CN109815267A - The branch mailbox optimization method and system, storage medium and terminal of feature in data modeling - Google Patents
The branch mailbox optimization method and system, storage medium and terminal of feature in data modeling Download PDFInfo
- Publication number
- CN109815267A CN109815267A CN201811567505.0A CN201811567505A CN109815267A CN 109815267 A CN109815267 A CN 109815267A CN 201811567505 A CN201811567505 A CN 201811567505A CN 109815267 A CN109815267 A CN 109815267A
- Authority
- CN
- China
- Prior art keywords
- branch mailbox
- feature
- data
- mailbox
- data modeling
- 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
- 238000000034 method Methods 0.000 title claims abstract description 38
- 238000005457 optimization Methods 0.000 title claims abstract description 29
- 238000004590 computer program Methods 0.000 claims description 12
- 230000015654 memory Effects 0.000 claims description 11
- 238000003066 decision tree Methods 0.000 claims description 9
- 230000005055 memory storage Effects 0.000 claims description 2
- 238000012545 processing Methods 0.000 description 13
- 238000005516 engineering process Methods 0.000 description 5
- 230000000875 corresponding effect Effects 0.000 description 4
- 238000007418 data mining Methods 0.000 description 4
- 230000000694 effects Effects 0.000 description 4
- 230000008859 change Effects 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 230000001174 ascending effect Effects 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 241000208340 Araliaceae Species 0.000 description 1
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 description 1
- 235000003140 Panax quinquefolius Nutrition 0.000 description 1
- 230000006978 adaptation Effects 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000002596 correlated effect Effects 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 235000008434 ginseng Nutrition 0.000 description 1
- 238000003709 image segmentation Methods 0.000 description 1
- 230000002265 prevention Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000012549 training Methods 0.000 description 1
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The present invention provides the branch mailbox optimization method and system, storage medium and terminal of feature in a kind of data modeling, comprising the following steps: selects corresponding branch mailbox algorithm based on data characteristics；Data characteristics is divided at least two continuous branch mailbox based on the branch mailbox algorithm；Branch mailbox result is adjusted based on preset condition.The branch mailbox optimization method of feature and system, storage medium and terminal can be according to specific business scenarios in data modeling of the invention, for the different branch mailbox method of different data feature selecting, branch mailbox adjustment is carried out further according to preselected conditions, so as to be switched fast branch mailbox method, and branch mailbox result is stable, has business interpretation.
Description
Technical field
The present invention relates to the technical fields of Modeling of Data Mining, excellent more particularly to a kind of branch mailbox of feature in data modeling
Change method and system, storage medium and terminal.
Background technique
The high speed development of finance, internet industry be unable to do without big data, artificial intelligence technology as support.In order to quickly,
Efficiently realize data realization, prevention and control risk, Modeling of Data Mining ability is more taken seriously.
In the prior art, one-stop Modeling of Data Mining process generally comprises business understanding, feature pretreatment, feature work
Journey, model training, the online and model iteration of model deployment and etc..Wherein Feature Engineering is taken over from the past and set a new course for the future in entire work flow, and one
It is directly the emphasis and difficult point of Modeling of Data Mining.In order to efficiently carry out feature discretization, feature derivative, the assessment of feature importance
Equal work, feature branch mailbox technology play very important effect again in Feature Engineering.
Feature branch mailbox technology includes following two major classes:
(1) unsupervised branch mailbox
Common unsupervised branch mailbox include wide branch mailbox, etc. deep branch mailbox etc..Wide branch mailbox is by the value model of characteristic variable
It encloses and is divided into several wide sections, each section is a branch mailbox.It is that the value of characteristic variable is ascending Deng depth branch mailbox
Sequence, determines branch mailbox cut point according to quantile and branch mailbox number, and the section of cut point composition is a branch mailbox two-by-two.
(2) there is supervision branch mailbox.
