Business notification method, system, device and storage medium

Abstract

The application relates to artificial intelligence and provides a business notification method, a system, equipment and a storage medium. The method comprises the following steps: acquiring a business amount, a risk coefficient, a risk probability and historical feedback information; performing product calculation processing on the business amount, the risk probability and the risk coefficient to obtain a business value; calculating a preset expected handling date based on the historical feedback information to determine a handling time period; judging whether the risk probability meets a preset handling mark probability range to determine a handling probability result; obtaining a single decision according to the business value, the handling time period and the handling probability result; and determining a mode of sending a business notification to a user according to the single decision. The business notification method provided by the application can make a single decision according to a business condition, reduce the occurrence of repeated notification, improve business notification efficiency and save operation cost.

Description

Technical Field

[0001] This invention relates to the field of artificial intelligence technology, and in particular to a business notification method, system, device, and storage medium. Background Technology

[0002] Currently, with the development of big data, cloud computing, and AI technologies, traditional human customer service is upgrading to intelligent customer service. Intelligent customer service, i.e., AI robots, can handle simple business notifications, while complex business notifications are handled by human customer service representatives. Currently, there is a large volume of business requiring customer notifications. To avoid omissions due to the massive volume, related technologies use both AI robots and human customer service representatives to simultaneously make phone calls to users. This not only repeatedly disturbs users and generates a large number of invalid notifications, making it difficult to improve the efficiency of business notifications, but also results in a significant investment of human customer service resources in simple, repetitive, and low-value labor, leading to high operating costs. Summary of the Invention

[0003] The following is an overview of the subject matter described in detail herein. This overview is not intended to limit the scope of the claims.

[0004] This invention provides a business notification method, system, device, and storage medium that can make order splitting decisions based on business conditions, reduce duplicate notifications, improve business notification efficiency, and save operating costs.

[0005] In a first aspect, embodiments of the present invention provide a business notification method, the method comprising:

[0006] Obtain business amount, risk coefficient, risk probability, and historical feedback information;

[0007] The business value is obtained by multiplying the business amount, the risk probability, and the risk coefficient.

[0008] Based on the historical feedback information, the preset expected processing date is calculated to determine the processing time period;

[0009] Determine whether the risk probability meets the preset processing mark probability range, and determine the processing probability result;

[0010] Based on the business value, the processing time period, and the processing probability results, an order allocation decision is obtained;

[0011] Based on the order allocation decision, the method for sending business notifications to users is determined.

[0012] The business notification method according to the embodiments of the present application has at least the following beneficial effects: basic data of a business to be notified is acquired, including a business amount, a risk coefficient, a risk probability and historical feedback information. The value lost due to non-notification of the business is calculated according to the business amount, the number of times of handling of the business by a customer and the risk probability, and the business value is obtained. In addition, the handling time period of the business is determined by calculating the historical feedback information and a preset expected handling date, i.e. the data of past notification and handling of the business. The handling probability result of the business is determined by comprehensively comparing the risk probability and the handling mark probability range preset for the business. The importance of the business is further judged according to the level of the business value, whether it exceeds the handling time period and the level of the handling probability result, and the corresponding single distribution decision is made, so that the business is distributed to intelligent customer service notification or manual customer service notification according to the single distribution decision. Therefore, the way of business notification is determined according to the relevant information of the business, which can avoid repeated notification of a user by manual customer service and intelligent customer service, and can distribute the business according to the importance of the business, so that manual resources can be separated from simple businesses, the efficiency of business notification is improved, resource waste caused by repeated notification is reduced, and operating costs are saved.

[0013] According to some embodiments of the present application, in the above business notification method, the product calculation and processing of the business amount, the risk probability and the risk coefficient to obtain the business value comprises:

[0014] acquiring the number of times of handling;

[0015] determining a proportion coefficient matched with the number of times of handling from a preset proportion coefficient set to obtain the risk coefficient.

[0016] The higher the historical number of times of handling of the business is, the higher the possibility of continuous handling of the business is, and the lower the historical number of times of handling is, the higher the possibility of termination of the business is. Therefore, the lower the historical number of times of handling is, the higher the preset proportion coefficient matched from the proportion coefficient set is, i.e. the risk coefficient is larger, and the value lost due to termination of the business is larger.

[0017] According to some embodiments of the present application, in the above business notification method, the historical feedback information comprises a historical handling date and a historical handling period, and the determination of the handling time period according to the historical feedback information comprises:

[0018] calculating the difference between the historical handling date and the historical handling period to obtain a historical lag duration;

[0019] superimposition calculating the expected handling date and the historical lag duration to obtain the handling time period.

[0020] The time length between the historical handling date and the historical handling period, i.e. the historical lag time length, is calculated to determine the historical handling habit of the service, such as early handling or late handling. According to the historical handling habit, the estimated handling time period of the current service is calculated by superimposing the expected handling date preset for the current service on the historical lag time length, and the customer service for service notification is allocated according to the historical handling habit of the customer.

[0021] According to some embodiments of the present application, in the above service notification method, the historical feedback information further includes a promised handling date; and the handling time period is calculated according to the historical lag time length and the expected handling date preset, including:

[0022] The expected handling date preset and the historical lag time length are superimposed to obtain a first handling date;

[0023] The promised handling date and the historical lag time length are superimposed to obtain a second handling date;

[0024] The handling time period is determined according to the first handling date and the second handling date.

[0025] The promised handling date is determined through historical communication with the customer, and the first handling date is calculated according to the promised handling date and the historical lag time length. The promised lag handling date, i.e. the second handling date, is calculated according to the promised handling date and the historical lag time length, so as to determine the time period between the first handling date and the second handling date as the handling time period. Therefore, the handling time period is determined in combination with the feedback information of the customer, the customer is notified in time, and the rationality of service notification is improved while avoiding advance notification to disturb the customer.

