Implementation method and system of sales task reminding and early warning mechanism

Abstract

The invention relates to an implementation method and system of a sales task reminding and early warning mechanism, and belongs to the technical field of task reminding and early warning. The method comprises the steps of obtaining sales opportunity data, and obtaining sales opportunity preprocessing data through a data preprocessing model according to the sales opportunity data; obtaining sales opportunity optimal data through a data mining model according to the sales opportunity preprocessing data; acquiring sales task data, and performing sales task reminding through a sales reminding model according to the sales task data and the sales opportunity optimal data to obtain a sales task priority; and performing sales task early warning through a sales early warning model according to the sales task data and the sales task priority. And the reminding and early warning of the sales task are realized.

Description

Technical Field

[0001] The present invention belongs to the technical field of task reminder and early warning, and specifically relates to a method and system for realizing a sales task reminder and early warning mechanism. Background Art

[0002] In modern enterprises, the timely completion and accurate execution of sales tasks are crucial to the operational efficiency and market competitiveness of enterprises. However, due to the complexity and large number of sales tasks, the status of sales tasks is not updated in a timely manner, making it difficult to grasp the progress of tasks in a timely manner; in addition, the tracking and risk assessment of sales tasks still rely heavily on manual experience, which makes it easy to ignore potential risk points and difficult to detect sales task delays or abnormal situations in a timely manner, resulting in the failure to complete sales targets on time.

[0003] Traditional sales management tools, including CRM systems and progress tracking software, provide task recording and status monitoring functions, but their task reminder and early warning capabilities are still significantly limited. These tools usually record the status of sales tasks in a passive manner, lack the ability to deeply analyze historical data, and are difficult to achieve data-driven risk prediction and early warning. As companies increase their demand for digital and intelligent sales management, how to establish an efficient and accurate sales task reminder and early warning method has become an urgent problem to be solved. Summary of the invention

[0004] In order to solve the above problems existing in the prior art, the present invention provides a method and system for realizing a sales task reminder and early warning mechanism. The purpose of the present invention can be achieved through the following technical solutions: A method for implementing a sales task reminder and early warning mechanism, comprising: Acquire sales opportunity data, and obtain sales opportunity preprocessing data through a data preprocessing model according to the sales opportunity data; Obtaining optimal sales opportunity data through a data mining model based on the sales opportunity preprocessing data; Acquire sales task data, and perform sales task reminders through a sales reminder model according to the sales task data and the sales opportunity optimal data to obtain sales task priorities; A sales task warning is performed according to the sales task data and the sales task priority through a sales warning model.

[0005] Preferably, obtaining sales opportunity preprocessing data through a data preprocessing model according to the sales opportunity data includes: Obtaining sales opportunity specification data through data specification processing according to the sales opportunity data; The data reduction processing includes standardization processing, deduplication processing, and normalization processing; The sales opportunity pre-processing data is obtained by filling missing values ​​with central values ​​according to the sales opportunity specification data.

[0006] Preferably, obtaining the optimal sales opportunity data through a data mining model based on the sales opportunity preprocessing data includes: The pre-processed data features of sales opportunities include real name, account activation, participation in official website activities, recharge amount, repetition frequency, and successful product purchases; Obtaining a CART classification tree by constructing a classification tree model according to the sales opportunity preprocessing data; Obtaining an optimal subtree according to the CART classification tree through a pruning model; Preset classification rules, and obtain the optimal data of the sales opportunity according to the optimal subtree through the classification rules; The classification rules include: Determine whether the recharge amount is greater than 0. If yes, the sales opportunity is classified as "high probability of successful product purchase"; if no, determine whether the real name is 0; If yes, the sales opportunity is classified as "low probability of successful product purchase"; if no, it is determined whether the repetition frequency is greater than 2; If yes, the sales opportunity is classified as "high probability of successful product purchase"; if no, it is determined whether the participation in the official website activity is 1; If yes, the sales opportunity is classified as "high probability of successful product purchase"; if no, it is determined whether the activation account is 1; If yes, the sales opportunity is classified as “higher probability of successful product purchase”; if no, the sales opportunity is classified as “lower probability of successful product purchase”.

