CN107967530A - Channel of disbursement based on data analysis elects method and its system - Google Patents
Channel of disbursement based on data analysis elects method and its system Download PDFInfo
- Publication number
- CN107967530A CN107967530A CN201710481257.7A CN201710481257A CN107967530A CN 107967530 A CN107967530 A CN 107967530A CN 201710481257 A CN201710481257 A CN 201710481257A CN 107967530 A CN107967530 A CN 107967530A
- Authority
- CN
- China
- Prior art keywords
- channel
- channels
- dimension
- success rate
- feasible
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; 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
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2415—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
- G06F18/24155—Bayesian classification
-
- 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
Abstract
The present invention relates to the channel of disbursement based on data analysis to elect method and its system, and this method includes the increment order data for obtaining magnanimity；Order data is analyzed, obtains the weighting success rate of each dimension weighted score of all channels of order data and the different channels of each dimension combination；Order accuracy matching is carried out according to weighting success rate for new order data, elects optimal channel.The present invention is by analyzing increment order data, obtain the weighting success rate of the different channels of each dimension of order data, for new order data, obtain channel, channel is obtained according to specific dimension and pays success rate, elect the highest channel of success rate, the historical data of magnanimity is analyzed in realization, analyzed for Operation Decision, for different provinces, analyze the order volume and success rate of different operators difference channel, ranked, make policymaker clear, select the channel of maximum success rate, solve the problems, such as low success rate of, improve the efficiency for electing channel and decision-making.
Description
Technical field
The present invention relates to channel of disbursement to elect method, more specifically refers to the channel of disbursement side of electing based on data analysis
Method and its system.
Background technology
Telecom operators pay the upstream operator channel that industry has accessed magnanimity in payment platform, there is provided charging support,
But the factors such as operator's air control, each operator's channel air control, each canalization quality and the limitation of each canalization are constrained to,
The success rate of pay invoice can be a greater impact, channel access and relatively low Order success rate facing to magnanimity, business
Personnel are more difficult to make a policy, for the province and channel for needing focus development, and how to make for various channels elect it is excellent
First level.
At present, for paying industry, also have by data analysis to recommend channel of disbursement, still, current number
It is this according to the main means of payment for being to make analysis for user's payment behavior, recommending its hobby according to analysis result of analysis
Mode, which is often limited to cause because of some special orders to pay unsuccessful phenomenon, to be existed, can not in time from order in itself
The dimensions of numerous channels analyzed, the success rate for pushing successfully channel is quite low.
Chinese patent 201510057059.9 discloses a kind of settlement method and settlement system for supporting different channel of disbursement.
Specific steps include：S1, by different channel of disbursement accept order, and reconciliation file is generated according to sequence information, and by reconciliation text
Part uploads to corresponding storage server；The storage that S2, reconciliation Documentation Processing Center are linked to from order channel of disbursement
Server downloads reconciliation file, and the reconciliation file of download is parsed, and the clearing basic data being resolved to is imported and is settled accounts
Center；The transaction risk of magnanimity order and be hidden among magnanimity order that S3, settlement center are related to clearing basic data
Trade company's commercial activity risk assessed；Such as the transaction risk assessment and the commercial activity of each trade company of each order
Risk assessment is passed through, then performs step S4, otherwise, performs step S5；S4, be related to all of same trade company in the reconciliation phase
Order carries out unified clearing, generates the first advice of settlement, and inter bank transfer operation is performed according to the first advice of settlement；S5, hand over existing
The order that the order of easy risk and/or commercial activity dishonour involved by trade company carries out manual examination and verification, to through manual examination and verification and being related to
All orders of same trade company carry out unified clearing, generate the second advice of settlement, and inter bank transfer behaviour is performed according to the second advice of settlement
Make.
Above-mentioned patent pushing efficiency is not high, and is analyzed according to reconciliation file, can not significantly be directed to different provinces
Part analyzes the order volume and success rate of different operators difference channel.
Therefore, it is necessary to which designing a kind of channel of disbursement based on data analysis elects method, realize the history number of magnanimity
According to being analyzed, analyzed for Operation Decision, for different provinces, analyze different operators difference channel order volume and into
Power, is ranked, and makes policymaker clear, selects the channel of maximum success rate, solve the problems, such as it is low success rate of, and improve elect
Channel and the efficiency of decision-making.
The content of the invention
The defects of it is an object of the invention to overcome the prior art, there is provided the channel of disbursement based on data analysis elects method
And its system.
To achieve the above object, the present invention uses following technical scheme：Channel of disbursement based on data analysis elects method,
The described method includes：
Obtain the increment order data of magnanimity；
The order data is analyzed, obtains each dimension weighted score of all channels of the order data and each
The weighting success rate of the different channels of dimension combination；
Order accuracy matching is carried out according to weighting success rate for new order data, elects optimal channel.
Its further technical solution is：The step of obtaining the increment order data of magnanimity, including step in detail below：
The order data of increment is obtained in setting time；
Store the order data of increment；
The magnanimity increment order data in certain period of time is obtained from the order data of storage.
Its further technical solution is：Analyze the order data, obtain the order data all channels it is each
The step of weighting success rate of dimension weighted score and the different channels of each dimension combination, including step in detail below：
Classify to all channels in the order data；
According to classification as a result, carrying out Bayesian analysis to every a kind of channel according to different dimensions；
According to Bayesian analysis as a result, each dimension of every a kind of channel is weighted, obtain per a kind of channel
Each dimension weighted score；
Arranged according to dimension combination, obtain dimension combination；
According to each dimension weighted score of every a kind of channel, the weighting success rate of the different channels of each dimension is obtained；
The weighting success rate of the different channels of each dimension is ranked up.
Its further technical solution is：Order accuracy matching is carried out according to weighting success rate for new order data,
The step of electing optimal channel, including step in detail below：
Obtain all available channels of new order data；
Rejecting, which reaches described, to use the channel for meeting air control requirement in channel, obtain feasible channel；
Judge whether feasible channel has corresponding dimension；
If so, obtaining the weighting success rate of all feasible channels under the corresponding dimension combination of new order data, push adds
Weigh the feasible channel of success rate maximum；
If no, channel is pushed according to the conventional payment behavior of user.
Its further technical solution is：Rejecting, which reaches described, can use the channel for meeting air control requirement in channel, obtain feasible
The step of channel, including step in detail below：
Reject all channels for the classification for meeting channel classification air control requirement；
The channel for reaching channel air control in remaining available channel is rejected, obtains feasible channel.
