CN107301493A - A kind of mutual golden business ratings model based on deep neural network - Google Patents
A kind of mutual golden business ratings model based on deep neural network Download PDFInfo
- Publication number
- CN107301493A CN107301493A CN201710361427.8A CN201710361427A CN107301493A CN 107301493 A CN107301493 A CN 107301493A CN 201710361427 A CN201710361427 A CN 201710361427A CN 107301493 A CN107301493 A CN 107301493A
- Authority
- CN
- China
- Prior art keywords
- data
- enterprises
- jin
- golden
- mutual
- 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
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
-
- 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
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Development Economics (AREA)
- Strategic Management (AREA)
- Human Resources & Organizations (AREA)
- Entrepreneurship & Innovation (AREA)
- Economics (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Marketing (AREA)
- Game Theory and Decision Science (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Educational Administration (AREA)
- Data Mining & Analysis (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a kind of mutual golden business ratings model based on deep neural network, it is related to Computer Applied Technology field, comprises the following steps：Excavate mutually golden company information collection of illustrative plates, manage collection of illustrative plates, public feelings information, Hu Jin enterprises and deposit pipe True Data, other channel data；Negative sample data are found out, according to sample data, with reference to the data excavated in S1 S5, each enterprise characteristic group of TOPN Hu Jin enterprises and inverse N Jia Hujin enterprises is extracted by CNN convolutional neural networks；By FNN forerunner's neutral net, validity feature is classified and scientific algorithm goes out shared weight；According to strong rule and model training, different mutual golden business ratings models are found out according to different demands；According to historical data, grading effect is verified, historical data is in the mutual through business ratings data of outside excavation；According to different application scenarios, the corresponding Hu Jin enterprises application model of output；Model needs moment supplementary data tuning, and artificial mark badcase methods carry out study tuning.
Description
Technical field
Commented the present invention relates to Computer Applied Technology field, more particularly to a kind of Hu Jin enterprises based on deep neural network
Level model.
Background technology
Hu Jin enterprises, i.e. internet financial company, no matter of the same trade or business from selection of the client to Hu Jin enterprises, or Hu Jin enterprises
Tear the business such as just borrow open, the grading of Hu Jin enterprises can play guiding effect.But, neither one is directed to interconnection on the market at present
The technical scheme and standards of grading of financial company's grading are netted, because present ten thousand network datas quality is not high, and is not true environment production
Raw data.
Existing Some Enterprises grading, is also simply walked according to the rule artificially fixed on the market, and the grading of this part is used
The grading factor arrived is within 100, and very superficial, effect is poor, it is impossible to which dynamic change, rating scale are consistent, it is impossible to thousand people
Thousand faces, are unscientific with reference to the mutual golden business ratings that these data are come.
CNN convolutional neural networks：CNN is a kind of multilayer neural network, based on artificial neural network, in artificial neural network
Before, feature extraction is carried out with wave filter, using convolution kernel as feature extractor, automatic training characteristics withdrawal device, that is volume
Product core and threshold parameter these be required for by network go study.
Feedforward neural network (feedforward neural network)：Abbreviation feedforward network, is artificial neural network
One kind.In such a neutral net, each neuron receives previous stage input since input layer, and is input to next stage, directly
To output layer.Feedback-less in whole network, can be represented with a directed acyclic graph.
NLP is neural LISP program LISP (Neuro-Linguistic Programming) english abbreviation.In Hong Kong,
It is translated into body and mind grammer formula intentionally.N (Neuro) refers to nervous system, including brain and thought process.L
(Linguistic) refer to language, more exact point is said, refer to the process of be input to the composition meaning from sensory signal.
Cosine similarity model：Cosine similarity, also known as cosine similarity.By calculating two vectorial included angle cosines
Value assesses their similarity.
The content of the invention
It is an object of the invention to：Mutually enter that business ratings effect is bad for solve current part, neither one is complete, accurate
True mutual golden business ratings model, the present invention provides a kind of mutual golden business ratings model based on deep neural network.
Technical scheme is as follows：
A kind of mutual golden business ratings model based on deep neural network, comprises the following steps：
S1：Excavate mutually golden company information data；Crawler technology, NLP technologies etc. can be mainly applied to, data source is mainly
Ten thousand network datas, and I manage it inside assets life data.