[0007] Preferably, obtaining a CART classification tree by constructing a classification tree model according to the sales opportunity preprocessing data comprises: Acquire the sales opportunity preprocessing data features, wherein the sales opportunity preprocessing data features include binary discrete features, common discrete features, and continuous distribution features; According to the binary discrete feature, a binary feature Gini value is calculated by using a binary Gini value; Obtaining a common feature Gini value by calculating a common Gini value according to the common discrete feature; Calculate the continuous feature Gini value by continuous Gini value according to the continuous distribution feature; Sorting the sales opportunity preprocessing data according to the binary feature Gini value, the common feature Gini value, and the continuous feature Gini value to obtain split sequence data; Preset a split threshold, and obtain the CART classification tree through threshold judgment according to the split threshold, the split sequence data, and the sales opportunity preprocessing data; The threshold judgment includes judging whether the sales opportunity pre-processed data is greater than the split threshold. If yes, the threshold judgment is repeated according to the split sequence data until all the split sequence data are judged; if no, the sales opportunity pre-processed data is a leaf node.

[0008] Preferably, obtaining the optimal subtree through a pruning model according to the CART classification tree includes: S202-1: Calculating the loss function reduction degree by node loss degree according to the CART classification tree; S202-2: obtaining a node with a minimum loss function reduction degree by sorting the loss function reduction degrees, and using the node with the minimum loss function reduction degree as a pruning point; S202-3: Pruning the CART classification tree according to the pruning point to obtain an initial optimal subtree; S202-4: Repeat steps S202-1 to S202-3 for the initial optimal subtree until the initial optimal subtree is the root node to obtain an initial optimal subtree sequence; S202-5: Obtain the optimal subtree through cross-validation according to the initial optimal subtree sequence.

[0009] Preferably, the obtaining of the sales task priority by performing sales task reminders through a sales reminder model according to the sales task data and the sales opportunity optimal data comprises: The sales task data includes basic task information, task status, important task node information, task association information, sales stage information, customer response data, and task priority level; Pushing the sales opportunity optimal data, the task related information, and the customer response data to the sales terminal for reminder; Preset task status threshold and task important node time threshold, and remind based on the task status threshold, the task important node time threshold, the task status, and the task important node information through threshold judgment; The threshold determination includes: Determine whether the task status exceeds the task status threshold. If yes, push the task status to the sales terminal for reminder and set the task priority level to the highest level; if no, do nothing; Determine whether the important node information of the task exceeds the important node time threshold of the task. If yes, push the important node information of the task to the sales terminal for reminder, and set the task priority level to the highest level; if no, do nothing; The sales stage information carries a stage update tag. If the stage update tag is 1, the sales stage information is pushed to the sales terminal for reminder, and the task priority level is set to the highest level; if the stage update tag is 0, no processing is performed.

[0010] Preferably, performing sales task warning through a sales warning model according to the sales task data and the sales task priority includes: Storing the sales task data and the sales task priority in a data set to obtain a sales task data set; Obtaining a sales task preprocessing data set through a sales task preprocessing model according to the sales task data set; The sales task preprocessing model includes standardization, normalization, and filling missing values ​​with center values; According to the sales task preprocessing data set, a sales task time series feature is obtained through a long short-term memory neural network; According to the sales task temporal features, a global correlation feature of the sales task is obtained through a multi-head attention mechanism of a Transformer network; Predicting the degree of completion of the sales task through a fully connected neural network according to the sales task time series characteristics and the sales task global correlation characteristics; A sales task warning is issued based on the degree of completion of the sales task.