Its further technical solution is：Judge whether feasible channel has the step of corresponding dimension, including in detail below
Step：
Judge whether all IMSI channels in feasible channel have corresponding dimension；
If all IMSI channels in feasible channel have corresponding dimension, feasible channel has corresponding dimension；
If all IMSI channels in feasible channel do not have corresponding dimension, other canals in all feasible channels are judged
Whether road has corresponding dimension；
If other channels in all feasible channels have corresponding dimension, feasible channel has corresponding dimension；
If other channels in all feasible channels do not have corresponding dimension, feasible channel does not have corresponding dimension.
Its further technical solution is：If so, obtain all feasible channels under the corresponding dimension combination of new order data
Weighting success rate, push the step of being weighted to prominent feasible channel, including step in detail below：
Obtain the weighting success rate of all feasible channels under the corresponding dimension combination of new order data；
Acquisition is weighted to prominent feasible channel；
Whether the number for judging to be weighted to prominent feasible channel is one；
If so, then push weights the feasible channel corresponding to success rate maximum；
If it is not, then randomly select a feasible channel in the feasible channel corresponding to weighting success rate maximum and go forward side by side
Row push.
Its further technical solution is：If no, the step of payment behavior push channel conventional according to user, including with
Lower specific steps：
Whether judge cell-phone number/IMSI of order has channel not have failure record in setting time；
Whether if there is channel not have failure record, judge cell-phone number/IMSI of order has channel to have into setting time
Work(records；
Successfully recorded if no channel has, rejecting the cell-phone number/IMSI has failure record in setting time and have into
The channel of work(record, rejects the channel of cell-phone number/IMSI channel classifications of success order in setting time, when remaining
When channel has various channels, using remaining channel as channel can be pushed, when remaining channel only has one, then by all canals
Road conduct can push channel；
If thering is channel to have successfully to record, using all channels as channel can be pushed；
If no channel does not have failure record, judge all channels whether all for IMSI channels；
If all channels are all to be IMSI channels, when the trade company corresponding to order only uses IMSI channels, by all canals
Road conduct can push channel, and when the trade company corresponding to order is not only to use IMSI channels, then output channel is elected unsuccessfully；
If all channels are not all to be IMSI channels, using all channels as channel can be pushed；
Judge whether the number that can push channel is more than three；
If so, then randomly selecting one of them in first three channel that can be pushed in channel, and push the channel of extraction；
If it is not, then randomly selecting one of them all pushed in channel, and push the channel of extraction.
Elect system present invention also offers the channel of disbursement based on data analysis, including mass data acquiring unit, into
Power obtaining unit and elect unit；
The mass data acquiring unit, for obtaining the increment order data of magnanimity；
The success rate acquiring unit, for analyzing the order data, obtains all channels of the order data
The weighting success rate of each dimension weighted score and the different channels of each dimension combination；
It is described to elect unit, for carrying out order accuracy matching according to weighting success rate for new order data, push away
Select optimal channel.
Its further technical solution is：The success rate acquiring unit includes sort module, Bayesian analysis module, weighting
Computing module, sorting module, be weighted to power acquisition module and sorting module；
The sort module, for classifying to all channels in the order data；
The Bayesian analysis module, for according to classification as a result, carrying out shellfish to every a kind of channel according to different dimensions
Ye Si is analyzed；
The weighted calculation module, for according to Bayesian analysis as a result, adding to each dimension of every a kind of channel
Power calculates, and obtains each dimension weighted score per a kind of channel；
The sorting module, for being arranged according to dimension combination, obtains dimension combination；
It is described to be weighted to power acquisition module, for according to each dimension weighted score per a kind of channel, obtaining each
The weighting success rate of the different channels of dimension；
The sorting module, the weighting success rate for the different channels to each dimension are ranked up.
Compared with the prior art, the invention has the advantages that：The channel of disbursement side of electing based on data analysis of the present invention
Method, by being analyzed for the increment order data in setting time, obtains the different channels of each dimension of the order data
Weighting success rate, for new order data, after first obtaining channel, further according to specific after rejecting processing is carried out to channel
Dimension, obtains channel and pays success rate, elect the highest channel of success rate, realize and analyzed the historical data of magnanimity, can
To be analyzed for Operation Decision, for different provinces, the order volume and success rate of different operators difference channel are analyzed, is carried out
Seniority among brothers and sisters, make policymaker clear, select the channel of maximum success rate, solve the problems, such as it is low success rate of, and improve elect channel and
The efficiency of decision-making.
The invention will be further described with specific embodiment below in conjunction with the accompanying drawings.
Brief description of the drawings
Fig. 1 is the flow chart that the channel of disbursement based on data analysis that the specific embodiment of the invention provides elects method；
Fig. 2 is the particular flow sheet of the increment order data for the acquisition magnanimity that the specific embodiment of the invention provides；
Fig. 3 is the idiographic flow of each dimension weighted score for the acquisition order data that the specific embodiment of the invention provides
Figure；
Fig. 4 is the specific frame diagram for each dimension weighted score of calculating that the specific embodiment of the invention provides；
Fig. 5 is the particular flow sheet for electing optimal channel that the specific embodiment of the invention provides；
Fig. 6 is the particular flow sheet for the feasible channel of acquisition that the specific embodiment of the invention provides；
Fig. 7 is the particular flow sheet of the channel for the push weighted score maximum that the specific embodiment of the invention provides；
Fig. 8 is the structure diagram that the channel of disbursement based on data analysis that the specific embodiment of the invention provides elects system；
Fig. 9 is the practical application that the channel of disbursement based on data analysis that the specific embodiment of the invention provides elects system
Organization Chart；
Figure 10 is the success that channel of disbursement of the use based on data analysis that the specific embodiment of the invention provides elects method
Rate compares figure one；
Figure 11 is the success that channel of disbursement of the use based on data analysis that the specific embodiment of the invention provides elects method
Rate compares figure two；
Figure 12 is that channel of disbursement of the use based on data analysis that the specific embodiment of the invention provides elects method for not
With the success rate distribution map in province；
Figure 13 is that channel of disbursement of the use based on data analysis that the specific embodiment of the invention provides elects method to be directed to four
The success rate distribution form of Chuan Sheng；
Figure 14 is that channel of disbursement of the use based on data analysis that the specific embodiment of the invention provides elects method to be directed to four
The success rate distribution form of river province difference channel.
Embodiment
In order to more fully understand the present invention technology contents, with reference to specific embodiment to technical scheme into
One step introduction and explanation, but not limited to this.
Specific embodiment as shown in Fig. 1~14, the channel of disbursement side of electing provided in this embodiment based on data analysis
Method, can be used in the payment process of any payment platform.
As shown in Figure 1, present embodiments providing the channel of disbursement based on data analysis elects method, this method includes：
S1, the increment order data for obtaining magnanimity；
S2, the analysis order data, obtain all channels of the order data each dimension weighted score and
The weighting success rate of the different channels of each dimension combination；
S3, for new order data according to weighting success rate carry out order accuracy matching, elect optimal channel.
