CN116385155A

CN116385155A - Risk list processing method, apparatus, device and storage medium

Info

Publication number: CN116385155A
Application number: CN202310429345.8A
Authority: CN
Inventors: 卢斯宇; 毛星越; 李婷
Original assignee: Shenzhen Pingan Integrated Financial Services Co ltd
Current assignee: Shenzhen Pingan Integrated Financial Services Co ltd
Priority date: 2023-04-14
Filing date: 2023-04-14
Publication date: 2023-07-04

Abstract

The invention relates to a data processing technology, and discloses a method for optimizing a shield risk list, which comprises the following steps: creating a risk tag for the interactive text of the incoming line corresponding to each customer ID in the customer list data; carrying out risk rating on each client ID in the client list with the risk tag through a preset rating model, and storing the rated client list into a preset risk framework library; according to a preset risk aging table, aging monitoring is carried out on risk labels in a rating client list in a preset risk framework library, invalid risk labels in the rating client list are removed, and a risk framework library in risk aging is obtained. The invention also relates to a blockchain technology, and related data of the client list is stored in the blockchain. The invention can solve the problems of single client single data category, difficult control of risk timeliness, difficult fitting of the actual application scene of the client, easy error shielding of normal clients, bad results affecting the business productivity and the like in the prior art.

Description

Risk list processing method, apparatus, device and storage medium

Technical Field

The present invention relates to the field of big data processing technologies, and in particular, to a method, an apparatus, a device, and a storage medium for optimizing a risk list.

Background

As laws and regulations in the financial and insurance industries become more sophisticated, the requirements for risk management and control are also gradually increasing. A wide variety of risk list products are generated on the market, and clients can use a risk list library to identify and shield risks of clients operating by themselves in the form of interface call. However, the list definition is different, the use mode is single, and the problems of unknown data sources and the like have serious influence on the use experience and the application effect.

On the one hand, if a client only calls a single category risk list, all the wind control demands of the client cannot be completely covered, and the business risk is difficult to obtain perfect early warning and countermeasure preparation. On the other hand, even if the mixed risk list library is used, the data in the library is often not available and can only be applied to the whole library, and the risk aging in the client list is difficult to control, so that the actual application scene of the client is difficult to attach, and the normal client can be shielded by mistake, thereby causing the bad effect of influencing the business productivity. In summary, the diversity of blacklist data is difficult to effectively match with the accuracy of customer needs.

In summary, the present client list data for evaluating the client risk has the problems of single category, difficult control of risk timeliness, difficulty in fitting the actual application scene of the client, easy error shielding of the normal client, bad results affecting the business productivity, and the like.

Disclosure of Invention

The invention provides a method, a device, equipment and a storage medium for optimizing a wind risk list, which mainly aim to solve the problems that in the prior art, the category of client single data used for evaluating the client risk is single, the risk timeliness is not easy to control, the actual application scene of a client is difficult to attach, normal clients are easy to shield by mistake, and bad results affecting the business productivity are caused.

In order to achieve the above object, the present invention provides a method for optimizing a wind shield risk list, the method comprising:

creating a risk tag for the incoming line interaction text corresponding to each client ID in the acquired client list data to obtain a client list with the risk tag; the client list with the risk tag comprises a client ID, incoming times corresponding to the client ID, incoming time of each time, an interactive text of each incoming time and the risk tag corresponding to the interactive text;

carrying out risk rating on each client ID in the client list with the risk tag through a preset rating model to obtain a rated client list, and storing the rated client list into a preset risk framework library; wherein, the preset grading model comprises:

The system comprises an input layer for inputting a customer list with risk labels, an incoming line risk grading layer for carrying out risk grade assessment on incoming line times in the customer list with risk labels according to a preset incoming line times risk assessment table, an incoming line time grading assessment layer for carrying out risk grade assessment on incoming line times in the customer list with risk labels according to a preset incoming line times risk assessment table, a label risk grading layer for carrying out risk grade assessment on risk labels corresponding to each incoming line in the customer list with risk labels according to a preset risk label risk grading assessment table, a customer risk grading layer for carrying out unified grading assessment on incoming line times risk grades obtained by the incoming line risk grading layer, incoming line time risk grades obtained by the time grading assessment layer and label risk grades obtained by the label risk grading layer according to a preset specific gravity table, and an output layer for outputting customer risk grading results obtained by the customer risk grading layer;

according to a preset risk aging table, performing aging monitoring on risk labels in a rating client list in the preset risk framework library, and removing invalid risk labels in the rating client list to obtain a risk framework library in risk aging;

Establishing a corresponding new risk tag according to an interactive text of a target client ID of a real-time incoming line, carrying out new risk rating on the target client according to the new risk tag and a risk tag of a client ID identical to the target client ID in a risk framework library in risk timeliness, and loading the new risk rating into a risk framework library of a rating client single risk framework library in which the target client ID is located.

In a second aspect, in order to solve the above-mentioned problems, the present invention also provides a wind shield risk list optimizing apparatus, the apparatus comprising:

the label creating module is used for creating a risk label for the incoming line interactive text corresponding to each client ID in the acquired client list data to obtain a client list with the risk label; the client list with the risk tag comprises a client ID, incoming times corresponding to the client ID, incoming time of each time, an interactive text of each incoming time and the risk tag corresponding to the interactive text;

the rating module is used for carrying out risk rating on each client ID in the client list with the risk tag through a preset rating model to obtain a rated client list, and storing the rated client list into a preset risk framework library; wherein, the preset grading model comprises:

the aging monitoring module is used for performing aging monitoring on risk labels in rated customer sheets in the preset risk framework library according to a preset risk aging table, and removing invalid risk labels in the rated customer sheets to obtain a risk framework library in risk aging;

The new rating module is used for establishing a corresponding new risk tag according to the interactive text of the target client ID of the real-time incoming line, carrying out new risk rating on the target client according to the new risk tag and the risk tag of the client ID which is the same as the target client ID in the risk framework library in the risk timeliness, and loading the new risk rating into the risk framework library of the rating client single risk framework library in which the target client ID is positioned.