[0011] A sales task reminder and early warning mechanism implementation system, including a data preprocessing module, a data mining module, a task reminder module, and a task early warning module; The data preprocessing module is used to obtain sales opportunity data, and obtain sales opportunity preprocessing data according to the sales opportunity data through a data preprocessing model; The data mining module is used to obtain the optimal data of sales opportunities through a data mining model according to the sales opportunity preprocessing data; The task reminder module is used to obtain sales task data, and to perform sales task reminders through a sales reminder model according to the sales task data and the sales opportunity optimal data to obtain sales task priorities; The task warning module is used to perform sales task warning according to the sales task data and the sales task priority through a sales warning model.

[0012] An electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, the method for implementing the above-mentioned sales task reminder and early warning mechanism is implemented.

[0013] A storage medium containing computer executable instructions, which, when executed by a computer processor, are used to execute the above-mentioned sales task reminder and early warning mechanism implementation method.

[0014] The beneficial effects of the present invention are: (1) Preprocess the data through standardization, deduplication, normalization and missing value filling to improve the standardization and consistency of the data source and provide a reliable data foundation for subsequent data mining models.

[0015] (2) The optimal data of sales opportunities is extracted through the CART classification tree combined with the pruning model, and sales opportunities are classified according to rules to obtain sales opportunities with a higher probability of successfully purchasing products. Sales personnel are then reminded to lock in high-value sales opportunities and improve their ability to convert potential customers.

[0016] (3) Through the sales reminder model, combined with the task status, important nodes and stage update labels in the task data, reminders are given for sales tasks to improve task management efficiency, provide real-time feedback, use dynamic changes in task processes, and reduce the risk of task delays.

[0017] (4) A sales early warning model is constructed through the long short-term memory network and the Transformer multi-head attention mechanism to fully extract the temporal characteristics and global correlation characteristics of the sales task data. Based on the temporal characteristics and global characteristics, the task completion is predicted through a fully connected neural network, the risk points of task completion are accurately identified, and early warnings are issued to improve the completion of sales tasks. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to facilitate understanding by those skilled in the art, the present invention is further described below with reference to the accompanying drawings.

[0019] Figure 1 The present invention is a flowchart of a method for implementing a sales task reminder and early warning mechanism. DETAILED DESCRIPTION

[0020] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the specific implementation methods, structures, features and effects of the present invention are described in detail below in conjunction with the accompanying drawings and preferred embodiments.

[0021] See also Figure 1 , a method for implementing a sales task reminder and early warning mechanism, comprising: S1: Acquire sales opportunity data, and obtain sales opportunity preprocessing data through a data preprocessing model according to the sales opportunity data; S2: obtaining optimal sales opportunity data through a data mining model based on the sales opportunity preprocessing data; S3: Acquire sales task data, and perform sales task reminders through a sales reminder model according to the sales task data and the optimal sales opportunity data to obtain sales task priorities; S4: Performing sales task warning through a sales warning model according to the sales task data and the sales task priority.

[0022] Specifically, the sales task data includes basic task information, task status, important task node information, task association information, sales stage information, customer response data, and task priority level.

[0023] In step S1, the data preprocessing model includes: S101: obtaining sales opportunity specification data through data specification processing according to the sales opportunity data; S102: Filling missing values ​​with central values ​​according to the sales opportunity specification data to obtain the sales opportunity preprocessing data.

[0024] In this embodiment, the sales opportunity pre-processing data features include real name, account activation, participation in official website activities, recharge amount, repetition frequency, and successful product purchase; the binary discrete features include the real name, the activation, and the participation in official website activities; the ordinary discrete features include the repetition frequency; and the continuous distribution features include the recharge amount.

[0025] Specifically, the data reduction processing includes standardization processing, deduplication processing, and normalization processing.

[0026] In step S2, the data mining model includes: S201: Obtaining a CART classification tree by constructing a classification tree model according to the sales opportunity preprocessing data; S202: Obtaining an optimal subtree according to the CART classification tree through a pruning model; S203: Preset classification rules, and obtain the optimal sales opportunity data according to the optimal subtree using the classification rules.