S2：Excavate mutually golden enterprise operation data；Data source is mainly access, and I manages it the number runed in the true line for deposit pipe
According to, the data that mutual enterprise externally issues, ten thousand network data gold etc..
S3：Excavate Hu Jin enterprises public feelings information；This partial information mostlys come from major regulators, major news media
Data mining.
S4：Excavate Hu Jin enterprises and deposit pipe True Data；Hu Jin enterprises, which are linked into me, manages it and deposits after guard system, can produce substantial amounts of
True generation data, containing user basic information, target essential information, cash flow, cash flow, information flow etc., this partial information is needed
Emphasis is wanted deeply to excavate, weight is larger in model construction.
S5：Excavate other channel data of Hu Jin enterprises；Bought containing data exchange, data etc..
S6：Find out positive and negative sample data；The conduct of TOPN Hu Jin enterprises is selected in existing enterprise's scoring according to disclosed in online
Positive sample；According to having closed down at present, close down, negative public sentiment enterprise, business circumstance reciprocal data, it is comprehensive that to extract N family reciprocal mutual
Golden enterprise's negative sense sample.
S7：According to sample data, binding time sequence, policy implication, management data, public sentiment data etc. pass through CNN convolution
Neutral net extracts each enterprise characteristic group of the TOPN Hu Jin enterprises and inverse N Jia Hujin enterprises in S6.
Details as Follows for calculating：
Correspondence validity feature can be extracted from sample data by calculating；X can enter Mobile state with the change of sample size
Adjustment.
S8：Neutral net is presented by FNN forerunner, validity feature is classified and scientific algorithm goes out shared weight.
In above formula, SjAnd θjRepresent validity feature vector, x=[x1, x2..., xn]TIt is input feature value, wjiIt is xiArrive
yjConnection weight, output quantity yj(j=1,2 ..., be m) classification results according to different characteristic, can calculate correspondence scoring, power
Tuple is according to as score data.
S9：Strong rule is used as with reference to different application scene, policy implication, blacklist, artificial bright-coloured knowledge etc..
S10：According to strong rule and model training, different mutual golden business ratings models are found out according to different demands；
S11：According to historical data, grading effect is verified, historical data is mutual being commented through enterprise in outside excavation
DBMS；
S12：According to different application scenarios, the corresponding Hu Jin enterprises application model of output.
Such as grading of enterprise operation performance assets, mutually business standing performance grading, golden enterprise investment value grading.
S13：Model needs moment supplementary data tuning, and artificial mark badcase methods carry out study tuning.
Here the method that machine learning can be used learns tuning automatically.
Further, in S8, calculating validity feature and calculating the method for shared weight can also use and be based on cosine phase
Like the classification Forecasting Methodology of degree, step is as follows：
S81：Sample data is extracted, including based on mutual golden enterprise operation data, public sentiment situation, industrial and commercial data, target data,
Corresponding scoring preference is extracted, calculation formula is as follows：
S82：Each feature preferences combination of Hu Jin enterprises is expressed as feature preferences vector：
v1=(p1, p2, p3..., pn)
S83:It is vector by business ratings character representation：
v2=(1,0,1 ..., 0)
Characteristic vector dimension values value is 0 or 1.
S84：Calculate cosine similarity：
Wherein simi values are bigger represents more similar.
S85：It is ranked up according to simi values, takes the business data most like with positive negative sample to be ranked up.
After such scheme, the beneficial effects of the present invention are：The business ratings model that mutually enters of the present invention instead of people
Work judges, and traditional rule model.The business ratings model set up using artificial intelligence related algorithm is more accurate
Really, and can real-time change, change at any time under different time points, different condition.
Brief description of the drawings
Fig. 1 is flow chart of the invention.
Embodiment
Embodiment 1
A kind of mutual golden business ratings model based on deep neural network, comprises the following steps：
S1：Excavate mutually golden company information collection of illustrative plates；Crawler technology, NLP technologies etc. can be mainly applied to, data source is mainly
Ten thousand network datas, and I manage it inside assets life data.
S2：Excavate mutually golden enterprise operation collection of illustrative plates；Data source is mainly access, and I manages it the number runed in the true line for deposit pipe
According to, the data that mutual enterprise externally issues, ten thousand network data gold etc..
S3：Excavate Hu Jin enterprises public feelings information；This partial information mostlys come from major regulators, major news media
Data mining.