[0026] According to some embodiments of the present application, in the above service notification method, the single distribution decision is obtained according to the service value, the handling time period and the handling probability result, including:

[0027] The service values are sorted according to the numerical values of the service values to obtain value sorting information;

[0028] The value gears corresponding to each of the service values are found from a preset gear parameter set;

[0029] The single distribution decision is obtained according to the value gears, the handling time period and the handling probability result.

[0030] The data of the plurality of services are acquired, and the notification mode of the plurality of services is allocated, so that the value ordering information of the plurality of services can be obtained based on the value of the plurality of services. The value ordering information and the preset gear parameter set are used to obtain the value gear corresponding to each service. Therefore, the importance of each service can be determined based on the value gear, the handling time period and the handling probability result, and the value degree, the expiration length and the difficulty degree of each service are determined, and the reasonable single distribution decision is made, and the service notification task is allocated to the intelligent customer service or the manual customer service, so that the rationality of the service distribution is improved, and the efficiency of the service notification is improved.

[0031] According to some embodiments of the present application, in the service notification method, the single distribution decision is obtained according to the service value, the handling time period and the handling probability result, and includes:

[0032] The current date is acquired.

[0033] The difference between the current date and the preset expected handling date is obtained, and the expiration length is obtained.

[0034] The single distribution decision is obtained according to the expiration length, the handling time period, the service value and the handling probability result.

[0035] The time difference between the current date and the expected handling date corresponding to the service is used to calculate the length of time that the service exceeds the expected handling date, that is, the expiration length of the service. When the expiration length of the service is long, it is considered that the importance of the service is higher, and the manual customer service is required to handle, so that the rationality of the service distribution is improved.

[0036] According to some embodiments of the present application, in the service notification method, the risk probability is obtained by the following steps:

[0037] The service information, the to-be-handled customer information and the service staff information are acquired.

[0038] The service information, the to-be-handled customer information and the service staff information are input into a preset probability model, and the risk probability is output.

[0039] The service information, the to-be-handled customer information and the service staff information corresponding to the service are input into a preset probability model, and the to-be-handled customer information, that is, the handling demand of the to-be-handled customer, the amount of money to be paid by the service and the expiration length, and the service ability of the service staff are comprehensively calculated, and the possibility that the service cannot be handled, that is, the risk probability, is estimated.

[0040] In a second aspect, the embodiments of the present application provide a service notification system, which includes:

[0041] The data acquisition module is configured to acquire a business amount, a risk coefficient, a risk probability and historical feedback information.

[0042] The value calculation module is configured to perform a product calculation on the business amount, the risk probability and the risk coefficient to obtain a business value.

[0043] The processing time calculation module is configured to calculate a preset expected processing day based on the historical feedback information to determine a processing time period.

[0044] The processing probability result calculation module is configured to determine a processing probability result by judging whether the risk probability meets a preset processing mark probability range.

[0045] The single distribution module is configured to obtain a single distribution decision according to the business value, the processing time period and the processing probability result.

[0046] The notification module is configured to determine a business notification sending mode to a user according to the single distribution decision.

[0047] In a third aspect, an electronic device is provided, which comprises a memory, a processor and a computer program stored in the memory and capable of running on the processor, and the processor implements the business notification method of the first aspect when executing the computer program.

[0048] In a fourth aspect, a computer readable storage medium is provided, which stores a computer program, and the computer program is executed by a processor to implement the business notification method of the first aspect.

[0049] Other features and advantages of the present application will be set forth in the following description, and in part will become apparent from the description, or can be learned by practice of the present application. The objects and other advantages of the present application will be realized and achieved by means of the structures particularly pointed out in the description and appended claims. BRIEF DESCRIPTION OF DRAWINGS

[0050] The accompanying drawings are included to provide a further understanding of the technical solutions of the present application, and constitute a part of the specification, and are used together with the embodiments of the present application to explain the technical solutions of the present application, and do not constitute a limitation on the technical solutions of the present application.

[0051] Figure 1 is a flowchart of the business notification method provided by the embodiments of the present application;

[0052] Figure 2 is a specific flowchart of the risk coefficient generation in the embodiments of the present application;

[0053] Figure 3 is Figure 1 is a specific implementation process diagram of step S300 in

[0054] Figure 4 is Figure 3 A specific implementation process diagram of step S320 in the method 1000 is shown in FIG. 11.

[0055] Figure 5 is Figure 1 A specific implementation process diagram of step S500 in the method 1000 is shown in FIG. 12.

[0056] Figure 6 is Figure 1 A specific implementation process diagram of step S500 in the method 1000 is shown in FIG. 12.

[0057] Figure 7 is a specific flowchart of the risk probability generation in the embodiment of the present application.

[0058] Figure 8 is a structural diagram of the service notification system provided by the embodiment of the present application.

[0059] Figure 9 is a structural diagram of the service notification system provided by another embodiment of the present application.

[0060] Figure 10 is a structural diagram of the electronic device provided by the embodiment of the present application. DETAILED DESCRIPTION

[0061] In order to make the objectives, technical solutions and advantages of the present application clearer, the present application will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and should not be used to limit the present application.

[0062] It should be noted that although the functional modules are divided in the module diagram, and the logical order is shown in the flowchart, in some cases, the steps shown or described in the modules or the order of execution in the flowchart can be different, and it is not necessary to describe a specific order or sequence.

