre-arranged code mode
As a result;
Input unit, for target code result to be input to the neural network model built in advance;Wherein, neural network
Model is used for:Determine the fraud detection result of target code result corresponding target electronic moneytary operations address;
Second obtaining unit, for obtaining the fraud detection of neural network model output as a result, as target electronic currency
The fraud detection result of transaction address.
Optionally, which can also include construction unit;
Construction unit, for input unit by target code result be input to the neural network model built in advance it
Before, build neural network model;
Correspondingly, construction unit may include:
Determination sub-module, for determining N number of preset electronic moneytary operations address, wherein N number of preset electronic moneytary operations
Location includes:Electronic money trading address with fraud, and/or, with not having the electronic money trading of fraud
Location;Wherein, N >=2;
Submodule is obtained, for obtaining friendship of each preset electronic moneytary operations address in the second preset time period respectively
Easily record, as the corresponding second target transaction record in each preset electronic moneytary operations address;
Encoding submodule is obtained for being encoded respectively to each second target transaction record using pre-arranged code mode
Corresponding coding result is recorded to each second target transaction;
Training submodule, for training to obtain neural network model using default neural network algorithm and N number of training sample,
Wherein, a training sample includes:One the second target transaction records corresponding coding result and fraud identification information;Fraud
Identification information is used for:It is fraud address or non-fraud to identify second target transaction with recording corresponding preset electronic moneytary operations
Address.
Optionally, submodule is obtained to be specifically used for:
First of the preset electronic moneytary operations address with fraud is obtained respectively extends transaction record and second
Transaction record in preset time period, and/or, the preset electronic moneytary operations address without fraud is obtained respectively the
Transaction record in two preset time periods, as the corresponding second target transaction record in corresponding preset electronic moneytary operations address;
First, which extends transaction record, includes:Receive the transaction note of the electronic money trading address of preset electronic moneytary operations address remittance
Record.
Optionally, first obtains unit is specifically used for:
Obtain target electronic moneytary operations address second extends transaction record and the transaction in the first preset time period
Record, as first object transaction record;Second, which extends transaction record, includes:Receive target electronic moneytary operations address remittance
The transaction record of electronic money trading address.
Optionally, when N number of preset electronic moneytary operations address includes:Electronic money trading address with fraud
When with electronic money trading address without fraud, the total number of the electronic money trading address with fraud with
The absolute difference of the total number of electronic money trading address without fraud is less than preset number.
Optionally, in embodiments of the present invention, which can also include:
Adjustment unit, after obtaining neural network model in training, using M optimization sample to neural network model
Tuning is carried out, tuning neural network model is obtained;Wherein, an optimization sample includes:Electronic money for Optimized model
The corresponding coding result of transaction address and fraud identification information;Wherein, M >=2;
Input unit is specifically used for:Target code result is input to tuning neural network model.
Optionally, default neural network algorithm may include:Recognition with Recurrent Neural Network algorithm, deep neural network algorithm and volume
Any one in product neural network algorithm.
Optionally, pre-arranged code mode includes:Term vector encodes item2vec modes or one-hot coding one-hot
Encoding modes.
Optionally, first object transaction record includes at least one sub- transaction record, and the first sub- transaction record includes:Target
Electronic money trading address, electronic money trading partner address, the revenue and expenditure type of target electronic moneytary operations address, trade gold
Volume, target electronic moneytary operations address transactions balances and exchange hour;First sub- transaction record is first object transaction record
In any one sub- transaction record;
Wherein, electronic money trading partner address is:It receives target electronic moneytary operations address remittance or remits money to target
The electronic money trading address of electronic money trading address
The third aspect, the embodiment of the present invention additionally provide a kind of electronic equipment, including processor, communication interface, memory
And communication bus, wherein processor, communication interface, memory complete mutual communication by communication bus;
Memory, for storing computer program;
Processor when for executing the program stored on memory, realizes the side of any of the above-described data detection method
Method step.
Fourth aspect, the embodiment of the present invention additionally provide a kind of readable storage medium storing program for executing, are stored in the readable storage medium storing program for executing
Computer program realizes the method and step of any of the above-described data detection method when computer program is executed by processor.
5th aspect, the embodiment of the present invention additionally provides a kind of computer program product including instruction, when it is in electronics
When being run in equipment so that electronic equipment executes:The method and step of any of the above-described data detection method.
In embodiments of the present invention, when it needs to be determined that whether pending transaction is fraudulent trading, it can first determine and wait for
The target electronic moneytary operations address being traded.Then, the target electronic moneytary operations address is obtained in the first preset time
Transaction record in section, and the transaction record of acquisition is denoted as first object transaction record.Later, using pre-arranged code mode pair
First object transaction record is encoded, and target code result is obtained.Target code result will be obtained again is input to neural network
In model.Wherein, since the neural network model is determined for taking advantage of for the corresponding electronic money trading address of coding result
Cheat testing result.Fraud that the neural network model exports, for the target electronic moneytary operations address can thus be obtained
Testing result.In addition, due to when an electronic money trading address be fraud address when, then with the electronic money trading address into
Capable transaction is possible for fraudulent trading, thus the fraud detection result of the target electronic moneytary operations address i.e. pending
Transaction fraud detection result.In this way, can determine whether pending transaction is fraud by the fraud detection result
Transaction, to reduce fraudulent trading, protects the assets security of user.
