CN111967790B - Credit scoring method capable of automatically calculating and terminal - Google Patents
Credit scoring method capable of automatically calculating and terminal Download PDFInfo
- Publication number
- CN111967790B CN111967790B CN202010882083.7A CN202010882083A CN111967790B CN 111967790 B CN111967790 B CN 111967790B CN 202010882083 A CN202010882083 A CN 202010882083A CN 111967790 B CN111967790 B CN 111967790B
- Authority
- CN
- China
- Prior art keywords
- data
- credit
- normative
- cleaning
- credit data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000013077 scoring method Methods 0.000 title claims description 9
- 238000004140 cleaning Methods 0.000 claims abstract description 117
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 107
- 238000001914 filtration Methods 0.000 claims abstract description 18
- 238000012795 verification Methods 0.000 claims abstract description 17
- 238000000034 method Methods 0.000 claims abstract description 9
- 238000012545 processing Methods 0.000 claims description 30
- 238000012986 modification Methods 0.000 claims description 19
- 230000004048 modification Effects 0.000 claims description 19
- 238000004364 calculation method Methods 0.000 claims description 10
- 238000012937 correction Methods 0.000 claims description 10
- 238000004590 computer program Methods 0.000 claims description 8
- 238000012544 monitoring process Methods 0.000 claims description 5
- 230000009286 beneficial effect Effects 0.000 description 3
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 239000003814 drug Substances 0.000 description 1
- 229940079593 drug Drugs 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000009499 grossing Methods 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/067—Enterprise or organisation modelling
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/21—Design, administration or maintenance of databases
- G06F16/215—Improving data quality; Data cleansing, e.g. de-duplication, removing invalid entries or correcting typographical errors
-
- 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
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/26—Government or public services
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Strategic Management (AREA)
- Human Resources & Organizations (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Tourism & Hospitality (AREA)
- Physics & Mathematics (AREA)
- Economics (AREA)
- Quality & Reliability (AREA)
- General Business, Economics & Management (AREA)
- Development Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Educational Administration (AREA)
- Marketing (AREA)
- Databases & Information Systems (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- General Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Health & Medical Sciences (AREA)
- Game Theory and Decision Science (AREA)
- Operations Research (AREA)
- Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)
Abstract
The invention relates to a credit score method and a terminal capable of automatically calculating, wherein a credit information list of each province is obtained according to a state-level credit information directory, original credit data of the credit information list is obtained, a first single-word-segment weight of the original credit data is configured, a first algorithm model is determined according to the first single-word-segment weight, the original credit data is substituted into the first algorithm model to obtain a first credit score, an actual credit score and actual policy information matched with the original credit data are obtained, the first algorithm model is corrected according to the actual credit score, the first credit score and the actual policy information to obtain a normative algorithm model, and the normative algorithm model is stored; cleaning and filtering the original credit data according to a preset rule to obtain cleaning credit data, verifying the cleaning credit data, and if verification is successful, taking the cleaning credit data as normative credit data; and calculating the normative credit data according to the normative algorithm model to obtain the normative credit score.
Description
Technical Field
The invention relates to the field of computer software, in particular to a credit scoring method capable of automatically calculating and a terminal.
Background
The collection of the corporate credit data is an important function in a corporate credit platform, and the collection of the corporate credit data is mainly used for pertinently extracting corporate credit data of different business types of 10 units of a city industry and commerce bureau, a city tax bureau, a city quality supervision bureau, a city food and drug supervision bureau, a city traffic bureau, a city government bureau, a city public security bureau, a city court and a city inspection and quarantine bureau.
The legal credit data acquisition is realized by relying on public information service platforms and public basic databases (hereinafter referred to as 'one database for short') of Fujian province and the province, the public information platforms are built one by one according to planning, the public information platforms are mainly responsible for gathering data of all committees and offices of various cities, meanwhile, the public basic databases, public services and the public service databases are built, and finally an information resource pool facing city sharing coordination is formed.
Due to various objective factors, credit data reported in various cities often have regional characteristics, collected data types, data dimensions, data frequency and the like can change due to specific conditions of the cities, and corresponding credit scores cannot be calculated by using the same algorithm model; in addition, aiming at the problems of ambiguous data, incompleteness, violation of business rules and the like which may occur in the public credit information of the legal person and the natural person reported in various places, the reasonable credit score cannot be calculated according to the collected credit data of the legal person.