S4：Excavate Hu Jin enterprises and deposit pipe True Data；Hu Jin enterprises, which are linked into me, manages it and deposits after guard system, can produce substantial amounts of
True generation data, containing user basic information, target essential information, cash flow, cash flow, information flow etc., this partial information is needed
Emphasis is wanted deeply to excavate, weight is larger in model construction.
S5：Excavate other channel data of Hu Jin enterprises；Bought containing data exchange, data etc..
S6：Find out positive and negative sample data；The conduct of TOPN Hu Jin enterprises is selected in existing enterprise's scoring according to disclosed in online
Positive sample；According to having closed down at present, close down, negative public sentiment enterprise, business circumstance reciprocal data, it is comprehensive that to extract N family reciprocal mutual
Golden enterprise's negative sense sample.
S7：According to sample data, binding time sequence, policy implication, management data, public sentiment data etc. pass through CNN convolution
Neutral net extracts each enterprise characteristic group of the TOPN Hu Jin enterprises and inverse N Jia Hujin enterprises in S6.
Details as Follows for calculating：
Correspondence validity feature can be extracted from sample data by calculating；X can enter Mobile state with the change of sample size
Adjustment.
S8：Neutral net is presented by FNN forerunner, validity feature is classified and scientific algorithm goes out shared weight.
In above formula, SjAnd θjRepresent validity feature vector, x=[x1, x2..., xn]TIt is input feature value, wjiIt is xiArrive
yjConnection weight, output quantity yj(j=1,2 ..., be m) classification results according to different characteristic, can calculate correspondence scoring, power
Tuple is according to as score data.
S9：Strong rule is used as with reference to different application scene, policy implication, blacklist, artificial bright-coloured knowledge etc..
S10：According to strong rule and model training, different mutual golden business ratings models are found out according to different demands；
S11：According to historical data, grading effect is verified, historical data is mutual being commented through enterprise in outside excavation
DBMS；
S12：According to different application scenarios, the corresponding Hu Jin enterprises application model of output.
Such as grading of enterprise operation performance assets, mutually business standing performance grading, golden enterprise investment value grading.
S13：Model needs moment supplementary data tuning, and artificial mark badcase methods carry out study tuning.
Here the method that machine learning can be used learns tuning automatically.
Further, in S8, calculating validity feature and calculating the method for shared weight can also use and be based on cosine phase
Like the classification Forecasting Methodology of degree, step is as follows：
S81：Sample data is extracted, including based on mutual golden enterprise operation data, public sentiment situation, industrial and commercial data, target data,
Corresponding scoring preference is extracted, calculation formula is as follows：
S82：Each feature preferences combination of Hu Jin enterprises is expressed as feature preferences vector：
v1=(p1, p2, p3..., pn)
S83：It is vector by business ratings character representation：
v2=(1,0,1 ..., 0)
Characteristic vector dimension values value is 0 or 1.
S84：Calculate cosine similarity：
Wherein simi values are bigger represents more similar.
S85：It is ranked up according to simi values, takes the business data most like with positive negative sample to be ranked up.
At present, the factor mutually used in golden business ratings model, that is, the packet used contains 2,090,000 web datas, 2,000,000
New network data, covers 4557 mutually 25 management datas of golden platform, and the group company information of 5 major class 16,850,000 company information is main
Type is wanted to have：
1. enterprise's industry and commerce information, such as " legal representative ", " number of registration ", " company's boudoir honey ", " affiliated industry ", " enterprise
Address " etc..
2. enterprise's shareholder's information, such as " owned enterprise ", " shareholder ", " holding ratio ", " subscribed capital contribution ", " shareholder's class
Type ", " subscribing the investment date " etc..
3. enterprise's change record information, such as " owned enterprise ", " change project ", " Date Of Change ".
4. enterprise annual reports information, such as " time ", " rank ", " finance ", " investor ", " source of news ".
5. enterprise trademark information, such as " trade mark ", " trade (brand) name ", " state " application time ".
6. top managers information, such as " name ", " position ", " photo ", " brief introduction ".
7. enterprise operation information, such as " grading earning rate ", " intended investment time limit ", " on-line time ", " registered capital ",
" registered place ", " state " etc..