Description of the drawings
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technology description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
Some embodiments of invention for those of ordinary skill in the art without creative efforts, can be with
Obtain other attached drawings according to these attached drawings.
Fig. 1 is a kind of flow chart of data detection method provided in an embodiment of the present invention;
Fig. 2 is a kind of flow chart of structure neural network model provided in an embodiment of the present invention;
Fig. 3 is a kind of structural schematic diagram of data detection device provided in an embodiment of the present invention;
Fig. 4 is the structural schematic diagram of a kind of electronic equipment provided in an embodiment of the present invention.
Specific implementation mode
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation describes, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts every other
Embodiment shall fall within the protection scope of the present invention.
In order to solve the problems, such as to be difficult to determine whether pending electronic money trading is fraudulent trading in the prior art, this
Inventive embodiments provide a kind of data detection method, device, electronic equipment and storage medium.
Data detection method provided in an embodiment of the present invention is illustrated first below.
Referring to Fig. 1, data detection method provided in an embodiment of the present invention may include steps of:
S101:Determine the target electronic moneytary operations address of pending transaction;
S102:Transaction record of the target electronic moneytary operations address in the first preset time period is obtained, as the first mesh
Mark transaction record;
S103:First object transaction record is encoded using pre-arranged code mode, obtains target code result;
S104:Target code result is input to the neural network model built in advance;Wherein, neural network model is used
In:Determine the fraud detection result of target code result corresponding target electronic moneytary operations address;
S105:The fraud detection of neural network model output is obtained as a result, taking advantage of as target electronic moneytary operations address
Cheat testing result.
In embodiments of the present invention, when it needs to be determined that whether pending transaction is fraudulent trading, it can first determine and wait for
The target electronic moneytary operations address being traded.Then, the target electronic moneytary operations address is obtained in the first preset time
Transaction record in section, and the transaction record of acquisition is denoted as first object transaction record.Later, using pre-arranged code mode pair
First object transaction record is encoded, and target code result is obtained.Target code result will be obtained again is input to neural network
In model.Wherein, since the neural network model is for determining target code result corresponding target electronic moneytary operations address
Fraud detection result.It can thus obtain that the neural network model exports, for the target electronic moneytary operations address
Fraud detection result.In addition, due to when an electronic money trading address is fraud address, then with the electronic money trading
The transaction that location carries out is possible for fraudulent trading, thus the fraud detection result of the target electronic moneytary operations address namely waits for
The fraud detection result of the transaction of progress.In this way, can by the fraud detection result come determine pending transaction whether be
Fraudulent trading protects the assets security of user to reduce fraudulent trading.
Wherein, the part electronic money in the embodiment of the present invention is the encryption currency based on block chain technology.Wherein, virtually
Coin is one kind in electronic money, and virtual coin includes but is not limited to bit coin, ether coin and Lai Te coin.In addition, of the invention
An electronic money trading address in embodiment is an electronic money trading account.
In addition, neural network model provided in an embodiment of the present invention can be input to the neural network model, it is arbitrary
The coding result of the corresponding transaction record in electronic money trading address is detected, to which the output phase answers electronic money trading address
Fraud detection result.
Ether coin is example below, and data detection method provided in an embodiment of the present invention is described in detail.
In the transaction of ether coin, when user A (assuming that the corresponding electronic money trading addresses user A be a) is to user B
Before (assuming that the corresponding electronic money trading addresses user B are b) payment ether coin, since user A can not know the body of user B
Part information leads to not determine whether user B is credible, thus can not determine whether pending transaction is fraudulent trading.
In order to determine whether the pending transaction is fraudulent trading, to avoid fraudulent trading from bringing assets to user A
Loss.Before user A executes the transaction, electronic equipment can determine the target electronic moneytary operations of the pending transaction of user A
Address, i.e. electronic money trading address b.
Then, electronic equipment can pass through ether mill API (Application Programming Interface, application
Program Interfaces), obtain transaction records of the electronic money trading address b in the first preset time period, and by the friendship of acquisition
Easily it is denoted as first object transaction record.Wherein, ether mill is an application platform based on block chain, decentralization,
This is existing concept, and it is not described here in detail.
Wherein, in order to obtain accurate fraud detection as a result, can be more as possible acquisition electronic money trading
The transaction record of address b.And in order to obtain more transaction record, when the first preset time period can be set corresponding to longer
It is long.Such as the first preset time period is set and is:Period before current time to 5 years corresponding to current time, certainly not office
It is limited to this.
In addition, in order to improve the speed for obtaining fraud detection result, the first preset time period can also be set as one
It in year (such as 2017 years), is not limited thereto certainly.Those skilled in the art can be arranged according to actual needs this
One preset time period, is not specifically limited herein.
For example, may include in a transaction record of the electronic money trading address b of acquisition:Electronic money trading
The revenue and expenditure type of address b, electronic money trading partner address c, electronic money trading address b:Expenditure, transaction amount:5 with
Too coin, target electronic moneytary operations address transactions balances:1 ether coin, exchange hour:20180316.
Wherein, electronic money trading partner address is in addition to that can be:Receive the electronics goods of electronic money trading address b remittances
Coin counterpart address c can also be:It remits money to the electronic money trading address e or f etc. of electronic money trading address b, this is
Reasonably.
