CN109829629B - Risk analysis report generation method, apparatus, computer device and storage medium - Google Patents
Risk analysis report generation method, apparatus, computer device and storage medium Download PDFInfo
- Publication number
- CN109829629B CN109829629B CN201910012944.3A CN201910012944A CN109829629B CN 109829629 B CN109829629 B CN 109829629B CN 201910012944 A CN201910012944 A CN 201910012944A CN 109829629 B CN109829629 B CN 109829629B
- Authority
- CN
- China
- Prior art keywords
- risk
- resource
- target
- case
- monitoring
- 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
Abstract
The application relates to a method, a device, computer equipment and a storage medium for generating a risk analysis report based on machine learning. The method comprises the following steps: receiving a risk analysis request sent by a terminal; the risk analysis request carries a target resource identifier of the selected virtual resource; determining a monitoring object corresponding to the target resource identifier, acquiring monitoring data of the monitoring object, and inputting the monitoring data into a risk analysis model to obtain a corresponding risk score; when the risk score exceeds a threshold, marking the target resource identification as a risk case; determining similar cases corresponding to the risk cases according to the monitoring data; identifying a plurality of risk points for the risk case based on the similar cases; connecting a plurality of risk points in series to generate a risk clue corresponding to a risk case; based on the risk scores, the similar cases and the risk clues, a risk analysis report corresponding to the target resource identification is generated, and the risk analysis report is fed back to the terminal. By adopting the method, the risk analysis efficiency and accuracy can be improved.
Description
Technical Field
The present application relates to the field of computer technologies, and in particular, to a method and apparatus for generating a risk analysis report, a computer device, and a storage medium.
Background
Risk monitoring is a necessary business link in a variety of industries. For example, in the financial industry, there is a need to monitor whether there is a risk of breach of the control object of a virtual resource. With the development of computing technology, some tools for risk monitoring appear on the market, but most of these tools only can provide mechanical comparison of risk indexes and a simple credit rating function, so that it is difficult to find risk points of monitored objects in real time.
Disclosure of Invention
In view of the foregoing, it is desirable to provide a risk analysis report generating method, apparatus, computer device, and storage medium that improve risk analysis efficiency and accuracy.
A method of generating a risk analysis report, the method comprising: receiving a risk analysis request sent by a terminal; the risk analysis request carries a target resource identifier of the selected virtual resource; determining a monitoring object corresponding to the target resource identifier, acquiring monitoring data of the monitoring object, and inputting the monitoring data into a risk analysis model to obtain a corresponding risk score; marking the target resource identification as a risk case when the risk score exceeds a threshold; determining similar cases corresponding to the risk cases according to the monitoring data; identifying a plurality of risk points for the risk case based on the similar case; the risk points are connected in series, and a risk clue corresponding to the risk case is generated; and generating a risk analysis report corresponding to the target resource identifier based on the risk score, the similar cases and the risk clues, and feeding back the risk analysis report to the terminal.
In one embodiment, before receiving the risk analysis request sent by the terminal, the method further includes: receiving a resource acquisition request sent by a terminal; the resource acquisition request carries a target attribute value; acquiring resource factors of a plurality of virtual resources, and inputting a target attribute value and the resource factors of each virtual resource into a regulation model to obtain a resource label of each virtual resource; selecting a corresponding expert model according to the resource label, and inputting the resource factors into the expert model to obtain a predicted attribute value of the corresponding virtual resource; screening virtual resources with predicted attribute values matched with target attribute values, and recording the virtual resources as target resources; determining factor characteristics of the target resources according to expert models corresponding to the target resources; acquiring resource information of the target resource, and returning the resource information and factor characteristics to a terminal; and the terminal further screens the pushed multiple target resources according to the factor characteristics.
In one embodiment, the inputting the monitoring data into the risk analysis model to obtain the corresponding risk score includes: acquiring a training sample, and screening a target sample from the training sample; training the first sub-model based on the training sample to obtain a health analysis model; training the second sub-model based on the target sample to obtain a bankruptcy analysis model; inputting the monitoring data into the health analysis model to obtain the health index of the monitored object; inputting the monitoring data into the bankruptcy analysis model to obtain a bankruptcy index of the monitoring object; the risk score is calculated based on the health index and the bankruptcy index.
In one embodiment, the determining, according to the monitoring data, a similar case corresponding to the risk case includes: determining the industry type of the monitoring object corresponding to the virtual resource; acquiring risk data of the risk case; the risk data comprises risk data of the virtual resource, industry type risk data and risk data of a monitoring object; extracting a risk label of the risk case from the risk data; and calculating the similarity between the risk label and the case labels of the prestored multiple historical cases, and marking the historical cases with the similarity exceeding a preset value as similar cases.
In one embodiment, the identifying the plurality of risk points for the risk case based on the similar case includes: acquiring risk indexes of a plurality of time nodes of the similar case; judging whether the risk cases have the same risk indexes, and judging whether the time sequence of the same risk indexes of the risk cases is consistent with that of the similar cases; if yes, determining the same risk index of the last time node, and marking the same risk index as a sign index; and marking different risk indexes of the time nodes after the symptom indexes as risk points of the risk cases.
In one embodiment, the step of concatenating the plurality of risk points to generate a risk cue corresponding to the risk case includes: determining a monitoring period of the risk case; when the monitoring period arrives, returning to the step of acquiring the monitoring data of the monitoring object, and analyzing the risk point of the risk case in the current monitoring period if the risk score still exceeds a threshold value; and connecting the risk points of the monitoring periods in series to generate a risk clue corresponding to the discovery subject.
A device for generating a risk analysis report, the device comprising: the risk scoring module is used for receiving a risk analysis request sent by the terminal; the risk analysis request carries a target resource identifier of the selected virtual resource; determining a monitoring object corresponding to the target resource identifier, acquiring monitoring data of the monitoring object, and inputting the monitoring data into a risk analysis model to obtain a corresponding risk score; a clue collection module for marking the target resource identification as a risk case when the risk score exceeds a threshold; determining similar cases corresponding to the risk cases according to the monitoring data; identifying a plurality of risk points for the risk case based on the similar case; the risk points are connected in series, and a risk clue corresponding to the risk case is generated; and the report generation module is used for generating a risk analysis report corresponding to the target resource identifier based on the risk scores, the similar cases and the risk clues, and feeding back the risk analysis report to the terminal.