[0063] At present, with the development of big data, cloud computing and AI technology, intelligent customer service, i.e. AI robot, is used to notify the business, for example, the AI robot notifies the business of premium collection by calling the customer. However, due to the low flexibility of AI robot business processing, it is difficult to handle complex business. Therefore, complex business needs artificial customer service for communication and notification.

[0064] At present, a large number of businesses need to be notified to customers, and it is difficult to distinguish the importance of the businesses and allocate the businesses, so as to avoid missing the notification due to a large amount of business, AI robots and artificial customer service are used to dial the phone to the user at the same time to make notification, that is, the AI robot dials the phone of most or all customers to make notification, and at the same time, the artificial customer service also makes phone notification again, which not only repeatedly disturbs the user and generates a large amount of invalid notification, but also the artificial customer service resource is greatly invested in simple, repetitive and low-value labor, and the operating cost is high.

[0065] The embodiment of the present application can acquire and process related data based on artificial intelligence technology. Artificial intelligence (AI) is the use of digital computers or digital computer-controlled machines to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results.

[0066] The present application relates to artificial intelligence, and provides a business notification method, which acquires a business amount, a risk coefficient, a risk probability and historical feedback information; performs product calculation processing on the business amount, the risk probability and the risk coefficient to obtain a business value; calculates a preset expected handling date based on the historical feedback information to determine a handling time period; determines a handling probability result by judging whether the risk probability meets a preset handling mark probability range; obtains a single allocation decision according to the business value, the handling time period and the handling probability result; and determines a business notification sending mode to a user according to the single allocation decision. According to the embodiment provided by the present application, after the basic data of the business, i.e., the business amount, the risk probability and the risk coefficient, are acquired, the business value of the business is obtained by performing product calculation on the business amount, the risk probability and the risk coefficient. The handling time period of the business is determined by calculating the preset expected handling date and the historical feedback information of the business. The handling probability result is obtained by judging whether the risk probability meets the handling mark probability range. Thus, the importance of the business is determined by the business value, the handling probability result and the length of time exceeding the handling time period, and the corresponding single allocation decision is made to allocate the business to intelligent customer service or artificial customer service for notification. The simple and low-importance business is allocated to intelligent customer service for notification, and the complex and high-importance business is allocated to artificial customer service for notification, which can avoid repeatedly notifying the customer, improve the business notification efficiency, and help the artificial customer service resource to be separated from simple business, improve the resource utilization rate and save the operating cost.

[0067] The embodiments of the present application will be described below with reference to the accompanying drawings.

[0068] In a first aspect, with reference to Figure 1 , Figure 1A flow chart of the service notification method provided by the embodiment of the present application is shown, which comprises but is not limited to the following steps:

[0069] In step S100, the service amount, risk coefficient, risk probability and historical feedback information are acquired.

[0070] In step S200, the service amount, risk probability and risk coefficient are multiplied to obtain the service value.

[0071] In step S300, the preset expected handling date is calculated based on the historical feedback information to determine the handling time period.

[0072] In step S400, it is determined whether the risk probability meets the preset handling mark probability range to determine the handling probability result.

[0073] In step S500, the single decision is obtained according to the service value, handling time period and handling probability result.

[0074] In step S600, the mode of sending the service notification to the user is determined according to the single decision.

[0075] It can be understood that, in order to improve the efficiency of the service notification and reasonably allocate the mode of the service notification, it is necessary to confirm the importance of the service to be notified, allocate the service notification task to the intelligent customer service or manual customer service according to the importance of the service, and avoid the intelligent customer service and manual customer service from synchronously notifying and repeatedly disturbing the customer.

[0076] The basic data of the to-be-notified business, i.e., the business amount, the risk coefficient, the risk probability and the historical feedback information, are acquired. For example, when the business notification is a notification of policy payment, the business amount can be the amount of the policy to be handled, and the business notification can also be a notification of loan payment, and the business amount can also be the remaining amount of the loan to be handled. The higher the business amount, the greater the loss caused by the client's failure to handle on time, and thus the higher the importance. The lower the business amount, the smaller the loss caused by the client's failure to handle on time, and thus the lower the importance of the business. The risk coefficient represents the weight value of the client's stop handling the business. According to the length of time of the client's handling of the business, i.e., the client's familiarity with the business, when the length of time of the client's handling of the business is short, it can be considered that the client's familiarity with the business and dependence are low, and the possibility of stopping handling is high, and thus the risk coefficient is higher. The risk probability can represent the probability of termination of the current business, or the proportion of the historical termination of the business of the corresponding client among all businesses, or it can be pre-set or calculated by a probability model based on the business information, the to-be-handled client information and the business agent information. The higher the risk probability, the higher the possibility of termination of the business, and the higher the importance. The lower the risk probability, the lower the possibility of termination of the business, and the lower the importance. Therefore, the business value that the business can possibly lose is calculated based on the business amount, the risk coefficient and the risk probability. When the business value is higher, it indicates that the importance of the business is higher, and thus the artificial customer service can be arranged to make the notification; when the business value is lower, it indicates that the importance of the business is lower, and thus the intelligent customer service can be arranged to make the notification.

[0077] The historical feedback information can be the information fed back by the client in the historical communication with the client, such as the promised handling date of the client. The historical feedback information can also include the handling date of the historical business of the client and the termination handling period of the historical business. The handling habit of the client and the expected handling time of the client can be determined based on the historical feedback information, and thus the handling time period, i.e., the handling time period, of the business can be determined based on the historical feedback information. When the business is not handled beyond the handling time period, and the longer the overdue time, the greater the loss and the higher the importance.