It include decision tree branch mailbox, card side's branch mailbox etc. there are commonly supervision branch mailbox.Decision tree branch mailbox is instructed by single features
Practice decision-tree model, according to the depth of branch mailbox number control tree, all values of feature is traversed, by minimizing all leaf sections
Total entropy of point obtains Image Segmentation Methods Based on Features point, and all ascending sequences of cut-point, the section of cut point composition is one two-by-two
Branch mailbox.Card side's branch mailbox will be merged with the adjacent interval of minimum X2 value on the basis of initial branch mailbox (such as wide branch mailbox),
Until meeting given stop condition, such as minimum X2 threshold value or maximum branch mailbox number.
However, branch mailbox technology in the prior art is generally basede on business and modeling personnel's artificial judgment, that there are methods is single,
Poor robustness, be affected by human subjective's factor, the branch mailbox result of variable do not have business interpretation the problems such as, thus
Model overall effect has been influenced to a certain extent.
Summary of the invention
In view of the foregoing deficiencies of prior art, the purpose of the present invention is to provide a kind of points of feature in data modeling
Case optimization method and system, storage medium and terminal, can be according to specific business scenario, not for different data feature selecting
Same branch mailbox method carries out branch mailbox adjustment further according to preselected conditions, and so as to be switched fast branch mailbox method, and branch mailbox result is steady
Determine, there is business interpretation.
In order to achieve the above objects and other related objects, the present invention provides a kind of branch mailbox optimization side of feature in data modeling
Method, comprising the following steps: corresponding branch mailbox algorithm is selected based on data characteristics；Data characteristics is drawn based on the branch mailbox algorithm
It is divided at least two continuous branch mailbox；Branch mailbox result is adjusted based on preset condition.
In one embodiment of the invention, the branch mailbox algorithm include wide branch mailbox, etc. deep branch mailbox, decision tree branch mailbox, card side
One of branch mailbox or multiple combinations.
In one embodiment of the invention, the preset condition include maximum branch mailbox number, sample number threshold value in case, in case just
Negative sample accounting, each branch mailbox WOE value meet one of monotonicity or multiple combinations.
In one embodiment of the invention, the data characteristics includes user's reference feature.
Accordingly, the present invention provides a kind of branch mailbox optimization system of feature in data modeling, including selecting module, branch mailbox mould
Block and adjustment module；
The selecting module is used to select corresponding branch mailbox algorithm based on data characteristics；
The branch mailbox module is used to that data characteristics to be divided at least two continuous branch mailbox based on the branch mailbox algorithm；
The adjustment module is used to be adjusted branch mailbox result based on preset condition.
In one embodiment of the invention, the branch mailbox algorithm include wide branch mailbox, etc. deep branch mailbox, decision tree branch mailbox, card side
One of branch mailbox or multiple combinations.
In one embodiment of the invention, the preset condition include maximum branch mailbox number, sample number threshold value in case, in case just
Negative sample accounting, each branch mailbox WOE value meet one of monotonicity or multiple combinations.
In one embodiment of the invention, the data characteristics includes user's reference feature.
The present invention provides a kind of storage medium, is stored thereon with computer program, and the computer program is held by processor
The branch mailbox optimization method of feature in above-mentioned data modeling is realized when row.
Finally, the present invention provides a kind of terminal, comprising: processor and memory；
The memory is for storing computer program；
The processor is used to execute the computer program of the memory storage, so that the terminal executes above-mentioned number
According to the branch mailbox optimization method of feature in modeling.
As described above, in data modeling of the invention feature branch mailbox optimization method and system, storage medium and terminal, tool
Have it is following the utility model has the advantages that
(1) different branch mailbox methods can be used for different data feature, and branch mailbox tune can be carried out according to preselected conditions
It is whole；
(2) support multiclass parameter setting, robustness stronger；
(3) it supports each branch mailbox evidence weight (Weight of Evidence, WOE) monotonicity detection, meets feature branch mailbox
As a result interpretation.