In one embodiment, the device further comprises a resource screening module, configured to receive a resource acquisition request sent by the terminal; the resource acquisition request carries a target attribute value; acquiring resource factors of a plurality of virtual resources, and inputting a target attribute value and the resource factors of each virtual resource into a regulation model to obtain a resource label of each virtual resource; selecting a corresponding expert model according to the resource label, and inputting the resource factors into the expert model to obtain a predicted attribute value of the corresponding virtual resource; screening virtual resources with predicted attribute values matched with target attribute values, and recording the virtual resources as target resources; determining factor characteristics of the target resources according to expert models corresponding to the target resources; acquiring resource information of the target resource, and returning the resource information and factor characteristics to a terminal; and the terminal further screens the pushed multiple target resources according to the factor characteristics.
A computer device comprising a memory storing a computer program and a processor implementing the steps of the method of generating a risk analysis report provided in any one of the embodiments of the application when the computer program is executed.
A computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements the steps of the risk analysis report generation method provided in any one of the embodiments of the present application.
According to the risk analysis report generation method, the risk analysis report generation device, the computer equipment and the storage medium, the monitoring object of the target resource selected by the user can be determined according to the risk analysis request sent by the terminal; acquiring monitoring data of a monitoring object, inputting the monitoring data into a risk analysis model, and obtaining a corresponding risk score; when the risk score exceeds a threshold, the target resource identification may be marked as a risk case; according to the monitoring data, determining similar cases corresponding to the risk cases; a plurality of risk points for obtaining the risk case can be identified based on the similar cases; the risk points are connected in series, so that a risk clue corresponding to the risk case can be generated; based on the risk scores, similar cases, and risk cues, a risk analysis report corresponding to the target resource identification may be generated. Because the risk analysis model can comprehensively consider various risk factors to carry out risk prediction, the risk analysis efficiency is improved; after the risk scores are obtained, similar cases of the risk cases are further determined, risk points, which are possibly appeared by the monitored object at a plurality of time nodes in the future, are predicted based on the similar cases, and the risk analysis report generated based on the information can be convenient for a user to comprehensively and rapidly know the risk condition of the selected virtual resources, so that the risk analysis accuracy is improved.
Drawings
FIG. 1 is an application scenario diagram of a method of generating a risk analysis report in one embodiment;
FIG. 2 is a flow diagram of a method of generating a risk analysis report in one embodiment;
FIG. 3 is a flow chart illustrating steps of target resource screening in one embodiment;
FIG. 4 is a block diagram of an apparatus for generating a risk analysis report in one embodiment;
fig. 5 is an internal structural diagram of a computer device in one embodiment.
Detailed Description
The present application will be described in further detail with reference to the drawings and examples, in order to make the objects, technical solutions and advantages of the present application more apparent. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the application.
The method for generating the risk analysis report can be applied to an application environment shown in fig. 1. Wherein the terminal 102 communicates with the server 104 via a network. The terminal 102 may be, but not limited to, various personal computers, notebook computers, smartphones, tablet computers, and portable wearable devices, and the server 104 may be implemented by a stand-alone server or a server cluster composed of a plurality of servers. When the user needs to perform risk analysis on the selected virtual resource, a risk analysis request may be sent to the server 104 through the terminal 102. The server 104 queries the monitored object corresponding to the target resource identifier selected by the user, and acquires the monitored data of the monitored object. And the server calls the risk analysis model, and inputs the monitoring data into the risk analysis model to obtain a corresponding risk score. The server compares whether the risk score exceeds a threshold. If so, the server 104 marks the target resource identification as a risk case. The server 104 determines similar cases corresponding to the risk cases according to the monitoring data, and identifies a plurality of risk points of the risk cases based on the similar cases. The server 104 connects a plurality of risk points in series to generate a risk cue corresponding to the risk case. Based on the risk scores, the similar cases and the risk cues, the server 104 generates a risk analysis report corresponding to the target resource identifier, and feeds back the risk analysis report to the terminal 102. According to the risk analysis report generation process, multiple risk factors can be comprehensively considered for risk prediction based on the risk analysis model, so that the risk analysis efficiency is improved; after the risk scores are obtained, similar cases of the risk cases are further determined, risk points, which are possibly appeared by the monitored object at a plurality of time nodes in the future, are predicted based on the similar cases, and the risk analysis report generated based on the information can be convenient for a user to comprehensively and rapidly know the risk condition of the selected virtual resources, so that the risk analysis accuracy is improved.
In one embodiment, as shown in fig. 2, a method for generating a risk analysis report is provided, and the method is applied to the server in fig. 1 for illustration, and includes the following steps:
step 202, receiving a risk analysis request sent by a terminal; the risk analysis request carries a target resource identification of the selected virtual resource.
The terminal is provided with a virtual resource acquisition platform. When the user needs to acquire the virtual resource, the virtual resource can be selected at the terminal based on the virtual resource acquisition platform, and risk analysis can be requested to be performed on the virtual resource. The virtual resource acquisition platform provides a plurality of analysis dimension options such as comprehensive analysis, financial analysis, public opinion analysis, analysis in the same party, analysis in the same region, and the like. And the terminal generates a risk analysis request according to the virtual resource and the analysis dimension selected by the user, and sends the risk analysis request to the server.
Step 204, determining a monitoring object corresponding to the target resource identifier, acquiring monitoring data of the monitoring object, and inputting the monitoring data into the risk analysis model to obtain a corresponding risk score.
The virtual resource may be a stock, bond, etc. The monitoring object corresponding to the virtual resource refers to a provider of the virtual resource. Because the importance degree to be considered by different risk monitoring indexes is different for monitoring objects of different industry types. And different attribute characteristics of monitoring objects in different industries are fully considered, the industries of the monitoring objects are distinguished, and different risk analysis models are built aiming at different industry types. The risk analysis model may be a machine learning model.