[0078] The handling probability result is the possibility of the continuation of the business, and the handling probability result is obtained by comprehensively processing the risk probability and the preset handling mark probability range, that is, judging whether the risk probability is within the handling mark probability range. The preset handling mark probability range can be obtained according to the actual situation of the customer, for example, can be obtained according to the number of businesses handled by the customer, or can be obtained according to the information fed back by the customer in the historical communication with the customer. The handling mark probability range can be set by manual, and the risk probability can be comprehensively processed by combining manual marking, so that a handling probability result with higher accuracy can be obtained. Among them, the business value can also be comprehensively calculated according to the handling probability result, the business amount and the handling times. For example, when the preset mark is the handling mark probability range of 0% to 30% and the risk probability is 25%, that is, the risk probability meets the handling mark probability range of the mark, the handling probability result can be determined as high. When the preset mark is the handling mark probability range of 70% to 100% of the termination handling, and the risk probability is 90%, that is, the risk probability meets the handling mark probability range of the mark of termination handling, the handling probability result can be determined as low. When the mark is the handling mark probability range of 30% to 70% of the uncertain mark, and the risk probability is 50%, that is, the risk probability meets the handling mark probability range of the mark of uncertain mark, the handling probability result can be determined as uncertain. Each business only has a corresponding handling mark probability range, when the risk probability of the business does not meet the preset handling mark probability range corresponding to the business, the handling probability result of the business is determined as uncertain.

[0079] Therefore, by obtaining the basic data of the to-be-notified business, the importance of the business is determined, that is, by obtaining the business amount, risk probability, risk coefficient and historical feedback information of the to-be-notified business, the importance and difficulty of the business are determined by determining the business value, whether it exceeds the handling period and the level of the handling probability result, so as to make corresponding single decision, for example, when the importance of the business is high, the difficulty is high, and the complexity is high, the business can be allocated to the artificial customer service for notification processing; when the importance of the business is low, the difficulty is low, the business can be allocated to the intelligent customer service for notification processing, so as to separate the artificial customer service resources from simple and repetitive low-value tasks, and put them into complex and high-difficulty tasks, improve the utilization rate of resources, and improve the efficiency of business notification.

[0080] It is worth noting that the single decision can be obtained according to the business value of the business, whether the handling period is reached, the probability result of the handling, and the pre-set decision rule, for example: all businesses that do not reach the handling time period and have high handling probability results can be considered to be low in importance and relatively simple, and the corresponding single decision can be intelligent customer service notification. All businesses that do not reach the handling time period, have low handling probability results, and have low business value can be considered to be low in importance, and the corresponding single decision can also be intelligent customer service notification. All businesses that reach or exceed the handling time period, have high handling probability results, and have low business value can be considered to be low in importance and relatively simple, and the corresponding single decision can also be intelligent customer service notification. When the handling time period is not reached, the business with high business value and low handling probability result can be considered to be high in importance, and the corresponding single decision is artificial customer service notification. After the handling time period is reached or exceeded, all businesses with low handling probability results can be considered to be high in importance, and the corresponding single decision is artificial customer service notification. In addition, after the handling time period is reached or exceeded, the business with high business value is high in importance, and the corresponding single decision is artificial customer service notification. In the case of uncertain handling probability result, the single decision corresponding to the business that does not reach the handling time period is intelligent customer service notification, and the single decision corresponding to the business that reaches or exceeds the handling time period is artificial customer service notification.

[0081] Referring to Figure 2 , Figure 2 The generation process of the risk coefficient is shown, and the risk coefficient generation process includes but is not limited to the following steps:

[0082] Step S210, obtaining the handling times;

[0083] Step S220, determining the proportion coefficient matched with the handling times from the pre-set proportion coefficient set to obtain the risk coefficient.

[0084] It can be understood that the handling times can be the historical times handled by the customer from the creation time of the current business to the current time, or the times handled by the customer for all businesses. The higher the handling times, the higher the credit of the customer, the higher the possibility of timely handling, and the lower the importance. The lower the handling times, the lower the credit of the customer, the lower the possibility of timely handling, and the higher the importance, which can cause loss. Therefore, the importance of the business can be judged by the handling times.

[0085] The risk coefficient is obtained by searching the preset set of proportional coefficients to match the handling times. The risk coefficient is calculated by the handling times and the determined proportional coefficient. For example, the matched proportional coefficient is 0.01, the handling times is multiplied by the proportional coefficient to obtain an intermediate coefficient, and the risk coefficient is obtained by subtracting the intermediate coefficient from 1.

[0086] When the handling times is higher, the customer's performance is good, the customer's credit is higher, and the possibility of timely handling is higher. Therefore, the higher the handling times, the smaller the risk coefficient obtained by the handling times and the corresponding proportional coefficient, and the lower the possible loss of the customer's business value. For example, when the handling times is 2, the intermediate coefficient is 0.02, and the risk coefficient is 0.98; when the handling times is 10, the intermediate coefficient is 0.1, and the risk coefficient is 0.9. Therefore, the lower the business value calculated by the risk coefficient, the business amount, and the risk probability, the lower the possible loss of the customer's business value, and the lower the importance.

[0087] In addition, the set of preset proportional coefficients is provided with a plurality of proportional coefficients, each handling times corresponds to each proportional coefficient, thereby obtaining the risk coefficient. For example, the handling times is 2, the corresponding proportional coefficient is -0.1, the intermediate coefficient is -0.2, and the obtained risk coefficient is 1.2; the handling times is 3, the corresponding proportional coefficient is -0.03, the intermediate coefficient is -0.1, and the obtained risk coefficient is 1.1. In addition, since the handling times of the new customer is less, it is easy to forget to handle, and the probability of not handling the business when it expires is higher. Therefore, the business with handling times of 0 to 3 times can match the corresponding proportional coefficient respectively, so that the risk coefficient is greater than 1, and the less the handling times, the higher the corresponding risk coefficient. When the handling times is 4 or more, a unified proportional coefficient, such as 0.01, can be obtained, so that the risk coefficient is less than 1. Thus, the difference between the risk coefficient of the business with handling times of 3 times or less and the risk coefficient of the business with handling times of 4 times or more is increased. Therefore, the business value calculated by the risk coefficient, the risk probability, and the business amount is greatly different between the business with handling times of 3 times or less and the business with handling times of 4 times or more, the business value and the importance of the business with handling times of 3 times or less are improved, and the artificial customer service is arranged in time to notify the new customer of the business, thereby improving the customer experience.