Detailed description of the invention
Fig. 1 is shown as flow chart of the branch mailbox optimization method of feature in data modeling of the invention in an embodiment；
User nearly 12 months non-overdue refund order numbers are gone back always when Fig. 2 is shown as not adjusting monotonicity in an embodiment
The accounting of money order numbers is with equal deep branch mailbox WOE trend graph；
User nearly 12 months non-overdue refund order numbers are refunded always when Fig. 3 is shown as adjusting monotonicity in an embodiment
The accounting of order numbers is with equal deep branch mailbox WOE trend graph；
User nearly 12 months non-overdue refund order numbers are gone back always when Fig. 4 is shown as not adjusting monotonicity in an embodiment
The accounting of money order numbers is with card branch mailbox WOE trend graph；
User nearly 12 months non-overdue refund order numbers are refunded always when Fig. 5 is shown as adjusting monotonicity in an embodiment
The accounting of order numbers is with card branch mailbox WOE trend graph；
Fig. 6 is shown as structural representation of the branch mailbox optimization system of feature in data modeling of the invention in an embodiment
Figure；
Fig. 7 is shown as the structural schematic diagram of terminal of the invention in an embodiment.
Component label instructions
61 selecting modules
62 branch mailbox modules
63 adjustment modules
71 processors
72 memories
Specific embodiment
Illustrate embodiments of the present invention below by way of specific specific example, those skilled in the art can be by this specification
Other advantages and efficacy of the present invention can be easily understood for disclosed content.The present invention can also pass through in addition different specific realities
The mode of applying is embodied or practiced, the various details in this specification can also based on different viewpoints and application, without departing from
Various modifications or alterations are carried out under spirit of the invention.
It should be noted that the basic conception that only the invention is illustrated in a schematic way is illustrated provided in the present embodiment,
Then only shown in schema with it is of the invention in related component rather than component count, shape and size when according to actual implementation draw
System, when actual implementation kenel, quantity and the ratio of each component can arbitrarily change for one kind, and its assembly layout kenel can also
It can be increasingly complex.
The branch mailbox optimization method of feature and system, storage medium and terminal can be according to specific in data modeling of the invention
Business scenario, for the different branch mailbox method of different data feature selecting, to select the branch mailbox algorithm of adaptation；Further according to pre-selection
Condition carries out branch mailbox adjustment, and to reach expected branch mailbox effect, so as to be switched fast branch mailbox method, and branch mailbox result is steady
Determine, there is business interpretation.
As shown in Figure 1, the branch mailbox optimization method of feature includes following step in data modeling of the invention in an embodiment
It is rapid:
Step S1, corresponding branch mailbox algorithm is selected based on data characteristics.
Specifically, the data characteristics of modeling data is analyzed, is adapted therewith according to data characteristics selection
Branch mailbox algorithm.Preferably, the data characteristics can be user's reference feature, so as to be carried out based on user's reference feature
Branch mailbox and data modeling.
In one embodiment of the invention, the branch mailbox algorithm include wide branch mailbox, etc. deep branch mailbox, decision tree branch mailbox, card side
One of branch mailbox or multiple combinations.
Step S2, data characteristics is divided by least two continuous branch mailbox based on the branch mailbox algorithm.
Specifically, it is determined that branch mailbox is carried out to the data characteristics based on the branch mailbox algorithm, to obtain after branch mailbox algorithm
Take at least two continuous branch mailbox.Wherein, the number of branch mailbox must be determined according to data characteristics.Branch mailbox number is less, Wu Faji
Data modeling is carried out in branch mailbox structure；Branch mailbox number is more, then data processing complexity is excessively high.
Step S3, branch mailbox result is adjusted based on preset condition.
Specifically, after branch mailbox, branch mailbox result can also further be adjusted based on user demand, it is a to meet
Property demand.
In one embodiment of the invention, the preset condition include maximum branch mailbox number, sample number threshold value in case, in case just
Negative sample accounting, each branch mailbox WOE value meet one of monotonicity or multiple combinations.
