CN110633316A - Multi-scene fusion double-random market supervision method - Google Patents
Multi-scene fusion double-random market supervision method Download PDFInfo
- Publication number
- CN110633316A CN110633316A CN201910745543.9A CN201910745543A CN110633316A CN 110633316 A CN110633316 A CN 110633316A CN 201910745543 A CN201910745543 A CN 201910745543A CN 110633316 A CN110633316 A CN 110633316A
- Authority
- CN
- China
- Prior art keywords
- supervision
- data
- market
- library
- scene
- 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
Images
Classifications
-
- 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
- G06F16/25—Integrating or interfacing systems involving database management systems
- G06F16/254—Extract, transform and load [ETL] procedures, e.g. ETL data flows in data warehouses
-
- 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
- G06F16/25—Integrating or interfacing systems involving database management systems
- G06F16/258—Data format conversion from or to a database
-
- 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
- G06F16/28—Databases characterised by their database models, e.g. relational or object models
- G06F16/284—Relational databases
- G06F16/288—Entity relationship models
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/31—Indexing; Data structures therefor; Storage structures
- G06F16/313—Selection or weighting of terms for indexing
-
- 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
Abstract
The invention discloses a multi-scene fusion double-random market supervision method, which relates to the technical field of internet big data and mainly comprises the following steps: constructing a supervisor library and a market subject library, wherein a plurality of supervisory scene models are constructed in the market subject library, and the supervisory scene models are adopted to process data in the market subject library so as to add scene codes to the data to form a dynamic scene library; the supervision personnel base is used for storing supervision personnel information and extracting supervision personnel from the supervision personnel base according to supervision tasks to form a target supervision personnel base; extracting target scene model data; fusing target scene model data; intelligently associating data in the target market subject library and the target supervisor library; by the multi-scene fusion double-random market supervision method of the double-random + supervision scene model, the problems of daily supervision, accurate supervision, credit supervision, risk supervision and key supervision in the integrated market supervision are solved, and therefore supervision efficiency is improved, supervision precision is improved, and supervision key points are mastered.
Description
Technical Field
The invention relates to the technical field of internet big data, in particular to a multi-scene fusion double-random market supervision method.
Background
In the aspect of market subject supervision, the market subject admission condition of 'wide-forward strict management' brings a new problem of market supervision, the traditional supervision means is difficult to meet the requirement of the existing market supervision, the market supervision mode is inevitably required to be innovated, the market subject data is subjected to supervision scene model analysis by relying on a big data technology, and the market supervision is carried out by processing the data according to the supervision scene model so as to conform to the direction and trend of the national policy.
The governments of people in provinces (regions and cities) need to comprehensively establish an inspection object directory library (market subject library) and a law enforcement inspector directory library (supervisor library) (collectively called 'two libraries') which can robustly cover all levels of the district and correspond to spot-check items. According to legal rules and departments, division of labor is carried out according to the principle of 'who examines and approves, who supervises, who manages and supervises', and through modes such as classification marking, batch import and the like, an inspection object name book library corresponding to department responsibilities is respectively established on a provincial platform, and supervision vacuum is avoided. The inspection object directory library (market subject library) may include market subjects such as enterprises and individual industrial and commercial businesses, and may also include products, items, behaviors, and the like.
The name list library (supervisor library) of the law enforcement inspectors comprises all relevant administrative law enforcement officers, staff with administrative enforcement qualification and staff engaged in daily supervision work, and classification and labeling are carried out according to law enforcement qualification and business expertise, so that the specialty of spot check inspection is improved. For the spot check in the specific field, on the basis of meeting the requirement of the number of the law enforcement inspectors, the spot check can be assisted by taking the participation of detection institutions, scientific research institutions, expert scholars and the like, and the professional spot check requirement can be met. The two libraries are dynamically managed according to the change of the inspected objects and law enforcement inspectors.
In the invention patent 'a double random extraction method' (application number: 201810879766X), how to perform over-balanced extraction in the double random extraction process is mainly proposed, so that the extracted probability of market subjects with less extracted times is increased, the goal that the extracted times of all market subjects tend to be balanced is reached, and the extreme condition of extraction results is avoided; in the process of extracting the law enforcement inspectors, the probability of being extracted of each expert is changed through the balance adjusting coefficient and the experience coefficient of the group leader expert or the group member expert, and finally the optimal effect of the double random-one open supervision and inspection work is achieved.
The existing double-random market supervision technology only performs random extraction on data in a market subject library and a supervisor library to form a double-random inspection task, and only performs random extraction according to market subject names in the process of extracting the data in the market subject library, so that the extracted market supervision subject data has universality and weak targeting property, and the problems of low supervision efficiency, poor supervision precision and difficult supervision control in market supervision cannot be solved in the whole double-random supervision process.
Disclosure of Invention
The invention provides a multi-scene fusion double-random market supervision method aiming at the problems of the background art, and solves the problems of low supervision efficiency, poor supervision precision and difficult supervision key in market supervision after the integration of three departments of industry and commerce, quality control and food and drug supervision.
In order to achieve the above object, the present invention provides a multi-scenario fusion dual-random market supervision method, which comprises the following steps:
constructing a supervisor library and a market subject library, wherein a plurality of supervisory scene models are constructed in the market subject library, and processing data in the market subject library by adopting the supervisory scene models to add scene codes to the data so as to form a dynamic scene library; the supervision personnel base is used for storing supervision personnel information and extracting supervision personnel from the supervision personnel base according to supervision tasks to form a target supervision personnel base;
extracting target scene model data from the dynamic scene library;
fusing the extracted target scene model data to construct a target market subject library;
and intelligently associating the data in the target market subject library and the target supervisor library.
Preferably, in step (ii): extracting target scene model data from the dynamic scene library, which previously further includes:
creating fusion supervision task information;
a desired supervisory scene model is selected.