The monitoring data includes data in multiple dimensions of finance, region, industry, law, public opinion, and the like. The different monitoring data respectively have corresponding data sources, acquisition time and data types. Data types include, but are not limited to, images, audio, text, and numbers. The server preprocesses the monitoring data of different data types. Specifically, the digital data, such as financial data of enterprises, can be used as a main data source for evaluating the quantitative index of the enterprise risk, and can be directly applied to the generation of the monitoring factors after simple processing. However, if the data of the text, image, audio and other data types are required to be subjected to extraction and quantization, the code table in the data is subjected to unified and standardized processing.
And the server calls a corresponding risk analysis model to perform risk scanning on the virtual resources.
In one embodiment, inputting the monitoring data into the risk analysis model to obtain a corresponding risk score includes: acquiring a training sample, and screening a target sample from the training sample; training the first sub-model based on the training sample to obtain a health analysis model; training the second sub-model based on the target sample to obtain a bankruptcy analysis model; inputting the monitoring data into a health analysis model to obtain the health index of the monitored object; inputting the monitoring data into a bankruptcy analysis model to obtain a bankruptcy index of the monitored object; a risk score is calculated based on the health index and the bankruptcy index.
If the analysis dimension selected by the user is 'financial analysis', the server calls a corresponding risk analysis model. The risk analysis model includes a health analysis model and a bankruptcy analysis model. The health analysis model comprises a health analysis sub-model of monitoring angles of multiple dimensions of debt repayment capability, operation capability, profitability, growth capability and profitability quality. The monitoring analysis sub-model may be trained based on LGBM models (Light gradient boosting machine, fast gradient lifting algorithm). The server extracts monitoring indexes of the monitoring object at a plurality of monitoring angles respectively from the monitoring data, and inputs the monitoring indexes of each monitoring angle into a corresponding health analysis sub-model to obtain corresponding health sub-scores. And the server inputs the corresponding health sub-scores of the monitoring angles into a score conversion model to obtain corresponding health indexes. The scoring transformation model can also be trained based on the LGBM model, or can be trained based on other models, without limitation.
The bankruptcy analysis model may be trained on a GBDT model (Gradient Boosting Decision Tree, gradient-lifting decision tree) based on bad sample data of bankruptcy enterprises. Compared with a health analysis model, the sample data adopted in the training of the bankruptcy analysis model are more concentrated. And the server inputs the extracted monitoring index into a bankruptcy analysis model to obtain a bankruptcy index of the monitored object. And the server calculates the financial score corresponding to the target resource identifier according to the health index and the bankruptcy index.
If the analysis dimension selected by the user is "comprehensive analysis", the server invokes the risk analysis model of each analysis dimension to calculate a score of the corresponding analysis dimension, such as a financial score, in the manner described above. A risk score for the monitored subject is calculated based on the plurality of score terms.
In step 206, the target resource identification is marked as a risk case when the risk score exceeds a threshold.
The server compares whether the risk score exceeds a threshold. If yes, the fact that the monitored object has higher default risk is indicated, and the server marks the target resource identifier as a risk case.
And step 208, determining similar cases corresponding to the risk cases according to the monitoring data.
The server pre-stores a plurality of historical cases and a risk profile (noted as bad sample profile) for each historical case in a case library. The bad sample representation includes a plurality of risk tags. The risk tags are used to characterize the questions of which subjects the historical cases are presented to. Over time, the risk means may also vary. In order to improve the matching accuracy of similar cases, the corresponding bad sample images can be dynamically updated.
The server determines one or more of the historical cases as a similar case of the risk case by calculating the similarity of the risk representation of the historical case to the risk representation of the current risk case. Similar cases refer to historical cases that have similar risk characteristics as the current risk case. The server generates a plurality of risk labels of the monitored object based on a plurality of monitoring indexes of the monitored object, and generates a risk portrait (named as a portrait to be matched) of the monitored object by using the plurality of risk labels. The risk analysis model also includes a whitewash analysis model. The rendering analysis model is not only used for predicting whether the monitored object has risk behaviors, but also used for predicting risk clues through similar case matching. And the server calls a rendering analysis model to calculate cosine similarity of the to-be-matched portrait and the bad sample portrait, so as to obtain similarity. If the similarity exceeds the threshold, the server marks the corresponding historical case as a similar case.
At step 210, a plurality of risk points for a risk case are identified based on similar cases.
Each bad sample portrait is associated with a risk indicator for a plurality of time nodes. And the server predicts the risk clues of the monitored objects according to the risk indexes of the plurality of time nodes associated with the matched bad sample portrait.
In one embodiment, identifying a plurality of risk points for a risk case based on similar cases includes: acquiring risk indexes of a plurality of time nodes of similar cases; judging whether the risk cases have the same risk indexes, and judging whether the time sequence of the same risk indexes appearing in the risk cases is consistent with that of similar cases; if yes, determining the same risk index of the last time node, and marking the same risk index as a sign index; the different risk indicators of the time node after the symptom indicator are marked as risk points of the risk case.
At different times the nodes may resemble risk patterns for different historical cases. In other words, similar cases to the current risk case may change over time. The server generates risk cues from both the perspective of "same risk tag as similar case" and "time sequence of same risk tag". Specifically, whether the risk case has the same risk index as the similar case or not and whether the time sequence of the same risk index is consistent with the similar case or not can be judged. If the same risk index exists as the similar case and the appearance time sequence of the same risk index is consistent with the similar case, the server marks the same risk index (marked as a sign index) of the last time node as a risk point. For example, a bad sample enterprise a has 6 abnormal indicators, and enterprise B has occurred 5 of the abnormal indicators and predicts that the 6 th abnormal indicator is likely to occur, so that the 6 th abnormal indicator can be marked as a risk point of enterprise B.
And 212, connecting a plurality of risk points in series to generate a risk clue corresponding to the risk case.