Disclosure of Invention
Technical problem to be solved
In order to solve the above problems in the prior art, the present invention provides an automatic calculation credit score method, which can establish a reasonable credit score algorithm model to ensure the reasonability of the calculated credit score.
(II) technical scheme
In order to achieve the purpose, the invention adopts a technical scheme that: an automatically calculable credit scoring method, comprising:
s1, obtaining a credit information list of each province according to a state-level credit information directory, obtaining original credit data of the credit information list, configuring first single-word-segment weights of the original credit data, determining a first algorithm model according to the first single-word-segment weights, substituting the original credit data into the first algorithm model to obtain a first credit score, obtaining actual credit scores and actual policy information matched with the original credit data, correcting the first algorithm model according to the actual credit scores, the first credit score and the actual policy information to obtain a normative algorithm model, and storing the normative algorithm model;
s2, cleaning and filtering the original credit data according to a preset rule to obtain cleaning credit data, verifying the cleaning credit data according to an integrity rule, a uniqueness rule, a consistency rule, a legality rule and an authority rule, and taking the cleaning credit data as normative credit data if verification is successful;
and S3, calculating the normative credit data according to the normative algorithm model to obtain the normative credit score.
The other technical scheme adopted by the invention is as follows: an automatically calculable credit terminal comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the following steps when executing the computer program:
s1, obtaining a credit information list of each province according to a state-level credit information directory, obtaining original credit data of the credit information list, configuring first single-word-segment weights of the original credit data, determining a first algorithm model according to the first single-word-segment weights, substituting the original credit data into the first algorithm model to obtain a first credit score, obtaining actual credit scores and actual policy information matched with the original credit data, correcting the first algorithm model according to the actual credit scores, the first credit score and the actual policy information to obtain a normative algorithm model, and storing the normative algorithm model;
s2, cleaning and filtering the original credit data according to a preset rule to obtain cleaning credit data, verifying the cleaning credit data according to an integrity rule, a uniqueness rule, a consistency rule, a legality rule and an authority rule, and taking the cleaning credit data as normative credit data if verification is successful;
and S3, calculating the normative credit data according to the normative algorithm model to obtain the normative credit score.
(III) advantageous effects
The invention has the beneficial effects that: configuring a first single-word-segment weight through original credit data to determine a first algorithm model, and correcting the first algorithm model by combining actual credit score and actual policy information to obtain a normative algorithm model, so that the accuracy and reliability of the normative algorithm model are ensured; in addition, the original credit data are cleaned and verified successfully to obtain the normative credit data, so that the problems of ambiguity, incompleteness, violation of business rules and the like of the original credit data are avoided, the validity of the data is guaranteed, and the rationality of the normative credit data can be guaranteed through the normative credit data and the normative credit score obtained by the normative algorithm model.
Drawings
FIG. 1 is a flow chart of an automatically calculable credit scoring method of the present invention;
FIG. 2 is a schematic diagram of a credit point terminal capable of automatic calculation according to the present invention;
[ description of reference ]
1. A credit sub-terminal capable of automatic calculation; 2. a memory; 3. a processor.
Detailed Description
For the purpose of better explaining the present invention and to facilitate understanding, the present invention will be described in detail by way of specific embodiments with reference to the accompanying drawings.
Referring to fig. 1, a credit scoring method capable of automatic calculation includes:
s1, obtaining a credit information list of each province according to a state-level credit information directory, obtaining original credit data of the credit information list, configuring first single-word-segment weights of the original credit data, determining a first algorithm model according to the first single-word-segment weights, substituting the original credit data into the first algorithm model to obtain a first credit score, obtaining actual credit scores and actual policy information matched with the original credit data, correcting the first algorithm model according to the actual credit scores, the first credit score and the actual policy information to obtain a normative algorithm model, and storing the normative algorithm model;
s2, cleaning and filtering the original credit data according to a preset rule to obtain cleaning credit data, verifying the cleaning credit data according to an integrity rule, a uniqueness rule, a consistency rule, a legality rule and an authority rule, and taking the cleaning credit data as normative credit data if verification is successful;
and S3, calculating the normative credit data according to the normative algorithm model to obtain the normative credit score.
From the above description, the beneficial effects of the present invention are: configuring a first single-word-segment weight through original credit data to determine a first algorithm model, and modifying the first algorithm model by combining actual credit score and actual policy information to obtain a normative algorithm model, so that the accuracy and reliability of the normative algorithm model can be ensured according to actual conditions; in addition, the original credit data are cleaned and verified successfully to obtain the normative credit data, so that the problems of ambiguity, incompleteness, violation of business rules and the like of the original credit data are avoided, the validity of the data is guaranteed, and the rationality of the normative credit data can be guaranteed through the normative credit data and the normative credit score obtained by the normative algorithm model.