Preferably, in step (ii): intelligently associating data in the target market subject library and the target supervisor library, and then further comprising:
fused supervisory task post-processing including, but not limited to:
and checking the supervision tasks, issuing the supervision tasks and feeding back the supervision tasks.
Preferably, the plurality of regulatory scenario models, including but not limited to: daily supervision model, accurate supervision model, credit supervision model, risk supervision model, and key supervision model.
Preferably, the daily supervision model is used for traversing and inquiring market main body data with zero annual check times in a market main body data information table, and storing the data into a daily supervision library;
the accurate supervision model is used for constructing and training an optimized self-learning model, predicting market subject data by using the self-learning model, and storing a prediction result into an accurate supervision scene library;
the credit supervision model is used for identifying a named entity of the complaint text content, associating and matching the named entity with data in the market subject, acquiring data in the market subject library and storing the data in the credit supervision scene library;
the risk supervision model is used for acquiring the data characteristics of the market main body through a prediction algorithm, determining the operation risk degree of the market main body, and storing enterprise information of which the operation risk degree exceeds a set threshold value into a risk supervision scene library;
the key supervision model is used for correspondingly marking enterprise fields in the market main body data information table according to the published key supervision enterprise list and storing the market main body data marked as key supervision enterprises in the key supervision scene library.
Preferably, the extracting target scene model data from the dynamic scene library includes but is not limited to:
extracting a scene model name, configuring a data source, a script interface, a script type, an automatic control type and parameter configuration.
Preferably, the fusing the extracted target scene model data to construct a target market subject library, including but not limited to:
data merging, data screening, data cleaning and data loading.
Preferably, the intelligence correlates data in the target market subject library with a target regulator library, including but not limited to:
selecting a target market subject library;
selecting a target supervisor library;
selecting a random matching rule;
and starting intelligent data association.
Preferably, the creating of the converged supervisory task information includes, but is not limited to:
task name, effective time, whether one team enters, inspection item list and inspection template.
Preferably, the selection of the required supervision scenario model supports single selection and multiple selection.
The invention provides a multi-scene fusion double-random market supervision method, which is characterized in that the multi-scene fusion double-random market supervision method of a double-random + supervision scene model is used for solving the problems of daily supervision, accurate supervision, credit supervision, risk supervision and key supervision in market supervision after integration of three departments of industry, quality control and food and drug supervision by carrying out supervision scene model processing on the market main body associated data in a market main body database and pasting scene labels on the market main body data through the supervision scene model processing, so that the market supervision scene application is increased, and further the supervision efficiency, the supervision precision and the supervision key are improved, so that the current market supervision technology and service problems can be solved in a targeted manner. The credit consciousness and the self-restraint force of market main bodies are enhanced, the interference on normal production and operation activities of the market main bodies is practically reduced, the responsibility of enterprise main bodies is strengthened, the conversion from government supervision to social co-management is realized, and the post-event supervision efficiency is improved in a supervision mode. Realizes 'double random, one public' supervision, full coverage and normalization. The novel supervision mechanism in the market supervision field is more perfect, and comprehensive supervision and intelligent supervision are realized.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the structures shown in the drawings without creative efforts.
FIG. 1 is a flow chart of a multi-scenario fusion dual-random market monitoring method according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of a multi-scenario fusion dual-random market monitoring process according to an embodiment of the present invention;
FIG. 3 is a schematic flow chart of building "two libraries" according to an embodiment of the present invention;
FIG. 4 is a flow diagram illustrating the creation of a converged policing task according to an embodiment of the invention;
FIG. 5 is an illustration of a selection supervision scenario in accordance with an embodiment of the present invention;
FIG. 6 is a flow chart illustrating the extraction of supervisory scenario data in accordance with an embodiment of the present invention;
FIG. 7 is a schematic diagram illustrating a multi-scenario data fusion process according to an embodiment of the present invention;
FIG. 8 is a schematic diagram illustrating an example data fusion process in one embodiment of the present invention;
FIG. 9 is a schematic diagram illustrating an intelligent data association process according to an embodiment of the present invention;
FIG. 10 is a diagram illustrating post-processing of a fusion supervision task according to an embodiment of the present invention;
the implementation, functional features and advantages of the objects of the present invention will be further explained with reference to the accompanying drawings.
Detailed Description
The technical solutions in the embodiments of the present invention will be described clearly and completely with reference to the accompanying drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be obtained by a person skilled in the art without inventive step based on the embodiments of the present invention, are within the scope of protection of the present invention.
It should be noted that, if directional indications (such as up, down, left, right, front, and back … …) are involved in the embodiment of the present invention, the directional indications are only used to explain the relative positional relationship between the components, the movement situation, and the like in a specific posture (as shown in the drawing), and if the specific posture is changed, the directional indications are changed accordingly.
In addition, if there is a description of "first", "second", etc. in an embodiment of the present invention, the description of "first", "second", etc. is for descriptive purposes only and is not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one such feature. In addition, the technical solutions in the embodiments may be combined with each other, but must be realized by a person skilled in the art, and when the technical solutions are contradictory or cannot be realized, such a combination of technical solutions should not be considered to exist, and is not within the protection scope of the present invention.