In one embodiment, a plurality of risk points are connected in series to generate a risk cue corresponding to a risk case, including: determining a monitoring period of a risk case; when the monitoring period arrives, returning to the step of acquiring the monitoring data of the monitoring object, and analyzing the risk point of the risk case in the current monitoring period if the risk score still exceeds the threshold value; and connecting the risk points of the monitoring periods in series to generate a risk clue corresponding to the discovery main body.
The monitoring period can be dynamically determined according to the risk score or the industry type of the monitored object, or can be a preset fixed value, and the monitoring period is not limited. The server determines the risk points of the monitoring object in each monitoring period according to the mode, and connects a plurality of risk points in series according to the time sequence to obtain a risk clue corresponding to the monitoring object.
And step 214, based on the risk scores, the similar cases and the risk clues, generating a risk analysis report corresponding to the target resource identifier, and feeding back the risk analysis report to the terminal.
The server can determine the risk category corresponding to the monitored object by clustering the risk scores. Each risk category has a corresponding category description. And carrying out risk prompt according to the risk score and the category description corresponding to the belonging risk category. Compared with the simple provision of a score, the risk prompt based on natural language can improve service interpretation.
The server generates a risk analysis report based on the risk cues, similar cases, and their associated bad sample portraits. In another embodiment, the risk analysis report also shows health sub-scores resulting from the risk analysis process. It is readily understood that the health sub-scores may be presented in the risk analysis report in the form of a chart such as a radar chart or a bar chart.
In this embodiment, according to a risk analysis request sent by a terminal, a monitoring object of a target resource selected by a user may be determined; acquiring monitoring data of a monitoring object, inputting the monitoring data into a risk analysis model, and obtaining a corresponding risk score; when the risk score exceeds a threshold, the target resource identification may be marked as a risk case; according to the monitoring data, determining similar cases corresponding to the risk cases; multiple risk points for obtaining the risk cases can be identified based on the similar cases; the risk points are connected in series, so that a risk clue corresponding to the risk case can be generated; based on the risk scores, similar cases and risk cues, a risk analysis report corresponding to the target resource identification can be generated. Because the risk analysis model can comprehensively consider various risk factors to carry out risk prediction, the risk analysis efficiency is improved; after the risk scores are obtained, similar cases of the risk cases are further determined, risk points, which are possibly appeared by the monitored object at a plurality of time nodes in the future, are predicted based on the similar cases, and the risk analysis report generated based on the information can be convenient for a user to comprehensively and rapidly know the risk condition of the selected virtual resources, so that the risk analysis accuracy is improved.
In one embodiment, as shown in fig. 3, before receiving the risk analysis request sent by the terminal, the method further includes a step of screening the target resource, specifically including:
step 302, receiving a resource acquisition request sent by a terminal; the resource acquisition request carries the target attribute value.
And step 304, acquiring resource factors of a plurality of virtual resources, and inputting the target attribute value and the resource factors of each virtual resource into a regulation model to obtain the resource label of each virtual resource.
And 306, selecting a corresponding expert model according to the resource label, and inputting the resource factors into the expert model to obtain the predicted attribute values of the corresponding virtual resources.
Step 308, screening the virtual resource with the predicted attribute value matched with the target attribute value, and recording as the target resource.
Step 310, determining factor characteristics of the target resource according to the expert model corresponding to the target resource.
Step 312, obtaining the resource information of the target resource, and returning the resource information and the factor characteristic to the terminal; and the terminal further screens the pushed multiple target resources according to the factor characteristics.
When the user needs to acquire the virtual resource, the target attribute value can be set in the virtual resource acquisition platform. The target attribute value may be an acquisition condition of the virtual resource set by the user. The corresponding target attribute values may be different depending on the virtual resource. For example, when the virtual resource is a financial product such as a stock or securities, the corresponding target attribute value may be an expected rate of return, a maximum risk rate, or the like. The target attribute value may be selected from a plurality of gear intervals or gear values given by the virtual resource acquisition platform.
The server prestores resource information of virtual resources in a plurality of transactions in a virtual resource pool. The server also pre-stores a hybrid expert model for screening virtual resources that meet the user-set target attribute values. The hybrid expert model includes a regulatory model and a plurality of expert models. The regulation and control model is used for determining which expert model is selected to calculate the predicted attribute value of the virtual resource. The regulatory model may be trained on the EM algorithm (Expectation Maximization Algorithm ) based on resource information of multiple virtual resources of the historical period.
The hypothetical hybrid expert model includes k expert models. Each expert model is a neural network model. Different expert models are adept at processing data from different data sources. The data source may be a virtual resource provider. Each expert model has a corresponding model number. If the resource label of one virtual resource is i (i is more than or equal to 1 and less than or equal to k), selecting an ith expert model. The resource labels can be data sources for identifying resource information, and corresponding expert models can be selected according to the data sources.
The server screens virtual resources with predicted attribute values matched with target attribute values and records the virtual resources as target resources. And the server determines the factor characteristic of the target resource according to the expert model corresponding to the target resource. Different expert models may reflect different factor characteristics of the virtual resource. The factor characteristic refers to a relationship between a resource factor and a predicted attribute value, for example, a linear relationship in which the larger the factor value of the resource factor is, the higher the predicted attribute value is, or a relationship in which the predicted attribute value exhibits a normal distribution with the resource factor. The server can judge which factor characteristic expression rule different virtual resources follow at different moments by adopting different expert models. The factor characteristic law can be characterized by using generalized linear regression, and can also be expanded into multiple factors, namely, the expression of the hyperplane serving as the multi-factor profit characteristic in space. The server acquires the resource information of the target resource and returns the resource information and the factor characteristic to the terminal. The terminal can further screen the pushed multiple target resources according to the factor characteristics.
In this embodiment, the user only needs to set the target attribute value of the virtual resource desired to be acquired, automatically screens the target resource set conforming to the target attribute value based on the preset virtual resource acquisition strategy, precisely reduces the virtual resource screening range, and improves the virtual resource acquisition efficiency. The factor characteristics of the target resources are further given, and the secondary screening can be assisted by the user according to personal preference, industry experience and the like, so that the virtual resources are obtained more personally, and the accuracy of virtual resource obtaining can be improved.