8. enterprise key management data, such as " average expectancy earning rate ", " intended investment time limit ", " exchange hand ", " conclusion of the business
Amount situation of change ", " investment number ", " investment number situation of change ", " day fund net inflow ", " day fund net inflow change feelings
Condition ", " day remaining sum situation of change to be changed " etc..
9. user's score information, such as " total score ", " comment number ", " withdrawing deposit as scoring ", " experience scoring ".
10. mutual Golden Customs key data information, such as " data month ", " development ranking ", " platform names ", " development index ",
" the relatively upper monthly variation of development index ", " conclusion of the business ", " mobility ", " transparency " etc..
11. other data, such as " enterprises recruitment information ", " enterprise web site record information ", " enterprise operation abnormal information ",
" enterprise's public feelings information ", " law court's notice information ", " region accounting data ", " platform quantity ", " Platform Type accounting information ",
" industry deal message ", " investment number classification ", " classification of loaning bill number ", " different target prospective earnings ", " different maturity periods target
Mark full used time " etc..
Embodiment 2
A kind of mutual golden business ratings model based on deep neural network, comprises the following steps：
S1：Excavate mutually golden company information collection of illustrative plates；Crawler technology, NLP technologies etc. can be mainly applied to, data source is mainly
Ten thousand network datas, and I manage it inside assets life data.
S2：Excavate mutually golden enterprise operation collection of illustrative plates；Data source is mainly access, and I manages it the number runed in the true line for deposit pipe
According to, the data that mutual enterprise externally issues, ten thousand network data gold etc..
S3：Excavate Hu Jin enterprises public feelings information；This partial information mostlys come from major regulators, major news media
Data mining.
S4：Excavate Hu Jin enterprises and deposit pipe True Data；Hu Jin enterprises, which are linked into me, manages it and deposits after guard system, can produce substantial amounts of
True generation data, containing user basic information, target essential information, cash flow, cash flow, information flow etc., this partial information is needed
Emphasis is wanted deeply to excavate, weight is larger in model construction.
S5：Excavate other channel data of Hu Jin enterprises；Bought containing data exchange, data etc..
S6：Find out positive and negative sample data；The conduct of TOPN Hu Jin enterprises is selected in existing enterprise's scoring according to disclosed in online
Positive sample；According to having closed down at present, close down, negative public sentiment enterprise, business circumstance reciprocal data, it is comprehensive that to extract N family reciprocal mutual
Golden enterprise's negative sense sample.
S7：According to sample data, binding time sequence, policy implication, management data, public sentiment data etc. pass through CNN convolution
Neutral net extracts each enterprise characteristic group of the TOPN Hu Jin enterprises and inverse N Jia Hujin enterprises in S6.
Details as Follows for calculating：
Correspondence validity feature can be extracted from sample data by calculating；X can enter Mobile state with the change of sample size
Adjustment.
S8：Validity feature is calculated using the classification Forecasting Methodology based on cosine similarity and the side of shared weight is calculated
Method, step is as follows：
S81：Sample data is extracted, including based on mutual golden enterprise operation data, public sentiment situation, industrial and commercial data, target data,
Corresponding scoring preference is extracted, calculation formula is as follows：
S82：Each feature preferences combination of Hu Jin enterprises is expressed as feature preferences vector：
v1=(p1, p2, p3..., pn)
S83:It is vector by business ratings character representation：
v2=(1,0,1 ..., 0)
Characteristic vector dimension values value is 0 or 1.
S84：Calculate cosine similarity：
Wherein simi values are bigger represents more similar.
S85：It is ranked up according to simi values, takes the business data most like with positive negative sample to be ranked up
S9：Strong rule is used as with reference to different application scene, policy implication, blacklist, artificial bright-coloured knowledge etc..
S10：According to strong rule and model training, different mutual golden business ratings models are found out according to different demands；
S11：According to historical data, grading effect is verified, historical data is mutual being commented through enterprise in outside excavation
DBMS；
S12：According to different application scenarios, the corresponding Hu Jin enterprises application model of output.
Such as grading of enterprise operation performance assets, mutually business standing performance grading, golden enterprise investment value grading.
S13：Model needs moment supplementary data tuning, and artificial mark badcase methods carry out study tuning.
Here the method that machine learning can be used learns tuning automatically.