The invention provides a multi-scene fusion double-random market supervision method;
in a first preferred embodiment of the present invention, as shown in fig. 1 and 2, the method comprises the following steps:
s10, constructing a supervisor library and a market subject library, wherein the market subject library is constructed with a plurality of supervisory scene models, and the supervisory scene models are adopted to process data in the market subject library so as to add scene codes to the data to form a dynamic scene library; the supervision personnel base is used for storing supervision personnel information and extracting supervision personnel from the supervision personnel base according to supervision tasks to form a target supervision personnel base;
in the embodiment of the present invention, as shown in fig. 3, a market subject library is constructed, a monitoring scene market subject library is formed through a monitoring scene model processing operation, and then an operation is extracted to form a target market subject library (steps S40 and S50 are described in detail); building a supervision personnel base, and starting an extraction task to form a target supervision personnel base after the configuration of an extraction rule;
in the embodiment of the invention, the market main body library comprises data of the whole life cycle of the market main body, the market main body has data information of different levels such as government supervision, enterprise operation, financial credit, market feedback, media evaluation and the like from registration, and the structured data extracted according to the information of different dimensions is processed by the supervision scene model to form the supervision scene library. The problems of fragmentization, decentralization and regionalization of market main body information are effectively solved;
in the embodiment of the invention, government regulatory data: such as license class data, including: license number, enterprise name, registration address, permission range, issuing organ, supervising organ and validity period; supervising inspection class data, comprising: checking plan information, checking process information and checking result information; administrative law enforcement class data, including: complaint report information, case source registration information, case transfer information, on-site inspection entry information, case investigation termination report information, administrative penalty advance notice information, and administrative penalty decision information.
In the embodiment of the invention, enterprise operation data: enterprise financing data comprises unit capital cost, comprehensive capital cost, total liability rate, mobile liability, long-term liability, and owner equity; the enterprise investment data comprises investment value-producing rate, investment profitability and investment recovery period; enterprise production and management result data including product yield, product type, product quality and market share; and analyzing the enterprise cost.
In the embodiment of the invention, the financial credit data is as follows: flow rate, snap rate, cash rate, asset liability rate, tangible asset liability rate, title rate, fold of interest gained, total asset liability rate, operating capital;
in the embodiment of the invention, market feedback data: employee evaluation data for the company; evaluation data of the partner to the company; customer evaluation data for the company; the user evaluates the data to the company product;
in the embodiment of the invention, the media evaluation data: the number of news related to the business; the total number of sentences in each news body; the number of negative emotion sentences in the body; the number of neutral sentiment sentences in the text; number of positive emotion sentences in the body.
In the embodiment of the present invention, the multiple supervisory scenario models include, but are not limited to: a daily supervision model, an accurate supervision model, a credit supervision model, a risk supervision model and a key supervision model;
in the embodiment of the invention, the daily supervision model is used for traversing and inquiring market main body data with zero annual checking times in a market main body data information table and storing the data into a daily supervision library;
the method specifically comprises the following steps: the daily supervision model data processing process comprises the following steps: storing market main data by adopting Mysql data, inquiring data in Mysql by adopting a shell script according to two fields of 'current annual check times' and 'historical check times' in a market main data information table, and storing the data with the current annual check times being zero in a daily supervisory library after the data is processed by the shell script;
in the embodiment of the invention, the accurate supervision model is used for constructing and training an optimized self-learning model, predicting market subject data by using the self-learning model, and storing a prediction result into an accurate supervision scene library; the data processing process of the accurate supervision model comprises the following steps: applying Neural Network to establish an artificial Neural model Ann.model for a market body data set, gradually improving the performance of the artificial Neural model by adjusting model parameters, and determining Ann.model model parameters: decay is 0.1, size is 20, maxit is 200, MaxNWts is 10000, where decay is decay rate, i.e. learning rate; size is the number of hidden neurons; maxit is the maximum iteration number of the algorithm; MaxNWts is the maximum weight number of runs. The data set is divided into two subsets according to the proportion of 7:3, wherein the training set accounts for 7, the testing set accounts for 3, and the data set is verified by adopting a ten-fold intersection method. Through calculation of each performance index of a fusion matrix and a ROC curve of an ann model, performance indexes such as Sensitivity, Specificity, Kappa, Accuracy, AUC and the like of each model are deeply analyzed and compared, and a model with the best performance is determined. Obtaining a classification model Ann.model, then predicting market main body data by using the model, and storing a prediction result in an accurate supervision scene library;
in the embodiment of the invention, the credit supervision model is used for identifying a named entity of the complaint text content, associating and matching the named entity with data in a market subject, acquiring the data in a market subject library and storing the data in a credit supervision scene library; credit supervision model data processing procedure: the credit supervision model adopts a BilSTM-CRF model to realize the named entity recognition of the content of the complaint text, and the model mainly comprises an Embedding layer (mainly comprising word vectors, word vectors and some additional characteristics), a bidirectional LSTM layer and a final CRF layer. Named entity recognition is a fundamental task in the analysis of the content of complaint texts, and refers to the recognition of names of people, places, companies, products, dates and times, proper nouns and the like from the complaint texts. Obtaining data in a market subject library through the association matching between the named entity identified in the credit supervision model and the data in the market subject, and storing the data in a credit supervision scene library;
in the embodiment of the invention, the risk supervision model is used for acquiring the data characteristics of a market main body through a prediction algorithm, determining the operation risk degree of the market main body, and storing enterprise information of which the operation risk degree exceeds a set threshold value into a risk supervision scene library; and (3) risk supervision model data processing: establishing a risk supervision model to process a market main body data set, adopting a random forest algorithm to obtain market main body data characteristics, reading in the market main body data by using a read function, realizing characteristic value selection by using a function Boruta () in a Boruta packet, setting a random seed number to be 1, and using a function parameter doTrace ═ T to track the algorithm process. Each iteration of the algorithm separates the remaining eigenvalues into three states, confirmed, rejected, and still. A new data set is created using the function getSelectedAttributes () to find those feature variables that are "validated" by the algorithm. And filtering the original 60 initial three-level variables by a random forest algorithm, and only selecting 32 variables from the original 60 initial three-level variables as the establishment of the enterprise risk model. The method comprises the steps of establishing a model model.fit by using an SEM structural equation as an architecture algorithm of the whole index system and relating to the methods of regression analysis, verification factors, path analysis, significance inspection, residual error analysis and the like, calculating the load of each variable factor of the model, and determining the operation risk score of a market main body according to the factor load. The model is evaluated for performance by adopting an evaluation standard of an international SEM structural equation, wherein each term of the parameter (para) means that x 2 is 12.88, RMR is 0.61, RMSEA is 0.023, PNFI is 0.973, NFI is 0.989, and CFI is 0.94; (ii) a PNFI ═ 0.973 indicates that model model.fit is simple and there is no complex situation; and CFI of model.fit is 0.94, which shows that the model does not lose the superiority of fitting on a concise basis, the final purpose of modeling the operation risk condition model of the market subject is to predict the operation risk of the market subject by combining the model according to the existing data (observation variables in an index system) of the market subject, and in n enterprises, the best credit quantification value is 100, and the worst is 0 through sort function sorting. In the embodiment, enterprises with the risk degree of more than 70 are selected and stored in a risk supervision scene library;