In one embodiment, determining similar cases corresponding to the risk cases based on the monitored data includes: determining the industry type of the monitoring object corresponding to the virtual resource; acquiring risk data of a risk case; the risk data comprises risk data of virtual resources, risk data of industry types and risk data of monitoring objects; extracting risk labels of risk cases from the risk data; and calculating the similarity between the risk label and the case labels of the prestored multiple historical cases, and marking the historical cases with the similarity exceeding a preset value as similar cases.
The server compares whether the risk score exceeds a first threshold. When the risk score is smaller than or equal to the first threshold, the risk of the virtual resource selected by the user is weak, and in order to improve the risk analysis efficiency, the server directly returns the risk score to the terminal. When the risk score exceeds a first threshold, the risk of the virtual resource selected by the user is relatively strong, and in order to improve the accuracy of risk analysis, the server marks the virtual resource selected by the user as a risk case and further identifies similar cases of the risk case.
The server determines the monitoring object corresponding to the virtual resource selected by the user and the industry type (recorded as the target industry type). The server may crawl risk data related to the virtual resource, risk data related to a target industry type to which the virtual resource belongs, and risk data related to the monitored object using a crawler technology. For example, the server may crawl news public opinion information including keywords from the target website by using the name of the virtual resource, the name of the target industry type, and the name of the monitoring object as crawling keywords, and perform denoising processing on the news public opinion information, and filter advertisement noise, dirty word noise, and the like in the news public opinion information one by one to obtain risk data including only news text content.
The risk data related to the virtual resource comprises the risk data of the virtual resource, the risk data of the issuing main body of the virtual resource and the risk data of the industry to which the virtual resource belongs, so that comprehensive risk identification of the virtual resource is realized, the risk identification is not performed only from the financial data and the credit rating of the virtual resource, and the accuracy rate of virtual resource default risk identification is obviously improved.
The server extracts risk labels of the risk cases from the risk data. Specifically, the data type of the risk data may be an image, audio, text, or number. When the risk data is text or text converted based on images or audios, the server splits the text according to the data amount and the separator. Specifically, the server calculates the data amount of the text, and detects whether the data amount exceeds a preset value. When the data volume exceeds a preset value, the server acquires a preset target data volume, and determines the splitting position of the text according to the target data volume. The server detects whether the split position is located between adjacent separators. When the splitting location is located at one separator, the server splits the text into a plurality of phrases at the splitting location. When the splitting location is located between adjacent separators, the server splits the text into a plurality of phrases at any one of the adjacent separators.
The server prestores regular expressions corresponding to various case types. The regular expression includes one or more risk keywords. And the server carries out regular matching on each split short sentence according to a plurality of preset regular expressions. And the server marks risk keywords contained in the regular expression which is successfully matched as risk labels respectively.
And the server calls a preset similarity evaluation model to calculate the similarity between the risk label and the case labels of the prestored multiple historical cases. Specifically, the server obtains a history case pool. The historical case pool includes a plurality of case types of historical cases. Each historical case has a different case label, each case label being provided with a risk level. The pre-case types may include, in particular, financial cases, legal cases, capital cases, business cases, and the like. Taking the example of a financial case type, the case labels corresponding to the financial case type may include labels for capital structure variation, poor liquidity, and performance loss. The server may analyze the monitoring data of different case types of the default virtual resource by taking the known default virtual resource as an example in advance, obtain case labels under different case types, and set a risk level for the case labels.
And the server matches the obtained risk label with the case label in the history case pool, and when the risk label is successfully matched with the case label, the case label successfully matched with the risk label and the risk grade thereof are written into a matching record. And the server calculates the similarity between each historical case and the current risk case according to the case label recorded in the matching record and the risk level of the case label.
In one embodiment, calculating the similarity of the risk tag to the pre-stored case tags of the plurality of historical cases includes: acquiring word vectors of the risk tags by using a preset word vector model; inputting all word vectors into a pre-trained SVM model (Support Vector Machine ), and calculating confidence coefficients of the word vectors and each case label; determining the case label with the highest confidence as the case label matched with the risk label; acquiring risk levels of the matched case labels; and calculating the similarity between the corresponding historical case and the risk case according to the risk level.
The server acquires the case labels corresponding to all the historical cases, removes the repeated case labels, and obtains a case label table. The server counts the number of times that the case label corresponding to each history case appears in the case label table to generate a case label matrix. After obtaining the case label matrix, the server calculates probability values of occurrence of each case label under the condition that the virtual resource has default risk, and quantifies the obtained probability values into risk levels corresponding to each case label. For example, if the probability value of occurrence of the case label "poor flowability" is 80% to 89%, the risk level corresponding to the case label "poor flowability" is set to 8.
The server acquires word vectors of the case keywords by using a preset word vector model. The preset word vector model may be a word2vec model; the server can acquire word vectors of all risk labels by using a word2vec model, input the acquired word vectors into a pre-trained SVM model, and calculate the confidence coefficient of matching the risk labels with different case labels. The SVM model may be trained using risk data of historical cases as training data. The server determines the case label with the highest confidence as the case label matched with the case keyword. And determining the case label matched with the risk label through the SVM model, so that the accuracy of user intention classification is effectively improved. The server marks the historical cases for which the similarity reaches the second threshold as similar cases.
In this embodiment, compared with the simple feedback of a score to the user, the method further performs deep risk analysis on the virtual resources, performs similar case pushing on the user, facilitates the user to know the risk of the virtual resources more clearly and accurately, and further facilitates more accurate screening of the virtual resources.
It should be understood that, although the steps in the flowcharts of fig. 2 and 3 are shown in order as indicated by the arrows, these steps are not necessarily performed in order as indicated by the arrows. The steps are not strictly limited to the order of execution unless explicitly recited herein, and the steps may be executed in other orders. Moreover, at least some of the steps in fig. 2 and 3 may include multiple sub-steps or stages that are not necessarily performed at the same time, but may be performed at different times, nor does the order in which the sub-steps or stages are performed necessarily occur in sequence, but may be performed alternately or alternately with at least a portion of the other steps or sub-steps of other steps.