[0027] Specifically, the construction of the classification tree model includes: S201-1: Acquire the sales opportunity preprocessing data features, where the sales opportunity preprocessing data features include binary discrete features, common discrete features, and continuous distribution features; S201-2: Obtaining a binary feature Gini value by calculating a binary Gini value according to the binary discrete feature; The binary Gini value calculation expression is: , in, Gini(X 1) is the Gini value of the binary feature, |D| Pre-process the total number of data for sales opportunities, |D 0 | Indicates the number of data with discrete features of 0, |D 1 | Indicates the number of data with discrete features of 1, D 0 Represents data with discrete features of 0, D represents the sales opportunity pre-processing data, D 1 Represents data with a discrete feature of 1, i represents data with discrete feature i, i The values ​​of are 0 and 1. k represents the kth successfully purchased product category, K Indicates the total number of categories of the successfully purchased products. p ki It represents the proportion of category k in the data with discrete feature i. Gini(D i ) Represents the Gini value of data with discrete feature i; S201-3: Obtaining a common feature Gini value by calculating a common Gini value according to the common discrete feature; The ordinary Gini value calculation expression is: , in, Gini(X 2 ) is the common feature Gini value, represents the minimum Gini value, |D| Pre-process the total number of data for sales opportunities, S represents random binary classification data features, |D L | Indicates the number of data whose data features belong to S, |D R | Indicates the number of data whose characteristics do not belong to S, D L Indicates that the data characteristics belong to S, D represents the sales opportunity pre-processing data, D R Data features do not belong to the data of S, i represents data with discrete feature i, k represents the kth successfully purchased product category, KIndicates the total number of categories of the successfully purchased products. p ki It represents the proportion of category k in the data with discrete feature i. Gini(S) represents the Gini value of random binary features, Gini(D i ) Denotes the calculation of Gini(D L ) and Gini(D R ) S201-4: Calculating a continuous feature Gini value based on the continuous distribution feature by using a continuous Gini value; The continuous Gini value calculation expression is: , in, Gini(X 3 ) is the continuous feature Gini value, x i and x i+1 For adjacent data, ti is the average value, the average value is the candidate segmentation point, T represents the total number of candidate segmentation points, min Indicates the minimum value operation. |D| Pre-process the total number of data for sales opportunities, |D m | Indicates the number of data that is less than or equal to ti. |D n | Indicates the number of data greater than ti. D m Indicates data that is less than or equal to ti, D represents the sales opportunity pre-processing data, D n Indicates that the data is greater than ti. i represents data with discrete feature i, k represents the kth successfully purchased product category, K Indicates the total number of categories of the successfully purchased products. p ki It represents the proportion of category k in the data with discrete feature i. Gini(t i ) The Gini value of the mean, Gini(D i ) represents the general formula for calculating Gini(Dm) and Gini(Dn); S201-5: Sort the sales opportunity pre-processed data according to the binary feature Gini value, the common feature Gini value, and the continuous feature Gini value to obtain split sequence data; S201-6: Preset a splitting threshold, and obtain the CART classification tree through threshold judgment according to the splitting threshold, the splitting sequence data, and the sales opportunity pre-processing data; The threshold judgment includes judging whether the sales opportunity pre-processed data is greater than the split threshold. If yes, the threshold judgment is repeated according to the split sequence data until all the split sequence data are judged; if no, the sales opportunity pre-processed data is a leaf node.

[0028] In this embodiment, the common discrete feature needs to be binary divided 10 times, and its Gini value is calculated respectively, and the smallest Gini value is taken as the Gini value of the common feature.

[0029] Specifically, the pruning model includes: S202-1: Calculating the loss function reduction degree by node loss degree according to the CART classification tree; The node loss calculation expression is: , in, g(t) is the loss function reduction degree, T t is the CART classification tree, t is the root node, c(t) is the training error of the root node, c(T t ) is the training error of the CART classification tree, |T t | is the number of leaf nodes in the CART classification tree; S202-2: obtaining a node with a minimum loss function reduction degree by sorting the loss function reduction degrees, and using the node with the minimum loss function reduction degree as a pruning point; S202-3: Pruning the CART classification tree according to the pruning point to obtain an initial optimal subtree; S202-4: Repeat steps S202-1 to S202-3 for the initial optimal subtree until the initial optimal subtree is the root node to obtain an initial optimal subtree sequence; S202-5: Obtain the optimal subtree through cross-validation according to the initial optimal subtree sequence.