in the embodiment of the invention, the key supervision model is used for correspondingly marking enterprise fields in the market main body data information table according to a published key supervision enterprise list and storing the market main body data marked as key supervision enterprises in a key supervision scene library; the data processing process of the key monitoring model comprises the following steps: marking a field of 'whether key enterprises are supervised' in a market main body data information table according to a key enterprise supervising list published in province, city, district and county, wherein the field is true; no is false. Adopting a shell script to inquire whether major supervising enterprises become major market main data and storing the major supervising enterprise data in a major supervising scene library;
in the embodiment of the invention, the supervisor library comprises all relevant administrative law enforcement officers, staff with administrative law enforcement qualification and staff engaged in daily supervision work, and classification and marking are carried out according to law enforcement qualification and business expertise, so that preparation can be made for subsequent supervisor extraction rule configuration, and the professional performance of spot check inspection can be improved. Meanwhile, for the spot check in the specific field, on the basis of meeting the requirement of the number of the law enforcement inspectors, the spot check can be assisted by taking the participation of detection institutions, scientific research institutions, expert scholars and the like, and the professional spot check requirement can be met. The two libraries are dynamically managed according to the change conditions of the inspection objects and law enforcement inspectors;
s20, creating fusion supervision task information;
in the embodiment of the present invention, as shown in fig. 4, the main work is to fill in basic information of the supervision task, where the basic information includes a task name, a task validity time, whether to enter a gate in a team, selecting an inspection item list, selecting an inspection template, and the like, and after the completion of the filling, the step S30 is performed;
in the embodiment of the invention, the task name is as follows: the method integrates the main names of supervision tasks, such as assault supervision and inspection on drug (development, production, operation and use) links, special supervision and inspection on cosmetic (production, operation and use) links and the like.
In the embodiment of the invention, the effective time is as follows: the start time and the end time of the supervision task. Such as: 2019-01-11-2019-03-15.
In the embodiment of the invention, a team enters a door: and singly selecting whether to adopt one team to enter to carry out the supervision and inspection task. If the selection is yes, the access of one team includes the step of fusing the original business, quality supervision, food and medicine and other departments to check the business list, so that the law enforcement cost is reduced, the enterprise burden is reduced, and the administrative efficiency is improved to provide a solid foundation. A team enters the gate for supervision, so that the fairness, the normalization and the effectiveness of supervision are improved, the problems of law enforcement arbitrariness, supervision disturbance of residents, law enforcement unfairness, low efficiency and the like in the market supervision work are solved, a fair and benign competitive development environment is created for the whole area, and the market activity is stimulated.
In the embodiment of the invention, the checking item list: and checking the item list of the market subject formed according to relevant laws and regulations. Such as: a business license (registration certificate) specifies checking of use, checking of business (standing) term, and the like.
In the embodiment of the invention, the template is checked: and the inspection table template corresponds to the inspection task. Such as: a daily supervision and inspection record table (food production link), an authentication and supervision and inspection condition record table and the like;
s30, selecting a needed supervision scene model;
in the embodiment of the invention, as shown in fig. 5, a supervision scene is selected, wherein the supervision scene comprises daily supervision, accurate supervision, credit supervision, risk supervision, key supervision and the like, and the combination selection among various supervision scenes is supported;
in the embodiment of the invention, market main body data related to each supervision scene is cleaned by a script and then stored in an elastic search engine, when a supervision task is carried out to extract a market main body, a person establishing a plan sets different screening conditions at the front end of a browser and transmits the screening conditions to a background interface, the background interface service acquires a DSL query condition (json format), a service inquires a main body suitable for the supervision scene in the elastic search through the DSL query condition, if various supervision scenes are fused, the market main body data are acquired from different indexes through multithreading query of the elastic search, and the market main body data are returned to the front end for display after a spark sql script is duplicated;
in the embodiment of the invention, daily supervision is as follows: the market subject supervision department starts the daily supervision and inspection work on the market subject;
in the embodiment of the invention, the precise supervision is as follows: the method has the advantages that the spot check proportion is improved or the probability of being spot checked is increased for the supervision objects with high risk and the records of losing confidence, the spot check proportion is properly reduced or the probability of being spot checked is reduced for the law-keeping confidence, but the 'drawing-free' and 'inspection-free' cannot be implemented for any supervision object;
in the embodiment of the invention, credit supervision: determining corresponding parameters (coefficients) and functional relations through individual difference analysis of data credibility, relevance, industry, attributes and the like, and establishing an enterprise credit classification evaluation model; the method comprises the steps of sorting and analyzing the existing market subject data and history data, judging the credit classification level of an enterprise according to a mathematical model, and supporting credit evaluation, level adjustment and credit restoration of the enterprise;
in the embodiment of the invention, risk supervision: and establishing a market order condition evaluation index system, analyzing the industrial classified risk monitoring, the operation category risk monitoring, the cluster registration risk monitoring, the commodity (food) spot inspection result risk and the like in the south sand area, and implementing online monitoring and early warning and risk prevention and control linkage. Collecting and sorting operation supervision data of enterprises, analyzing production and operation conditions of the enterprises by using patterns such as charts, and analyzing supervision risks of the enterprises in a multi-dimensional way according to tax payment conditions, employment conditions, water use conditions, power utilization conditions, government department supervision and inspection conditions and the like. Establishing cluster enterprise risk analysis indexes, mining and analyzing government department involved information and enterprise dynamic information generated in production and operation activities of cluster registered enterprises, and performing risk early warning;
in the embodiment of the invention, the key supervision is as follows: the key fields of medical treatment, medicine, health-care food, financial investment and the like and the key areas of rural areas, urban and rural junctions and the like are highlighted, and the serious false illegal advertisements such as the life and property safety of people in the affairs are mainly investigated and treated.