Certainly, it in this transaction record can also include transaction total degree, such as 1000 times, include in certain transaction record
Element be not limited thereto.Wherein, which refers to:5 ether coin are being imported electricity by electronic money trading address b
Performed total number of transactions number before sub- moneytary operations partner address c.
After obtaining the corresponding first object transaction records of electronic money trading address b, term vector can be utilized to encode
Item2vec modes or one-hot coding one-hot encoding modes, encode the first object transaction record, will compile
Code result is denoted as target code as a result, obtaining target code result.
Wherein, one-hot coding one-hot encoding modes can indicate the member in transaction record in the form of vectors
Element a, you can element is indicated with the vector using a various dimensions.Wherein, the value of multiple dimensions in the vector is 0, only
The value of one dimension is 1.It is encoded using term vector coding item2vec modes, dimension can be reduced, and can excavate
The context relation of each element in transaction record, to improve the accuracy of vector semantically.
After obtaining target code result, neural network model can be built, and the target code result is input to
It builds in obtained neural network model, since the neural network model can determine the corresponding electronic money trading of coding result
The fraud detection of address as a result, thus neural network model output, taking advantage of for electronic money trading address b can be obtained
Cheat testing result.It is laid out, the mode for building the neural network model is described in detail for clarity subsequently.
Wherein, in one implementation, which can specifically include:Confidence level is very high, it is with a high credibility,
One kind during confidence level is general, with a low credibility and confidence level is very low.In another implementation, which also may be used
Think:Risk of fraud is high or risk of fraud is low.Certainly it is not limited thereto.
Assuming that neural network model output, be for the fraud detection result of electronic money trading address b:Confidence level
Low, i.e. risk of fraud degree is:It is with a low credibility.At this point, showing trading activity and the fraud address of electronic money trading address b
Trading activity similarity is higher, or relatively low with the trading activity similarity of non-fraud address.That is, electronic money trading
Location b is possible for fraud address.At this point, the transaction pending with electronic money trading address b is possible for fraudulent trading, in order to
Ensure the assets security of user, fraud detection result that can be by (i.e. risk of fraud) with a low credibility as pending transaction.
In this way, when user A learn the pending transaction it is with a low credibility after, can shut the book, to ensure the assets of user A
Safety.
In addition, the fraud detection result as electronic money trading address b is:When with a high credibility, show electronic money trading
The trading activity of address b and the trading activity similarity of fraud address are relatively low, or the trading activity similarity with non-fraud address
It is higher.In this way, when user A learn the pending transaction it is with a high credibility after, can be traded, ensure that user A's
Assets security.
The building process of neural network model provided in an embodiment of the present invention is described in detail below.
Referring to Fig. 2, the construction step of the neural network model may include:
S201:Determine N number of preset electronic moneytary operations address, wherein N number of preset electronic moneytary operations address includes:
Electronic money trading address with fraud, and/or, do not have the electronic money trading address of fraud;Wherein, N
≥2;
Wherein, in one implementation, the N number of preset electronic moneytary operations address determined is with fraud
Electronic money trading address.In another implementation, the N number of preset electronic moneytary operations address determined is not have to take advantage of
The electronic money trading address of swindleness behavior.In another realization method, the N number of preset electronic moneytary operations address determined was both wrapped
The electronic money trading address with fraud is included, and includes the electronic money trading address without fraud.Again
In a kind of realization method, the N number of preset electronic moneytary operations address determined had both included the electronic money trading for having fraud
Address, and include the first extension transaction address of the electronic money trading address with fraud.This is all reasonable.
Wherein, the first extension transaction address of the electronic money trading address with fraud refers to:Receiving this has
The electronic money trading address of the electronic money trading address remittance of fraud.
It is understood that each the preset electronic moneytary operations address determined is corresponding, there are one fraud mark letters
Breath.The fraud identification information is also non-fraud address for being fraud address with identifying preset electronic moneytary operations.Such as with 1
It is identified as fraud address, with the 0 non-fraud address of mark, this is all reasonable.Certainly it is not limited thereto.
When N number of preset electronic moneytary operations address includes:Electronic money trading address with fraud and do not have
When having the electronic money trading address of fraud, in order to avoid the testing result inaccuracy for the model that training obtains, Ke Yishe
The total number for setting the electronic money trading address with fraud and the electronic money trading address without fraud
The absolute difference of total number is less than preset number.
Wherein, the preset number can be arranged according to specific requirements in those skilled in the art, do not limit herein.
S202:Transaction record of each preset electronic moneytary operations address in the second preset time period is obtained respectively, is made
For the corresponding second target transaction record in each preset electronic moneytary operations address;
Wherein, obtain accurately detecting the neural network model of fraud in order to training, acquisition that can be more as possible
The transaction record of each preset electronic moneytary operations address.And in order to obtain more transaction record, it is default that second can be arranged
Period corresponds to longer duration.Such as the first preset time period is set and is:Before current time to 10 years corresponding to current time
Period, be not limited thereto certainly.Second preset time can be arranged in those skilled in the art according to actual needs
Section, is not specifically limited herein.
Since the electronic money trading address (cheating address) with fraud is after receiving remittance, can usually give
Other N number of electronic money trading addresses are transferred accounts, and this N number of electronic money trading address can execute similar transfer operation, finally
It transfers accounts into the wallet address convenient for embodiment.