Carry out the branch mailbox optimization method of feature in the data modeling that the present invention is further explained below by specific embodiment.
In this embodiment, it needs to evaluate user's reference and carries out data modeling.Wherein, with continuous feature, " user nearly 12
For accounting of a month non-overdue refund order numbers in total refund order numbers ", from business scenario and characteristic variable branch mailbox result
A possibility that interpretation is set out, and this feature should be handy family with user (probability) is negatively correlated.I.e. user nearly 12 months non-to exceed
A possibility that phase refund order numbers are lower in the accounting of total refund order numbers, and user is handy family is higher；User nearly 12 months non-
A possibility that overdue refund order numbers are higher in the accounting of total refund order numbers, and user is handy family is lower.Therefore, such as Fig. 2 and
Shown in Fig. 4, it can be based on the branch mailbox methods such as equal deep branch mailbox, card side's branch mailbox first, branch mailbox is carried out to the continuous feature；Such as Fig. 3 and Fig. 5
It is shown, it is then based on each branch mailbox WOE value and meets the preset condition of monotonicity branch mailbox result is adjusted, to obtain each branch mailbox
WOE value meets the branch mailbox result of monotonicity.
As shown in fig. 6, the branch mailbox optimization system of feature includes selection in the data modeling that the present invention mentions in an embodiment
Module 61, branch mailbox module 62 and adjustment module 63.
Selecting module 61 is used to select corresponding branch mailbox algorithm based on data characteristics.
Specifically, the data characteristics of modeling data is analyzed, is adapted therewith according to data characteristics selection
Branch mailbox algorithm.Preferably, the data characteristics can be user's reference feature, so as to be carried out based on user's reference feature
Branch mailbox and data modeling.
In one embodiment of the invention, the branch mailbox algorithm include wide branch mailbox, etc. deep branch mailbox, decision tree branch mailbox, card side
One of branch mailbox or multiple combinations.
Branch mailbox module 62 is connected with selecting module 61, for data characteristics to be divided at least two based on the branch mailbox algorithm
A continuous branch mailbox.
Specifically, it is determined that branch mailbox is carried out to the data characteristics based on the branch mailbox algorithm, to obtain after branch mailbox algorithm
Take at least two continuous branch mailbox.Wherein, the number of branch mailbox must be determined according to data characteristics.Branch mailbox number is less, Wu Faji
Data modeling is carried out in branch mailbox structure；Branch mailbox number is more, then data processing complexity is excessively high.
Adjustment module 63 is connected with branch mailbox module 62, for being adjusted based on preset condition to branch mailbox result.
Specifically, after branch mailbox, branch mailbox result can also further be adjusted based on user demand, it is a to meet
Property demand.
In one embodiment of the invention, the preset condition include maximum branch mailbox number, sample number threshold value in case, in case just
Negative sample accounting, each branch mailbox WOE value meet one of monotonicity or multiple combinations.
It should be noted that it should be understood that the modules of apparatus above division be only a kind of logic function division,
It can completely or partially be integrated on a physical entity in actual implementation, it can also be physically separate.And these modules can be with
All realized by way of processing element calls with software；It can also all realize in the form of hardware；It can also part mould
Block realizes that part of module passes through formal implementation of hardware by way of processing element calls software.For example, x module can be
The processing element individually set up also can integrate and realize in some chip of above-mentioned apparatus, in addition it is also possible to program generation
The form of code is stored in the memory of above-mentioned apparatus, is called by some processing element of above-mentioned apparatus and is executed the above x mould
The function of block.The realization of other modules is similar therewith.Furthermore these modules completely or partially can integrate together, can also be only
It is vertical to realize.Processing element described here can be a kind of integrated circuit, the processing capacity with signal.During realization,
Each step of the above method or the above modules can be by the integrated logic circuits of the hardware in processor elements or soft
The instruction of part form is completed.