For above-mentioned S1 steps, the step of obtaining the increment order data of magnanimity, including step in detail below：
The order data of increment is obtained in S11, setting time；
S12, the order data for storing increment；
S13, the magnanimity increment order data from the order data of storage in acquisition certain period of time.
Above-mentioned S11 steps, are that client database extracts the previous day backward at least one time daily by the payment platform of front end
The order data of increment, and by internet HTTP protocol interaction technique, md5 encryption is carried out using irreversible HASH algorithms
Data be transmitted and push to mass data platform, magnanimity order data is obtained with this；Encryption format：MD5
(userAccount+userPwd+ file name .txt) .toLowerCase (), then base64 encryptions.In the process,
Result data mode is pulled according to trade company and obtains data, in access, it is necessary to which the url addresses of active obtaining data, are actively obtained
Access according to when, it is necessary to submit encryption data.
The initial data of big data analysis is derived from the form ordering system of small amount payment, and data are analyzed once daily, for
The History Order data of nearly half a year, at the appointed time import order data carry out big data analysis daily.Increment order data
Refer to and data of the small amount payment system for the previous day of current time are only directed into big data analysis system daily, be not required to
Will all data all import because the data before one day have existed.The time of system introducing is so saved, order data is all
Original order data.
For above-mentioned S12 steps, specifically order data is stored by the way of HDFS, HDFS is
The document storage system that Haddoop is carried, can be uploaded to HAFS, file divides on HAFS in Hadoop clients by instruction
Catalogue is stored, and realizes distributed storage, when data volume increase, direct heap adds machine, without any code development.
For above-mentioned S13 steps, in the present embodiment, the magnanimity increment order numbers in the previous moon are mainly obtained
According to, the data in nearest a period of time are analyzed, it is more representative.
As shown in figure 9, in above-mentioned data acquisition, request method：post；Its parameter declaration is as follows：
Request response：{"code":"0","msg":" there are for Test.txt files " }；Its parameter declaration is as follows：
Result data after data analysis is pushed in the loopback address of trade company's offer, and trade company receives file data.
Further, above-mentioned S2 steps, analyze the order data, obtain all channels of the order data
The step of weighting success rate of each dimension weighted score and the different channels of each dimension combination, including walk in detail below
Suddenly：
S21, classify all channels in the order data；
S22, according to classification as a result, carrying out Bayesian analysis to every a kind of channel according to different dimensions；
S23, according to Bayesian analysis as a result, each dimension of every a kind of channel is weighted, obtain per a kind of
Each dimension weighted score of channel；
S24, arranged according to dimension combination, obtains dimension combination；
S25, each dimension weighted score according to every a kind of channel, obtain the weighting success of the different channels of each dimension
Rate；
S26, the weighting success rate to the different channels of each dimension are ranked up.
For above-mentioned S2 steps, weighted score is specifically calculated by the way of Mapreduce, changes Mapreduce and exists
The upper Distributed Calculations of Hadoop, first carry out map, i.e. separate computations, then carry out reduce, i.e., are counted result of calculation defeated
Go out.Data analysis is built with Hadoop, and has done secondary development on this basis, including is uniformly accessed into data-interface, data point
Analysis monitoring, and the monitoring of data conversion ratio.Data results at the same time, are returned by the interface of secondary development.
For above-mentioned S21 steps, due to regional, lower single time of each order, face amount, air control size attribute not
Together, therefore all channels have each different dimensions in each order data, are classified to channel and calculated again, easy to follow-up
The channel dimension given priority to needs makes analysis.
Above-mentioned S22 steps, S23 steps, S24 steps, S25 steps and S26 steps, makes full use of and is calculated using Bayes
Method, decision tree, logic analysis, association rules mining algorithm, Sequential Pattern Mining Algorithm are inferred, to the data of payment process
Analyzed and calculated, be conducive to that new order is accomplished precisely to predict, and the channel dimension given priority to needs is made
Analysis, solves the problems, such as low success rate of.
For above-mentioned S22 steps, realize that success rate prejudges using bayesian algorithm for each dimension, Bayes' theorem
Claim Bayesian inference, early in 18th century, British scholar Bayes (1702~1763) it is proposed that the formula of design conditions probability is used
To solve the problems, such as following one kind：Assuming that H [1], H [2] ..., one complete event of H [n] mutual exclusions and composition, it is known that their probability P
(H [i]), i=1,2 ..., n, now observe that certain event A and H [1], and H [2] ..., H [n] are random together to be occurred, and known conditions is general
Rate P (A/H [i]), seeks P (H [i]/A).In every a kind of channel, bayesian algorithm analysis is carried out according to different dimensions, is specifically
Analyzed for order dimension, including the dimension such as province, face amount, time, prejudge per class channel under different dimensions into
Power.Data analysis makes weighted analysis on this basis based on bayesian algorithm, calculates the success rate to dimension, together
When by continuously attempting to, setting weighting change value rule, in units of day, system sets daily weighted value automatically, passes through weighting
The channel that value carries out new order is recommended.
For above-mentioned S23 steps, according to after above-mentioned Bayesian analysis as a result, every class channel can be obtained in difference
Dimension under success rate, in the weight according to shared by each dimension, be weighted, so as to obtain each of every a kind of channel
A dimension weighted score.
For above-mentioned S24 steps, dimension combination is obtained using the mode of combination, it is each under same dimension in order to obtain
The distribution situation of channel.
For above-mentioned S25 steps, the weighting success rate of the different channels of the extremely each dimension of above-mentioned weighted score, this
In be to start with obtaining optimal channel in terms of dimension for the ease of new order.
For above-mentioned S26 steps, specifically by the weighting success rate of the different channels of each dimension according to from high to low
Order sequence, easy to follow-up new order elect channel condition meet weighting success rate elect mode when, quickly learn plus
The high channel of success rate is weighed, and is pushed, improves push success rate and efficiency.
Certainly, in other embodiment, the channel after sequence and dimension can also be carried out regular.
As shown in figure 4, for regional, lower single time of every order, face amount, air control size attribute, it is divided into variant dimension
Degree, while each channel is directed to, analyzed with reference to the success or failure of order, analyze the group of each dimension of each channel
Corresponding weighted score is closed, and by weighting algorithm, calculates weighting success of each channel under the combination of each dimension
Rate.
Further, for above-mentioned S3 steps, order standard is carried out according to weighting success rate for new order data
The step of true property matches, elects optimal channel, including step in detail below：
S31, all available channels for obtaining new order data；
S32, rejecting, which reach described, to use the channel for meeting air control requirement in channel, obtain feasible channel；
S33, judge whether feasible channel has corresponding dimension；
S34, if so, obtaining the weighting success rate of all feasible channels under the combination of new order data corresponding dimension, push away
Send and be weighted to prominent feasible channel；
If S35, do not have, channel is pushed according to the conventional payment behavior of user.