In order to solve the above-mentioned problems, the present invention also provides an electronic device including:

at least one processor; the method comprises the steps of,

a memory communicatively coupled to the at least one processor; wherein, the liquid crystal display device comprises a liquid crystal display device,

the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the steps of the wind shield risk list optimization method as described above.

In a fourth aspect, in order to solve the above-described problems, the present invention also provides a computer-readable storage medium storing a computer program which, when executed by a processor, implements the wind shield risk list optimization method as described above.

According to the method, the device, the equipment and the storage medium for optimizing the wind risk list, risk labels are created through the interactive text of each incoming line of the client list, then risk ratings are carried out according to the risk labels, rated client lists are obtained, then time-effect control is carried out on the risk labels stored in rated client lists of a preset risk framework library, invalid risk labels are removed in time, then new risk labels are added according to the interactive text of the real-time incoming line, risk ratings are carried out on the client lists again, so that each client list has risk data input and output, and the defects that the body quantity of an original risk list library is increased increasingly and a release mechanism is lacked are overcome; the risk assessment of attaching the businesses in different scenes through the effective risk labels gets rid of the fixed definition of other risk list library products on the risk categories of the lists, and an application scheme of the special risk labels is built for the business volume; the risk labels in the client list for evaluating the clients are all in the effective period, so that the problem that the normal clients are shielded by mistake and bad consequences affecting the business productivity are avoided.

Drawings

FIG. 1 is a flowchart illustrating a method for optimizing a wind shield risk list according to an embodiment of the present invention;

FIG. 2 is a schematic diagram of a device for optimizing a wind shield risk list according to an embodiment of the present invention;

FIG. 3 is a schematic diagram illustrating an internal structure of an electronic device for implementing a method for optimizing a wind shield risk list according to an embodiment of the present invention;

the achievement of the objects, functional features and advantages of the present invention will be further described with reference to the accompanying drawings, in conjunction with the embodiments.

Detailed Description

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 invention.

The embodiment of the application can acquire and process the related data based on the artificial intelligence technology. Among these, artificial intelligence (Artificial Intelligence, AI) is the theory, method, technique and application system that uses a digital computer or a digital computer-controlled machine to simulate, extend and extend human intelligence, sense the environment, acquire knowledge and use knowledge to obtain optimal results.

Artificial intelligence infrastructure technologies generally include technologies such as sensors, dedicated artificial intelligence chips, cloud computing, distributed storage, big data processing technologies, operation/interaction systems, mechatronics, and the like. The artificial intelligence software technology mainly comprises a computer vision technology, a robot technology, a biological recognition technology, a voice processing technology, a natural language processing technology, machine learning/deep learning and other directions.

The invention provides a method for optimizing a wind shield risk list. Referring to fig. 1, a flowchart of a method for optimizing a wind shield risk list according to an embodiment of the invention is shown. The method may be performed by an apparatus, which may be implemented in software and/or hardware.

In this embodiment, the method for optimizing the wind shield risk list includes:

step S110, creating a risk tag for an incoming line interactive text corresponding to each client ID in the acquired client list data to obtain a client list with the risk tag; the client list with the risk labels comprises a client ID, incoming times corresponding to the client ID, incoming time of each time, interactive text of each incoming time and the risk labels corresponding to the interactive text.

Specifically, each incoming line interactive text of each client ID retains interactive words among each other, and a risk label corresponding to the interactive text, such as complaints, abuse and the like, is created by identifying risk prone words in the interactive text, so that a client list with the risk label is obtained. The client list with the risk labels comprises clients, the incoming times of the clients, incoming times of each time, interactive texts of each incoming time and risk labels corresponding to the interactive texts.

As an optional embodiment of the present invention, the client list data is stored in a blockchain, creating a risk tag for the incoming line interactive text corresponding to each client ID in the acquired client list data, and obtaining the client list with the risk tag includes:

carrying out data cleaning treatment on the client list data to unify the structural form of the related data of the client list data and obtain a structured client list; the structured client list comprises a client ID and an incoming line interaction text corresponding to the client ID;

and creating a risk tag for the incoming line interactive text in the structured client list to obtain a client list with the risk tag.

Specifically, relevant data of a client list is collected from a risk list library, including risk complaint clients of various professional companies, high Guan Bai list clients, overdue default clients and the like, the clients generally perform incoming line consultation or incoming line complaint services, for example, a client A incoming line, consults a certain service, complaints on a certain service and the like, and finally, certain default conditions appear on the service performed by the client A, all the clients are recorded in the client list, and the data generated by each client serve as relevant data of the corresponding client in the client list. Because the sources of the client list data collection are different, for example, from insurance business or from bank loan business, etc., the storage form (such as date format) of the relevant data of each client list, etc., will be different, so the structural form of the relevant data of the client list needs to be unified, thereby obtaining the structured client list. The structured customer list must include interactive text for the customer and incoming lines corresponding to the customer.

As an optional embodiment of the present invention, performing data cleaning processing on client list data to unify structural forms of relevant data of the client list data, and obtaining a structured client list includes:

according to the preset time limit, eliminating the expiration data in the related data of the client list to obtain the related data meeting the preset time limit;

and carrying out format unification processing on the related data meeting the preset time limit according to a preset format to obtain a structured client list with unified data format.