[0030] Specifically, the classification rules include: Determine whether the recharge amount is greater than 0. If yes, the sales opportunity is classified as "high probability of successful product purchase"; if no, determine whether the real name is 0; If yes, the sales opportunity is classified as "low probability of successful product purchase"; if no, it is determined whether the repetition frequency is greater than 2; If yes, the sales opportunity is classified as "high probability of successful product purchase"; if no, it is determined whether the participation in the official website activity is 1; If yes, the sales opportunity is classified as "high probability of successful product purchase"; if no, it is determined whether the activation account is 1; If yes, the sales opportunity is classified as “higher probability of successful product purchase”; if no, the sales opportunity is classified as “lower probability of successful product purchase”.

[0031] In step S3, the sales reminder model includes: S301: Pushing the sales opportunity optimal data, the task related information, and the customer response data to the sales terminal for reminder; S302: Preset a task status threshold and a task important node time threshold, and make a reminder based on the task status threshold, the task important node time threshold, the task status, and the task important node information through threshold judgment; The threshold determination includes: Determine whether the task status exceeds the task status threshold. If yes, push the task status to the sales terminal for reminder and set the task priority level to the highest level; if no, do nothing; Determine whether the important node information of the task exceeds the important node time threshold of the task. If yes, push the important node information of the task to the sales terminal for reminder, and set the task priority level to the highest level; if no, do nothing; S303: The sales stage information carries a stage update tag. If the stage update tag is 1, the sales stage information is pushed to the sales terminal for reminder, and the task priority level is set to the highest level; if the stage update tag is 0, no processing is performed.

[0032] In step S4, the sales early warning model includes: S401: Storing the sales task data and the sales task priority in a data set to obtain a sales task data set; S402: Obtaining a sales task preprocessing data set through a sales task preprocessing model according to the sales task data set; S403: Obtaining sales task time series features through a long short-term memory neural network according to the sales task preprocessing data set; S404: Obtaining global correlation features of the sales tasks through a multi-head attention mechanism of a Transformer network according to the sales task temporal features; S405: Predicting the degree of completion of the sales task through a fully connected neural network according to the sales task time series features and the sales task global correlation features; S406: Providing sales task warning according to the sales task completion degree.

[0033] Specifically, the sales task preprocessing model includes standardization, normalization, and filling missing values ​​with center values.

[0034] A sales task reminder and early warning mechanism implementation system, including a data preprocessing module, a data mining module, a task reminder module, and a task early warning module; The data preprocessing module is used to obtain sales opportunity data, and obtain sales opportunity preprocessing data according to the sales opportunity data through a data preprocessing model; The data mining module is used to obtain the optimal data of sales opportunities through a data mining model according to the sales opportunity preprocessing data; The task reminder module is used to obtain sales task data, and to perform sales task reminders through a sales reminder model according to the sales task data and the sales opportunity optimal data to obtain sales task priorities; The task warning module is used to perform sales task warning according to the sales task data and the sales task priority through a sales warning model.

[0035] The computer storage medium of the embodiment of the present invention may adopt any combination of one or more computer-readable media. The computer-readable medium may be a computer-readable signal medium or a computer-readable storage medium. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples (non-exhaustive list) of computer-readable storage media include: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this document, a computer-readable storage medium may be any tangible medium containing or storing a program that may be used by or in combination with an instruction execution system, device or device.