For above-mentioned S31 steps, all available channels of new order data are obtained, then are selected from available channel optimal
Channel is elected, and can avoid the occurrence of omission.For above-mentioned S32 steps, specifically meet in order to avoid can use in channel
The channel of air control requirement is picked as risk occur during optimal channel.It is in order to feasible channel pair for above-mentioned S33 steps
The dimension answered, which whether there is, to be judged, to judge whether optimal channel can be pushed from data results.For above-mentioned S34
Step, in the case that data results can push optimal channel, push is weighted to prominent feasible channel, to improve
Pay success rate.For above-mentioned S35 steps, in the case where data results do not elect optimal channel, then original is carried out
There is logic to carry out electing optimal channel.
Original logic and data analysis are combined together, payment success rate can be improved, operation can also be directed to
Analysis of Policy Making, for different provinces, analyzes the order volume and success rate of different operators difference channel, is ranked, make to determine
Plan person is clear, selects the channel of maximum success rate, solves the problems, such as low success rate of, and improves and elects the effect of channel and decision-making
Rate.
Above-mentioned S31 steps are to S35 steps, primarily directed to recommending after data analysis as a result, reverse conversion, analyzes
Under the combination of each dimension, the chance of success score value sequence of all channels is recorded, the high channel of push chance of success score value is optimal
Channel, and pushed, interacted using the channel of push.
As shown in Figure 10 and Figure 11, the analysis result recommended for channel, is divided into two parts, one of top by daily order
Curve is the order for according to data results recommend channel, a curve of top for not by data results into
Row recommends the order of channel.Test 90 days is carried out, daily success rate is all the height of mass data, mean height 5% or so.Separately
Outside, as shown in Figure 12 to Figure 14, the order volume and success rate of the different channels of the different operators in different provinces can consult,
It can be seen from the above that the channel of disbursement based on data analysis elects method that policymaker can be made clear, the channel of maximum success rate is selected, is solved
Certainly low success rate of problem, and improve the efficiency for electing channel and decision-making.
Further, for above-mentioned S32 steps, rejecting, which reaches described, can use the canal for meeting air control requirement in channel
Road, the step of obtaining feasible channel, including step in detail below：
S321, rejecting meet all channels of the classification of channel classification air control requirement；
S322, reject the channel for reaching channel air control in remaining available channel, obtains feasible channel.
Certain undesirable a kind of or some classes channel first are rejected from big classification, then from the channel of remaining classification
Undesirable channel is picked out, is rejected, you can gets all feasible channels, that is, meets the requirements, do not deposit
Meeting the channel of air control requirement.
Further, for above-mentioned S33 steps, judge whether feasible channel has the step of corresponding dimension, wrap
Include step in detail below：
S331, judge whether all IMSI channels in feasible channel have corresponding dimension；
If all IMSI channels in feasible channel have corresponding dimension, S332, feasible channel have corresponding dimension；
If all IMSI channels in S333, feasible channel do not have corresponding dimension, judge in all feasible channels
Whether other channels have corresponding dimension；
If other channels in all feasible channels have corresponding dimension, S332, feasible channel have corresponding dimension；
If other channels in S334, all feasible channels do not have corresponding dimension, feasible channel is not corresponding
Dimension.
By first obtaining channel, reversely judging whether channel has corresponding dimension, if so, then having in feasible channel opposite
Result after the dimension answered, i.e. data analysis can elect optimal channel.
Further, above-mentioned S34 steps, if so, obtain under the combination of new order data corresponding dimension it is all can
The step of weighting success rate of row channel, push is weighted to prominent feasible channel, including step in detail below：
The weighting success rate of all feasible channels under the corresponding dimension combination of S341, acquisition new order data；
S342, acquisition are weighted to prominent feasible channel；
Whether the number that S343, judgement are weighted to prominent feasible channel is one；
S344, if so, then push weighting success rate maximum corresponding to feasible channel；
S345, if it is not, then weighting success rate maximum corresponding to feasible channel in, randomly select a feasible channel
And pushed.
Above-mentioned S341 steps, are the weighting success rates of all feasible channels of acquisition first under the combination of specific dimension,
It can obtain in a certain province, sometime or under the dimension of a certain face amount, the feasible respective success rate of channel.
Above-mentioned S342 steps, the bigger channel of weighting success rate, the probability that success is paid are higher.
Above-mentioned S343 steps to S345 steps, judgement be weighted to prominent feasible channel number whether be it is multiple,
If multiple, need to randomly select one of them and pushed, if only one, directly push.
Further, above-mentioned S35 steps, if not having, the step of according to user's conventional payment behavior push channel,
Including step in detail below：
S351, judge whether cell-phone number/IMSI of order has channel not have failure record in setting time；
Whether if S352, have channel not have failure record, judge cell-phone number/IMSI of order has channel in setting time
Have and successfully record；
Successfully recorded if S353, no channel have, reject the cell-phone number/IMSI have in setting time failure record and
There is the channel successfully recorded, the channel of cell-phone number/IMSI channel classifications of success order in setting time is rejected, when surplus
Under channel when having various channels, using remaining channel as channel can be pushed, when remaining channel only has one, then by institute
There is channel conduct to push channel, and enter S357 steps；
If thering is channel to have successfully to record, S354, using all channels as can push channel, and enter S357 steps；
If S355, no channel do not have failure record, judge all channels whether all for IMSI channels；
If S356, all channels are all to be IMSI channels, when the trade company corresponding to order only uses IMSI channels, by institute
There is channel conduct to push channel, and enter S357 steps, when the trade company corresponding to order is not only to use IMSI channels, then export
Channel is elected unsuccessfully；
If it is all IMSI channels that all channels, which are not, S354, using all channels as channel can be pushed, and enter S357
Step；
Whether the number that S357, judgement can push channel is more than three；
S358, if so, then randomly select one of them in first three channel that can be pushed in channel, and push extraction
Channel；
S359, if it is not, then randomly select one of them all pushed in channel, and push the channel of extraction.
Above-mentioned S351 to S359 steps, be when data results do not push any channel, then use with
Past logic is pushed, that is, the custom used in the past by analyzing user is pushed.
The technology that channel is pushed after data analysis and the technology of original logic push channel are combined together, to improve
The success rate that channel is paid.