In one embodiment, as shown in fig. 4, there is provided a risk analysis report generating apparatus, including: a risk scoring module 402, a cue collection module 404, and a report generation module 406, wherein:
a risk scoring module 402, configured to receive a risk analysis request sent by a terminal; the risk analysis request carries a target resource identifier of the selected virtual resource; and determining a monitoring object corresponding to the target resource identifier, acquiring monitoring data of the monitoring object, and inputting the monitoring data into a risk analysis model to obtain a corresponding risk score.
A thread collection module 404 for marking the target resource identification as a risk case when the risk score exceeds a threshold; determining similar cases corresponding to the risk cases according to the monitoring data; identifying a plurality of risk points for the risk case based on the similar cases; and connecting a plurality of risk points in series to generate a risk clue corresponding to the risk case.
The report generating module 406 is configured to generate a risk analysis report corresponding to the target resource identifier based on the risk score, the similar case and the risk clue, and feed back the risk analysis report to the terminal.
In one embodiment, the apparatus further includes a resource screening module 408, configured to receive a resource acquisition request sent by the terminal; the resource acquisition request carries a target attribute value; acquiring resource factors of a plurality of virtual resources, and inputting a target attribute value and the resource factors of each virtual resource into a regulation model to obtain a resource label of each virtual resource; selecting a corresponding expert model according to the resource annotation, and inputting the resource factors into the expert model to obtain a predicted attribute value of the corresponding virtual resource; screening virtual resources with predicted attribute values matched with target attribute values, and recording the virtual resources as target resources; determining factor characteristics of the target resources according to expert models corresponding to the target resources; acquiring resource information of a target resource, and returning the resource information and factor characteristics to the terminal; and the terminal further screens the pushed multiple target resources according to the factor characteristics.
In one embodiment, risk scoring module 402 is further configured to obtain training samples, and screen target samples from the training samples; training the first sub-model based on the training sample to obtain a health analysis model; training the second sub-model based on the target sample to obtain a bankruptcy analysis model; inputting the monitoring data into a health analysis model to obtain the health index of the monitored object; inputting the monitoring data into a bankruptcy analysis model to obtain a bankruptcy index of the monitored object; a risk score is calculated based on the health index and the bankruptcy index.
In one embodiment, the thread collection module 404 is further configured to determine an industry type to which the monitored object corresponding to the virtual resource belongs; acquiring risk data of a risk case; the risk data comprises risk data of virtual resources, risk data of industry types and risk data of monitoring objects; extracting risk labels of risk cases from the risk data; and calculating the similarity between the risk label and the case labels of the prestored multiple historical cases, and marking the historical cases with the similarity exceeding a preset value as similar cases.
In one embodiment, the thread collection module 404 is further configured to obtain risk indicators for a plurality of time nodes of similar cases; judging whether the risk cases have the same risk indexes, and judging whether the time sequence of the same risk indexes appearing in the risk cases is consistent with that of similar cases; if yes, determining the same risk index of the last time node, and marking the same risk index as a sign index; the different risk indicators of the time node after the symptom indicator are marked as risk points of the risk case.
In one embodiment, the thread collection module 404 is further configured to determine a monitoring period for the risk case; when the monitoring period arrives, returning to the step of acquiring the monitoring data of the monitoring object, and analyzing the risk point of the risk case in the current monitoring period if the risk score still exceeds the threshold value; and connecting the risk points of the monitoring periods in series to generate a risk clue corresponding to the discovery main body.
For specific limitations on the means for generating the risk analysis report, reference may be made to the above limitations on the method for generating the risk analysis report, and will not be described in detail herein. The respective modules in the above-described risk analysis report generation apparatus may be implemented in whole or in part by software, hardware, and combinations thereof. The above modules may be embedded in hardware or may be independent of a processor in the computer device, or may be stored in software in a memory in the computer device, so that the processor may call and execute operations corresponding to the above modules.
In one embodiment, a computer device is provided, which may be a server, the internal structure of which may be as shown in fig. 5. The computer device includes a processor, a memory, a network interface, and a database connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The database of the computer device is used for storing a plurality of historical cases and a risk portrait corresponding to each historical case. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program is executed by a processor to implement a method of generating a risk analysis report.
It will be appreciated by those skilled in the art that the structure shown in FIG. 5 is merely a block diagram of some of the structures associated with the present inventive arrangements and is not limiting of the computer device to which the present inventive arrangements may be applied, and that a particular computer device may include more or fewer components than shown, or may combine some of the components, or have a different arrangement of components.
A computer-readable storage medium, on which a computer program is stored, which when executed by a processor implements the steps of the risk analysis report generation method provided in any one of the embodiments of the present application.
Those skilled in the art will appreciate that implementing all or part of the above-described methods in accordance with the embodiments may be accomplished by way of a computer program stored on a non-transitory computer readable storage medium, which when executed may comprise the steps of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in embodiments provided herein may include non-volatile and/or volatile memory. The nonvolatile memory can include Read Only Memory (ROM), programmable ROM (PROM), electrically Programmable ROM (EPROM), electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double Data Rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous Link DRAM (SLDRAM), memory bus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), among others.
The technical features of the above embodiments may be arbitrarily combined, and all possible combinations of the technical features in the above embodiments are not described for brevity of description, however, as long as there is no contradiction between the combinations of the technical features, they should be considered as the scope of the description.
The foregoing examples illustrate only a few embodiments of the application, which are described in detail and are not to be construed as limiting the scope of the application. It should be noted that it will be apparent to those skilled in the art that several variations and modifications can be made without departing from the spirit of the application, which are all within the scope of the application. Accordingly, the scope of protection of the present application is to be determined by the appended claims.