Further, the S3 includes before:
judging whether the normative credit data is missing or not, if so, sending the normative credit data to a modification terminal, configuring a second single-word segment weight of the normative credit data, determining a second algorithm model according to the second single-word segment weight, calculating the normative credit data according to the second algorithm model to obtain a second credit score, sending the second credit score to an application end, returning a modification suggestion returned by the application end, and modifying the normative algorithm model according to the modification suggestion and the second single-word segment weight to obtain a final normative algorithm model.
The S3 specifically comprises the following steps:
and S3, calculating the normative credit data according to the final normative algorithm model to obtain the normative credit score.
As can be seen from the above description, because there is a difference in data amount between the original credit data of the credit information lists of each city and each city, there may be a situation that some of the original credit data of the legal person in the city and some of the original credit data may have partial data missing, and the original credit data cannot be calculated by using the general normative algorithm model, when the normative credit data are missing, a second single-field weight needs to be configured correspondingly to obtain a second algorithm model, and the second algorithm model is used to obtain a second credit score for the normative credit data, and the second algorithm model needs to be corrected according to a modification opinion returned by an application end (for example, when the second algorithm model is used by a bank in the city to calculate the normative credit data missing in the city and the second credit score is not within a reasonable range, the modification opinion is sent), so as to obtain a final normative algorithm model, thereby ensuring reliability of the algorithm model.
Further, the S2 further includes:
if the verification is unsuccessful, analyzing the cleaning credit data, and judging whether the cleaning credit data is data capable of automatically correcting errors;
if so, automatically correcting the cleaning credit data to obtain error correction data, and taking the error correction data as normative credit data;
if not, classifying and storing according to the zone attribution of the cleaning credit data, and simultaneously storing the error description and the error data field of the cleaning credit data.
As can be seen from the above description, after the verification of the cleaning credit data is unsuccessful, whether the data is automatically error-correctable or not is analyzed, if so, the cleaning credit data can be automatically error-corrected and then stored, the degree of automation is high, and the manpower consumption is reduced, and if not, the cleaning credit data is classified and stored according to the division attribution of the cleaning credit data (namely, the cleaning credit data is classified according to the specific city), and the error description and the error data field of the cleaning credit data are simultaneously stored, so that the cleaning credit data with problems can be classified, and the manual verification of the cleaning credit data by a worker can be facilitated according to the error description and the error data field of the cleaning credit data.
If not, classifying and storing according to the zone attribution of the cleaning credit data, and simultaneously storing the error description and the error data field of the cleaning credit data, the method further comprises the following steps:
and acquiring processing data obtained by processing the cleaning credit data according to the error description and the error data field of the cleaning credit data, and taking the processing data as normative credit data if the processing data is successfully processed.
From the above description, it can be known that, when the processed data after manual processing is acquired, whether the processing is successful or not is judged, and the processed data is used as normative credit data after the processing is successful, so that the data quality is ensured.
Further, the preset rules comprise basic preset rules and newly added preset rules;
counting and monitoring the data quality and the data quantity of cleaning credit data obtained by cleaning and filtering the original credit data according to a preset rule, respectively obtaining the front data quality, the front data quantity, the rear data quality and the rear data quantity of the cleaning credit data obtained by cleaning and filtering the original credit data within a preset time period before and after adding the newly added preset rule in a basic preset rule, respectively comparing the front data quality and the front data quantity with the rear data quality and the rear data quantity to obtain a comparison result, and judging whether to alarm or not according to the comparison result.
From the above description, after the preset rule is newly added to the basic preset rule, the change of the subsequent data quality and data quantity is counted to judge whether to alarm, so that the preset rule is convenient to adjust.
Referring to fig. 2, a credit score method capable of automatic calculation and a terminal thereof include a memory, a processor and a computer program stored in the memory and running on the processor, wherein the processor implements the following steps when executing the computer program:
s1, obtaining a credit information list of each province according to a state-level credit information directory, obtaining original credit data of the credit information list, configuring first single-word-segment weights of the original credit data, determining a first algorithm model according to the first single-word-segment weights, substituting the original credit data into the first algorithm model to obtain a first credit score, obtaining actual credit scores and actual policy information matched with the original credit data, correcting the first algorithm model according to the actual credit scores, the first credit score and the actual policy information to obtain a normative algorithm model, and storing the normative algorithm model;
s2, cleaning and filtering the original credit data according to a preset rule to obtain cleaning credit data, verifying the cleaning credit data according to an integrity rule, a uniqueness rule, a consistency rule, a legality rule and an authority rule, and taking the cleaning credit data as normative credit data if verification is successful;
and S3, calculating the normative credit data according to the normative algorithm model to obtain the normative credit score.