Therefore, the neural network model for obtaining accurately detecting fraud address in order to training, can also be in determination
Preset electronic moneytary operations address when being electronic money trading address with fraud, obtain has fraud respectively
Preset electronic moneytary operations address first extend transaction record and the transaction record in the second preset time period, as phase
Answer the corresponding second target transaction record in preset electronic moneytary operations address.It is not in determining preset electronic moneytary operations address
When electronic money trading address with fraud, the preset electronic moneytary operations address without fraud is obtained respectively
Transaction record in the second preset time period, as the corresponding second target transaction note in corresponding preset electronic moneytary operations address
Record.It is electronic money trading with fraud and without fraud in determining preset electronic moneytary operations address
When location, first of the preset electronic moneytary operations address with fraud is obtained respectively and extends transaction record and is preset second
Transaction record in period, and the preset electronic moneytary operations address without fraud is obtained respectively when second is default
Between transaction record in section, as the corresponding second target transaction record in corresponding preset electronic moneytary operations address.
Wherein, the first extension transaction record includes:The electronic money for receiving preset electronic moneytary operations address remittance is handed over
The transaction record of easy address.
Correspondingly, in order to ensure the accuracy of testing result, pending transaction is being detected using the neural network model
During whether being fraudulent trading, obtains the second of target electronic moneytary operations address and extend transaction record and preset first
Transaction record in period, as first object transaction record.Wherein, the second extension transaction record includes:Receive the target
The transaction record of the electronic money trading address of electronic money trading address remittance.This is reasonable.
S203:Each second target transaction record is encoded respectively using pre-arranged code mode, obtains each second
Target transaction records corresponding coding result;
Wherein, pre-arranged code mode may include:Term vector encodes item2vec modes or one-hot coding one-hot
Encoding modes.Certainly it is not limited thereto.
S204:It trains to obtain neural network model using default neural network algorithm and N number of training sample, wherein one
Training sample includes:One the second target transaction records corresponding coding result and fraud identification information;Cheat identification information
For:It is fraud address or non-fraud address to identify second target transaction with recording corresponding preset electronic moneytary operations.
Wherein, default neural network algorithm includes:Recognition with Recurrent Neural Network algorithm (Recurrent neural networks,
RNN), deep neural network algorithm (Deep Neural Networks, DNN) and convolutional neural networks algorithm
Any one in (Convolutional Neural Network, CNN).Correspondingly, when utilization Recognition with Recurrent Neural Network algorithm
When, it can train to obtain Recognition with Recurrent Neural Network model;When using deep neural network algorithm, it can train to obtain depth nerve
Network model;When using convolutional neural networks algorithm, it can train to obtain convolutional neural networks model.
These neural network models can calculate the target code result for being input to the neural network model and fraud address
Or the similarity of non-fraud address, to export the fraud detection of target code result corresponding target electronic moneytary operations address
As a result.It is thus possible to according to the fraud detection result of the determination and the pending transaction in the target electronic moneytary operations address.
In addition, the accuracy of the fraud detection result in order to improve neural network model output, nerve net is obtained in training
After network model, tuning can also be carried out to the neural network model using M optimization sample, wherein M >=2.Wherein, one
A optimization sample includes:The corresponding coding result in electronic money trading address and fraud identification information for Optimized model.
In this way, can be optimized to the parameter in neural network model, to improve the accuracy of output result.
To sum up, the neural network model that can be provided through the embodiment of the present invention determines whether pending transaction is fraud
Transaction, to reduce fraudulent trading, protects the assets security of user.
Corresponding to above method embodiment, the embodiment of the present invention additionally provides a kind of data detection device, should referring to Fig. 3
Device may include:
Determination unit 301, the target electronic moneytary operations address for determining pending transaction;
First obtains unit 302, for obtaining transaction of the target electronic moneytary operations address in the first preset time period
Record, as first object transaction record;
Coding unit 303 obtains target volume for being encoded to first object transaction record using pre-arranged code mode
Code result;
Input unit 304, for target code result to be input to the neural network model built in advance;Wherein, neural
Network model is used for:Determine the fraud detection result of target code result corresponding target electronic moneytary operations address;
Second obtaining unit 305, for obtaining the fraud detection of neural network model output as a result, as target electronic goods
The fraud detection result of coin transaction address.
It can when it needs to be determined that whether pending transaction is fraudulent trading using device provided in an embodiment of the present invention
First to determine the target electronic moneytary operations address of pending transaction.Then, the target electronic moneytary operations address is obtained
Transaction record in one preset time period, and the transaction record of acquisition is denoted as first object transaction record.Later, using default
Coding mode encodes first object transaction record, obtains target code result.The input of target code result will be obtained again
Into neural network model.Wherein, it is handed over since the neural network model is determined for the corresponding electronic money of coding result
The fraud detection result of easy address.Thus can obtain the neural network model output, for the target electronic moneytary operations
The fraud detection result of address.In addition, due to when an electronic money trading address be fraud address when, then with the electronic money
The transaction that transaction address carries out is possible for fraudulent trading, thus the fraud detection result of the target electronic moneytary operations address
It is exactly the fraud detection result of pending transaction.In this way, pending transaction can be determined by the fraud detection result
Whether it is fraudulent trading, to reduce fraudulent trading, protects the assets security of user.