Specifically, relevant data with long date may exist in the acquired client list, for example, interactive data before ten years, and in order to more rationalize the data according to which the clients evaluate the risk level, the relevant data with long date needs to be deleted, so that the expiration data exceeding the preset time period is cleared through the preset time period, for example, within two years in one year, so as to obtain relevant data meeting the preset time period; and then, carrying out format unification processing on the related data meeting the preset time limit according to a preset format, so that the formats of the related data meeting the preset time limit in the client list are unified, and obtaining a structured client list with unified data format.

As an optional embodiment of the present invention, creating a risk tag for incoming line interactive text in a structured client list, obtaining a client list with risk tags includes:

extracting a risk tendency speaking operation from the interactive text in the structured client list through a preset natural language understanding model, and taking the risk tendency speaking operation as a risk basis speaking operation; the method comprises the steps that a natural language understanding model is preset, wherein the natural language understanding model comprises an input layer for inputting interactive texts, an identification layer for identifying risk prone languages in the interactive texts and a language outputting layer for outputting the risk prone languages;

creating a risk label corresponding to the risk-based conversation based on a preset risk label list according to the risk-based conversation to obtain a client list with the risk label; the preset risk tag list comprises a risk basis phone and risk tags corresponding to the risk basis phone.

Specifically, a client text semantic understanding technology of public hawk is introduced (which is a voice understanding technology currently existing in the field and is not described in detail here), a natural language understanding model is preset, and a risk trend speaking operation can be extracted from an interactive text through the preset natural language understanding model and is used as a risk basis speaking operation; and then, according to the risk-based conversation, finding out a corresponding risk label from a preset risk label list, thereby obtaining a client list with the risk label. The preset natural language understanding model can be obtained by collecting a large number of risk prone speaking samples for model training.

S120, carrying out risk rating on each client ID in the client list with the risk tag through a preset grading model to obtain a rated client list, and storing the rated client list into a preset risk framework library; the preset grading model comprises the following steps: the system comprises an input layer for inputting a customer list with risk labels, an incoming line risk grading layer for carrying out risk grade evaluation on incoming line times in the customer list with risk labels according to a preset incoming line times risk evaluation table, an incoming line time grading evaluation layer for carrying out risk grade evaluation on incoming line times in the customer list with risk labels according to a preset incoming line times risk evaluation table, a label risk grading layer for carrying out risk grade evaluation on risk labels corresponding to incoming lines in the customer list with risk labels according to a preset risk label risk grading evaluation table, a customer risk grading layer for carrying out unified grading evaluation on incoming line times risk grades obtained by the incoming line risk grading layer, incoming line time risk grades obtained by the time grading evaluation layer and label risk grades obtained by the label risk grading layer according to a preset proportion table, and an output layer for outputting customer risk grading results obtained by the customer risk grading layer.

Specifically, risk rating is carried out on each client ID in the client list with the risk tag through a preset grading model, so that risk rating efficiency of the client IDs in the client list is improved. Inputting the client list with the risk tag into a preset grading model, and carrying out risk grading on the corresponding client according to the content such as the risk tag in the client list with the risk tag, so as to obtain a graded client list; wherein the rating client list comprises a client ID and a rating corresponding to the client; and then storing the rating client list into a preset risk framework library. The preset grading model is preferably an extremely gradient lifting tree model, and can refer to multiple factors for carrying out risk grading on clients in multiple dimensions, wherein the model comprises an input layer, an incoming line risk grading layer, an incoming line time grading evaluation layer, a tag risk grading layer, a client risk grading layer and an output layer; corresponding risk assessment tables are respectively preset in the incoming line risk classification layer, the incoming line time classification assessment layer and the tag risk classification layer and are used as risk assessment bases of the factors or dimensions; and finally comprehensively evaluating the risk grades of all the dimensions through a preset proportion table, thereby obtaining the final customer risk grade. The preset proportion table comprises incoming frequency risk levels, reference proportion corresponding to the incoming frequency risk levels, incoming time risk levels, time reference proportion corresponding to the incoming time risk levels, label risk levels and label reference proportion corresponding to the label risk levels.

As an optional embodiment of the present invention, performing risk rating on each client ID in the client list with the risk tag through a preset rating model, to obtain a rated client list, and storing the rated client list in a preset risk framework library includes:

inputting the client list with the risk labels into a preset grading model, and respectively carrying out risk grade assessment on the preset grading model according to the incoming times, incoming times and risk labels corresponding to the incoming times of each client in the client list with the risk labels to respectively obtain incoming times risk grades, incoming times risk grades and label risk grades;

and carrying out comprehensive risk level assessment on the clients according to the specific gravity corresponding to each incoming line times, incoming line times and label risk levels to obtain rated client orders, and storing the rated client orders into a preset risk framework library.

Specifically, the client list with the risk tag may include a plurality of clients, each client corresponds to the incoming line times, the incoming line time of each time, the interactive text corresponding to each incoming line and the risk tag corresponding to the interactive text, after the client list with the risk tag is input into a preset grading model, the model carries out risk grade assessment on each client according to the incoming line times, the incoming line time of each time and the risk tag corresponding to each incoming line, and obtains the incoming line times risk grade, the incoming line time risk grade and the tag risk grade of each client respectively; and then, carrying out comprehensive risk level assessment on the clients according to the respective proportions of the incoming line times risk level, the incoming line times risk level and the label risk level, thereby obtaining the risk rating of each client in the whole client list, namely, a rating client list, wherein the rating client list comprises the clients and the risk levels corresponding to the clients.

S130, according to a preset risk aging table, aging monitoring is conducted on risk labels in a rating client list in a preset risk framework library, invalid risk labels in the rating client list are removed, and a risk framework library in risk aging is obtained.

Specifically, according to a preset risk aging table pre-stored in a preset risk framework library, aging monitoring is performed on risk labels in the rating client list, and invalid risk labels in the rating client list are removed, for example, the overdue record of credit investigation is invalid for 5 years and the aging of insurance banning is 2 years; when the risk label corresponding to a certain incoming line of a certain customer is a credit overdue, deleting the risk label when the incoming line time reaches five years, so that data in a preset risk framework library are input and output, and all the data are valid data, and the mistaken shielding of the customer is avoided.

As an optional embodiment of the present invention, according to a preset risk aging table, performing aging monitoring on risk labels in a rating client list in a preset risk framework library, removing risk labels that fail in the rating client list, and obtaining a risk framework library in risk aging includes:

optionally selecting a risk label from a rating client list in a preset risk framework library as a risk label for undetermined aging; wherein, each risk tag in the rating client list corresponds to the incoming line time of the client;

Taking the incoming line time of a customer corresponding to the risk tag to be aged as the starting time of the risk tag to be aged;

acquiring a risk label which is the same as the risk label to be subjected to the undetermined aging from a preset risk aging table as a target risk label; the preset risk aging table comprises risk labels and aging corresponding to the risk labels;

and removing the invalid risk label subjected to the fixed aging according to the aging corresponding to the target risk label based on the starting time of the risk label subjected to the fixed aging, so as to obtain a risk framework library in the risk aging.

Specifically, a risk label is selected from a rating client list in a preset risk framework library as a risk label for undetermined ageing, and because the risk labels are obtained according to an interactive text of incoming lines, the incoming line time of a client corresponding to the risk label for undetermined ageing is taken as the starting time of the risk label for undetermined ageing, then the risk label which is the same as the risk label for undetermined ageing is obtained from the preset risk ageing table as a target risk label, and based on the starting time of the risk label for undetermined ageing, the risk label for undetermined ageing is removed according to ageing corresponding to the target risk label, so that a risk framework library in risk ageing is obtained.

S140, establishing a corresponding new risk label according to the interactive text of the target client ID of the real-time incoming line, carrying out new risk rating on the target client according to the new risk label and the risk label of the client ID which is the same as the target client ID in a risk framework library in risk timeliness, and loading the new risk rating to a rating client list where the target client ID is located.

Specifically, incoming lines of clients are received in real time, the clients of the incoming lines in real time are used as target clients, new risk labels are created for interaction texts of staff and target users according to the risk label creation method, then rated client lists corresponding to the target clients are obtained from a risk framework library in risk timeliness, the interaction texts of the target clients and the new risk labels are correspondingly added to corresponding incoming line positions of the corresponding users in the rated client lists, incoming line time is marked, incoming line times are updated, new risk ratings are carried out on the target clients according to the new risk labels and the risk labels in the timeliness of the target clients in the rated client lists, and the new risk ratings are loaded into the rated client lists of the target clients.

As an optional embodiment of the present invention, establishing a corresponding new risk tag according to an interactive text of a target client ID of a real-time incoming line, and performing a new risk rating on the target client ID according to the new risk tag and a risk tag of a client identical to the target client ID in a risk framework library within a risk timeframe, and loading the new risk rating into a rating client list where the target client ID is located includes:

The interactive voice of the target client ID is obtained in real time, the interactive voice is converted into an interactive text through a voice conversion technology, and the interactive text of the target client ID of the real-time incoming line is obtained;

creating a corresponding risk tag for the interactive text of the target client ID to obtain a new risk tag of the target client;

acquiring a risk label of a client with the same ID as a target client from a risk framework library in risk timeliness as a historical risk label of the target client;

and carrying out new risk rating on the target client according to the new risk label of the target client and the historical risk label of the target client, and loading the new risk rating into a rating client list of the target client.

Specifically, the interactive voice of the target client is converted into an interactive text through a voice conversion technology, so that the interactive text of the target client of the real-time incoming line is obtained; then acquiring a risk label of the same customer as the target customer from a rating customer list of a risk framework library in risk timeliness as a historical risk label of the target customer; and then, carrying out new risk rating on the target client according to the new risk label of the target client and the historical risk label of the target client by a previous risk rating method, and loading the new risk rating into a rating client list of the target client.

Fig. 2 is a functional block diagram of a wind shield risk list optimizing apparatus according to an embodiment of the present invention.

The wind damage risk list optimizing apparatus 200 of the present invention may be installed in an electronic device. The wind risk list optimizing apparatus may include a tag creation module 210, a rating module 220, an aging monitoring module 230, and a new rating module 240 according to the implemented functions. The module of the present invention may also be referred to as a unit, meaning a series of computer program segments capable of being executed by the processor of the electronic device and of performing fixed functions, stored in the memory of the electronic device.

In the present embodiment, the functions concerning the respective modules/units are as follows:

the tag creation module 210 is configured to create a risk tag for the incoming line interaction text corresponding to each client ID in the obtained client list data, so as to obtain a client list with risk tags; the client list with the risk labels comprises a client ID, incoming times corresponding to the client ID, incoming time of each time, interactive text of each incoming time and the risk labels corresponding to the interactive text.

As an alternative embodiment of the present invention, the tag creation module 210 further includes: a data cleansing unit and a label creation unit (not shown in the figure). Wherein, the liquid crystal display device comprises a liquid crystal display device,

the data cleaning unit is used for performing data cleaning processing on the client list data to unify the structural forms of the related data of the client list data and obtain a structured client list; the structured client list comprises a client ID and an incoming line interaction text corresponding to the client ID;

and the label creating unit is used for creating a risk label for the incoming line interactive text in the structured client list to obtain a client list with the risk label.

Specifically, relevant data of a client list is collected from a risk list library, including risk complaint clients of various professional companies, high Guan Bai list clients, overdue default clients and the like, the clients generally perform incoming line consultation or incoming line complaint services, for example, a client A incoming line, consults a certain service, complaints on a certain service and the like, and finally, certain default conditions appear on the service performed by the client A, all the clients are recorded in the client list, and the data generated by each client serve as relevant data of the corresponding client in the client list. Because the sources of the client list data collection are different, for example, from insurance business or from bank loan business, etc., the storage form (such as date format) of the relevant data of each client list, etc., will be different, so the structural form of the relevant data of the client list needs to be unified through the data cleaning unit, thereby obtaining the structured client list. And creating a risk tag for the incoming line interactive text in the structured client list by a tag creation unit to obtain a client list with the risk tag.

As an alternative embodiment of the present invention, the data cleansing unit further comprises: a data-cleaning subunit and a structural unifying subunit (not shown). Wherein, the liquid crystal display device comprises a liquid crystal display device,

the data clearing subunit is used for clearing the expired data in the related data of the client list according to the preset time limit to obtain the related data meeting the preset time limit;

and the structure unifying subunit is used for performing format unification processing on the related data meeting the preset time limit according to a preset format to obtain a structured client list with uniform data format.

Specifically, relevant data with long date may exist in the acquired client list, for example, interactive data before ten years, and in order to more rationalize the data according to which the clients evaluate the risk level, the relevant data with long date needs to be deleted, so that the data clearing subunit clears the expiration data exceeding the preset time period according to the preset time period, for example, within two years in one year, so as to obtain relevant data meeting the preset time period; and then, carrying out format unification processing on the related data meeting the preset time limit according to a preset format through a structure unification subunit, so that the formats of the related data meeting the preset time limit in the client list are unified, and obtaining the structured client list with unified data format.

As an alternative embodiment of the present invention, the label creation unit further comprises: the risk is based on the extraction sub-unit and the label creation sub-unit (not shown in the figure). Wherein, the liquid crystal display device comprises a liquid crystal display device,

the risk basis extraction subunit is used for extracting risk tendency dialogs from the interactive texts in the structured client list through a preset natural language understanding model, and taking the risk tendency dialogs as risk basis dialogs; the method comprises the steps that a natural language understanding model is preset, wherein the natural language understanding model comprises an input layer for inputting interactive texts, an identification layer for identifying risk prone languages in the interactive texts and a language outputting layer for outputting the risk prone languages;

the label creation subunit is used for creating a risk label corresponding to the risk-based conversation based on a preset risk label list according to the risk-based conversation to obtain a client list with the risk label; the preset risk tag list comprises a risk basis phone and risk tags corresponding to the risk basis phone.

Specifically, a client text semantic understanding technology of public hawk is introduced (which is a voice understanding technology currently existing in the field and is not described in detail here), a natural language understanding model is preset, and a risk-prone speaking operation can be extracted from the interactive text by using the preset natural language understanding model through a risk-based extraction subunit and is used as a risk-based speaking operation; and then, according to the risk basis, the label creation subunit finds out a corresponding risk label from a preset risk label list, so that a client list with the risk label is obtained. The preset natural language understanding model can be obtained by collecting a large number of risk prone speaking samples for model training.

The rating module 220 is configured to perform risk rating on each client ID in the client list with the risk tag through a preset rating model, obtain a rated client list, and store the rated client list into a preset risk framework library; the preset grading model comprises the following steps: the system comprises an input layer for inputting a customer list with risk labels, an incoming line risk grading layer for carrying out risk grade evaluation on incoming line times in the customer list with risk labels according to a preset incoming line times risk evaluation table, an incoming line time grading evaluation layer for carrying out risk grade evaluation on incoming line times in each time in the customer list with risk labels according to a preset incoming line times risk evaluation table, a label risk grading layer for carrying out risk grade evaluation on risk labels corresponding to each incoming line in the customer list with risk labels according to a preset risk label risk grading evaluation table, a customer risk grading layer for carrying out unified grading evaluation on incoming line times risk grades obtained by the incoming line risk grading layer, incoming line time risk grades obtained by the time grading evaluation layer and label risk grades obtained by the label risk grading layer according to a preset proportion table, and an output layer for outputting customer risk grading results obtained by the customer risk grading layer.

As an alternative embodiment of the present invention, the rating module 220 further includes a risk assessment unit and an assessment integration unit (not shown). Wherein, the liquid crystal display device comprises a liquid crystal display device,

the risk assessment unit is used for inputting the client list with the risk tag into a preset grading model, and the preset grading model carries out risk grade assessment according to the incoming number of each client in the client list with the risk tag, incoming time of each time and the risk tag corresponding to each incoming time to obtain incoming number risk grade, incoming time risk grade and tag risk grade respectively;

and the evaluation comprehensive unit is used for evaluating the comprehensive risk level of the client according to the specific gravity corresponding to each of the incoming times risk level, the incoming times risk level and the label risk level to obtain a rated client list, and storing the rated client list into a preset risk framework library.

Specifically, the client list with the risk tag may include a plurality of clients, each client corresponds to the incoming line times, the incoming line time of each time, the interactive text corresponding to each incoming line and the risk tag corresponding to the interactive text, after the client list with the risk tag is input into a preset classification model through a risk assessment unit, the model carries out risk level assessment on each client according to the incoming line times, the incoming line time of each time and the risk tag corresponding to each incoming line, and obtains the incoming line times risk level, the incoming line time risk level and the tag risk level of each client respectively; and then, carrying out comprehensive risk level assessment on the clients through an assessment comprehensive unit according to the respective proportions of the incoming number risk level, the incoming time risk level and the tag risk level, thereby obtaining the risk rating of each client in the whole client list, namely, a rating client list, wherein the rating client list comprises the clients and the risk levels corresponding to the clients.

The aging monitoring module 230 is configured to perform aging monitoring on risk labels in a rating client list in a preset risk framework library according to a preset risk aging table, and reject invalid risk labels in the rating client list to obtain a risk framework library in risk aging.

As an alternative embodiment of the present invention, the aging monitor module 230 further includes: a tag selection unit, a start time determination unit, a target tag determination unit, and a culling unit (not shown in the figure).

Wherein, the liquid crystal display device comprises a liquid crystal display device,

the label selecting unit is used for selecting a risk label from rating client sheets in a preset risk framework library as a risk label for undetermined aging; wherein, each risk tag in the rating client list corresponds to the incoming line time of the client;

The starting time determining unit is used for taking the incoming time of the customer corresponding to the risk tag to be subjected to the predetermined ageing as the starting time of the risk tag to be subjected to the predetermined ageing;

the target label determining unit is used for acquiring the risk label which is the same as the risk label to be subjected to the predetermined aging from the preset risk aging table as a target risk label; the preset risk aging table comprises risk labels and aging corresponding to the risk labels;

the eliminating unit is used for eliminating the invalid risk label of the undetermined ageing according to the ageing corresponding to the target risk label based on the starting time of the risk label of the undetermined ageing, and obtaining a risk framework library in the risk ageing.

Specifically, the risk labels are selected from a rating customer list in a preset risk framework library by a label selection unit to serve as risk labels for undetermined ageing, and because the risk labels are obtained according to interactive texts of incoming wires, incoming wire time of a corresponding customer is used as starting time of the risk labels for undetermined ageing by a starting time determination unit, then the risk labels which are the same as the risk labels for undetermined ageing are obtained from a preset risk ageing table by a target label determination unit to serve as target risk labels, and the failure risk labels for undetermined ageing are removed by a removal unit based on the starting time of the risk labels for undetermined ageing according to ageing corresponding to the target risk labels, so that the risk framework library in risk ageing is obtained.

The new rating module 240 is configured to establish a corresponding new risk tag according to the interactive text of the target client ID of the real-time incoming line, and perform new risk rating on the target client according to the new risk tag and the risk tag of the client ID identical to the target client ID in the risk framework library in the risk timeframe, and load the new risk rating into the rating client list where the target client ID is located.

As an alternative embodiment of the present invention, the new rating module 240 further includes a text conversion unit, a tag creation unit, a historical risk tag determination unit, and a rating unit (not shown in the figure). Wherein, the liquid crystal display device comprises a liquid crystal display device,

The text conversion unit is used for acquiring the interactive voice of the target client ID in real time, converting the interactive voice into an interactive text through a voice conversion technology, and obtaining the interactive text of the target client ID of the real-time incoming line;

the label creating unit is used for creating a corresponding risk label for the interactive text of the target client ID to obtain a new risk label of the target client;

the historical risk tag determining unit is used for acquiring a risk tag of a client with the same ID as the target client from a risk framework library in risk timeliness and taking the risk tag as a historical risk tag of the target client;

and the rating unit is used for carrying out new risk rating on the target client according to the new risk label of the target client and the historical risk label of the target client, and loading the new risk rating into the rating client list of the target client.

Specifically, the interactive voice of the target client is converted into an interactive text by a text conversion unit through a voice conversion technology, so that the interactive text of the target client of the real-time incoming line is obtained; then creating a corresponding risk tag for an interactive text of a target client ID through a tag creation unit to obtain a new risk tag of the target client, and acquiring a risk tag of the client identical to the target client from a rating client list of a risk framework library in risk timeliness through a history risk tag determination unit to serve as a history risk tag of the target client; and finally, carrying out new risk rating on the target client according to the new risk tag of the target client and the historical risk tag of the target client by a rating unit according to the previous risk rating method, and loading the new risk rating into a rating client list of the target client.

Fig. 3 is a schematic structural diagram of an electronic device for implementing a method for optimizing a wind shield risk list according to an embodiment of the present invention.

The electronic device 1 may include a processor 10, a memory 11, and a bus, and may further include a computer program, such as a wind shield risk list optimizing program 12, stored in the memory 11 and executable on the processor 10.

The memory 11 includes at least one type of readable storage medium, including flash memory, a mobile hard disk, a multimedia card, a card memory (e.g., SD or DX memory, etc.), a magnetic memory, a magnetic disk, an optical disk, etc. The memory 11 may in some embodiments be an internal storage unit of the electronic device 1, such as a removable hard disk of the electronic device 1. The memory 11 may also be an external storage device of the electronic device 1 in other embodiments, for example, a plug-in mobile hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card) or the like, which are provided on the electronic device 1. Further, the memory 11 may also include both an internal storage unit and an external storage device of the electronic device 1. The memory 11 may be used not only for storing application software installed in the electronic device 1 and various kinds of data, such as codes of a wind shield risk list optimization program, etc., but also for temporarily storing data that has been output or is to be output.

The processor 10 may be comprised of integrated circuits in some embodiments, for example, a single packaged integrated circuit, or may be comprised of multiple integrated circuits packaged with the same or different functions, including one or more central processing units (Central Processing unit, CPU), microprocessors, digital processing chips, graphics processors, various control chips, and the like. The processor 10 is a Control Unit (Control Unit) of the electronic device, connects various components of the entire electronic device using various interfaces and lines, and executes various functions of the electronic device 1 and processes data by running or executing programs or modules (e.g., a wind shield risk list optimizing program, etc.) stored in the memory 11, and calling data stored in the memory 11.