From the above description, the beneficial effects of the present invention are: configuring a first single-word-segment weight through original credit data to determine a first algorithm model, and modifying the first algorithm model by combining actual credit score and actual policy information to obtain a normative algorithm model, so that the accuracy and reliability of the normative algorithm model are guaranteed; in addition, the original credit data are cleaned and verified successfully to obtain the normative credit data, so that the problems of ambiguity, incompleteness, violation of business rules and the like of the original credit data are avoided, the validity of the data is guaranteed, and the rationality of the normative credit data can be guaranteed through the normative credit data and the normative credit score obtained by the normative algorithm model.
Further, the S3 includes before:
judging whether the normative credit data is missing or not, if so, sending the normative credit data to a modification terminal, configuring a second single-word segment weight of the normative credit data, determining a second algorithm model according to the second single-word segment weight, calculating the normative credit data according to the second algorithm model to obtain a second credit score, sending the second credit score to an application end, returning a modification suggestion returned by the application end, and modifying the normative algorithm model according to the modification suggestion and the second single-word segment weight to obtain a final normative algorithm model.
The S3 specifically comprises the following steps:
and S3, calculating the normative credit data according to the final normative algorithm model to obtain the normative credit score.
As can be seen from the above description, because there is a difference in data amount between the original credit data of the credit information lists of each city and each city, there may be a situation that some of the original credit data of the legal person in the city and some of the original credit data may have partial data missing, and the original credit data cannot be calculated by using the general normative algorithm model, when the normative credit data are missing, a second single-field weight needs to be configured correspondingly to obtain a second algorithm model, and the second algorithm model is used to obtain a second credit score for the normative credit data, and the second algorithm model needs to be corrected according to a modification opinion returned by an application end (for example, when the second algorithm model is used by a bank in the city to calculate the normative credit data missing in the city and the second credit score is not within a reasonable range, the modification opinion is sent), so as to obtain a final normative algorithm model, thereby ensuring reliability of the algorithm model.
Further, the S2 further includes:
if the verification is unsuccessful, analyzing the cleaning credit data, and judging whether the cleaning credit data is data capable of automatically correcting errors;
if so, automatically correcting the cleaning credit data to obtain error correction data, and taking the error correction data as normative credit data;
if not, classifying and storing according to the zone attribution of the cleaning credit data, and simultaneously storing the error description and the error data field of the cleaning credit data.
As can be seen from the above description, after the verification of the cleaning credit data is unsuccessful, whether the data is automatically error-correctable or not is analyzed, if so, the cleaning credit data can be automatically error-corrected and then stored, the degree of automation is high, and the manpower consumption is reduced, and if not, the cleaning credit data is classified and stored according to the division attribution of the cleaning credit data (namely, the cleaning credit data is classified according to the specific city), and the error description and the error data field of the cleaning credit data are simultaneously stored, so that the cleaning credit data with problems can be classified, and the manual verification of the cleaning credit data by a worker can be facilitated according to the error description and the error data field of the cleaning credit data.
If not, classifying and storing according to the zone attribution of the cleaning credit data, and simultaneously storing the error description and the error data field of the cleaning credit data, the method further comprises the following steps:
and acquiring processing data obtained by processing the cleaning credit data according to the error description and the error data field of the cleaning credit data, and if the processing data is successfully processed, taking the processing data as normative credit data.
From the above description, it can be known that, when the processed data after manual processing is acquired, whether the processing is successful or not is judged, and the processed data is used as normative credit data after the processing is successful, so that the data quality is ensured.
Further, the preset rules comprise basic preset rules and newly added preset rules;
counting and monitoring the data quality and the data quantity of cleaning credit data obtained by cleaning and filtering the original credit data according to a preset rule, respectively obtaining the front data quality, the front data quantity, the rear data quality and the rear data quantity of the cleaning credit data obtained by cleaning and filtering the original credit data within a preset time period before and after adding the newly added preset rule in a basic preset rule, respectively comparing the front data quality and the front data quantity with the rear data quality and the rear data quantity to obtain a comparison result, and judging whether to alarm or not according to the comparison result.