For example, the above module can be arranged to implement one or more integrated circuits of above method, such as:
One or more specific integrated circuits (Application Specific Integrated Circuit, abbreviation ASIC), or,
One or more microprocessors (Digital Singnal Processor, abbreviation DSP), or, one or more scene can compile
Journey gate array (Field Programmable Gate Array, abbreviation FPGA) etc..For another example, when some above module passes through place
When managing the form realization of element scheduler program code, which can be general processor, such as central processing unit
(Central Processing Unit, abbreviation CPU) or it is other can be with the processor of caller code.For another example, these modules
It can integrate together, realized in the form of system on chip (system-on-a-chip, abbreviation SOC).
It is stored with computer program on storage medium of the invention, is realized when the computer program is executed by processor
The branch mailbox optimization method of feature in the data modeling stated.
As shown in fig. 7, terminal of the invention includes: processor 71 and memory 72 in an embodiment.
The memory 72 is for storing computer program.
The memory 72, which includes: that ROM, RAM, magnetic disk, USB flash disk, storage card or CD etc. are various, can store program generation
The medium of code.
The processor 71 is connected with the memory 72, the computer program stored for executing the memory 72,
So that the terminal executes the branch mailbox optimization method of feature in above-mentioned data modeling.
Preferably, the processor 71 can be general processor, including central processing unit (Central Processing
Unit, abbreviation CPU), network processing unit (Network Processor, abbreviation NP) etc.；It can also be digital signal processor
(Digital Signal Processor, abbreviation DSP), specific integrated circuit (Application Specific
Integrated Circuit, abbreviation ASIC), field programmable gate array (Field Programmable Gate Array,
Abbreviation FPGA) either other programmable logic device, discrete gate or transistor logic, discrete hardware components.
In conclusion in data modeling of the invention feature branch mailbox optimization method and system, storage medium and terminal needle
Different branch mailbox methods can be used to different data feature, and branch mailbox adjustment can be carried out according to preselected conditions；Support multiclass ginseng
Number setting, robustness are stronger；It supports each branch mailbox WOE monotonicity to detect, meets the interpretation of feature branch mailbox result.So this
Invention effectively overcomes various shortcoming in the prior art and has high industrial utilization value.
The above-described embodiments merely illustrate the principles and effects of the present invention, and is not intended to limit the present invention.It is any ripe
The personage for knowing this technology all without departing from the spirit and scope of the present invention, carries out modifications and changes to above-described embodiment.Cause
This, institute is complete without departing from the spirit and technical ideas disclosed in the present invention by those of ordinary skill in the art such as
At all equivalent modifications or change, should be covered by the claims of the present invention.
Claims (10)
1. the branch mailbox optimization method of feature in a kind of data modeling, it is characterised in that: the following steps are included:
Corresponding branch mailbox algorithm is selected based on data characteristics；
Data characteristics is divided at least two continuous branch mailbox based on the branch mailbox algorithm；
Branch mailbox result is adjusted based on preset condition.
2. the branch mailbox optimization method of feature in data modeling according to claim 1, it is characterised in that: the branch mailbox algorithm
Including wide branch mailbox, etc. one of deep branch mailbox, decision tree branch mailbox, card side's branch mailbox or multiple combinations.
3. the branch mailbox optimization method of feature in data modeling according to claim 1, it is characterised in that: the preset condition
Meet one in monotonicity including positive and negative sample accounting, each branch mailbox WOE value in sample number threshold value in maximum branch mailbox number, case, case
Kind or multiple combinations.
4. the branch mailbox optimization method of feature in data modeling according to claim 1, it is characterised in that: the data characteristics
Including user's reference feature.
5. the branch mailbox optimization system of feature in a kind of data modeling, it is characterised in that: including selecting module, branch mailbox module and adjustment
Module；The selecting module is used to select corresponding branch mailbox algorithm based on data characteristics；
The branch mailbox module is used to that data characteristics to be divided at least two continuous branch mailbox based on the branch mailbox algorithm；
The adjustment module is used to be adjusted branch mailbox result based on preset condition.