Optionally, in embodiments of the present invention, which can also include construction unit;
Construction unit be used for input unit 304 target code result is input to the neural network model that builds in advance it
Before, build the neural network model;
The construction unit can specifically include:
Determination sub-module, for determining N number of preset electronic moneytary operations address, wherein N number of preset electronic moneytary operations
Location includes:Electronic money trading address with fraud, and/or, with not having the electronic money trading of fraud
Location;Wherein, N >=2;
Submodule is obtained, for obtaining friendship of each preset electronic moneytary operations address in the second preset time period respectively
Easily record, as the corresponding second target transaction record in each preset electronic moneytary operations address;
Encoding submodule is obtained for being encoded respectively to each second target transaction record using pre-arranged code mode
Corresponding coding result is recorded to each second target transaction;
Training submodule, for training to obtain neural network model using default neural network algorithm and N number of training sample,
Wherein, a training sample includes:One the second target transaction records corresponding coding result and fraud identification information;Fraud
Identification information is used for:It is fraud address or non-fraud to identify second target transaction with recording corresponding preset electronic moneytary operations
Address.
Optionally, submodule is obtained to be specifically used for:
First of the preset electronic moneytary operations address with fraud is obtained respectively extends transaction record and second
Transaction record in preset time period, and/or, the preset electronic moneytary operations address without fraud is obtained respectively the
Transaction record in two preset time periods, as the corresponding second target transaction record in corresponding preset electronic moneytary operations address;
First, which extends transaction record, includes:Receive the transaction note of the electronic money trading address of preset electronic moneytary operations address remittance
Record.
Optionally, in embodiments of the present invention, first obtains unit 302 specifically can be used for:
Obtain target electronic moneytary operations address second extends transaction record and the transaction in the first preset time period
Record, as first object transaction record;Second, which extends transaction record, includes:Receive target electronic moneytary operations address remittance
The transaction record of electronic money trading address.
Optionally, when N number of preset electronic moneytary operations address includes:Electronic money trading address with fraud
When with electronic money trading address without fraud, the total number of the electronic money trading address with fraud with
The absolute difference of the total number of electronic money trading address without fraud is less than preset number.
Optionally, in embodiments of the present invention, which can also include:
Adjustment unit, after obtaining neural network model in training, using M optimization sample to neural network model
Tuning is carried out, tuning neural network model is obtained;Wherein, an optimization sample includes:Electronic money for Optimized model
The corresponding coding result of transaction address and fraud identification information;Wherein, M >=2;
Input unit is specifically used for:The target code result is input to the tuning neural network model.
Optionally, in embodiments of the present invention, default neural network algorithm may include:Recognition with Recurrent Neural Network algorithm, depth
Spend any one in neural network algorithm and convolutional neural networks algorithm.
Optionally, in embodiments of the present invention, pre-arranged code mode includes:Term vector encodes item2vec modes or solely heat
Encode one-hot encoding modes.
Optionally, first object transaction record includes at least one sub- transaction record, and the first sub- transaction record includes:Target
Electronic money trading address, electronic money trading partner address, the revenue and expenditure type of target electronic moneytary operations address, trade gold
Volume, target electronic moneytary operations address transactions balances and exchange hour;First sub- transaction record is first object transaction record
In any one sub- transaction record;
Wherein, electronic money trading partner address includes:It receives target electronic moneytary operations address remittance or remits money to mesh
Mark the electronic money trading address of electronic money trading address.
Corresponding to above method embodiment, the embodiment of the present invention additionally provides a kind of electronic equipment, referring to Fig. 4, the electronics
Equipment includes processor 401, communication interface 402, memory 403 and communication bus 404, wherein processor 401, communication interface
402, memory 403 completes mutual communication by communication bus 404;
Memory 403, for storing computer program;
Processor 401 when for executing the program stored on memory 403, realizes any of the above-described Data Detection side
The method and step of method.
When it needs to be determined that whether pending transaction is fraudulent trading, electronic equipment provided in an embodiment of the present invention can be with
First determine the target electronic moneytary operations address of pending transaction.Then, the target electronic moneytary operations address is obtained first
Transaction record in preset time period, and the transaction record of acquisition is denoted as first object transaction record.Later, using default volume
Code mode encodes first object transaction record, obtains target code result.Target code result will be obtained again to be input to
In neural network model.Wherein, since the neural network model is determined for the corresponding electronic money trading of coding result
The fraud detection result of address.Thus, it is possible to obtain the neural network model output, for the target electronic moneytary operations
The fraud detection result of location.In addition, due to when an electronic money trading address is fraud address, then handed over the electronic money
The transaction that easy address carries out is possible for fraudulent trading, thus the fraud detection result of the target electronic moneytary operations address is also
It is the fraud detection result of pending transaction.In this way, can be to determine pending transaction by the fraud detection result
It is no to protect the assets security of user for fraudulent trading to reduce fraudulent trading.
Corresponding to above method embodiment, the embodiment of the present invention additionally provides a kind of readable storage medium storing program for executing, the readable storage
Dielectric memory contains computer program, and the side of any of the above-described data detection method is realized when computer program is executed by processor
Method step.