From the above description, after the preset rule is newly added to the basic preset rule, the change of the subsequent data quality and data quantity is counted to judge whether to alarm, so that the preset rule is convenient to adjust.
Example one
Referring to fig. 1, a credit scoring method capable of automatic calculation includes:
s1, obtaining a credit information list of each province according to a state-level credit information directory, obtaining original credit data of the credit information list, configuring first single-word-segment weights of the original credit data, determining a first algorithm model according to the first single-word-segment weights, substituting the original credit data into the first algorithm model to obtain a first credit score, obtaining actual credit scores and actual policy information matched with the original credit data, correcting the first algorithm model according to the actual credit scores, the first credit score and the actual policy information to obtain a normative algorithm model, and storing the normative algorithm model;
s2, cleaning and filtering the original credit data according to a preset rule to obtain cleaning credit data, verifying the cleaning credit data according to an integrity rule, a uniqueness rule, a consistency rule, a legality rule and an authority rule, and taking the cleaning credit data as normative credit data if verification is successful;
and S3, calculating the normative credit data according to the normative algorithm model to obtain the normative credit score.
Specifically, acquiring the original credit data of the credit information list comprises: and acquiring original credit data of the credit information list, and formatting the original credit data, so that the disordered format is avoided after format conversion, and the uniformity of the data format is ensured.
Specifically, the following preset rules may be configured for the integrity rule, the uniqueness rule, the consistency rule, the validity rule, and the authority rule, respectively:
1. the preset rules for completeness, such as partial data missing in the original credit data, can be complemented by the following preset rules:
(1) Completing by other information, for example, the original credit data lacks the administrative division code, and automatically filling the administrative division code according to the administrative division based on the administrative division code comparison table; if the time format is not correct, the error can be automatically corrected according to the time bits in the code.
(2) Through the completion of the previous and subsequent data, for example, the time series in the original credit data has short data, the average value before and after the data is used, the short data is more, the data can be omitted by using the smoothing treatment, and the cleaning of other data is not influenced.
2. The preset rules aiming at the legality rules can be processed through the following preset rules:
(1) Setting a forced legal rule, and forcibly setting the rule to be a maximum value or judging the rule to be invalid and rejecting the rule if the rule is not in the rule range;
(2) Field type legal rules: the date field format is "2010-10-10";
(3) The legal rule of the field content is as follows: the sex of the corporate legal person is 'male or female or unknown'; birth date < = today;
(4) Setting a warning rule, warning if the rule is not in the rule range, and then manually processing;
(5) And (4) carrying out manual special treatment on the outlier, and finding the outlier by using modes of box separation, clustering, regression and the like.
3. The preset rule for data uniqueness, i.e. removing duplicate records, only keeps one, for example:
(1) Pressing a main key to remove duplicate, and triggering an instruction of 'removing duplicate records' in the sql or excel;
(2) And (4) removing duplication according to rules, and compiling a corresponding series of rules according to complex data with repeated conditions to remove duplication. For example, data from different channels can be matched through the same key information, and the duplication is removed through combination.
4. Aiming at the preset rule of data consistency, for example, corresponding names in each table are kept consistent, for example, the existence names of the household locations in the same type of information submitted in each city are different, some are the household locations, and other are the household mouths and the native place, and the names are cleaned and filtered uniformly to be the household locations; for example, in the same type of information filling form submitted by each city, if the name of "the xi information filling form of the province of Fujian province" is different, if some are "xx information filling forms of the province of Fujian province", and if some are "xx information filling forms of the Fujian province", these names are cleaned and filtered uniformly to be "xx information filling forms of the province of Fujian province".
5. According to the preset rule of data authority, original credit data come from different channels, different authority levels are set for the different channels, and when the original credit data input by the channel with lower authority level and the original credit data input by the channels with higher authority level are the same group of data but have differences, the original credit data input by the channel with lower authority level are cleaned and filtered.
Wherein S3 is preceded by:
judging whether the normative credit data is missing or not, if so, sending the normative credit data to a modification terminal, configuring a second single-word segment weight of the normative credit data, determining a second algorithm model according to the second single-word segment weight, calculating the normative credit data according to the second algorithm model to obtain a second credit score, sending the second credit score to an application end, returning a modification suggestion returned by the application end, and modifying the normative algorithm model according to the modification suggestion and the second single-word segment weight to obtain a final normative algorithm model.