The bus may be a peripheral component interconnect standard (peripheral component interconnect, PCI) bus or an extended industry standard architecture (extended industry standard architecture, EISA) bus, among others. The bus may be classified as an address bus, a data bus, a control bus, etc. The bus is arranged to enable a connection communication between the memory 11 and at least one processor 10 etc.

Fig. 3 shows only an electronic device with components, it being understood by a person skilled in the art that the structure shown in fig. 3 does not constitute a limitation of the electronic device 1, and may comprise fewer or more components than shown, or may combine certain components, or may be arranged in different components.

For example, although not shown, the electronic device 1 may further include a power source (such as a battery) for supplying power to each component, and preferably, the power source may be logically connected to the at least one processor 10 through a power management device, so that functions of charge management, discharge management, power consumption management, and the like are implemented through the power management device. The power supply may also include one or more of any of a direct current or alternating current power supply, recharging device, power failure detection circuit, power converter or inverter, power status indicator, etc. The electronic device 1 may further include various sensors, bluetooth modules, wi-Fi modules, etc., which will not be described herein.

Further, the electronic device 1 may also comprise a network interface, optionally the network interface may comprise a wired interface and/or a wireless interface (e.g. WI-FI interface, bluetooth interface, etc.), typically used for establishing a communication connection between the electronic device 1 and other electronic devices.

The electronic device 1 may optionally further comprise a client interface, which may be a Display, an input unit, such as a Keyboard (Keyboard), or a standard wired interface, a wireless interface. Alternatively, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, an OLED (Organic Light-Emitting Diode) touch, or the like. The display may also be referred to as a display screen or display unit, as appropriate, for displaying information processed in the electronic device 1 and for displaying a visual client interface.

It should be understood that the embodiments described are for illustrative purposes only and are not limited to this configuration in the scope of the patent application.

The wind-shield risk list optimizing program 12 stored in the memory 11 of the electronic device 1 is a combination of a plurality of instructions, which when executed in the processor 10, can realize:

creating a risk tag for the interactive text of the incoming line corresponding to each client ID in the acquired client list data to obtain a client list with the risk tag; the client list with the risk tag comprises a client ID, incoming times corresponding to the client ID, incoming time of each time, interactive text of each incoming time and the risk tag corresponding to the interactive text;

Carrying out risk rating on each client ID in the client list with the risk tag through a preset rating model to obtain a rated client list, and storing the rated client list into a preset risk framework library; the preset grading model comprises the following steps:

the system comprises an input layer for inputting a customer list with risk labels, an incoming line risk grading layer for carrying out risk grade evaluation on incoming line times in the customer list with risk labels according to a preset incoming line times risk evaluation table, an incoming line time grading evaluation layer for carrying out risk grade evaluation on incoming line times in each time in the customer list with risk labels according to a preset incoming line times risk evaluation table, a label risk grading layer for carrying out risk grade evaluation on risk labels corresponding to each incoming line in the customer list with risk labels according to a preset risk label risk grading evaluation table, a customer risk grading layer for carrying out unified grading evaluation on incoming line times risk grades obtained by the incoming line risk grading layer, incoming line time risk grades obtained by the time grading evaluation layer and label risk grades obtained by the label risk grading layer according to a preset proportion table, and an output layer for outputting customer risk grading results obtained by the customer risk grading layer;

According to a preset risk aging table, performing aging monitoring on risk labels in a rating client list in a preset risk framework library, and removing invalid risk labels in the rating client list to obtain a risk framework library in risk aging;

establishing a corresponding new risk label according to an interactive text of a target client ID of a real-time incoming line, carrying out new risk rating on a target client according to the new risk label and a risk label of a client ID which is the same as the target client ID in a risk framework library in risk timeliness, and loading the new risk rating to a rating client list where the target client ID is located.

Specifically, the specific implementation method of the above instructions by the processor 10 may refer to the description of the relevant steps in the corresponding embodiment of fig. 1, which is not repeated herein. It should be emphasized that, to further ensure the privacy and security of the relevant data of the client list, the relevant data of the client list may also be stored in a blockchain node.

Further, the modules/units integrated in the electronic device 1 may be stored in a computer readable storage medium if implemented in the form of software functional units and sold or used as separate products. The computer readable medium may include: any entity or device capable of carrying the computer program code, a recording medium, a U disk, a removable hard disk, a magnetic disk, an optical disk, a computer Memory, a Read-Only Memory (ROM).

In the several embodiments provided in the present invention, it should be understood that the disclosed apparatus, device and method may be implemented in other manners. For example, the above-described apparatus embodiments are merely illustrative, and for example, the division of the modules is merely a logical function division, and there may be other manners of division when actually implemented.

The modules described as separate components may or may not be physically separate, and components shown as modules may or may not be physical units, may be located in one place, or may be distributed over multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

In addition, each functional module in the embodiments of the present invention may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit. The integrated units can be realized in a form of hardware or a form of hardware and a form of software functional modules.

It will be evident to those skilled in the art that the invention is not limited to the details of the foregoing illustrative embodiments, and that the present invention may be embodied in other specific forms without departing from the spirit or essential characteristics thereof.

The present embodiments are, therefore, to be considered in all respects as illustrative and not restrictive, the scope of the invention being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein. Any reference signs in the claims shall not be construed as limiting the claim concerned.

The blockchain is a novel application mode of computer technologies such as distributed data storage, point-to-point transmission, consensus mechanism, encryption algorithm and the like. The Blockchain (Blockchain), which is essentially a decentralised database, is a string of data blocks that are generated by cryptographic means in association, each data block containing a batch of information of network transactions for verifying the validity of the information (anti-counterfeiting) and generating the next block. The blockchain may include a blockchain underlying platform, a platform product services layer, an application services layer, and the like.