The computer program stored in readable storage medium storing program for executing provided in an embodiment of the present invention is held by the processor of electronic equipment
After row, electronic equipment can first determine the target electronic moneytary operations address of pending transaction.Then, the target electronic goods is obtained
Transaction record of the coin transaction address in the first preset time period, and the transaction record of acquisition is denoted as first object transaction note
Record.Later, first object transaction record is encoded using pre-arranged code mode, obtains target code result.It will obtain again
Target code result is input in neural network model.Wherein, since the neural network model is determined for coding result
The fraud detection result of corresponding electronic money trading address.Thus, it is possible to obtain the neural network model output, for this
The fraud detection result of target electronic moneytary operations address.In addition, due to being fraud address when an electronic money trading address
When, then the transaction carried out with the electronic money trading address is possible for fraudulent trading, thus the target electronic moneytary operations
The fraud detection result of location i.e. the fraud detection result of pending transaction.In this way, the fraud detection result can be passed through
It determines whether pending transaction is fraudulent trading, to reduce fraudulent trading, protects the assets security of user.
Corresponding to above method embodiment, the embodiment of the present invention additionally provides a kind of computer program production comprising instruction
Product, when it runs on an electronic device so that electronic equipment executes:The method and step of any of the above-described data detection method.
Computer program product provided in an embodiment of the present invention comprising instruction makes when it runs on an electronic device
The target electronic moneytary operations address of pending transaction can first be determined by obtaining electronic equipment.Then, the target electronic currency is obtained
Transaction record of the transaction address in the first preset time period, and the transaction record of acquisition is denoted as first object transaction record.
Later, first object transaction record is encoded using pre-arranged code mode, obtains target code result.Target will be obtained again
Coding result is input in neural network model.Wherein, it is corresponded to since the neural network model is determined for coding result
Electronic money trading address fraud detection result.Thus, it is possible to obtain the neural network model output, for the target
The fraud detection result of electronic money trading address.In addition, due to when an electronic money trading address is fraud address, then
The transaction carried out with the electronic money trading address is possible for fraudulent trading, thus the target electronic moneytary operations address is taken advantage of
Cheat testing result i.e. the fraud detection result of pending transaction.In this way, can be determined by the fraud detection result
Whether pending transaction is fraudulent trading, to reduce fraudulent trading, protects the assets security of user.
The communication bus that above-mentioned electronic equipment is mentioned can be Peripheral Component Interconnect standard (Peripheral Component
Interconnect, PCI) bus or expanding the industrial standard structure (Extended Industry Standard
Architecture, EISA) bus etc..The communication bus can be divided into address bus, data/address bus, controlling bus etc..For just
It is only indicated with a thick line in expression, figure, it is not intended that an only bus or a type of bus.
Communication interface is for the communication between above-mentioned electronic equipment and other equipment.
Memory may include random access memory (Random Access Memory, RAM), can also include non-easy
The property lost memory (Non-Volatile Memory, NVM), for example, at least a magnetic disk storage.Optionally, memory may be used also
To be at least one storage device for being located remotely from aforementioned processor.
Above-mentioned processor can be general processor, including central processing unit (Central Processing Unit,
CPU), network processing unit (Network Processor, NP) etc.;It can also be digital signal processor (Digital Signal
Processing, DSP), it is application-specific integrated circuit (Application Specific Integrated Circuit, ASIC), existing
It is field programmable gate array (Field-Programmable Gate Array, FPGA) or other programmable logic device, discrete
Door or transistor logic, discrete hardware components.
The term used in the embodiment of the present application is the purpose only merely for description specific embodiment, is not intended to be limiting
The application.In the embodiment of the present application and "an" of singulative used in the attached claims, " described " and "the"
It is also intended to including most forms, unless context clearly shows that other meanings.It is also understood that term used herein
"and/or" refer to and include one or more associated list items purposes any or all may combine.
It will be appreciated that though may be described in the embodiment of the present application using term " first ", " second ", " third " etc.
Various connectivity ports and identification information etc., but these connectivity ports and identification information etc. should not necessarily be limited by these terms.These terms
Only it is used for connectivity port and identification information etc. being distinguished from each other out.For example, in the case where not departing from the embodiment of the present application range,
First connectivity port can also be referred to as second connection end mouth, and similarly, second connection end mouth can also be referred to as the first connection
Port.
Depending on context, word as used in this " if " can be construed to " ... when " or " when ...
When " or " in response to determination " or " in response to detection ".Similarly, depend on context, phrase " if it is determined that " or " if detection
(condition or event of statement) " can be construed to " when determining " or " in response to determination " or " when the detection (condition of statement
Or event) when " or " in response to detection (condition or event of statement) ".
Through the above description of the embodiments, it is apparent to those skilled in the art that, for description
It is convenienct and succinct, only the example of the division of the above functional modules, in practical application, can as needed and will be upper
It states function distribution to be completed by different function modules, i.e., the internal structure of device is divided into different function modules, to complete
All or part of function described above.The specific work process of the system, apparatus, and unit of foregoing description, before can referring to
The corresponding process in embodiment of the method is stated, details are not described herein.
In several embodiments provided herein, it should be understood that disclosed electronic equipment, device and method,
It may be implemented in other ways.For example, the apparatus embodiments described above are merely exemplary, for example, the mould
The division of block or unit, only a kind of division of logic function, formula that in actual implementation, there may be another division manner, for example (,) it is multiple
Unit or component can be combined or can be integrated into another system, or some features can be ignored or not executed.It is another
Point, shown or discussed mutual coupling, direct-coupling or communication connection can be by some interfaces, device or
The INDIRECT COUPLING of unit or communication connection can be electrical, machinery or other forms.