[0036] Computer-readable signal media may include data signals propagated in baseband or as part of a carrier wave, which carry computer-readable program code. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium other than a computer-readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0037] The program code included on the computer readable medium can be transmitted with any appropriate medium, including but not limited to wireless, electric wire, optical cable, RF, etc., or any suitable combination of the above. The computer program code for performing the operation of the present invention can be written in one or more programming languages ​​or their combinations, and the programming language includes object-oriented programming languages-such as Java, Smalltalk, C++, and also includes conventional procedural programming languages-such as "C" language or similar programming languages. The program code can be executed completely on the user's computer, partially on the user's computer, as an independent software package, partially on the user's computer and partially on the remote computer, or completely on the remote computer or server. In the case of a remote computer, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computer (for example, using an Internet service provider to connect through the Internet).

[0038] The above description is only a preferred embodiment of the present invention and does not limit the present invention in any form. Although the present invention has been disclosed as a preferred embodiment as above, it is not used to limit the present invention. Any technical personnel in this field can make some changes or modify the technical contents disclosed above into equivalent embodiments without departing from the scope of the technical solution of the present invention. However, any simple modification, equivalent change and modification made to the above embodiments according to the technical essence of the present invention without departing from the content of the technical solution of the present invention still fall within the scope of the technical solution of the present invention.

Claims

1. A method for implementing a sales task reminder and early warning mechanism, characterized in that: The following steps are involved: Acquire sales opportunity data, and obtain sales opportunity preprocessing data through a data preprocessing model according to the sales opportunity data; Obtaining optimal sales opportunity data through a data mining model based on the sales opportunity preprocessing data; Acquire sales task data, and perform sales task reminders through a sales reminder model according to the sales task data and the sales opportunity optimal data to obtain sales task priorities; A sales task warning is performed according to the sales task data and the sales task priority through a sales warning model.

2. The method for implementing the sales task reminder and early warning mechanism according to claim 1, characterized in that: The obtaining of sales opportunity preprocessing data by using a data preprocessing model according to the sales opportunity data comprises: Obtaining sales opportunity specification data through data specification processing according to the sales opportunity data; The data reduction processing includes standardization processing, deduplication processing, and normalization processing; The sales opportunity pre-processing data is obtained by filling missing values ​​with central values ​​according to the sales opportunity specification data.

3. The method for implementing the sales task reminder and early warning mechanism according to claim 1, characterized in that: The obtaining of the optimal sales opportunity data by using a data mining model based on the sales opportunity preprocessing data includes: The pre-processed data features of sales opportunities include real name, account activation, participation in official website activities, recharge amount, repetition frequency, and successful product purchases; Obtaining a CART classification tree by constructing a classification tree model according to the sales opportunity preprocessing data; Obtaining an optimal subtree according to the CART classification tree through a pruning model; Preset classification rules, and obtain the optimal data of the sales opportunity according to the optimal subtree through the classification rules; The classification rules include: Determine whether the recharge amount is greater than 0. If yes, the sales opportunity is classified as "high probability of successful product purchase"; if no, determine whether the real name is 0; If yes, the sales opportunity is classified as "low probability of successful product purchase"; if no, it is determined whether the repetition frequency is greater than 2; If yes, the sales opportunity is classified as "high probability of successful product purchase"; if no, it is determined whether the participation in the official website activity is 1; If yes, the sales opportunity is classified as "high probability of successful product purchase"; if no, it is determined whether the activation account is 1; If yes, the sales opportunity is classified as "higher probability of successful product purchase"; if no, the sales opportunity is classified as "lower probability of successful product purchase".

4. The method for implementing the sales task reminder and early warning mechanism according to claim 3, characterized in that: The obtaining of a CART classification tree by constructing a classification tree model according to the sales opportunity preprocessing data comprises: Acquire the sales opportunity preprocessing data features, wherein the sales opportunity preprocessing data features include binary discrete features, common discrete features, and continuous distribution features; According to the binary discrete feature, a binary feature Gini value is calculated by using a binary Gini value; Obtaining a common feature Gini value by calculating a common Gini value according to the common discrete feature; Calculate the continuous feature Gini value by continuous Gini value according to the continuous distribution feature; Sorting the sales opportunity preprocessing data according to the binary feature Gini value, the common feature Gini value, and the continuous feature Gini value to obtain split sequence data; Preset a split threshold, and obtain the CART classification tree through threshold judgment according to the split threshold, the split sequence data, and the sales opportunity preprocessing data; The threshold judgment includes judging whether the sales opportunity pre-processed data is greater than the split threshold. If yes, the threshold judgment is repeated according to the split sequence data until all the split sequence data are judged; if no, the sales opportunity pre-processed data is a leaf node.