Claims (8)
1. A method of generating a risk analysis report, the method comprising:
receiving a risk analysis request sent by a terminal; the risk analysis request carries a target resource identifier of the selected virtual resource;
determining a monitoring object corresponding to the target resource identifier, acquiring monitoring data of the monitoring object,
acquiring a training sample, and screening a target sample from the training sample;
Training the first sub-model based on the training sample to obtain a health analysis model;
training the second sub-model based on the target sample to obtain a bankruptcy analysis model;
inputting the monitoring data into the health analysis model to obtain the health index of the monitored object;
inputting the monitoring data into the bankruptcy analysis model to obtain a bankruptcy index of the monitoring object;
calculating a risk score based on the health index and the bankruptcy index;
marking the target resource identification as a risk case when the risk score exceeds a threshold;
determining the industry type of the monitoring object corresponding to the virtual resource;
acquiring risk data of the risk case; the risk data comprises risk data of the virtual resource, industry type risk data and risk data of a monitoring object;
extracting a risk label of the risk case from the risk data;
calculating the similarity between the risk label and case labels of a plurality of prestored historical cases, and marking the historical cases with the similarity exceeding a preset value as similar cases;
acquiring risk indexes of a plurality of time nodes of the similar case;
Judging whether the risk cases have the same risk indexes, and judging whether the time sequence of the same risk indexes of the risk cases is consistent with that of the similar cases;
if yes, determining the same risk index of the last time node, and marking the same risk index as a sign index;
marking different risk indexes of the time node after the symptom indexes as risk points of the risk case;
the risk points are connected in series, and a risk clue corresponding to the risk case is generated;
and generating a risk analysis report corresponding to the target resource identifier based on the risk score, the similar cases and the risk clues, and feeding back the risk analysis report to the terminal.
2. The method of claim 1, further comprising, prior to receiving the risk analysis request sent by the terminal:
receiving a resource acquisition request sent by a terminal; the resource acquisition request carries a target attribute value;
acquiring resource factors of a plurality of virtual resources, and inputting a target attribute value and the resource factors of each virtual resource into a regulation model to obtain a resource label of each virtual resource;
selecting a corresponding expert model according to the resource label, and inputting the resource factors into the expert model to obtain a predicted attribute value of the corresponding virtual resource;
Screening virtual resources with predicted attribute values matched with target attribute values, and recording the virtual resources as target resources;
determining factor characteristics of the target resources according to expert models corresponding to the target resources;
acquiring resource information of the target resource, and returning the resource information and factor characteristics to a terminal; and the terminal further screens the pushed multiple target resources according to the factor characteristics.
3. The method of claim 1, wherein the concatenating the plurality of risk points to generate the risk cue corresponding to the risk case comprises:
determining a monitoring period of the risk case;
when the monitoring period arrives, returning to the step of acquiring the monitoring data of the monitoring object, and analyzing the risk point of the risk case in the current monitoring period if the risk score still exceeds a threshold value;
and connecting the risk points of the monitoring periods in series to generate a risk clue corresponding to the discovery main body.
4. A device for generating a risk analysis report, the device comprising:
the risk scoring module is used for receiving a risk analysis request sent by the terminal; the risk analysis request carries a target resource identifier of the selected virtual resource; determining a monitoring object corresponding to the target resource identifier, acquiring monitoring data of the monitoring object, acquiring a training sample, and screening a target sample from the training sample; training the first sub-model based on the training sample to obtain a health analysis model; training the second sub-model based on the target sample to obtain a bankruptcy analysis model; inputting the monitoring data into the health analysis model to obtain the health index of the monitored object; inputting the monitoring data into the bankruptcy analysis model to obtain a bankruptcy index of the monitoring object; calculating a risk score based on the health index and the bankruptcy index;
A clue collection module for marking the target resource identification as a risk case when the risk score exceeds a threshold; determining the industry type of the monitoring object corresponding to the virtual resource; acquiring risk data of the risk case; the risk data comprises risk data of the virtual resource, industry type risk data and risk data of a monitoring object; extracting a risk label of the risk case from the risk data; calculating the similarity between the risk label and case labels of a plurality of prestored historical cases, and marking the historical cases with the similarity exceeding a preset value as similar cases; acquiring risk indexes of a plurality of time nodes of the similar case; judging whether the risk cases have the same risk indexes, and judging whether the time sequence of the same risk indexes of the risk cases is consistent with that of the similar cases; if yes, determining the same risk index of the last time node, and marking the same risk index as a sign index; marking different risk indexes of the time node after the symptom indexes as risk points of the risk case; the risk points are connected in series, and a risk clue corresponding to the risk case is generated;
And the report generation module is used for generating a risk analysis report corresponding to the target resource identifier based on the risk scores, the similar cases and the risk clues, and feeding back the risk analysis report to the terminal.
5. The apparatus of claim 4, wherein the apparatus further comprises:
the resource screening module is used for receiving a resource acquisition request sent by the terminal; the resource acquisition request carries a target attribute value; acquiring resource factors of a plurality of virtual resources, and inputting a target attribute value and the resource factors of each virtual resource into a regulation model to obtain a resource label of each virtual resource; selecting a corresponding expert model according to the resource label, and inputting the resource factors into the expert model to obtain a predicted attribute value of the corresponding virtual resource; screening virtual resources with predicted attribute values matched with target attribute values, and recording the virtual resources as target resources; determining factor characteristics of the target resources according to expert models corresponding to the target resources; acquiring resource information of the target resource, and returning the resource information and factor characteristics to a terminal; and the terminal further screens the pushed multiple target resources according to the factor characteristics.
6. The apparatus of claim 4, wherein the thread collection module is further configured to determine a monitoring period of the risk case; when the monitoring period arrives, returning to the step of acquiring the monitoring data of the monitoring object, and analyzing the risk point of the risk case in the current monitoring period if the risk score still exceeds a threshold value; and connecting the risk points of the monitoring periods in series to generate a risk clue corresponding to the discovery main body.
7. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor implements the steps of the method of any of claims 1 to 3 when the computer program is executed.
8. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the steps of the method of any of claims 1 to 3.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910012944.3A CN109829629B (en) | 2019-01-07 | 2019-01-07 | Risk analysis report generation method, apparatus, computer device and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910012944.3A CN109829629B (en) | 2019-01-07 | 2019-01-07 | Risk analysis report generation method, apparatus, computer device and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109829629A CN109829629A (en) | 2019-05-31 |
| CN109829629B true CN109829629B (en) | 2023-08-22 |
Family
ID=66860768
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910012944.3A Active CN109829629B (en) | 2019-01-07 | 2019-01-07 | Risk analysis report generation method, apparatus, computer device and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109829629B (en) |
Families Citing this family (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110377519B (en) * | 2019-07-22 | 2023-06-23 | 中国工商银行股份有限公司 | Performance capacity test method, device and equipment of big data system and storage medium |
| CN110659799A (en) * | 2019-08-14 | 2020-01-07 | 深圳壹账通智能科技有限公司 | Attribute information processing method and device based on relational network, computer equipment and storage medium |
| CN110619568A (en) * | 2019-09-17 | 2019-12-27 | 王文斌 | Risk assessment report generation method, device, equipment and storage medium |
| CN110766559B (en) * | 2019-10-30 | 2023-09-29 | 海腾保险代理有限公司 | Label configuration method and device |
| CN111198945A (en) * | 2019-12-03 | 2020-05-26 | 泰康保险集团股份有限公司 | Data processing method, device, medium and electronic equipment |
| CN111736953A (en) * | 2020-06-23 | 2020-10-02 | 深圳市云智融科技有限公司 | Virtual resource delivery method and device, computer equipment and storage medium |
| CN111770095B (en) * | 2020-06-29 | 2023-04-18 | 百度在线网络技术（北京）有限公司 | Detection method, device, equipment and storage medium |
| CN112434082A (en) * | 2020-11-25 | 2021-03-02 | 平安普惠企业管理有限公司 | Operation and maintenance resource management method, device, equipment and medium |
| CN112613569B (en) * | 2020-12-29 | 2024-04-09 | 北京百度网讯科技有限公司 | Image recognition method, training method and device for image classification model |
| CN113283760B (en) * | 2021-05-31 | 2023-04-18 | 浙江环玛信息科技有限公司 | Case flow analysis report generation method and system |
| CN115168686B (en) * | 2022-07-25 | 2023-05-12 | 江苏金农股份有限公司 | Public opinion clue processing method based on feature words |
| CN115964582B (en) * | 2022-11-03 | 2023-09-19 | 太平洋电信股份有限公司 | Network security risk assessment method and system |
| CN116401679B (en) * | 2023-06-08 | 2023-09-05 | 张家港金典软件有限公司 | Security management method and system based on enterprise software vulnerability recognition |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106681910A (en) * | 2016-12-07 | 2017-05-17 | 中国工商银行股份有限公司 | Method and device for generating test cases on basis of mainframe code analysis |
| CN106952155A (en) * | 2017-03-08 | 2017-07-14 | 深圳前海纵腾金融科技服务有限公司 | A kind of collection method and device based on credit scoring |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8751375B2 (en) * | 2009-08-31 | 2014-06-10 | Bank Of America Corporation | Event processing for detection of suspicious financial activity |
| US9324119B2 (en) * | 2013-03-15 | 2016-04-26 | Alert Enterprise | Identity and asset risk score intelligence and threat mitigation |
| US10650460B2 (en) * | 2014-11-04 | 2020-05-12 | Hartford Fire Insurance Company | System to administer risk transfer and property salvage recovery |
-
2019
- 2019-01-07 CN CN201910012944.3A patent/CN109829629B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106681910A (en) * | 2016-12-07 | 2017-05-17 | 中国工商银行股份有限公司 | Method and device for generating test cases on basis of mainframe code analysis |
| CN106952155A (en) * | 2017-03-08 | 2017-07-14 | 深圳前海纵腾金融科技服务有限公司 | A kind of collection method and device based on credit scoring |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109829629A (en) | 2019-05-31 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109829629B (en) | Risk analysis report generation method, apparatus, computer device and storage medium | |
| CN109165840B (en) | Risk prediction processing method, risk prediction processing device, computer equipment and medium | |
| CN108876600B (en) | Early warning information pushing method, device, computer equipment and medium | |
| CN110765244B (en) | Method, device, computer equipment and storage medium for obtaining answering operation | |
| CN109272396B (en) | Customer risk early warning method, device, computer equipment and medium | |
| US20230334254A1 (en) | Fact checking | |
| CN109829628A (en) | Method for prewarning risk, device and computer equipment based on big data | |
| CN109543925B (en) | Risk prediction method and device based on machine learning, computer equipment and storage medium | |
| CN110751533B (en) | Product portrait generation method and device, computer equipment and storage medium | |
| CN110674131A (en) | Financial statement data processing method and device, computer equipment and storage medium | |
| CN110377631B (en) | Case information processing method, device, computer equipment and storage medium | |
| CN112035611B (en) | Target user recommendation method, device, computer equipment and storage medium | |
| US20180336272A1 (en) | Generation of natural language processing events using machine intelligence | |
| CN109801151B (en) | Financial falsification risk monitoring method, device, computer equipment and storage medium | |
| CN114218958A (en) | Work order processing method, device, equipment and storage medium | |
| CN110362798B (en) | Method, apparatus, computer device and storage medium for judging information retrieval analysis | |
| CN111190946A (en) | Report generation method and device, computer equipment and storage medium | |
| CN110377618B (en) | Method, device, computer equipment and storage medium for analyzing decision result | |
| CN112651236A (en) | Method and device for extracting text information, computer equipment and storage medium | |
| CN117196322B (en) | Intelligent wind control method, intelligent wind control device, computer equipment and storage medium | |
| CN112380346B (en) | Financial news emotion analysis method and device, computer equipment and storage medium | |
| CN110362592B (en) | Method, device, computer equipment and storage medium for pushing arbitration guide information | |
| CN112288279A (en) | Business risk assessment method and device based on natural language processing and linear regression | |
| CN111104422A (en) | Training method, device, equipment and storage medium of data recommendation model | |
| CN113256383B (en) | Recommendation method and device for insurance products, electronic equipment and storage medium |
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 |