The S3 specifically comprises the following steps:
and S3, calculating the normative credit data according to the final normative algorithm model to obtain the normative credit score.
Wherein the S2 further comprises:
if the verification is unsuccessful, analyzing the cleaning credit data, and judging whether the cleaning credit data is data capable of automatically correcting errors;
if so, automatically correcting the cleaning credit data to obtain error correction data, and taking the error correction data as normative credit data;
if not, classifying and storing according to the zone attribution of the cleaning credit data, and simultaneously storing the error description and the error data field of the cleaning credit data.
If not, classifying and storing according to the zone attribution of the cleaning credit data, and simultaneously storing the error description and the error data field of the cleaning credit data, and then the method further comprises the following steps:
acquiring processing data obtained by processing the cleaning credit data according to the error description and the error data field of the cleaning credit data, and taking the processing data as normative credit data if the processing data is successfully processed; if the processing data processing is unsuccessful, returning to the step of storing the error description and the error data field of the cleaning credit data to wait for the next data processing.
The preset rules comprise basic preset rules and newly added preset rules;
counting and monitoring the data quality and the data quantity of cleaning credit data obtained by cleaning and filtering the original credit data according to a preset rule, respectively obtaining the front data quality, the front data quantity, the rear data quality and the rear data quantity of the cleaning credit data obtained by cleaning and filtering the original credit data within a preset time period before and after adding the newly added preset rule in a basic preset rule, respectively comparing the front data quality and the front data quantity with the rear data quality and the rear data quantity to obtain a comparison result, and judging whether to alarm or not according to the comparison result.
Example two
Referring to fig. 2, a credit terminal 1 capable of automatic calculation includes a memory 2, a processor 3 and a computer program stored on the memory 2 and capable of running on the processor 3, wherein the processor 3 implements the steps of the first embodiment when executing the computer program.
In summary, according to the credit score method and the terminal capable of automatic calculation provided by the invention, the first algorithm model is determined by configuring the first single-field weight through the original credit data, and the first algorithm model is corrected by combining the actual credit score and the actual policy information to obtain the normative algorithm model, so that the accuracy and reliability of the normative algorithm model can be ensured according to the actual situation; in addition, the original credit data are cleaned and verified successfully to obtain the normative credit data, so that the problems of ambiguity, incompleteness, violation of business rules and the like of the original credit data are avoided, the validity of the data is guaranteed, and the rationality of the normative credit data can be guaranteed through the normative credit data and the normative credit score obtained by the normative algorithm model.
The above description is only an embodiment of the present invention, and not intended to limit the scope of the present invention, and all equivalent changes made by using the contents of the present specification and the drawings, or applied directly or indirectly to the related technical fields, are included in the scope of the present invention.
Claims (4)
1. An automatically calculable credit scoring method, comprising:
s1, obtaining a credit information list of each province according to a state-level credit information directory, obtaining original credit data of the credit information list, configuring first single-word-segment weights of the original credit data, determining a first algorithm model according to the first single-word-segment weights, substituting the original credit data into the first algorithm model to obtain a first credit score, obtaining actual credit scores and actual policy information matched with the original credit data, correcting the first algorithm model according to the actual credit scores, the first credit score and the actual policy information to obtain a normative algorithm model, and storing the normative algorithm model;
s2, cleaning and filtering the original credit data according to a preset rule to obtain cleaning credit data, verifying the cleaning credit data according to an integrity rule, a uniqueness rule, a consistency rule, a legality rule and an authority rule, and taking the cleaning credit data as normative credit data if verification is successful;
judging whether the normative credit data is missing or not, if so, sending the normative credit data to a modification terminal, configuring a second single-word segment weight of the normative credit data, determining a second algorithm model according to the second single-word segment weight, calculating the normative credit data according to the second algorithm model to obtain a second credit score, sending the second credit score to an application end, returning a modification suggestion returned by the application end, and modifying the normative algorithm model according to the modification suggestion and the second single-word segment weight to obtain a final normative algorithm model;
s3, calculating normative credit data according to the final normative algorithm model to obtain normative credit scores;
the S2 further comprises:
if the verification is unsuccessful, analyzing the cleaning credit data, and judging whether the cleaning credit data is data capable of automatically correcting errors;
if so, automatically correcting the cleaning credit data to obtain error correction data, and taking the error correction data as normative credit data;
if not, classifying and storing according to the region attribution of the cleaning credit data, and simultaneously storing the error description and the error data field of the cleaning credit data;
if not, classifying and storing according to the zone attribution of the cleaning credit data, and simultaneously storing the error description and the error data field of the cleaning credit data, and then the method further comprises the following steps:
and acquiring processing data obtained by processing the cleaning credit data according to the error description and the error data field of the cleaning credit data, and taking the processing data as normative credit data if the processing data is successfully processed.