The unit illustrated as separating component may or may not be physically separated, aobvious as unit
The component shown may or may not be physical unit, you can be located at a place, or may be distributed over multiple
In network element.Some or all of unit therein can be selected according to the actual needs to realize the mesh of this embodiment scheme
's.
In addition, each functional unit in each embodiment of the application can be integrated in a processing unit, it can also
It is that each unit physically exists alone, it can also be during two or more units be integrated in one unit.Above-mentioned integrated list
The form that hardware had both may be used in member is realized, can also be realized in the form of SFU software functional unit.
If the integrated unit is realized in the form of SFU software functional unit and sells or use as independent product
When, it can be stored in a computer read/write memory medium.Based on this understanding, the technical solution of the application is substantially
The all or part of the part that contributes to existing technology or the technical solution can be in the form of software products in other words
It embodies, which is stored in a storage medium, including some instructions are used so that a computer
It is each that equipment (can be personal computer, server or the network equipment etc.) or processor (processor) execute the application
The all or part of step of embodiment the method.And storage medium above-mentioned includes:USB flash disk, mobile hard disk, read-only memory
(Read Only Memory;Hereinafter referred to as:ROM), random access memory (Random Access Memory;Hereinafter referred to as:
RAM), the various media that can store program code such as magnetic disc or CD.
The above, the only specific implementation mode of the application, but the protection domain of the application is not limited thereto, it is any
Those familiar with the art can easily think of the change or the replacement in the technical scope that the application discloses, and should all contain
It covers within the protection domain of the application.Therefore, the protection domain of the application should be based on the protection scope of the described claims.
Claims (10)
1. a kind of data detection method, which is characterized in that the method includes:
Determine the target electronic moneytary operations address of pending transaction;
Transaction record of the target electronic moneytary operations address in the first preset time period is obtained, is merchandised as first object
Record;
The first object transaction record is encoded using pre-arranged code mode, obtains target code result;
The target code result is input to the neural network model built in advance;Wherein, the neural network model is used for:
Determine the fraud detection result of target code result corresponding target electronic moneytary operations address;
The fraud detection of neural network model output is obtained as a result, fraud as target electronic moneytary operations address
Testing result.
2. according to the method described in claim 1, it is characterized in that, the target code result is input to advance structure described
Before the neural network model built, the method further includes:
Build the neural network model;
The structure neural network model, including:
Determine N number of preset electronic moneytary operations address, wherein N number of preset electronic moneytary operations address includes:With taking advantage of
The electronic money trading address of swindleness behavior, and/or, do not have the electronic money trading address of fraud;Wherein, the N >=
2;
Transaction record of each preset electronic moneytary operations address in the second preset time period is obtained respectively, as each default
The corresponding second target transaction record in electronic money trading address;
Each second target transaction record is encoded respectively using the pre-arranged code mode, each second target is obtained and hands over
Easily record corresponding coding result;
It trains to obtain the neural network model using default neural network algorithm and N number of training sample, wherein a trained sample
This includes:One the second target transaction records corresponding coding result and fraud identification information;The fraud identification information is used
In:It is fraud address or non-fraud address to identify second target transaction with recording corresponding preset electronic moneytary operations.
3. according to the method described in claim 2, it is characterized in that, described obtain each preset electronic moneytary operations address respectively
Transaction record in the second preset time period, as the corresponding second target transaction note in each preset electronic moneytary operations address
Record, including:
First of the preset electronic moneytary operations address with fraud is obtained respectively extends transaction record and described second
Transaction record in preset time period, and/or, the preset electronic moneytary operations address without fraud is obtained respectively in institute
The transaction record in the second preset time period is stated, as the corresponding second target transaction note in corresponding preset electronic moneytary operations address
Record;Described first, which extends transaction record, includes:With receiving the electronic money trading of preset electronic moneytary operations address remittance
The transaction record of location.
4. according to the method described in claim 3, it is characterized in that, the acquisition target electronic moneytary operations address is
Transaction record in one preset time period, the step of as first object transaction record, including:
It obtains the second of target electronic moneytary operations address and extends transaction record and in first preset time period
Transaction record, as first object transaction record;Described second, which extends transaction record, includes:The target electronic currency is received to hand over
The transaction record of the electronic money trading address of easy address remittance.
5. according to the method described in claim 2, it is characterized in that, when N number of preset electronic moneytary operations address includes:
When electronic money trading address with fraud and the electronic money trading address without fraud, it is described have take advantage of
The sum of the total number of the electronic money trading address of swindleness behavior and the electronic money trading address without fraud
Purpose absolute difference is less than preset number.
6. according to the method described in claim 2, it is characterized in that, training obtain the neural network model after, it is described
Method further includes:
Tuning is carried out to the neural network model using M optimization sample, obtains tuning neural network model;Wherein, one
Optimizing sample includes:The corresponding coding result in electronic money trading address and fraud identification information for Optimized model;Its
In, M >=2;
It is described that the target code result is input to the neural network model built in advance, including:
The target code result is input to the tuning neural network model.
7. according to the method described in claim 2, it is characterized in that, the default neural network algorithm includes:Recycle nerve net
Any one in network algorithm, deep neural network algorithm and convolutional neural networks algorithm.