In the embodiment of the invention, an enterprise panoramic image and enterprise operational behavior relationship network is established through data integration and association, the market subject condition is mastered in time, and subject risk monitoring is carried out, so that supervision is more accurate, the realization of' law keeping without disturbance and law violation must be promoted, and powerful support is provided for continuously optimizing an operator environment and building a good market order; the data integration comprises the following steps: a nosql non-relational database and a full-text search engine elastic search are adopted to store, index and divide words for data associated with market main bodies, a manual or machine-divided directory structure is stored in a mysql database, and finally, related main bodies are searched from the elastic search through data elements, namely keywords such as enterprise names and registration numbers; the enterprise portrait: high-risk enterprises, medium-risk enterprises, low-risk enterprises, abnormal-operation enterprises, high-quality enterprises, high-dispute enterprises, innovative enterprises, high-tax enterprises and the like.
S40, extracting target scene model data from the dynamic scene library;
in the embodiment of the invention, as shown in fig. 6, corresponding monitoring scene data is extracted mainly according to the monitoring scene selected in the third step and the monitoring scene model, the extraction step is as follows, the name of the monitoring scene model is selected, a data source with extraction is configured, the data source is a market subject library, a script interface is connected, and a corresponding script type is selected, the three types of script types can be selected to be Python script, Scala script and Sql script respectively, and a model script execution control mode is selected. If 'no' is selected, the supervision scene model script is executed once; if 'yes' is selected, the method is in an automatic control mode, script execution time is selected, the script execution time is executed regularly every day or every week or every month, finally script parameters are configured, a supervision scene model is started, and corresponding scene library data are generated.
In the embodiment of the invention, the name of a scene model is as follows: the supervision scene model comprises a daily supervision model, an accurate supervision model, a credit supervision model, a risk supervision model and a key supervision model; configuring a data source: configuring data source information of the extraction task, wherein the data source information comprises information such as node names, node aliases, node types, IP addresses, port numbers, accounts, passwords, examples, target users, scanning periods and the like; script interface: an HTTP interface automation script is adopted, such as HTTP:// localhost:8000/api/v 1/lda; script type: the three types of scripts can be selected as Python scripts, Scala scripts and Sql scripts respectively; automatic control: whether binary selection is carried out or not, if the scene model script is not selected to be executed once, if the selection is in an automatic control mode, the execution time of the script is selected, and the script is executed regularly every day or every week or every month; parameter configuration: configuring script parameters;
s50, fusing the extracted target scene model data to construct a target market subject library;
in the embodiment of the invention, as shown in fig. 7, a target market subject library is finally formed through multi-scene data fusion, and the preparation work for intelligent data association is provided; the multi-scene data fusion is mainly realized by four steps of data merging, data screening, data cleaning and data loading, and the data are fused to form a target market subject library through the final data loading;
in the embodiment of the present invention, as shown in fig. 8, data merging: the extracted supervised scene data is merged into the same temporary table. And (3) screening data: and screening out repeated market subject data in different scene codes through two fields of the market subject name and the social credit uniform code. Data cleaning: and cleaning the repeated data, and removing the repeated market main data according to a data cleaning rule (combining scene codes, adding a piece of combined data, keeping the added data, and deleting a plurality of pieces of combined data). Loading data: and loading the cleaned market subject data into a target market subject library. Target market subject library: namely, a database is formed after multi-scene data fusion, the database is a basic data source for the six-step intelligent data fusion, and the enterprise information to be detected by the double random tasks is stored.
S60, intelligently associating the data in the target market subject library and the target supervisor library;
in the embodiment of the invention, as shown in fig. 9, the intelligent data association mainly realizes data in a target market subject library and a target supervisor library, the database stores areas and grid identifications of different market subjects, the personnel library also stores different identifications, in the directional spot check, a background queries a batch of market subjects according to spot check conditions and spot check quantity, queries a batch of personnel from the personnel library according to the spot check quantity of the spot check conditions, groups the batch of personnel sets through a random function, polls the market subject sets at the same time, assigns a group of personnel to each market subject, performs intelligent association according to a random matching rule, and performs data association operation after starting the intelligent data association operation. Data matching between the market main body to be inspected and the supervisor can be completed through the sixth step, namely the supervisor can inspect which market main bodies (enterprises) through the supervision task;
in the embodiment of the invention, a target market subject library is selected: the data stored in the target market subject library is data after multi-scene data fusion, namely to-be-inspected market subject information of the multi-scene fusion double-random inspection task; selecting a target supervisor library: the data stored in the target supervisor database is the supervisor information participating in the scene fusion double random inspection task; selecting a random matching rule: rules for matching between a supervisor and a regulatory object (market subject). If in the process of secondary supervision task development, a certain supervisor checks 3 enterprises, a certain key person checks 4 enterprises, and then market subjects in the target market subject library are randomly matched to the supervisor; and (3) starting intelligent data association: after a target market main body library, a target supervisor library and a random matching rule are selected, intelligent data association is started to generate the double random inspection task information;
s70, fusing the post-processing of the supervision task;
in the embodiment of the invention, as shown in fig. 10, the post-processing of the fused supervision task mainly completes the functions of three aspects of supervision task viewing, supervision task issuing and supervision task feedback;
in the embodiment of the invention, the functions are realized based on the B/S architecture, a user accesses a static page through the browser end, the page acquires data to the background interface server, the interface server inquires the supervision task and the personnel information related to the supervision task from the cache and the database, and finally, the information is responded to the browser end.