The above-mentioned channel of disbursement based on data analysis elects method, by for the increment order data in setting time
Analyzed, obtain the weighting success rate of the different channels of each dimension of the order data, for new order data, first obtain
After channel, further according to specific dimension after rejecting processing is carried out to channel, obtain channel and pay success rate, elect success rate most
High channel, realizes and is analyzed the historical data of magnanimity, can be directed to Operation Decision and analyze, for different provinces, analysis
Go out the order volume and success rate of different operators difference channel, ranked, make policymaker clear, select the canal of maximum success rate
Road, solves the problems, such as low success rate of, and improves and elects the efficiency of channel and decision-making.
As shown in figure 8, the present embodiment, which additionally provides the channel of disbursement based on data analysis, elects system, it includes magnanimity number
According to acquiring unit 1, success rate acquiring unit 2 and elect unit 3.
Mass data acquiring unit 1, for obtaining the increment order data of magnanimity.
Success rate acquiring unit 2, for analyzing the order data, obtain the order data all channels it is each
The weighting success rate of dimension weighted score and the different channels of each dimension combination.
Unit 3 is elected, for carrying out order accuracy matching according to weighting success rate for new order data, is elected most
Excellent channel.
Include data acquisition module, memory module and abstraction module for above-mentioned mass data acquiring unit 1.
Acquisition module, for obtaining the order data of increment in setting time；It is specifically daily by the payment platform of front end
At least once backward client database extract the previous day increment order data, and by internet HTTP protocol interaction technique,
It is transmitted using the data of irreversible HASH algorithms progress md5 encryption and pushes to mass data platform, is obtained with this
Magnanimity order data.
Memory module, for storing the order data of increment；Specifically order data is deposited by the way of HDFS
Storage, HDFS is the document storage system that Haddoop is carried, and HAFS, file can be uploaded to by instruction in Hadoop clients
Sectional lists is stored on HAFS, realizes distributed storage, when data volume increase, direct heap adds machine, without any generation
Code exploitation.
Abstraction module, it is main for obtaining the magnanimity increment order data in certain period of time from the order data of storage
If obtaining the magnanimity increment order data in the previous moon, the data in nearest a period of time are analyzed, relatively there is representative
Property.
Above-mentioned success rate acquiring unit 2 is specifically that weighted score is calculated by the way of Mapreduce, is changed
Mapreduce Distributed Calculations on Hadoop, first carry out map, i.e. separate computations；Reduce is carried out again, i.e., by result of calculation
Counted and exported.
Further, above-mentioned success rate acquiring unit 2 includes sort module, Bayesian analysis module, weighted calculation
Module, sorting module, be weighted to power acquisition module and sorting module.
Sort module, for classifying to all channels in the order data.Due to the area of each order, place an order
Time, face amount, the difference of air control size attribute, therefore all channels have each different dimensions in each order data, it is right
Channel is classified to be calculated again, and the channel dimension easy to subsequently be given priority to needs makes analysis.
Above-mentioned Bayesian analysis module, weighted calculation module, sorting module, be weighted to power acquisition module and sequence
Module is made full use of using bayesian algorithm, decision tree, logic analysis, association rules mining algorithm, Sequential Pattern Mining Algorithm
Inferred, the data of payment process are analyzed and calculated, be conducive to that new order is accomplished precisely to predict, and to needing
The channel dimension to be given priority to makes analysis, solves the problems, such as low success rate of.
Bayesian analysis module, for according to classification as a result, carrying out Bayes to every a kind of channel according to different dimensions
Analysis.Realize that success rate prejudges using bayesian algorithm for each dimension, Bayes' theorem is also referred to as Bayesian inference, early in 18 generation
Record, British scholar Bayes (1702~1763) is it is proposed that the formula of design conditions probability is used for solving the problems, such as following one kind：It is false
If H [1], H [2] ..., one complete event of H [n] mutual exclusions and composition, it is known that their probability P (H [i]), i=1,2 ..., n are existing
It was observed that certain event A and H [1], H [2] ..., H [n] are random together to be occurred, and known conditions probability P (A/H [i]), ask P (H [i]/
A).In every a kind of channel, bayesian algorithm analysis is carried out according to different dimensions, particularly directed to order dimension, including province,
The dimensions such as face amount, time are analyzed, and are prejudged per success rate of the class channel under different dimensions.
Weighted calculation module, by according to Bayesian analysis as a result, based on being weighted to each dimension of every a kind of channel
Calculate, obtain each dimension weighted score per a kind of channel.According to after above-mentioned Bayesian analysis as a result, every class can be obtained
Success rate of the channel under different dimensions, in the weight according to shared by each dimension, is weighted, so as to obtain each
Each dimension weighted score of class channel.
Sorting module, for being arranged according to dimension combination, obtains dimension combination.Obtained using the mode of combination
Dimension combines, in order to obtain the distribution situation of each channel under same dimension.
Power acquisition module is weighted to, for according to each dimension weighted score per a kind of channel, obtaining each dimension
Different channels weighting success rate.The weighting success rate of the different channels of the extremely each dimension of above-mentioned weighted score, here
It is to start with obtaining optimal channel in terms of dimension for the ease of new order.
Sorting module, the weighting success rate for the different channels to each dimension are ranked up.Specifically by each dimension
The weighting success rate of the different channels of degree sorts according to order from high to low, easy to the condition for electing channel of follow-up new order
When meeting that weighting success rate elects mode, the high channel of weighting success rate is quickly learnt, and pushed, improve push success rate
And efficiency.
Certainly, in other embodiment, the channel after sequence and dimension can also be carried out regular.
For regional, lower single time of every order, face amount, air control size attribute, it is divided into variant dimension, is directed at the same time
Each channel, is analyzed with reference to the success or failure of order, analyze each dimension of each channel combination it is corresponding plus
Score value is weighed, and by weighting algorithm, calculates weighting success rate of each channel under the combination of each dimension.
Further, it is above-mentioned to elect unit 3 to include available channel acquisition module, feasible channel acquisition module, dimension
Judgment module, analysis pushing module and existing channel pushing module.
Channel acquisition module can be used, for obtaining all available channels of new order data.
Feasible channel acquisition module, for reject reach it is described can with the channel for meeting air control requirement in channel, acquisition
Row channel.Specifically meet that the channel of air control requirement risk occurs when being picked as optimal channel in channel in order to avoid can use
Dimension judgment module, for judging whether feasible channel has corresponding dimension.In order to corresponding to feasible channel
Dimension, which whether there is, to be judged, to judge whether optimal channel can be pushed from data results
Pushing module is analyzed, for if so, obtaining all feasible channels under the corresponding dimension combination of new order data
Success rate is weighted, push is weighted to prominent feasible channel.In the case that data results can push optimal channel,
Push is weighted to prominent feasible channel, to improve payment success rate.
Existing channel pushing module, if for not having, channel is pushed according to the conventional payment behavior of user.In data analysis
As a result in the case of not electing optimal channel, then carry out original logic and carry out electing optimal channel.