6. the branch mailbox optimization system of feature in data modeling according to claim 5, it is characterised in that: the branch mailbox algorithm
Including wide branch mailbox, etc. one of deep branch mailbox, decision tree branch mailbox, card side's branch mailbox or multiple combinations.
7. the branch mailbox optimization system of feature in data modeling according to claim 5, it is characterised in that: the preset condition
Meet one in monotonicity including positive and negative sample accounting, each branch mailbox WOE value in sample number threshold value in maximum branch mailbox number, case, case
Kind or multiple combinations.
8. the branch mailbox optimization system of feature in data modeling according to claim 5, it is characterised in that: the data characteristics
Including user's reference feature.
9. a kind of storage medium, is stored thereon with computer program, which is characterized in that the computer program is executed by processor
In Shi Shixian Claims 1-4 in any data modeling feature branch mailbox optimization method.
10. a kind of terminal characterized by comprising processor and memory；
The memory is for storing computer program；
The processor is used to execute the computer program of the memory storage, so that the terminal perform claim requires 1 to 4
In in any data modeling feature branch mailbox optimization method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811567505.0A CN109815267A (en) | 2018-12-21 | 2018-12-21 | The branch mailbox optimization method and system, storage medium and terminal of feature in data modeling |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811567505.0A CN109815267A (en) | 2018-12-21 | 2018-12-21 | The branch mailbox optimization method and system, storage medium and terminal of feature in data modeling |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109815267A true CN109815267A (en) | 2019-05-28 |
Family
ID=66601826
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811567505.0A Pending CN109815267A (en) | 2018-12-21 | 2018-12-21 | The branch mailbox optimization method and system, storage medium and terminal of feature in data modeling |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109815267A (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110704535A (en) * | 2019-09-26 | 2020-01-17 | 深圳前海微众银行股份有限公司 | Data binning method, device, equipment and computer readable storage medium |
| CN110766167A (en) * | 2019-10-29 | 2020-02-07 | 深圳前海微众银行股份有限公司 | Interactive feature selection method, device and readable storage medium |
| CN110807072A (en) * | 2019-10-30 | 2020-02-18 | 深圳前海微众银行股份有限公司 | Interactive feature binning method and device and readable storage medium |
| CN110837894A (en) * | 2019-10-28 | 2020-02-25 | 腾讯科技（深圳）有限公司 | Feature processing method and device and storage medium |
| CN110909085A (en) * | 2019-11-25 | 2020-03-24 | 深圳前海微众银行股份有限公司 | Data processing method, device, equipment and storage medium |
| CN111079283A (en) * | 2019-12-13 | 2020-04-28 | 四川新网银行股份有限公司 | Method for processing information saturation unbalanced data |
| CN111506485A (en) * | 2020-04-15 | 2020-08-07 | 深圳前海微众银行股份有限公司 | Feature binning method, device, equipment and computer-readable storage medium |
| CN111507822A (en) * | 2020-04-13 | 2020-08-07 | 深圳微众信用科技股份有限公司 | Enterprise risk assessment method based on feature engineering |
| CN117992883A (en) * | 2024-02-21 | 2024-05-07 | 广东圣千科技有限公司 | User behavior processing method and system based on business data abnormality |
-
2018
- 2018-12-21 CN CN201811567505.0A patent/CN109815267A/en active Pending
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110704535A (en) * | 2019-09-26 | 2020-01-17 | 深圳前海微众银行股份有限公司 | Data binning method, device, equipment and computer readable storage medium |
| CN110704535B (en) * | 2019-09-26 | 2023-10-24 | 深圳前海微众银行股份有限公司 | Data binning method, device, equipment and computer readable storage medium |
| CN110837894A (en) * | 2019-10-28 | 2020-02-25 | 腾讯科技（深圳）有限公司 | Feature processing method and device and storage medium |
| CN110837894B (en) * | 2019-10-28 | 2024-02-13 | 腾讯科技（深圳）有限公司 | Feature processing method, device and storage medium |
| CN110766167A (en) * | 2019-10-29 | 2020-02-07 | 深圳前海微众银行股份有限公司 | Interactive feature selection method, device and readable storage medium |
| CN110807072A (en) * | 2019-10-30 | 2020-02-18 | 深圳前海微众银行股份有限公司 | Interactive feature binning method and device and readable storage medium |
| CN110909085A (en) * | 2019-11-25 | 2020-03-24 | 深圳前海微众银行股份有限公司 | Data processing method, device, equipment and storage medium |
| CN111079283A (en) * | 2019-12-13 | 2020-04-28 | 四川新网银行股份有限公司 | Method for processing information saturation unbalanced data |
| CN111507822A (en) * | 2020-04-13 | 2020-08-07 | 深圳微众信用科技股份有限公司 | Enterprise risk assessment method based on feature engineering |
| CN111506485A (en) * | 2020-04-15 | 2020-08-07 | 深圳前海微众银行股份有限公司 | Feature binning method, device, equipment and computer-readable storage medium |
| CN117992883A (en) * | 2024-02-21 | 2024-05-07 | 广东圣千科技有限公司 | User behavior processing method and system based on business data abnormality |
| CN117992883B (en) * | 2024-02-21 | 2024-07-19 | 广东圣千科技有限公司 | User behavior processing method and system based on business data abnormality |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109815267A (en) | The branch mailbox optimization method and system, storage medium and terminal of feature in data modeling | |
| CN110956272B (en) | Method and system for realizing data processing | |
| US7324363B2 (en) | SPICE optimized for arrays | |
| CN106600067B (en) | Method and device for optimizing multidimensional cube model | |
| US6295517B1 (en) | Method and apparatus for adaptively or selectively choosing event-triggered cycle-based simulation or oblivious-triggered cycle-based simulation on a cluster-by-cluster basis | |
| EP0733984A1 (en) | Cycle-based event-driven simulator | |
| CN108491226B (en) | Spark configuration parameter automatic tuning method based on cluster scaling | |
| CN110187965A (en) | The running optimizatin and data processing method of neural network, equipment and storage medium | |
| CN110058882B (en) | OPU instruction set definition method for CNN acceleration | |
| US11288266B2 (en) | Candidate projection enumeration based query response generation | |
| CN113807046B (en) | Test excitation optimization regression verification method, system and medium | |
| CN109992866A (en) | Training method, device, readable medium and the electronic equipment of load forecasting model | |
| CN110399182B (en) | CUDA thread placement optimization method | |
| CN112347722A (en) | Method and device for efficiently evaluating chip Feed-through flow stage number | |
| CN116432570A (en) | Method and device for generating test case of chip and storage medium | |
| WO2020113526A1 (en) | Chip verification method and device | |
| CN108509723A (en) | LRU Cache based on artificial neural network prefetch mechanism performance income evaluation method | |
| CN113886092A (en) | Computation graph execution method and device and related equipment | |
| CN107315863B (en) | Layout optimization method and device, terminal and storage medium | |
| CN110245978B (en) | Method and device for evaluating and selecting policies in policy group | |
| US6370494B1 (en) | Simulator and computer-readable recordable medium having program for execution on computer realizing the simulator recorded thereon | |
| CN112257368A (en) | Clock layout method, device, EDA tool and computer readable storage medium | |
| CN111045959B (en) | Complex algorithm variable mapping method based on storage optimization | |
| CN112560376A (en) | Method for optimizing model parameters of semiconductor device | |
| CN110443346A (en) | A kind of model explanation method and device based on input feature vector importance |
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 |
Application publication date: 20190528 |
|
| RJ01 | Rejection of invention patent application after publication |