In the embodiment of the invention, the supervision task is checked: the method realizes the viewing of the contents of the four aspects of basic information, supervisor information, overall progress and task list. The basic information comprises a plan name, a starting time, an ending time, a checking table, an extraction rule, a checking item and a plan remark; the supervisor information comprises personnel names, law enforcement certificate numbers, affiliated units, affiliated departments and mobile phone numbers; the overall progress comprises the overall progress condition of the current inspection task, and the task is displayed in a bar form; the task list comprises information such as spot check objects, addresses, inspectors, supervision departments, check time arrangement, task states and the like. And (3) issuing a supervision task: and selecting the double random inspection tasks in batches, and issuing the double random inspection tasks to corresponding units, corresponding departments and corresponding personnel. And (3) supervision task feedback: after the supervision task for which the supervision personnel is responsible, the supervision task information can be fed back.
The invention innovatively provides that a double-random + supervision scene model is applied in market supervision by taking a supervision scene as a support to develop accurate market supervision, so that the supervision efficiency is further improved, and the market supervision service level is improved. And selecting a supervision scene, wherein five supervision scenes comprise daily supervision, accurate supervision, credit supervision, risk supervision and key supervision. And according to the supervision scene, according to the supervision scene data extraction flow, namely selecting a supervision scene model, configuring a data source, accessing a scene model script in an HTTP mode, selecting the type of the supervision scene model script, namely a script operation control mode, and configuring script parameters, and completing the extraction of the scene model data in the whole flow. And then, carrying out multi-scene data fusion, namely, fusing through data merging, data screening, data cleaning and data loading to form a target market subject library.
Compared with the common double-water-machine market supervision, the multi-scene fusion double-random market supervision method has the greatest characteristic that the application of supervision scenes is added according to the characteristics of the market supervision, data in a market subject library are extracted according to the supervision scenes, a target market subject library is formed after scene model processing and multi-scene data fusion, and a supervision task is formed according to the fact that the target market subject library and a target supervision personnel library develop intelligent data association. According to the whole process, the scene label is attached to the market main body data by adding the supervision scene, so that the problems of accurate supervision, credit supervision, risk supervision and key supervision in market supervision after integration of three departments of industry and commerce, quality inspection and food and drug supervision are solved, and the problem of current market supervision business is solved in a targeted manner. Thereby enhancing the credit consciousness and self-restraint force of market main bodies and realizing 'limp and high suspension' for the offenders; the intervention on the normal production and operation activities of the market main body is practically reduced, and the law keeper is 'undisturbed'. The responsibility of enterprise subjects is strengthened, the conversion from government supervision to social co-treatment is realized, and the post-affair supervision efficiency is improved in a supervision mode innovation mode. The method realizes that the proposal of ' double random and one open ' supervision ' of the national institute about the comprehensive implementation department in the market supervision field is definitely proposed in ' No. 5 of the national issue 2019 ', takes ' double random and one open ' supervision as a basic means, takes key supervision as supplement and takes credit supervision as a basis in the market supervision field, and further creates a fair competitive market environment and a legal and convenient operator environment.
In describing embodiments of the present invention, it should be noted that any process or method descriptions in flow charts or otherwise described herein may be understood as representing modules, segments, or portions of code which include one or more executable instructions for implementing specific logical functions or steps of the process, and that the scope of the preferred embodiments of the present invention includes additional implementations in which functions may be executed out of order from that shown or discussed, including substantially concurrently or in reverse order, depending on the functionality involved, as would be understood by those reasonably skilled in the art of the present invention.
The logic and/or steps represented in the flowcharts or otherwise described herein, such as an ordered listing of executable instructions that can be considered to implement logical functions, can be embodied in any computer-readable medium for use by or in connection with an instruction execution system, apparatus, or device, such as a computer-based system, processing module-containing system, or other system that can fetch the instructions from the instruction execution system, apparatus, or device and execute the instructions. For the purposes of this description, a "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of the computer-readable medium would include the following: an electrical connection (electronic device) having one or more wires, a portable computer diskette (magnetic device), a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). Additionally, the computer-readable medium could even be paper or another suitable medium upon which the program is printed, as the program can be electronically captured, via for instance optical scanning of the paper or other medium, then compiled, interpreted or otherwise processed in a suitable manner if necessary, and then stored in a computer memory.
The above description is only a preferred embodiment of the present invention, and is not intended to limit the scope of the present invention, and all modifications and equivalents of the present invention, which are made by the contents of the present specification and the accompanying drawings, or directly/indirectly applied to other related technical fields, are included in the scope of the present invention.
Claims (10)
1. A multi-scene fusion double-random market supervision method is characterized by comprising the following steps:
constructing a supervisor library and a market subject library, wherein a plurality of supervisory scene models are constructed in the market subject library, and processing data in the market subject library by adopting the supervisory scene models to add scene codes to the data so as to form a dynamic scene library; the supervision personnel base is used for storing supervision personnel information and extracting supervision personnel from the supervision personnel base according to supervision tasks to form a target supervision personnel base;
extracting target scene model data from the dynamic scene library;
fusing the extracted target scene model data to construct a target market subject library;
and intelligently associating the data in the target market subject library and the target supervisor library.
2. The multi-scenario fusion dual-stochastic market regulation method according to claim 1, characterized in that at step: extracting target scene model data from the dynamic scene library, which previously further includes:
creating fusion supervision task information;
a desired supervisory scene model is selected.