[0088] Therefore, the product of the business amount, the risk probability and the risk coefficient is the business value, and the higher the risk coefficient, the higher the business value, the higher the value that the business can lose, and the more important the business is, so that the artificial customer service can be arranged in time to notify the business; and the lower the risk coefficient, the lower the business value, the higher the value that the business can lose, and the less important the business is, so that the intelligent customer service can be arranged to notify the business. Therefore, the business value is adjusted by using the number of handling times, the business values of various businesses are distinguished, the notification mode of each business is allocated and adjusted, the artificial customer service resources are separated from simple and repetitive low-value businesses, and the utilization rate of artificial customer service resources is improved.

[0089] Referring to Figure 3 , Figure 1 The step S300 in the embodiment shown includes but is not limited to the following steps:

[0090] Step S310, calculating the difference between the historical handling date and the historical handling period to obtain the historical lag time length;

[0091] Step S320, superimposing the preset expected handling date and the historical lag time length to obtain the handling time period.

[0092] It can be understood that the historical feedback information can include the historical handling date and the historical handling period. The historical handling date can be the historical date of the current business handled by the customer, or the historical date of all businesses handled by the customer. The historical handling period corresponds to the historical handling date. If the historical handling date is the historical date for the current business, the historical handling period is the historical date of the current business that is not renewed and terminated. If the historical handling date is the historical date for all businesses, the historical handling period is the historical date of all businesses that are not renewed and terminated. By calculating the time length between the historical handling date and the corresponding historical handling period, the historical lag time length is obtained, that is, the historical overage handling time length of the customer. The historical lag time length can determine the handling habit of the customer. The difference between the historical handling date and the historical handling period is obtained. If the historical lag time length is negative, it can be considered that the customer usually handles in advance. If the historical lag time length is positive, it can be considered that the customer usually handles late.

[0093] Therefore, the handling time period is obtained by superimposing the historical lag time and the preset expected handling day. The preset handling time period is the date required for the current business to be handled, and the handling time period that the customer will handle the current business is calculated by predicting the customer's historical handling habits. The expected handling date is obtained by adding the historical lag time and the preset expected handling day. If the expected handling date is earlier than the expected handling day, the time between the expected handling date and the expected handling day is the handling time period. If the expected handling date is later than the expected handling day, the time between the expected handling day and the expected handling date is the handling time period. When the customer still does not handle the business after exceeding the handling time period, it is considered that the business has a high possibility of loss, therefore, the business that has not received the fee after exceeding the handling time period is important, and the artificial customer service needs to be arranged in time to notify the business, and the customer needs to be communicated and reminded in time. When the handling time period is not exceeded, it is considered that the business will be handled within the handling time period, and the intelligent customer service needs to be arranged in time to notify the business and remind the customer.

[0094] Referring to Figure 4 , Figure 3 The step S320 in the embodiment shown includes but is not limited to the following steps:

[0095] Step S321, superimposing the preset expected handling day and the historical lag time to obtain a first handling day;

[0096] Step S322, superimposing the promised handling date and the historical lag time to obtain a second handling day;

[0097] Step S323, determining the handling time period according to the first handling day and the second handling day.

[0098] It can be understood that the historical feedback information can also include the promised handling date fed back by the customer in the historical communication, that is, the customer promises to handle on the corresponding date. The promised handling date and the preset expected handling day are respectively predicted and calculated based on the historical handling habits of the customer, the historical lag time is added to the expected handling day to obtain the expected handling date, and the first handling day is obtained. The historical lag time is added to the promised handling date to obtain the expected promised date, that is, the second handling day; thus, the time between the first handling day and the second handling day is marked as the handling time period, that is, the customer has a greater possibility to handle the business within the handling time period. Therefore, the handling time period is adjusted by the promised handling date to avoid that the handling time period is earlier than the customer's expected handling time, and when the handling time period is exceeded and the customer's expected handling time is not reached, the artificial customer service and the intelligent customer service disturb the customer multiple times, which not only affects the customer but also wastes resources.

[0099] Referring to Figure 5 , Figure 4The step S500 in the illustrated embodiment includes but is not limited to the following steps:

[0100] The step S510 sorts the service values according to the numerical values of the service values to obtain value sorting information.

[0101] The step S520 finds the value range corresponding to each service value from the preset range parameter set.

[0102] The step S530 obtains a single distribution decision according to the value range, the service time period and the service probability result.

[0103] It can be understood that in the process of distributing the service notification, multiple services can be distributed at the same time, that is, the basic data of multiple services is obtained at the same time to obtain one-to-one service values. The higher the service value is, the more important the service is. The multiple service values are sorted according to the numerical values of the service values, that is, the multiple services are sorted according to the numerical values of the service values of the multiple services to obtain value sorting information. The value range of the preset range parameter set is used to sort the service values in the value sorting information, for example, the first 30% of the service values in the value sorting information are classified as high value range, the last 30% of the service values in the value sorting information are classified as low value range, and the service values of the remaining value sorting information are classified as medium value range. The higher the value range is, the more important the service is, and the obtained single distribution decision is more likely to be artificial customer service notification. The services in the high value range are important, and when the service time period is reached or exceeded, the single distribution decision of the service in the high value range is artificial customer service notification, while the single distribution decision of the service in the medium and low value range with high service probability result is intelligent customer service notification. The single distribution decision of the service in the medium and low value range is intelligent customer service notification when the service time period is not reached. Thus, a large amount of artificial customer service resources is separated from low value services and invested in high value services, improving the utilization rate of artificial customer service resources and saving operation costs.

[0104] The service values sorted can be the service values corresponding to the current service to be notified, or the service values of all services including the service to be notified and the service notified.

[0105] Referring to Figure 6 , Figure 1 The step S500 in the illustrated embodiment also includes but is not limited to the following steps:

[0106] The step S540 obtains the current date.

[0107] The step S550 obtains the overdue time length by subtracting the difference between the current date and the preset expected service date.

[0108] Step S560, according to the overdue time length, the handling time period, the business value and the handling probability result, obtaining the single distribution decision.

[0109] It can be understood that in the case that the handling time period is far away from the preset expected handling day, for example, the handling time period is 30 days or more away from the preset expected handling day, in order to avoid missing the notification due to calculation error, the overdue time length is obtained by calculating the time difference between the current date and the preset expected handling day corresponding to the current business. When the overdue time length exceeds the preset overdue time, for example, the overdue time length has accumulated to 11 days, exceeding the preset overdue time of 10 days, it can be considered that the importance of the business is high, and the obtained single distribution decision can be artificial customer service notification. When the overdue time length does not exceed the preset overdue time, and the single distribution decision obtained by the business with low business value or high handling probability result can be intelligent customer service notification. Therefore, in the case that the handling time period is far away from the preset expected handling day, the single distribution decision is adjusted by the overdue time length, the business notification task is intelligently distributed, and the business notification efficiency is improved.

[0110] Referring to Figure 7 , Figure 7 The risk probability generation specific flowchart in the embodiment of the application is shown, and the specific process of risk probability generation includes but is not limited to the following steps:

[0111] Step S700, obtaining business information, to-be-handled customer information and salesperson information;

[0112] Step S800, importing the business information, to-be-handled customer information and salesperson information into a preset probability model, and outputting to obtain the risk probability.

[0113] It can be understood that the pre-registered to-be-handled customer information, business information and salesperson information are obtained through the business system, for example, the to-be-handled customer information includes the handling history information of the to-be-handled customer for the current business or historical business, the overdue period, and the credit rating of the to-be-handled customer by the business system, which is used to evaluate the handling demand of the to-be-handled customer. The salesperson information includes the length of service of the employee, the business processing experience and the recent business success rate, which is used to evaluate the business ability of the salesperson. The business information includes the business type, the business quantity, the business duration, which is used to evaluate the importance of the business.

[0114] The to-be-handled customer, i.e. the customer's handling demand, the business amount and the due date, and the business ability of the salesperson are comprehensively calculated by importing the business corresponding business information, to-be-handled customer information and salesperson information into the preset probability model, and the risk probability that the business cannot be handled is estimated. Therefore, according to the risk probability, the corresponding single distribution decision can be made, so that the artificial customer service resources are concentrated in the difficult collection business, and the utilization rate of artificial customer service resources is improved.

[0115] It should be noted that in various specific embodiments of the present application, when relevant processing needs to be performed on data related to the identity or characteristics of the user, such as user information, user behavior data, user history data, and user location information, the user's permission or consent will be obtained first, and the collection, use, and processing of such data will comply with relevant laws, regulations, and standards of the country or region. In addition, when the embodiments of the present application need to obtain sensitive personal information of the user, the user's separate permission or separate consent will be obtained through a pop-up window or by jumping to a confirmation page, and after obtaining the user's separate permission or separate consent, the necessary user-related data for enabling the embodiments of the present application to function normally will be obtained.

[0116] In a second aspect, referring to Figure 8 , Figure 8 A structural schematic diagram of a service notification system 800 provided by an embodiment of the present application is shown.

[0117] The data acquisition module 810 is configured to acquire a service amount, a risk coefficient, a risk probability, and historical feedback information.

[0118] The value calculation module 820 is configured to perform product calculation processing on the service amount, the risk probability, and the risk coefficient to obtain a service value.

[0119] The handling time calculation module 830 is configured to calculate a preset expected handling day based on the historical feedback information to determine a handling time period.

[0120] The handling probability result calculation module 840 is configured to determine a handling probability result by judging whether the risk probability meets a preset handling mark probability range.

[0121] The single distribution module 850 is configured to obtain a single distribution decision according to the service value, the handling time period, and the handling probability result.

[0122] The notification module 860 is configured to determine a manner of sending a service notification to a user according to the single distribution decision.

[0123] The business notification system 800 obtains the basic data of the business to be notified through the data acquisition module 810, including the business amount, the number of handling times, the risk probability and the historical feedback information. The value calculation module 820 calculates the value of the business not notified by the business amount, the number of handling times of the customer to the business and the risk probability, and obtains the business value. In addition, the handling time calculation module 830 determines the estimated handling time period of the business according to the historical feedback information, i.e. the past notification and handling data of the business. The handling probability result calculation module 840 compares the risk probability and the handling mark probability range set in advance for the business, and determines the handling probability result of the business. Then, the single distribution module 850 judges the importance of the business according to the business value, whether it exceeds the handling time period and the handling probability result, and makes the corresponding single distribution decision, so that the notification module 860 distributes the business to the intelligent customer service notification or the manual customer service notification according to the single distribution decision. Therefore, the business notification system 800 determines the business notification mode according to the related information of the business, which can avoid the repeated notification of the user by the manual customer service and the intelligent customer service, and can distribute the business according to the importance of the business, so that the manual resources can be separated from the simple business, the efficiency of the business notification is improved, the resource waste of repeated notification is reduced, and the operation cost is saved.