Original logic and data analysis are combined together, payment success rate can be improved, operation can also be directed to
Analysis of Policy Making, for different provinces, analyzes the order volume and success rate of different operators difference channel, is ranked, make to determine
Plan person is clear, selects the channel of maximum success rate, solves the problems, such as low success rate of, and improves and elects the effect of channel and decision-making
Rate.
For recommending after data analysis as a result, reverse conversion, analyzes under the combination of each dimension, record all channels
Chance of success score value sorts, and the high channel of push chance of success score value is optimal channel, and is pushed.
As shown in Figure 10 and Figure 11, the analysis result recommended for channel, is divided into two parts, one of top by daily order
Curve is the order for according to data results recommend channel, a curve of top for not by data results into
Row recommends the order of channel.Test 90 days is carried out, daily success rate is all the height of mass data, mean height 5% or so.Separately
Outside, as shown in Figure 12 to Figure 14, the order volume and success rate of the different channels of the different operators in different provinces can consult,
It can be seen from the above that the channel of disbursement based on data analysis elects method that policymaker can be made clear, the channel of maximum success rate is selected, is solved
Certainly low success rate of problem, and improve the efficiency for electing channel and decision-making.
Above-mentioned feasible channel acquisition module includes classification and rejects submodule and channel rejecting submodule.
Classification rejects submodule, for rejecting all channels for the classification for meeting channel classification air control requirement.
Channel rejects submodule, reaches the channel of channel air control in remaining available channel for rejecting, obtains feasible channel.
Certain undesirable a kind of or some classes channel first are rejected from big classification, then from the channel of remaining classification
Undesirable channel is picked out, is rejected, you can gets all feasible channels, that is, meets the requirements, do not deposit
Meeting the channel of air control requirement.
Above-mentioned dimension judgment module includes IMSI channels judging submodule and all channel judging submodules.IMSI canals
Road judging submodule is used to judge whether all IMSI channels in feasible channel have corresponding dimension, if the institute in feasible channel
There are IMSI channels to have corresponding dimension, then feasible channel has corresponding dimension.If all channel judging submodules are used for feasible
All IMSI channels in channel do not have corresponding dimension, then it is corresponding to judge whether other channels in all feasible channels have
Dimension, if other channels in all feasible channels have corresponding dimension, feasible channel has corresponding dimension, if all can
Other channels in row channel do not have corresponding dimension, then feasible channel does not have corresponding dimension.
By first obtaining channel, reversely judging whether channel has corresponding dimension, if so, then having in feasible channel opposite
Result after the dimension answered, i.e. data analysis can elect optimal channel.
Further, probability acquisition submodule is included for above-mentioned analysis pushing module, corresponding channel obtains submodule
Block and number judging submodule.
Probability acquisition submodule, for obtaining the weighting of all feasible channels under the corresponding dimension combination of new order data
Success rate.First under the combination of specific dimension, the weighting success rate of all feasible channels is obtained, can also be obtained in a certain province
Part, sometime or under the dimension of a certain face amount, the feasible respective success rate of channel.Corresponding channel acquisition submodule is used to obtain
Take and be weighted to prominent feasible channel.The bigger channel of success rate is weighted, the probability that success is paid is higher.Number judges
Whether the number that submodule is used to judge to be weighted to prominent feasible channel is one, if so, then push weighting success rate
Feasible channel corresponding to maximum；If it is not, then one is randomly selected in the feasible channel corresponding to weighting success rate maximum
Feasible channel is simultaneously pushed.
Further, above-mentioned existing channel pushing module includes failure record acquisition submodule, successfully record obtains
Submodule, IMSI judging submodules and channel number judging submodule.
Whether failure record acquisition submodule, cell-phone number/IMSI for judging order have channel not have in setting time
There is failure record.
Success records acquisition submodule, if for there is channel not have failure record, judges that cell-phone number/IMSI of order is being set
Whether there is channel to have in fixing time successfully to record, successfully record if no channel has, reject the cell-phone number/IMSI in setting
It is interior to have failure record and have the channel successfully recorded, reject the canal of cell-phone number/IMSI success orders in setting time
The channel of road classification, when remaining channel has various channels, using remaining channel as channel can be pushed, when remaining channel
At only one, then using all channels as channel can be pushed, successfully recorded if there is channel to have, using all channels as can push away
Send channel.
Whether IMSI judging submodules, if not having failure record for no channel, judge all channels all for IMSI
Channel；If all channels are all to be IMSI channels, when the trade company corresponding to order only uses IMSI channels, all channels are made
For that can push channel, when the trade company corresponding to order is not only to use IMSI channels, then output channel is elected unsuccessfully；If all channels
It is all IMSI channels not to be, then using all channels as can push channel.
Channel number judging submodule, for judging whether the number that can push channel is more than three；If so, then random take out
One of them in first three channel that can be pushed in channel is taken, and pushes the channel of extraction；If it is not, then randomly select all
One of them in channel can be pushed, and pushes the channel of extraction.
When data results do not push any channel, then pushed using conventional logic, that is,
The custom used in the past by analyzing user is pushed.The technology that channel is pushed after data analysis is pushed with original logic
The technology of channel is combined together, to improve the success rate of channel payment.
The above-mentioned channel of disbursement based on data analysis elects system, by for the increment order data in setting time
Analyzed, obtain the weighting success rate of the different channels of each dimension of the order data, for new order data, first obtain
After channel, further according to specific dimension after rejecting processing is carried out to channel, obtain channel and pay success rate, elect success rate most
High channel, realizes and is analyzed the historical data of magnanimity, can be directed to Operation Decision and analyze, for different provinces, analysis
Go out the order volume and success rate of different operators difference channel, ranked, make policymaker clear, select the canal of maximum success rate
Road, solves the problems, such as low success rate of, and improves and elects the efficiency of channel and decision-making.
The above-mentioned technology contents that the present invention is only further illustrated with embodiment, in order to which reader is easier to understand, but not
Represent embodiments of the present invention and be only limitted to this, any technology done according to the present invention extends or recreation, by the present invention's
Protection.Protection scope of the present invention is subject to claims.
Claims (10)
1. the channel of disbursement based on data analysis elects method, it is characterised in that the described method includes：
Obtain the increment order data of magnanimity；
The order data is analyzed, obtains each dimension weighted score of all channels of the order data and each dimension
The weighting success rate of the different channels of combination；
Order accuracy matching is carried out according to weighting success rate for new order data, elects optimal channel.
2. the channel of disbursement according to claim 1 based on data analysis elects method, it is characterised in that obtains magnanimity
The step of increment order data, including step in detail below：
The order data of increment is obtained in setting time；
Store the order data of increment；
The magnanimity increment order data in certain period of time is obtained from the order data of storage.
3. the channel of disbursement according to claim 2 based on data analysis elects method, it is characterised in that is ordered described in analysis
Forms data, obtains each dimension weighted score of all channels of the order data and the different channels of each dimension combination
Weighting success rate the step of, including step in detail below：
Classify to all channels in the order data；
According to classification as a result, carrying out Bayesian analysis to every a kind of channel according to different dimensions；
According to Bayesian analysis as a result, each dimension of every a kind of channel is weighted, obtain per each of a kind of channel
A dimension weighted score；
Arranged according to dimension combination, obtain dimension combination；
According to each dimension weighted score of every a kind of channel, the weighting success rate of the different channels of each dimension is obtained；
The weighting success rate of the different channels of each dimension is ranked up.
4. the channel of disbursement according to any one of claims 1 to 3 based on data analysis elects method, it is characterised in that
The step of carrying out order accuracy matching according to weighting success rate for new order data, elect optimal channel, including it is following
Specific steps：
Obtain all available channels of new order data；
Rejecting, which reaches described, to use the channel for meeting air control requirement in channel, obtain feasible channel；
Judge whether feasible channel has corresponding dimension；
If so, obtaining the weighting success rate of all feasible channels under the corresponding dimension combination of new order data, push is weighted to
Prominent feasible channel；
If no, channel is pushed according to the conventional payment behavior of user.
5. the channel of disbursement according to claim 4 based on data analysis elects method, it is characterised in that rejecting reaches institute
The step of stating the channel for meeting air control requirement in available channel, obtaining feasible channel, including step in detail below：
Reject all channels for the classification for meeting channel classification air control requirement；
The channel for reaching channel air control in remaining available channel is rejected, obtains feasible channel.
6. the channel of disbursement according to claim 4 based on data analysis elects method, it is characterised in that judges feasible canal
Whether road has the step of corresponding dimension, including step in detail below：
Judge whether all IMSI channels in feasible channel have corresponding dimension；
If all IMSI channels in feasible channel have corresponding dimension, feasible channel has corresponding dimension；
If all IMSI channels in feasible channel do not have corresponding dimension, judge that other channels in all feasible channels are
It is no to have corresponding dimension；
If other channels in all feasible channels have corresponding dimension, feasible channel has corresponding dimension；
If other channels in all feasible channels do not have corresponding dimension, feasible channel does not have corresponding dimension.
7. the channel of disbursement according to claim 4 based on data analysis elects method, it is characterised in that if so, obtaining
The weighting success rate of all feasible channels under the corresponding dimension combination of new order data, push is weighted to prominent feasible
The step of channel, including step in detail below：
Obtain the weighting success rate of all feasible channels under the corresponding dimension combination of new order data；
Acquisition is weighted to prominent feasible channel；
Whether the number for judging to be weighted to prominent feasible channel is one；
If so, then push weights the feasible channel corresponding to success rate maximum；
If it is not, then randomly select a feasible channel in the feasible channel corresponding to weighting success rate maximum and pushed away
Send.
8. the channel of disbursement according to claim 4 based on data analysis elects method, it is characterised in that if not having,
The step of payment behavior push channel conventional according to user, including step in detail below：
Whether judge cell-phone number/IMSI of order has channel not have failure record in setting time；
If there is channel there is no failure record, judge whether cell-phone number/IMSI of order has channel to have in setting time and successfully remember
Record；
Successfully recorded if no channel has, reject the cell-phone number/IMSI has failure record and having successfully to remember in setting time
The channel of record, rejects the channel of cell-phone number/IMSI channel classifications of success order in setting time, when remaining channel
When having various channels, using remaining channel as channel can be pushed, when remaining channel only has one, then all channels are made
For channel can be pushed；
If thering is channel to have successfully to record, using all channels as channel can be pushed；
If no channel does not have failure record, judge all channels whether all for IMSI channels；
If all channels are all to be IMSI channels, when the trade company corresponding to order only uses IMSI channels, all channels are made
For that can push channel, when the trade company corresponding to order is not only to use IMSI channels, then output channel is elected unsuccessfully；
If all channels are not all to be IMSI channels, using all channels as channel can be pushed；
Judge whether the number that can push channel is more than three；
If so, then randomly selecting one of them in first three channel that can be pushed in channel, and push the channel of extraction；
If it is not, then randomly selecting one of them all pushed in channel, and push the channel of extraction.
9. the channel of disbursement based on data analysis elects system, it is characterised in that is obtained including mass data acquiring unit, success rate
Take unit and elect unit；
The mass data acquiring unit, for obtaining the increment order data of magnanimity；
The success rate acquiring unit, for analyzing the order data, obtain the order data all channels it is each
The weighting success rate of dimension weighted score and the different channels of each dimension combination；
It is described to elect unit, for carrying out order accuracy matching according to weighting success rate for new order data, elect most
Excellent channel.
10. the channel of disbursement according to claim 9 based on data analysis elects system, it is characterised in that the success
Rate acquiring unit includes sort module, Bayesian analysis module, weighted calculation module, sorting module, weighting success rate and obtains mould
Block and sorting module；
The sort module, for classifying to all channels in the order data；
The Bayesian analysis module, for according to classification as a result, carrying out Bayes to every a kind of channel according to different dimensions
Analysis；
The weighted calculation module, by according to Bayesian analysis as a result, based on being weighted to each dimension of every a kind of channel
Calculate, obtain each dimension weighted score per a kind of channel；
The sorting module, for being arranged according to dimension combination, obtains dimension combination；
It is described to be weighted to power acquisition module, for according to each dimension weighted score per a kind of channel, obtaining each dimension
Different channels weighting success rate；
The sorting module, the weighting success rate for the different channels to each dimension are ranked up.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710481257.7A CN107967530A (en) | 2017-06-22 | 2017-06-22 | Channel of disbursement based on data analysis elects method and its system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710481257.7A CN107967530A (en) | 2017-06-22 | 2017-06-22 | Channel of disbursement based on data analysis elects method and its system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107967530A true CN107967530A (en) | 2018-04-27 |
Family
ID=61997424
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710481257.7A Pending CN107967530A (en) | 2017-06-22 | 2017-06-22 | Channel of disbursement based on data analysis elects method and its system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107967530A (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108764863A (en) * | 2018-05-24 | 2018-11-06 | 腾讯科技（深圳）有限公司 | A kind of virtual resource transfer method, device, server and storage medium |
| CN108880829A (en) * | 2018-05-16 | 2018-11-23 | 北京搜狐新动力信息技术有限公司 | A kind of payment member mall system control method and device |
| CN109447598A (en) * | 2018-09-25 | 2019-03-08 | 平安科技（深圳）有限公司 | Payment data processing method, device, computer equipment and storage medium |
| CN109670797A (en) * | 2018-09-11 | 2019-04-23 | 深圳平安财富宝投资咨询有限公司 | Pay route selecting method, apparatus, equipment and storage medium |
| CN112633743A (en) * | 2020-12-30 | 2021-04-09 | 北京嘀嘀无限科技发展有限公司 | Method, apparatus, computer device, medium, and product for adjusting resource allocation |
| CN113762819A (en) * | 2020-06-18 | 2021-12-07 | 北京沃东天骏信息技术有限公司 | Channel scheduling method and device |
| US11282052B2 (en) * | 2019-05-06 | 2022-03-22 | Advanced New Technologies Co., Ltd. | Payment channel recommendation |
| CN117495504A (en) * | 2023-11-20 | 2024-02-02 | 广州市翎泷科技有限公司 | Order data analysis method, system, equipment and storage medium for digital commodity |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103577985A (en) * | 2012-07-26 | 2014-02-12 | 阿里巴巴集团控股有限公司 | Data channel selection method and data processing platform |
| US20150095990A1 (en) * | 2011-03-09 | 2015-04-02 | Ebay Inc. | Device reputation |
| CN105354313A (en) * | 2015-11-11 | 2016-02-24 | 南京安讯科技有限责任公司 | Method for carrying out credit assessment by big data |
| CN105809502A (en) * | 2014-12-30 | 2016-07-27 | 阿里巴巴集团控股有限公司 | Transaction risk detection method and apparatus |
| CN106022758A (en) * | 2016-04-29 | 2016-10-12 | 北京小米移动软件有限公司 | Wireless router smart home managing method and wireless router |
| CN106127480A (en) * | 2016-06-16 | 2016-11-16 | 上海携程商务有限公司 | transaction payment method and system |
| CN106557956A (en) * | 2016-11-29 | 2017-04-05 | 国网山东省电力公司电力科学研究院 | A kind of method with regard to configuring client's paying service information pushing strategy |
-
2017
- 2017-06-22 CN CN201710481257.7A patent/CN107967530A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150095990A1 (en) * | 2011-03-09 | 2015-04-02 | Ebay Inc. | Device reputation |
| CN103577985A (en) * | 2012-07-26 | 2014-02-12 | 阿里巴巴集团控股有限公司 | Data channel selection method and data processing platform |
| CN105809502A (en) * | 2014-12-30 | 2016-07-27 | 阿里巴巴集团控股有限公司 | Transaction risk detection method and apparatus |
| CN105354313A (en) * | 2015-11-11 | 2016-02-24 | 南京安讯科技有限责任公司 | Method for carrying out credit assessment by big data |
| CN106022758A (en) * | 2016-04-29 | 2016-10-12 | 北京小米移动软件有限公司 | Wireless router smart home managing method and wireless router |
| CN106127480A (en) * | 2016-06-16 | 2016-11-16 | 上海携程商务有限公司 | transaction payment method and system |
| CN106557956A (en) * | 2016-11-29 | 2017-04-05 | 国网山东省电力公司电力科学研究院 | A kind of method with regard to configuring client's paying service information pushing strategy |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108880829A (en) * | 2018-05-16 | 2018-11-23 | 北京搜狐新动力信息技术有限公司 | A kind of payment member mall system control method and device |
| CN108764863A (en) * | 2018-05-24 | 2018-11-06 | 腾讯科技（深圳）有限公司 | A kind of virtual resource transfer method, device, server and storage medium |
| CN108764863B (en) * | 2018-05-24 | 2021-10-29 | 腾讯科技（深圳）有限公司 | Data transfer method, device, server and storage medium |
| CN109670797A (en) * | 2018-09-11 | 2019-04-23 | 深圳平安财富宝投资咨询有限公司 | Pay route selecting method, apparatus, equipment and storage medium |
| CN109447598A (en) * | 2018-09-25 | 2019-03-08 | 平安科技（深圳）有限公司 | Payment data processing method, device, computer equipment and storage medium |
| US11282052B2 (en) * | 2019-05-06 | 2022-03-22 | Advanced New Technologies Co., Ltd. | Payment channel recommendation |
| CN113762819A (en) * | 2020-06-18 | 2021-12-07 | 北京沃东天骏信息技术有限公司 | Channel scheduling method and device |
| CN112633743A (en) * | 2020-12-30 | 2021-04-09 | 北京嘀嘀无限科技发展有限公司 | Method, apparatus, computer device, medium, and product for adjusting resource allocation |
| CN117495504A (en) * | 2023-11-20 | 2024-02-02 | 广州市翎泷科技有限公司 | Order data analysis method, system, equipment and storage medium for digital commodity |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107967530A (en) | Channel of disbursement based on data analysis elects method and its system | |
| Bi et al. | A big data clustering algorithm for mitigating the risk of customer churn | |
| CN108932585B (en) | Merchant operation management method and equipment, storage medium and electronic equipment thereof | |
| CN103778548B (en) | Merchandise news and key word matching method, merchandise news put-on method and device | |
| WO2019037202A1 (en) | Method and apparatus for recognising target customer, electronic device and medium | |
| CN108665366A (en) | Determine method, terminal device and the computer readable storage medium of consumer's risk grade | |
| US20110093457A1 (en) | Method for calculating resource points of resource information and distributing points | |
| WO2011047474A1 (en) | Systems and methods for social graph data analytics to determine connectivity within a community | |
| CN102081774A (en) | Card-raising identification method and system | |
| CN108537542B (en) | Data processing method for social network | |
| CN107038511A (en) | A kind of method and device for determining risk assessment parameter | |
| CN108629379A (en) | A kind of individual's reference appraisal procedure and system | |
| CN113537807A (en) | Enterprise intelligent wind control method and device | |
| CN112116103A (en) | Method, device and system for evaluating personal qualification based on federal learning and storage medium | |
| CN109117501A (en) | Science data modeling and storage method based on block chain | |
| KR20210069611A (en) | Method, platform and computer program for estimating outcome of predictive proposition | |
| CN113011966A (en) | Credit scoring method and device based on deep learning | |
| Rasolinezhad | Evaluating electronic banking systems in developing nations through analytic hierarchy process model: a case study | |
| Chowdary et al. | Analogous approach towards performance analysis for software defect prediction and prioritization | |
| He et al. | Blockchain Operations in the Presence of Security Concerns | |
| CN110288334A (en) | Project charging method, calculates equipment and computer readable storage medium at device | |
| Petrov | Combined criteria weighting in MCDM: AHP in blocks with traditional Entropy and novel Hierarchy in TOPSIS evaluation of Cloud Services | |
| CN110781384B (en) | Priority-based content recommendation method, device, equipment and medium | |
| CN110175271B (en) | Case random ordering method and device | |
| Purohit et al. | Replaceability based web service selection approach |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180427 |