3. The multi-scenario fusion dual-stochastic market regulation method according to claim 1, characterized in that at step: intelligently associating data in the target market subject library and the target supervisor library, and then further comprising:
fused supervisory task post-processing including, but not limited to:
and checking the supervision tasks, issuing the supervision tasks and feeding back the supervision tasks.
4. The multi-scenario fusion dual-stochastic market regulation method of claim 1, wherein the plurality of regulatory scenario models include, but are not limited to: daily supervision model, accurate supervision model, credit supervision model, risk supervision model, and key supervision model.
5. The multi-scenario fusion dual-random market regulation method of claim 4,
the daily supervision model is used for traversing and inquiring market main body data with zero annual check times in a market main body data information table and storing the data into a daily supervision library;
the accurate supervision model is used for constructing and training an optimized self-learning model, predicting market subject data by using the self-learning model, and storing a prediction result into an accurate supervision scene library;
the credit supervision model is used for identifying a named entity of the complaint text content, associating and matching the named entity with data in the market subject, acquiring the data in the market subject library and storing the data in the credit supervision scene library;
the risk supervision model is used for acquiring the data characteristics of the market main body through a prediction algorithm, determining the operation risk degree of the market main body, and storing enterprise information of which the operation risk degree exceeds a set threshold value into a risk supervision scene library;
the key supervision model is used for correspondingly marking enterprise fields in the market main body data information table according to the published key supervision enterprise list and storing the market main body data marked as key supervision enterprises in the key supervision scene library.
6. The multi-scenario fusion dual-stochastic market regulation method of claim 1, wherein the extracting target scenario model data from the dynamic scenario library comprises but is not limited to:
extracting a scene model name, configuring a data source, a script interface, a script type, an automatic control type and parameter configuration.
7. The multi-scenario fusion dual-stochastic market regulation method according to claim 1, wherein the fusion of the extracted target scenario model data to construct a target market subject library comprises but is not limited to:
data merging, data screening, data cleaning and data loading.
8. The multi-scenario fusion dual-stochastic market regulation method according to claim 1, wherein the intelligent association of data in the target market subject library and target regulator library includes but is not limited to:
selecting a target market subject library;
selecting a target supervisor library;
selecting a random matching rule;
and starting intelligent data association.
9. The multi-scenario converged dual random market regulation method according to claim 2, wherein the creating converged regulatory task information includes but is not limited to:
task name, effective time, whether one team enters, inspection item list and inspection template.
10. The multi-scenario fusion dual-stochastic market regulation method according to claim 2, wherein the selection of the required regulatory scenario model supports single selection and multiple selection.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910745543.9A CN110633316A (en) | 2019-08-13 | 2019-08-13 | Multi-scene fusion double-random market supervision method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910745543.9A CN110633316A (en) | 2019-08-13 | 2019-08-13 | Multi-scene fusion double-random market supervision method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110633316A true CN110633316A (en) | 2019-12-31 |
Family
ID=68970395
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910745543.9A Pending CN110633316A (en) | 2019-08-13 | 2019-08-13 | Multi-scene fusion double-random market supervision method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110633316A (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111523830A (en) * | 2020-07-03 | 2020-08-11 | 北京梦天门科技股份有限公司 | Method, device, equipment and medium for guiding market supervision based on multi-dimensional data |
| CN111552703A (en) * | 2020-05-25 | 2020-08-18 | 支付宝(杭州)信息技术有限公司 | Data processing method and device |
| CN111563687A (en) * | 2020-03-31 | 2020-08-21 | 北京梦天门科技股份有限公司 | Double random health supervision law enforcement method and terminal |
| CN111951141A (en) * | 2020-07-09 | 2020-11-17 | 广东港鑫科技有限公司 | Double-random supervision method and system based on big data intelligent analysis and terminal equipment |
| CN112381407A (en) * | 2020-11-16 | 2021-02-19 | 浪潮软件股份有限公司 | Credit weighting double-random supervision method based on random algorithm |
| CN112926781A (en) * | 2021-03-02 | 2021-06-08 | 浪潮云信息技术股份公司 | Market subject behavior map intelligent analysis method and system based on supervision field |
| CN114036347A (en) * | 2021-11-18 | 2022-02-11 | 北京中关村软件园发展有限责任公司 | Cloud platform supporting digital fusion service and working method |
| CN115511392A (en) * | 2022-11-16 | 2022-12-23 | 北京信城通数码科技有限公司 | Task matching method and device, electronic equipment and storage medium |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20020169658A1 (en) * | 2001-03-08 | 2002-11-14 | Adler Richard M. | System and method for modeling and analyzing strategic business decisions |
| US20030172017A1 (en) * | 2002-03-11 | 2003-09-11 | Vincent Feingold | High performance multi-dimensional risk engines for enterprise wide market risk management |
| CN101551882A (en) * | 2008-12-31 | 2009-10-07 | 东莞市天新企业管理咨询有限公司 | Engineering management monitoring system and method |
| CN105893526A (en) * | 2016-03-30 | 2016-08-24 | 上海坤士合生信息科技有限公司 | Multi-source data fusion system and method |