[0124] Referring to Figure 9 , Figure 9 The structure schematic diagram of the business notification system 800 provided by another embodiment of the application is shown.

[0125] The value calculation module 820 further includes a coefficient calculation module 821.

[0126] The coefficient calculation module 821 is configured to determine the proportion coefficient matched with the number of handling times from the preset proportion coefficient set, and obtain the risk coefficient.

[0127] In addition, the handling time calculation module 830 includes a lag duration calculation module 831.

[0128] The lag duration calculation module 831 is configured to calculate the difference between the historical handling date and the historical handling period, and obtain the historical lag duration.

[0129] In addition, the handling time calculation module 830 is further configured to superimpose the preset expected handling date and the historical lag duration, and obtain the handling time period; superimpose the preset expected handling date and the historical lag duration, and obtain the first handling date; superimpose the promised handling date and the historical lag duration, and obtain the second handling date; and determine the handling time period according to the first handling date and the second handling date.

[0130] In addition, the order sorting module 850 includes a value sorting module 851 and a tier allocation module 852.

[0131] The value ranking module 851 sorts business values ​​according to their numerical values ​​to obtain value ranking information. The tier allocation module 852 searches for and matches the value tiers corresponding to each business value from a preset tier parameter set. The order allocation module 850 also makes order allocation decisions based on the value tier, processing time period, and processing probability results.

[0132] In addition, the order module 850 also includes an overdue period calculation module 853.

[0133] The overdue duration calculation module 853 is used to calculate the overdue duration by subtracting the preset expected processing date from the current date.

[0134] The order allocation module 850 is also used to make order allocation decisions based on overdue duration, processing time period, business value, and processing probability results.

[0135] In addition, the business notification system 800 also includes a probability model module 870.

[0136] The probability model module 870 is used to acquire business information, pending customer information, and salesperson information, and import the business information, pending customer information, and salesperson information into a preset probability model to output the risk probability.

[0137] Thirdly, referring to Figure 10 , Figure 10 An electronic device 1000 provided in an embodiment of the present invention is shown. The electronic device 1000 includes a memory 1010, a processor 1020, and a computer program stored in the memory 1010 and executable on the processor 1020. When the processor 1020 executes the computer program, it implements the service notification method as described in the above embodiment.

[0138] The memory 1010, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs, such as the service notification method in the above embodiments of the present invention. The processor 1020 implements the service notification method in the above embodiments of the present invention by running the non-transitory software program and instructions stored in the memory 1010.

[0139] The memory 1010 can include a program storage area and a data storage area, where the program storage area can store an operating system, at least one application required by a function, and the data storage area can store data required for executing the service notification method in the above embodiments and the like. In addition, the memory 1010 can include a high-speed random access memory, and can also include a non-transitory memory such as at least one magnetic disk storage device, a flash memory device, or other non-transitory solid-state memory device. It should be noted that the memory 1010 can optionally include a memory disposed remotely with respect to the processor 1020, and these remote memories can be connected to the terminal through a network. Examples of the above network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof.

[0140] The non-transitory software programs and instructions required to implement the service notification method in the above embodiments are stored in the memory, and when executed by one or more processors, the service notification method in the above embodiments is executed, for example, the method steps S100 to S600 in the above described Figure 1 , the method steps S2100 to S220 in the above described Figure 2 , the method steps S310 to S320 in the above described Figure 3 , the method steps S321 to S323 in the above described Figure 4 , the method steps S510 to S530 in the above described Figure 5 , the method steps S540 to S560 in the above described Figure 6 , and the method steps S700 to S800 in the above described Figure 7 .

[0141] In a fourth aspect, the present application further provides a computer readable storage medium, which stores computer executable instructions for causing a computer to execute the service notification method in the above embodiments, for example, to execute the method steps S100 to S600 in the above described Figure 1 , the method steps S2100 to S220 in the above described Figure 2 , the method steps S310 to S320 in the above described Figure 3 , the method steps S321 to S323 in the above described Figure 4 , the method steps S510 to S530 in the above described Figure 5 , the method steps S540 to S560 in the above described Figure 6 , and the method steps S700 to S800 in the above described Figure 7 .

[0142] The apparatus embodiments described above are merely illustrative, wherein the units described as separate components can or can not be physically separate, i.e., can be located in one place, or can also be distributed to multiple network units. Part or all of the modules can be selected according to actual needs to achieve the purpose of the present embodiment scheme.

[0143] Those of ordinary skill in the art understand that all or some steps in the method disclosed above can be implemented as software, firmware, hardware and appropriate combinations thereof. Some or all physical components can be implemented as software executed by a processor, such as a central processing unit, a digital signal processor or a microprocessor, or as hardware, or as an integrated circuit, such as an application specific integrated circuit. Such software can be distributed on a computer readable medium, which can include computer storage media (or non-transitory media) and communication media (or transitory media). As known to those of ordinary skill in the art, the term computer storage media includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information such as computer readable instructions, data structures, program modules or other data. Computer storage media includes, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital versatile disk (DVD) or other optical disk storage, magnetic cassettes, magnetic tapes, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to store desired information and can be accessed by a computer. In addition, as known to those of ordinary skill in the art, communication media generally includes computer readable instructions, data structures, program modules or other data in modulated data signals such as carrier waves or other transmission mechanisms, and can include any information delivery medium.

[0144] It should be noted that the server can be a standalone server, or a cloud server providing cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDN), and basic cloud computing services such as big data and artificial intelligence platforms.