Claims (2)
1. a kind of mutual golden business ratings model based on deep neural network, it is characterised in that comprise the following steps：
S1：Excavate mutually golden company information data；
S2：Excavate mutually golden enterprise operation data；
S3：Excavate Hu Jin enterprises public feelings information；
S4：Excavate Hu Jin enterprises and deposit pipe True Data；
S5：Excavate other channel data of Hu Jin enterprises；
S6：Find out positive and negative sample data；Existing enterprise's scoring selects TOPN Hu Jin enterprises as forward direction according to disclosed in online
Sample；According to having closed down at present, close down, negative public sentiment enterprise, business circumstance reciprocal data, it is comprehensive to extract N family reciprocal mutually gold is looked forward to
Industry negative sense sample.
S7：According to sample data, with reference to the data excavated in S1-S5, the TOPN in S6 is extracted by CNN convolutional neural networks
Each enterprise characteristic group of Hu Jin enterprises and inverse N Jia Hujin enterprises；
Details as Follows for calculating：
Correspondence validity feature can be extracted from sample data by calculating；X can enter Mobile state adjustment with the change of sample size；
S8：By FNN forerunner's neutral net, validity feature is classified and scientific algorithm goes out shared weight；
In above formula, SjAnd θjRepresent validity feature vector, x=[x1, x2..., xn]TIt is input feature value, wjiIt is xiTo yj's
Connection weight, output quantity yj(j=1,2 ..., be m) classification results according to different characteristic, can calculate correspondence scoring, weight number
According to as score data；
S9：Strong rule is used as with reference to different application scene, policy implication, blacklist, artificial bright-coloured knowledge etc.；
S10：According to strong rule and model training, different mutual golden business ratings models are found out according to different demands；
S11：According to historical data, grading effect is verified, historical data is in the mutual through business ratings number of outside excavation
According to；
S12：According to different application scenarios, the corresponding Hu Jin enterprises application model of output；
Such as grading of enterprise operation performance assets, mutually business standing performance grading, golden enterprise investment value grading；
S13：Model needs moment supplementary data tuning, and artificial mark badcase methods carry out study tuning；
Here the method that machine learning can be used learns tuning automatically.
2. a kind of mutual golden business ratings model based on deep neural network according to claim 1, it is characterised in that S8
In, the classification Forecasting Methodology based on cosine similarity can also be used by calculating validity feature and calculating the method for shared weight,
Step is as follows：
S81：Sample data is extracted, including based on mutual golden enterprise operation data, public sentiment situation, industrial and commercial data, target data, is extracted
Corresponding scoring preference, calculation formula is as follows：
S82：Each feature preferences combination of Hu Jin enterprises is expressed as feature preferences vector：
v1=(p1, p2, p3..., pn)
S83:It is vector by business ratings character representation：
v2=(1,0,1 ..., 0)
Characteristic vector dimension values value is 0 or 1；
S84：Calculate cosine similarity：
Wherein simi values are bigger represents more similar.
S85：It is ranked up according to simi values, takes the business data most like with positive negative sample to be ranked up.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710361427.8A CN107301493A (en) | 2017-05-19 | 2017-05-19 | A kind of mutual golden business ratings model based on deep neural network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710361427.8A CN107301493A (en) | 2017-05-19 | 2017-05-19 | A kind of mutual golden business ratings model based on deep neural network |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107301493A true CN107301493A (en) | 2017-10-27 |
Family
ID=60137536
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710361427.8A Pending CN107301493A (en) | 2017-05-19 | 2017-05-19 | A kind of mutual golden business ratings model based on deep neural network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107301493A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107909274A (en) * | 2017-11-17 | 2018-04-13 | 平安科技（深圳）有限公司 | Enterprise investment methods of risk assessment, device and storage medium |
| CN107918644A (en) * | 2017-10-31 | 2018-04-17 | 北京锐思爱特咨询股份有限公司 | News subject under discussion analysis method and implementation system in reputation Governance framework |