5. The method for implementing the sales task reminder and early warning mechanism according to claim 3, characterized in that: The obtaining of the optimal subtree by a pruning model according to the CART classification tree comprises: S202-1: Calculating the loss function reduction degree by node loss degree according to the CART classification tree; S202-2: obtaining a node with a minimum loss function reduction degree by sorting the loss function reduction degrees, and using the node with the minimum loss function reduction degree as a pruning point; S202-3: Pruning the CART classification tree according to the pruning point to obtain an initial optimal subtree; S202-4: Repeat steps S202-1 to S202-3 for the initial optimal subtree until the initial optimal subtree is the root node to obtain an initial optimal subtree sequence; S202-5: Obtain the optimal subtree through cross-validation according to the initial optimal subtree sequence.

6. The method for implementing the sales task reminder and early warning mechanism according to claim 1, characterized in that: The obtaining of the sales task priority by performing sales task reminders through a sales reminder model according to the sales task data and the sales opportunity optimal data includes: The sales task data includes basic task information, task status, important task node information, task association information, sales stage information, customer response data, and task priority level; Pushing the sales opportunity optimal data, the task related information, and the customer response data to the sales terminal for reminder; Preset task status threshold and task important node time threshold, and remind based on the task status threshold, the task important node time threshold, the task status, and the task important node information through threshold judgment; The threshold determination includes: Determine whether the task status exceeds the task status threshold. If yes, push the task status to the sales terminal for reminder and set the task priority level to the highest level; if no, do nothing; Determine whether the important node information of the task exceeds the important node time threshold of the task. If yes, push the important node information of the task to the sales terminal for reminder, and set the priority level of the task to the highest level; if no, do nothing; The sales stage information carries a stage update tag. If the stage update tag is 1, the sales stage information is pushed to the sales terminal for reminder, and the task priority level is set to the highest level; if the stage update tag is 0, no processing is performed.

7. The method for implementing the sales task reminder and early warning mechanism according to claim 1, characterized in that: The performing of sales task warning by using a sales warning model according to the sales task data and the sales task priority includes: Storing the sales task data and the sales task priority in a data set to obtain a sales task data set; Obtaining a sales task preprocessing data set through a sales task preprocessing model according to the sales task data set; The sales task preprocessing model includes standardization, normalization, and filling missing values ​​with center values; According to the sales task preprocessing data set, a sales task time series feature is obtained through a long short-term memory neural network; According to the sales task temporal features, a global correlation feature of the sales task is obtained through a multi-head attention mechanism of a Transformer network; Predicting the degree of completion of the sales task through a fully connected neural network according to the sales task time series characteristics and the sales task global correlation characteristics; A sales task warning is issued based on the degree of completion of the sales task.

8. A sales task reminder and early warning mechanism implementation system, characterized in that: Including data preprocessing module, data mining module, task reminder module, and task warning module; The data preprocessing module is used to obtain sales opportunity data, and obtain sales opportunity preprocessing data according to the sales opportunity data through a data preprocessing model; The data mining module is used to obtain the optimal data of sales opportunities through a data mining model according to the sales opportunity preprocessing data; The task reminder module is used to obtain sales task data, and to perform sales task reminders through a sales reminder model according to the sales task data and the sales opportunity optimal data to obtain sales task priorities; The task warning module is used to perform sales task warning according to the sales task data and the sales task priority through a sales warning model.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the method for implementing the sales task reminder and early warning mechanism as described in any one of claims 1-7 is implemented.

10. A storage medium containing computer executable instructions, characterized in that: The computer executable instructions, when executed by a computer processor, are used to execute the method for implementing the sales task reminder and early warning mechanism as described in any one of claims 1-7.

Images