2. The automatically calculable credit scoring method according to claim 1, wherein the preset rules include basic preset rules and additional preset rules;
counting and monitoring the data quality and the data quantity of cleaning credit data obtained by cleaning and filtering the original credit data according to a preset rule, respectively obtaining the front data quality, the front data quantity, the rear data quality and the rear data quantity of the cleaning credit data obtained by cleaning and filtering the original credit data within a preset time period before and after adding the newly added preset rule in a basic preset rule, respectively comparing the front data quality and the front data quantity with the rear data quality and the rear data quantity to obtain a comparison result, and judging whether to alarm or not according to the comparison result.
3. An automatically calculable credit terminal comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor when executing the computer program implements the steps of:
s1, obtaining a credit information list of each province according to a state-level credit information directory, obtaining original credit data of the credit information list, configuring first single-word-segment weights of the original credit data, determining a first algorithm model according to the first single-word-segment weights, substituting the original credit data into the first algorithm model to obtain a first credit score, obtaining actual credit scores and actual policy information matched with the original credit data, correcting the first algorithm model according to the actual credit scores, the first credit score and the actual policy information to obtain a normative algorithm model, and storing the normative algorithm model;
s2, cleaning and filtering the original credit data according to a preset rule to obtain cleaning credit data, verifying the cleaning credit data according to an integrity rule, a uniqueness rule, a consistency rule, a legality rule and an authority rule, and taking the cleaning credit data as normative credit data if verification is successful;
judging whether the normative credit data is missing or not, if so, sending the normative credit data to a modification terminal, configuring a second single-word segment weight of the normative credit data, determining a second algorithm model according to the second single-word segment weight, calculating the normative credit data according to the second algorithm model to obtain a second credit score, sending the second credit score to an application end, returning a modification suggestion returned by the application end, and modifying the normative algorithm model according to the modification suggestion and the second single-word segment weight to obtain a final normative algorithm model;
s3, calculating normative credit data according to the final normative algorithm model to obtain normative credit scores;
the S2 further comprises:
if the verification is unsuccessful, analyzing the cleaning credit data, and judging whether the cleaning credit data is data capable of automatically correcting errors;
if so, automatically correcting the cleaning credit data to obtain error correction data, and taking the error correction data as normative credit data;
if not, classifying and storing according to the region attribution of the cleaning credit data, and simultaneously storing the error description and the error data field of the cleaning credit data;
if not, classifying and storing according to the zone attribution of the cleaning credit data, and simultaneously storing the error description and the error data field of the cleaning credit data, and then the method further comprises the following steps:
and acquiring processing data obtained by processing the cleaning credit data according to the error description and the error data field of the cleaning credit data, and taking the processing data as normative credit data if the processing data is successfully processed.
4. The credit terminal capable of automatic calculation according to claim 3, wherein the preset rules include basic preset rules and new preset rules;
counting and monitoring the data quality and the data quantity of cleaning credit data obtained by cleaning and filtering the original credit data according to a preset rule, respectively obtaining the front data quality, the front data quantity, the back data quality and the back data quantity of the cleaning credit data obtained by cleaning and filtering the original credit data in a preset time period before and after adding the newly added preset rule in a basic preset rule, respectively comparing the front data quality and the front data quantity with the back data quality and the back data quantity to obtain a comparison result, and judging whether to alarm or not according to the comparison result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010882083.7A CN111967790B (en) | 2020-08-28 | 2020-08-28 | Credit scoring method capable of automatically calculating and terminal |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010882083.7A CN111967790B (en) | 2020-08-28 | 2020-08-28 | Credit scoring method capable of automatically calculating and terminal |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111967790A CN111967790A (en) | 2020-11-20 |