| CN107727411A (en) * | 2017-10-30 | 2018-02-23 | 青岛慧拓智能机器有限公司 | A kind of automatic driving vehicle test and appraisal scene generation system and method |
| CN109033409A (en) * | 2018-08-03 | 2018-12-18 | 华北水利水电大学 | A kind of pair is randomly selected method |
| CN109472489A (en) * | 2018-11-06 | 2019-03-15 | 江苏怡通数码科技有限公司 | Urban Event management platform and implementation method based on mobile interchange |
| CN109784643A (en) * | 2018-12-14 | 2019-05-21 | 新疆维吾尔自治区产品质量监督检验研究院 | A kind of double random administration system and methods, information data processing terminal |
| CN109934718A (en) * | 2019-02-25 | 2019-06-25 | 国网安徽省电力有限公司铜陵供电公司 | A kind of electric grid investment method considering the market risk |
-
2019
- 2019-08-13 CN CN201910745543.9A patent/CN110633316A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20020169658A1 (en) * | 2001-03-08 | 2002-11-14 | Adler Richard M. | System and method for modeling and analyzing strategic business decisions |
| US20030172017A1 (en) * | 2002-03-11 | 2003-09-11 | Vincent Feingold | High performance multi-dimensional risk engines for enterprise wide market risk management |
| CN101551882A (en) * | 2008-12-31 | 2009-10-07 | 东莞市天新企业管理咨询有限公司 | Engineering management monitoring system and method |
| CN105893526A (en) * | 2016-03-30 | 2016-08-24 | 上海坤士合生信息科技有限公司 | Multi-source data fusion system and method |
| CN107727411A (en) * | 2017-10-30 | 2018-02-23 | 青岛慧拓智能机器有限公司 | A kind of automatic driving vehicle test and appraisal scene generation system and method |
| CN109033409A (en) * | 2018-08-03 | 2018-12-18 | 华北水利水电大学 | A kind of pair is randomly selected method |
| CN109472489A (en) * | 2018-11-06 | 2019-03-15 | 江苏怡通数码科技有限公司 | Urban Event management platform and implementation method based on mobile interchange |
| CN109784643A (en) * | 2018-12-14 | 2019-05-21 | 新疆维吾尔自治区产品质量监督检验研究院 | A kind of double random administration system and methods, information data processing terminal |
| CN109934718A (en) * | 2019-02-25 | 2019-06-25 | 国网安徽省电力有限公司铜陵供电公司 | A kind of electric grid investment method considering the market risk |
Non-Patent Citations (1)
| Title |
|---|
| 国市监信〔2019〕38号: "市场监管总局关于全面推进"双随机、一公开"监管工作的通知", 《国务院公报》 * |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111563687A (en) * | 2020-03-31 | 2020-08-21 | 北京梦天门科技股份有限公司 | Double random health supervision law enforcement method and terminal |
| CN111552703A (en) * | 2020-05-25 | 2020-08-18 | 支付宝(杭州)信息技术有限公司 | Data processing method and device |
| CN111552703B (en) * | 2020-05-25 | 2023-11-21 | 支付宝(杭州)信息技术有限公司 | Data processing method and device |
| CN111523830A (en) * | 2020-07-03 | 2020-08-11 | 北京梦天门科技股份有限公司 | Method, device, equipment and medium for guiding market supervision based on multi-dimensional data |
| CN111523830B (en) * | 2020-07-03 | 2021-02-19 | 北京梦天门科技股份有限公司 | Method, device, equipment and medium for guiding market supervision based on multi-dimensional data |
| CN111951141A (en) * | 2020-07-09 | 2020-11-17 | 广东港鑫科技有限公司 | Double-random supervision method and system based on big data intelligent analysis and terminal equipment |
| CN112381407A (en) * | 2020-11-16 | 2021-02-19 | 浪潮软件股份有限公司 | Credit weighting double-random supervision method based on random algorithm |
| CN112926781A (en) * | 2021-03-02 | 2021-06-08 | 浪潮云信息技术股份公司 | Market subject behavior map intelligent analysis method and system based on supervision field |
| CN114036347A (en) * | 2021-11-18 | 2022-02-11 | 北京中关村软件园发展有限责任公司 | Cloud platform supporting digital fusion service and working method |
| CN115511392A (en) * | 2022-11-16 | 2022-12-23 | 北京信城通数码科技有限公司 | Task matching method and device, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Zhang et al. | The impact of artificial intelligence and blockchain on the accounting profession | |
| CN110633316A (en) | Multi-scene fusion double-random market supervision method | |
| Ayling et al. | Putting AI ethics to work: are the tools fit for purpose? | |
| CN110796470B (en) | Data analysis system for market subject supervision and service | |
| Akhtar et al. | Detecting fake news and disinformation using artificial intelligence and machine learning to avoid supply chain disruptions | |
| CN112182246B (en) | Method, system, medium, and application for creating an enterprise representation through big data analysis | |
| Lane | Democratizing our data: A manifesto | |
| CN113011185A (en) | Legal field text analysis and identification method, system, storage medium and terminal | |
| KR102121901B1 (en) | System for online public fund investment management assessment service | |
| Plaksiy et al. | Applying big data technologies to detect cases of money laundering and counter financing of terrorism | |
| Hamzehi et al. | Business intelligence using machine learning algorithms | |
| Franchina et al. | High quality dataset for machine learning in the business intelligence domain | |
| Liu et al. | Analysis of Beijing Tianjin Hebei regional credit system from the perspective of big data credit reporting | |
| Mitsuzuka et al. | Analysis of CSR activities affecting corporate value using machine learning | |
| Tundis et al. | Limits in the data for detecting criminals on social media | |
| Chun | An AI framework for the automatic assessment of e-government forms | |
| Bhardwaj et al. | Decision-making optimisation in insurance market using big data analytics survey | |
| Johns | Data mining as global governance | |
| Dorofeev et al. | Application of machine analysis algorithms to automate implementation of tasks of combating criminal money laundering | |
| Williams | Data collection, sharing and analysis to tackle undeclared work: a toolkit | |
| Rahayu | Utilization of artificial intelligence in tax audit in Indonesia | |
| KR20220108970A (en) | Method of risk analysis and technology transfer opportunity recommendation by technology-product-material based on machine learning or AI | |
| CN115345401A (en) | Six-dimensional analysis method for finding enterprise financial risk | |
| Vydra et al. | Big data ethics: A life cycle perspective | |
| Mortaz Hejri et al. | Factors affecting the readiness of green business process management in the Iranian banking industry |
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 |