[0145] It is noted that all or some of the steps of the methods disclosed above can be utilized in a number of computer system environments or configurations, such as, for example, a personal computer, a server computer, a hand-held device or portable device, a tablet device, a multiprocessor system, a microprocessor-based system, a set top box, programmable consumer electronics, network PC, minicomputer, mainframe computer, distributed computing environments that include any of the above systems or devices, and the like. The present application can be described in the general context of computer-executable instructions, such as program modules, being executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, and the like, that perform particular tasks or implement particular abstract data types. The present application can also be practiced in distributed computing environments where tasks are performed by remote processing devices that are linked through a communications network. In a distributed computing environment, program modules can be located in local and remote computer storage media including memory storage devices.

[0146] The above detailed description of the embodiments of the present application is made with reference to the accompanying drawings, but the present application is not limited to the above embodiments, and various changes can be made within the knowledge of those skilled in the art without departing from the spirit of the present application.

Claims

1. A service notification method characterized by, The method comprises: obtaining a business amount, a risk coefficient, a risk probability and historical feedback information; productively calculating the business amount, the risk probability and the risk coefficient to obtain a business value; based on the historical feedback information, calculating a preset expected handling date to determine a handling time period; judging whether the risk probability meets a preset handling mark probability range to determine a handling probability result; obtaining a single decision according to the business value, the handling time period and the handling probability result; determining a way of sending a business notice to a user according to the single decision; wherein the risk coefficient is obtained through the following steps, comprising: obtaining a handling frequency; determining a proportional coefficient matching the handling frequency from a preset proportional coefficient set to obtain a risk coefficient, which is obtained by multiplying the handling frequency and the proportional coefficient to obtain an intermediate coefficient, and then subtracting 1 from the intermediate coefficient; wherein the risk probability is obtained through the following steps: obtaining business information, to-be-handled customer information and business staff information, the to-be-handled customer information including to-be-handled customer handling history information for current business or historical business, overdue period number, and credit rating of the to-be-handled customer by a business system, for evaluating handling demand of the to-be-handled customer, and the business staff information including employee service length, business processing experience and recent business processing success rate, for evaluating business ability of the business staff; inputting the business information, the to-be-handled customer information and the business staff information into a preset probability model to output a risk probability, wherein the risk probability is obtained by comprehensively calculating the handling demand of the to-be-handled customer, business amount and due date of the business, and the business ability of the business staff.

2. The service notification method of claim 1, wherein, The historical feedback information includes historical handling date and historical handling period; based on the historical feedback information, the preset expected handling date is calculated to determine the handling time period, comprising: calculating a difference between the historical handling date and the historical handling period to obtain a historical lag time; superimposition calculating the preset expected handling date and the historical lag time to obtain the handling time period.

3. The service notification method according to claim 2, characterized by, The historical feedback information further includes a promised handling date; the preset expected handling date and the historical lag time are superimposition calculated to obtain the handling time period, comprising: superimposition calculating the preset expected handling date and the historical lag time to obtain a first handling date; superimposition calculating the promised handling date and the historical lag time to obtain a second handling date; determining the handling time period according to the first handling date and the second handling date.

4. The service notification method of claim 1, wherein, The single decision is obtained according to the business value, the handling time period and the handling probability result, comprising: sorting the business values according to their numerical values to obtain value sorting information; finding a value gear corresponding to each business value from a preset gear parameter set; obtaining the single decision according to the value gear, the handling time period and the handling probability result.

5. The service notification method of claim 1, wherein, The single processing decision is obtained according to the service value, the processing time period and the processing probability result. The current date is obtained; The difference between the current date and the preset expected processing date is obtained to obtain the overdue time length; The single processing decision is obtained according to the overdue time length, the processing time period, the service value and the processing probability result.

6. A service notification system, characterized by The service notification system is used to implement the service notification method in any one of claims 1 to 5, and the system comprises: A data acquisition module is configured to acquire a service amount, a risk coefficient, a risk probability and historical feedback information; A value calculation module is configured to multiply the service amount, the risk probability and the risk coefficient to obtain a service value; A processing time calculation module is configured to calculate a preset expected processing date based on the historical feedback information to determine a processing time period; A processing probability result calculation module is configured to determine a processing probability result by judging whether the risk probability meets a preset processing mark probability range; A single processing module is configured to obtain a single processing decision according to the service value, the processing time period and the processing probability result; A notification module is configured to determine a service notification sending mode to a user according to the single processing decision; The risk coefficient is obtained by the following steps, comprising: The number of processing times is obtained; A proportion coefficient matched with the number of processing times is determined from a preset proportion coefficient set to obtain a risk coefficient, the risk coefficient is obtained by multiplying the number of processing times and the proportion coefficient to obtain an intermediate coefficient, and the risk coefficient is obtained by subtracting the intermediate coefficient from 1; The risk probability is obtained by the following steps: Business information, to-be-processed client information and business staff information are obtained, the to-be-processed client information comprises to-be-processed client processing history information for a current service or a historical service, overdue period, and a credit rating of the to-be-processed client by a business system, and is used to evaluate to-be-processed client processing demand, and the business staff information comprises employee service length, business processing experience and recent business processing success rate, and is used to evaluate business staff business ability; The business information, the to-be-processed client information and the business staff information are input into a preset probability model to output a risk probability, wherein the risk probability is obtained by comprehensively calculating the to-be-processed client processing demand, service amount and due date, and the business staff business ability.

7. An electronic device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, The processor executes the computer program to implement the service notification method in any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that, The computer program is stored in the memory and is executed by the processor to implement the service notification method in any one of claims 1 to 5.

Images