Furthermore, it is evident that the word "comprising" does not exclude other elements or steps, and that the singular does not exclude a plurality. A plurality of units or means recited in the system claims can also be implemented by means of software or hardware by means of one unit or means. The terms second, etc. are used to denote a name, but not any particular order.

Finally, it should be noted that the above-mentioned embodiments are merely for illustrating the technical solution of the present invention and not for limiting the same, and although the present invention has been described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that modifications and equivalents may be made to the technical solution of the present invention without departing from the spirit and scope of the technical solution of the present invention.

Claims

1. A risk list optimization method applied to an electronic device, the method comprising:

creating a risk tag for the interactive text of the incoming line corresponding to each client ID in the acquired client list data to obtain a client list with the risk tag; the client list with the risk tag comprises a client ID, incoming times corresponding to the client ID, incoming time of each time, an interactive text of each incoming time and the risk tag corresponding to the interactive text;

establishing a corresponding new risk tag according to an interactive text of a target client ID of a real-time incoming line, carrying out new risk rating on the target client according to the new risk tag and a risk tag of the client ID which is the same as the target client ID in a risk framework library in risk timeliness, and loading the new risk rating into a rating client list where the target client ID is located.

2. The method for optimizing a risk list according to claim 1, wherein the client list data is stored in a blockchain, the creating a risk tag for incoming line interactive text corresponding to each client ID in the obtained client list data, and obtaining a client list with risk tags includes:

performing data cleaning processing on the client list data to unify the structural form of the related data of the client list data and obtain a structured client list; the structured client list comprises a client ID and an incoming line interaction text corresponding to the client ID;

3. The method for optimizing a risk list according to claim 2, wherein the performing data cleaning processing on the client list data to unify structural forms of relevant data of the client list data, and obtaining a structured client list includes:

according to a preset time limit, eliminating the expiration data in the related data of the client list to obtain the related data meeting the preset time limit;

and carrying out format unification processing on the related data meeting the preset time limit according to a preset format to obtain a structured client list with uniform data format.

4. The method of claim 2, wherein creating risk tags for the incoming line interactive text in the structured client list, and obtaining a client list with risk tags comprises:

extracting a risk tendency speaking operation from the interactive text in the structured client list through a preset natural language understanding model, and taking the risk tendency speaking operation as a risk basis speaking operation; the preset natural language understanding model comprises an input layer for inputting interactive text, an identification layer for identifying risk prone telephone in the interactive text and a telephone output layer for outputting the risk prone telephone;

Creating a risk label corresponding to the risk-based conversation based on a preset risk label list according to the risk-based conversation to obtain a client list with the risk label; the preset risk tag list comprises a risk-based conversation and risk tags corresponding to the risk-based conversation.

5. The method for optimizing a risk list according to claim 1, wherein the risk rating each client ID in the client list with risk tag by a preset rating model to obtain a rated client list, and storing the rated client list in a preset risk framework library comprises:

inputting the client list with the risk labels into a preset grading model, and respectively carrying out risk grade assessment on the preset grading model according to the incoming times of each client in the client list with the risk labels, the incoming times of each time and the risk labels corresponding to each incoming time to respectively obtain incoming times risk grade, incoming time risk grade and label risk grade;

and carrying out comprehensive risk level evaluation on the clients according to the incoming times risk level, the incoming times risk level and the label risk level according to the corresponding specific gravity to obtain rated client orders, and storing the rated client orders into a preset risk framework library.

6. The method for optimizing a risk list according to claim 1, wherein the step of performing aging monitoring on risk labels in a rating client list in the preset risk framework library according to a preset risk aging table, and removing risk labels that fail in the rating client list to obtain a risk framework library in risk aging comprises:

optionally selecting a risk tag from the rating client list in the preset risk framework library as a risk tag for undetermined aging; wherein, each risk tag in the rating client list corresponds to the incoming line time of the client;

taking the incoming line time of a customer corresponding to the risk tag to be subjected to the ageing treatment as the starting time of the risk tag to be subjected to the ageing treatment;

acquiring a risk label which is the same as the risk label of the undetermined aging from the preset risk aging table as a target risk label; the preset risk aging table comprises risk labels and aging corresponding to the risk labels;

and removing the invalid risk label of the pending aging according to the aging corresponding to the target risk label based on the starting time of the risk label of the pending aging to obtain a risk framework library in the risk aging.

7. The risk list optimization method according to claim 1, wherein the establishing a corresponding new risk tag according to the interactive text of the target client ID of the real-time incoming line, and performing new risk rating on the target client ID according to the new risk tag and risk tags of clients identical to the target client ID in a risk framework library in the risk timeframe, and loading the new risk rating into a rating client list where the target client ID is located includes:

the method comprises the steps of acquiring interactive voice of a target client ID in real time, and converting the interactive voice into an interactive text through a voice conversion technology to obtain the interactive text of the target client ID of the real-time incoming line;

acquiring a risk label of a client with the same ID as the target client from a risk framework library in the risk timeframe as a historical risk label of the target client;

8. A risk list optimization apparatus, the apparatus comprising:

the new rating module is used for establishing a corresponding new risk tag according to the interactive text of the target client ID of the real-time incoming line, carrying out new risk rating on the target client according to the new risk tag and the risk tag of the client ID which is the same as the target client ID in a risk framework library in the risk timeliness, and loading the new risk rating into a rating client list where the target client ID is located.

Risk framework library.

9. An electronic device, the electronic device comprising:

at least one processor; the method comprises the steps of,

the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the steps of the risk list optimization method of any one of claims 1 to 7.

10. A computer readable storage medium storing a computer program, wherein the computer program when executed by a processor implements the risk list optimization method according to any one of claims 1 to 7.