| CN108073692A (en) * | 2017-12-06 | 2018-05-25 | 国云科技股份有限公司 | A kind of enterprise's ranking system and its implementation |
| CN109615232A (en) * | 2018-12-12 | 2019-04-12 | 税友软件集团股份有限公司 | A kind of method, system and the relevant apparatus of credit score prediction |
| CN114611972A (en) * | 2022-03-21 | 2022-06-10 | 广东贤能数字科技有限公司 | Merchant credit rating system and method based on artificial intelligence |
-
2017
- 2017-05-19 CN CN201710361427.8A patent/CN107301493A/en active Pending
Non-Patent Citations (1)
| Title |
|---|
| 朱智恒: "基于BP神经网络的互联网金融风险预警研究", 《中国优秀硕士学位论文全文数据库信息科技辑》 * |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107918644A (en) * | 2017-10-31 | 2018-04-17 | 北京锐思爱特咨询股份有限公司 | News subject under discussion analysis method and implementation system in reputation Governance framework |
| CN107918644B (en) * | 2017-10-31 | 2020-12-08 | 北京锐思爱特咨询股份有限公司 | News topic analysis method and implementation system in reputation management framework |
| CN107909274A (en) * | 2017-11-17 | 2018-04-13 | 平安科技（深圳）有限公司 | Enterprise investment methods of risk assessment, device and storage medium |
| CN107909274B (en) * | 2017-11-17 | 2023-02-28 | 平安科技（深圳）有限公司 | Enterprise investment risk assessment method and device and storage medium |
| CN108073692A (en) * | 2017-12-06 | 2018-05-25 | 国云科技股份有限公司 | A kind of enterprise's ranking system and its implementation |
| CN108073692B (en) * | 2017-12-06 | 2021-09-21 | 国云科技股份有限公司 | Method for implementing enterprise ranking system |
| CN109615232A (en) * | 2018-12-12 | 2019-04-12 | 税友软件集团股份有限公司 | A kind of method, system and the relevant apparatus of credit score prediction |
| CN114611972A (en) * | 2022-03-21 | 2022-06-10 | 广东贤能数字科技有限公司 | Merchant credit rating system and method based on artificial intelligence |
| CN114611972B (en) * | 2022-03-21 | 2023-01-10 | 广东贤能数字科技有限公司 | Merchant credit rating system and method based on artificial intelligence |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107301493A (en) | A kind of mutual golden business ratings model based on deep neural network | |
| Lee et al. | The double-edged effects of the corporate venture capital unit's structural autonomy on corporate investors' explorative and exploitative innovation | |
| Sharchilev et al. | Web-based startup success prediction | |
| Chen et al. | A new measure of disclosure quality: The level of disaggregation of accounting data in annual reports | |
| Moudud-Ul-Huq | The Role of Artificial Intelligence in the Development of Accounting Systems: A Review. | |
| Kogut et al. | Real options pricing and organizations: The contingent risks of extended theoretical domains | |
| CN107464037A (en) | Enterprise's portrait method and system based on multi objective dimensional model | |
| CN107563630A (en) | Enterprise's methods of marking and its system based on various dimensions | |
| CN109859052A (en) | A kind of intelligent recommendation method, apparatus, storage medium and the server of investment tactics | |
| CN110096631B (en) | Stock market emotion report generation method based on postings text analysis of stock forum | |
| AU2019204988B2 (en) | Determination of a response to a query | |
| CN115456584A (en) | Similar JD recall and recommendation method based on deep learning model and expert system | |
| Engström et al. | Succeeding at home and abroad: accounting for the international spillovers of cities’ SDG actions | |
| CN108446291A (en) | The real-time methods of marking and points-scoring system of user credit | |
| CN107481135A (en) | A kind of personal credit evaluation method and system based on BP neural network | |
| Wu et al. | A TOPSIS data mining demonstration and application to credit scoring | |
| KR20230089513A (en) | Investment attraction flatform system based on artificial intelligence | |
| Huynh et al. | Stock price prediction leveraging reddit: The role of trust filter and sliding window | |
| Li et al. | Application of XGBoost in P2P default prediction | |
| TWM583089U (en) | Smart credit risk assessment system | |
| Fatima et al. | Current state and future directions of sustainability and innovation in finance: a bibliometric review | |
| Füss et al. | Investing in your alumni: endowments’ investment choices in private equity | |
| CN108009927A (en) | One B shareB methods of marking and platform | |
| Zhang | RETRACTED ARTICLE: COVID-19 Forecast and Bank Credit Decision Model Based on BiLSTM-Attention Network | |
| Lu et al. | Enhancing Financial Risk Prediction for Listed Companies: A Catboost-Based Ensemble Learning 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 |
Application publication date: 20171027 |
|
| RJ01 | Rejection of invention patent application after publication |