8. a kind of data detection device, which is characterized in that described device includes:
Determination unit, the target electronic moneytary operations address for determining pending transaction;
First obtains unit, for obtaining transaction note of the target electronic moneytary operations address in the first preset time period
Record, as first object transaction record;
Coding unit obtains target code for being encoded to the first object transaction record using pre-arranged code mode
As a result;
Input unit, for the target code result to be input to the neural network model built in advance;Wherein, the nerve
Network model is used for:Determine the fraud detection result of target code result corresponding target electronic moneytary operations address;
Second obtaining unit, for obtaining the fraud detection of the neural network model output as a result, as the target electronic
The fraud detection result of moneytary operations address.
9. a kind of electronic equipment, which is characterized in that including processor, communication interface, memory and communication bus, wherein processing
Device, communication interface, memory complete mutual communication by communication bus;
Memory, for storing computer program;
Processor when for executing the program stored on memory, realizes the method step described in any one of claim 1-7
Suddenly.
10. a kind of readable storage medium storing program for executing, which is characterized in that be stored with computer program, the meter in the readable storage medium storing program for executing
The method and step described in any one of claim 1-7 is realized when calculation machine program is executed by processor.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810276790.4A CN108510281B (en) | 2018-03-30 | 2018-03-30 | Data detection method and device, electronic equipment and storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810276790.4A CN108510281B (en) | 2018-03-30 | 2018-03-30 | Data detection method and device, electronic equipment and storage medium |
Publications (2)
Publication Number | Publication Date |
---|---|
CN108510281A true CN108510281A (en) | 2018-09-07 |
CN108510281B CN108510281B (en) | 2021-05-28 |
Family
ID=63379377
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810276790.4A Expired - Fee Related CN108510281B (en) | 2018-03-30 | 2018-03-30 | Data detection method and device, electronic equipment and storage medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108510281B (en) |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR20150061541A (en) * | 2013-11-26 | 2015-06-04 | 주식회사 씽크풀 | Providing method and system for preventing fraud trading |
CN106447333A (en) * | 2016-11-29 | 2017-02-22 | 中国银联股份有限公司 | Fraudulent trading detection method and server |
CN107085812A (en) * | 2016-12-06 | 2017-08-22 | 雷盈企业管理(上海)有限公司 | The anti money washing system and method for block chain digital asset |
-
2018
- 2018-03-30 CN CN201810276790.4A patent/CN108510281B/en not_active Expired - Fee Related
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR20150061541A (en) * | 2013-11-26 | 2015-06-04 | 주식회사 씽크풀 | Providing method and system for preventing fraud trading |
CN106447333A (en) * | 2016-11-29 | 2017-02-22 | 中国银联股份有限公司 | Fraudulent trading detection method and server |
CN107085812A (en) * | 2016-12-06 | 2017-08-22 | 雷盈企业管理(上海)有限公司 | The anti money washing system and method for block chain digital asset |
Also Published As
Publication number | Publication date |
---|---|
CN108510281B (en) | 2021-05-28 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Li et al. | The technology and economic determinants of cryptocurrency exchange rates: The case of Bitcoin | |
Kwon et al. | Time series classification of cryptocurrency price trend based on a recurrent LSTM neural network | |
CN104915879B (en) | The method and device that social relationships based on finance data are excavated | |
Fan et al. | The elements of financial econometrics | |
CN109598331A (en) | A kind of fraud identification model training method, fraud recognition methods and device | |
CN108734338A (en) | Credit risk forecast method and device based on LSTM models | |
CN108564386A (en) | Trade company's recognition methods and device, computer equipment and storage medium | |
Yang et al. | Retailer’s optimal order and credit policies when a supplier offers either a cash discount or a delay payment linked to order quantity | |
CN103400446A (en) | System and method capable of enabling cigarette vending machine to recognize adult identity of customer | |
CN108765144A (en) | Processing method, terminal device and the medium of e-sourcing packet | |
JP2024512076A (en) | Asset class backed tokenization platform | |
CN107798457A (en) | Investment combination proposal recommending method, device, computer equipment and storage medium | |
Chen et al. | RETRACTED ARTICLE: An intelligent trading mechanism based on the group trading strategy portfolio to reduce massive loss by the grouping genetic algorithm | |
CN110033165A (en) | The recommended method of overdue loaning bill collection mode, device, medium, electronic equipment | |
CN112598311A (en) | Risk operation identification model construction method and risk operation identification method | |
CN108229964A (en) | Trading activity profile is built and authentication method, system, medium and equipment | |
CN108510281A (en) | Data detection method and device, electronic equipment and storage medium | |
CN110598918A (en) | Aquatic product price prediction method and device and computer storage medium | |
CN111144899A (en) | Method and device for identifying false transactions and electronic equipment | |
CN115601034A (en) | Attack detection method for decentralized finance | |
JP5433947B2 (en) | Electronic wallet system and payment processing method | |
Xie et al. | [Retracted] Financial Default Risk Prediction Algorithm Based on Neural Network under the Background of Big Data | |
Sinha et al. | Consumers' Attitude and Perception towards Doing Cashless Transactions: An Empirical Study in Vadodara. | |
Wamwea et al. | A Multicurve Cross‐Currency LIBOR Market Model | |
Yang | Security Evaluation of Financial and Insurance and Ruin Probability Analysis Integrating Deep Learning Models |
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 | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20210528 |
|
CF01 | Termination of patent right due to non-payment of annual fee |