| CN111967790B true CN111967790B (en) | 2023-04-07 |
Family
ID=73401072
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010882083.7A Active CN111967790B (en) | 2020-08-28 | 2020-08-28 | Credit scoring method capable of automatically calculating and terminal |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111967790B (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2017148269A1 (en) * | 2016-02-29 | 2017-09-08 | 阿里巴巴集团控股有限公司 | Method and apparatus for acquiring score credit and outputting feature vector value |
| CN107545356A (en) * | 2017-06-29 | 2018-01-05 | 惠国征信服务股份有限公司 | Personal social credibility methods of marking based on government data and application |
| CN108256993A (en) * | 2017-12-29 | 2018-07-06 | 浪潮天元通信信息系统有限公司 | A kind of credit score appraisal procedure and credit score Evaluation Platform |
| CN108564286A (en) * | 2018-04-19 | 2018-09-21 | 天合泽泰（厦门）征信服务有限公司 | A kind of artificial intelligence finance air control credit assessment method and system based on big data reference |
| CN109636181A (en) * | 2018-12-11 | 2019-04-16 | 北京首汽智行科技有限公司 | A kind of user credit divides calculation method and system |
| CN110264340A (en) * | 2019-06-12 | 2019-09-20 | 重庆无界领智普惠商务信息咨询有限公司 | A kind of P2P net loan customers' credit methods of marking and system based on machine learning |
-
2020
- 2020-08-28 CN CN202010882083.7A patent/CN111967790B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2017148269A1 (en) * | 2016-02-29 | 2017-09-08 | 阿里巴巴集团控股有限公司 | Method and apparatus for acquiring score credit and outputting feature vector value |
| CN107545356A (en) * | 2017-06-29 | 2018-01-05 | 惠国征信服务股份有限公司 | Personal social credibility methods of marking based on government data and application |
| CN108256993A (en) * | 2017-12-29 | 2018-07-06 | 浪潮天元通信信息系统有限公司 | A kind of credit score appraisal procedure and credit score Evaluation Platform |
| CN108564286A (en) * | 2018-04-19 | 2018-09-21 | 天合泽泰（厦门）征信服务有限公司 | A kind of artificial intelligence finance air control credit assessment method and system based on big data reference |
| CN109636181A (en) * | 2018-12-11 | 2019-04-16 | 北京首汽智行科技有限公司 | A kind of user credit divides calculation method and system |
| CN110264340A (en) * | 2019-06-12 | 2019-09-20 | 重庆无界领智普惠商务信息咨询有限公司 | A kind of P2P net loan customers' credit methods of marking and system based on machine learning |
Non-Patent Citations (2)
| Title |
|---|
| 基于优化CBR的个人信用评分研究;姜明辉等;《中国软件学》;20141231(第12期);全文 * |
| 基于数据的信用评级处理和分析系统的设计与实现;程冠皓;《中国优秀硕士学位论文全文数据库（电子期刊）信息科技辑》;20160215(第02期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111967790A (en) | 2020-11-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US8316114B2 (en) | Updating IP address database based on transaction information | |
| CN107832444B (en) | Event discovery method and device based on search log | |
| CN107909516A (en) | A kind of problem source of houses recognition methods and system | |
| CN113420026A (en) | Database table structure changing method, device, equipment and storage medium | |
| CN113010903A (en) | Catering industry oil smoke online monitoring method and system based on block chain | |
| CN113535689A (en) | Database establishment management system for digital therapy | |
| CN116012178A (en) | Automatic financial accounting method based on artificial intelligence | |
| CN111428095A (en) | Graph data quality verification method and graph data quality verification device | |
| CN111967790B (en) | Credit scoring method capable of automatically calculating and terminal | |
| US20070130316A1 (en) | System and method to generate hosting company statistics | |
| CN117611008A (en) | Data quality evaluation method and device | |
| CN111125069A (en) | Data cleaning and fusing system | |
| CN114049016A (en) | Index similarity judgment method, system, terminal device and computer storage medium | |
| CN113918705A (en) | Contribution auditing method and system with early warning and recommendation functions | |
| CN116319815B (en) | Cloud data placement policy management system introducing SaaS features | |
| EP2776954A1 (en) | A method of analysing data | |
| CN115330497A (en) | Data processing method, system and computer readable storage medium | |
| CN111966675A (en) | Fixed asset investment project data cleaning method and terminal | |
| CN111413952A (en) | Robot fault detection method and device, electronic equipment and readable storage medium | |
| CN113553336B (en) | Talent resume database updating system and method based on Internet | |
| CN112162991B (en) | Intelligent management method and device for data | |
| CN112232962A (en) | Transaction index processing method, device and equipment | |
| CN111632384B (en) | Game online number detection method, device, equipment and storage medium | |
| WO2021129024A1 (en) | Method for recognizing state of communication device, and communication system and storage medium | |
| CN112199396A (en) | Industrial Internet identification query method and system facing MES |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |