Insurance salesman recommendation method and system based on hybrid recommendation algorithm

Abstract

The invention relates to an insurance salesman recommendation method and system based on a hybrid recommendation algorithm, and the method comprises the steps: enabling client equipment to generate an insurance service request according to an insurance service demand, and transmitting the insurance service request to a salesman recommendation platform; the salesperson recommendation platform obtains salesperson information of all salespersons from the insurance company service platform according to the insurance service request, and calculates a single score value of each salesperson according to the insurance service request and the salesperson information; the salesperson recommendation platform calculates a comprehensive recommendation value of each salesperson based on a mixed recommendation algorithm and the single score values; and the salesperson recommendation platform selects the salesperson with the maximum comprehensive recommendation value as the optimal salesperson, and sends a service instruction to the sales end equipment corresponding to the optimal salesperson, so that the optimal salesperson handles an insurance service for the target customer according to the service instruction. And the quality of the insurance service process and the customer satisfaction are improved, so that the customer conversion rate and the sales efficiency are improved.

Description

Technical Field

[0001] The present invention belongs to the field of insurance business, and in particular relates to an insurance salesperson recommendation method and system based on a hybrid recommendation algorithm. Background Art

[0002] In the current sales scenario of the insurance industry, when potential customers first access the insurance company system and seek to establish contact with a salesperson, the traditional recommendation mechanism often relies on a single geographic location data and adopts a strategy of proximity allocation or random matching to recommend salespeople. Although this approach simplifies the initial contact process to a certain extent and facilitates salespeople to respond quickly to customer needs, it exposes significant limitations and shortcomings in in-depth customer relationship management and long-term services. Existing practices ignore the historical interactions and potential connections that may exist between customers and other salespeople within the company. This past relationship is not only an important foundation for building trust and promoting conversions, but also a key factor in improving customer satisfaction and enhancing customer stickiness. However, the current system fails to effectively capture and utilize this relationship information, resulting in a lack of personalization and depth in the salesperson recommendation process and limited customer experience.

[0003] In addition, the salesperson's professional ability score, current customer load, and the degree of match between his or her areas of expertise and customer needs are another important factor affecting sales efficiency and customer conversion rate. The salesperson's professional quality, experience accumulation, and expertise in specific fields directly affect whether he or she can accurately understand and respond to the personalized needs of customers, and the salesperson's customer load determines the timeliness of the salesperson's response to customer needs. However, traditional recommendation mechanisms often ignore these soft power considerations and are unable to effectively identify and recommend salespeople who best meet customer needs, resulting in low matching during the service process, reduced customer satisfaction, and difficulty in improving conversion rates.

[0004] In summary, the existing technology over-relies on hard indicators such as geographic location in the initial matching between customers and salespeople, while ignoring soft indicators such as historical interactions, professional ability scores, customer carrying capacity, and demand matching. This not only affects the customer's first contact experience, but also limits the customer conversion rate and sales efficiency in the subsequent service process. Summary of the invention

[0005] In order to remedy the deficiencies of the prior art, the present invention provides an insurance salesperson recommendation method and system based on a hybrid recommendation algorithm.

[0006] In order to solve the above technical problems, the technical solution adopted by the present invention is:

[0007] In a first aspect, a method for recommending insurance salespersons based on a hybrid recommendation algorithm is provided, which is applied to an insurance salesperson recommendation system based on a hybrid recommendation algorithm. The insurance salesperson recommendation system based on the hybrid recommendation algorithm includes a salesperson recommendation platform, a sales end device, a client device, and an insurance company business platform. The method includes:

[0008] When the target customer initiates an insurance business demand through the client device, the client device generates an insurance business request according to the insurance business demand and sends the insurance business request to the salesperson recommendation platform;

[0009] The salesperson recommendation platform obtains the salesperson information of all salespersons from the insurance company's business platform according to the insurance business request, and calculates the individual score of each salesperson according to the insurance business request and the salesperson information. The individual score includes the distance score, the previous relationship score, the business demand matching score, the sales professional score and the customer carrying score.

[0010] The salesperson recommendation platform calculates the comprehensive recommendation value of each salesperson based on the hybrid recommendation algorithm and the single score value;

[0011] The salesperson recommendation platform selects the salesperson with the highest comprehensive recommendation value as the best salesperson, and sends business instructions to the sales-end device corresponding to the best salesperson, so that the best salesperson can handle insurance business for the target customer according to the business instructions.

[0012] Furthermore, the insurance business request includes the expected insurance type information, the target customer location information and the target customer unique identifier.

[0013] The salesperson recommendation platform obtains the salesperson information of all salespersons from the insurance company's business platform according to the insurance business request, and calculates the individual score value of each salesperson based on the hybrid recommendation algorithm according to the insurance business request and the salesperson information, including:

[0014] The salesperson recommendation platform analyzes the insurance business request to obtain the expected insurance type information, target customer location information and target customer unique identifier;

[0015] The salesperson recommendation platform obtains the salesperson information of all salespersons from the insurance company's business platform; the salesperson information includes the salesperson's unique identification, salesperson's location information, insurance types that can be undertaken, sales performance information, and the number of customers who have not yet issued a policy;

[0016] The salesperson recommendation platform calculates the distance score value based on the target customer location information and the salesperson location information;

[0017] The salesperson recommendation platform calculates the past relationship score based on the salesperson's unique ID and the target customer's unique ID;

[0018] The salesperson recommendation platform calculates the demand matching score based on the information of the insurance types that can be undertaken and the expected insurance types;

[0019] The salesperson recommendation platform calculates the sales professional score based on sales performance information;

[0020] The salesperson recommendation platform calculates the customer carrying score based on the number of customers who have not placed an order.

[0021] Furthermore, the salesperson recommendation platform calculates the distance score value based on the target customer location information and the salesperson location information, including:

[0022] The salesperson recommendation platform uses the preset map navigation software to obtain the physical distance value between the target customer location information and the salesperson location information;

[0023] The salesperson recommendation platform selects salespersons whose physical distance value is not greater than the preset distance value as candidate salespersons to form a candidate salesperson set;

[0024] The salesperson recommendation platform normalizes the physical distance values ​​corresponding to all candidate salespersons to obtain a distance score value for each candidate salesperson. The distance score value ranges from [0,1]. A larger value indicates a longer distance, and a smaller value indicates a longer distance.

[0025] Furthermore, the salesperson recommendation platform calculates the past relationship score value based on the salesperson's unique identifier and the target customer's unique identifier, including:

[0026] The salesperson recommendation platform searches the insurance company's business platform database based on the target customer's unique identifier to determine whether the target customer already has an association with a salesperson;

[0027] If there is no relationship with any salesperson, the previous relationship score of all the salespersons to be selected will be 0 points;

[0028] If there is an association relationship with the first salesperson, the salesperson recommendation platform determines whether the first salesperson is in the set of salespersons to be selected based on the salesperson unique identifier of the first salesperson;

[0029] If the target customer is in the set of salespeople to be selected, the salesperson recommendation platform determines whether the relationship between the target customer and the first salesperson is a communication relationship or a business relationship; if it is a communication relationship, the previous relationship score of the first salesperson is counted as 0.5 points; if it is a business relationship, the previous relationship score of the first salesperson is counted as 1 point; the previous relationship scores of other salespeople in the set of salespeople to be selected except the first salesperson are counted as 0 points;

[0030] If the first salesperson is not in the set of salespeople to be selected, the salesperson recommendation platform will count the past relationship scores of all salespeople to be selected in the set as 0, and add the first salesperson as a candidate salesperson to the set of salespeople to be selected.

[0031] Furthermore, after adding the first salesperson as a candidate salesperson to the set of salespersons to be selected, the method further includes:

[0032] Assign the distance score value corresponding to the candidate salesperson to 0 points.

[0033] Furthermore, the salesperson recommendation platform calculates the demand matching score based on the insurance types that can be undertaken and the expected insurance types, including:

[0034] The salesperson recommendation platform determines whether the insurance types that can be undertaken by all salespersons in the salesperson set to be selected include the expected insurance types;

[0035] If the insurance types that the second salesperson can undertake include the expected insurance types, the salesperson recommendation platform will count the demand matching score of the second salesperson as 1 point;

[0036] If the insurance types that the third salesperson can undertake do not include the expected insurance types, the salesperson recommendation platform extracts insurance type keywords related to the expected insurance types from the sales performance information of the third salesperson, and analyzes whether the salesperson has the ability to undertake the expected insurance types corresponding to the expected insurance types;

[0037] If the fourth salesperson is able to take the job, the salesperson recommendation platform will assign the fourth salesperson's corresponding demand matching score to 0.5 points;

[0038] If the fifth salesperson is unable to take the job, the salesperson recommendation platform will count the demand matching score of the fifth salesperson as 0.

[0039] Furthermore, the salesperson recommendation platform calculates the sales professional score based on the sales performance information, including:

[0040] The salesperson recommendation platform analyzes sales performance information to obtain job grade information, cumulative sales data, claim amounts, and abnormal operation behaviors, and obtains sales professional scores of all salespeople in the set of candidate salespeople based on preset sales professional ability scoring rules; the sales professional score is a normalized value, and the sales professional score range is [0,1]. The larger the value, the stronger the professional ability, and the smaller the value, the lower the professional ability.

[0041] Furthermore, the insurance business request also includes the business demand time period.

[0042] The salesperson recommendation platform calculates the customer carrying score based on the number of customers who have not placed an order, including:

[0043] The salesperson recommendation platform determines whether the number of customers who have not placed orders for all salespersons in the salesperson set to be selected is less than the customer carrying capacity threshold;

[0044] If there is a salesperson whose customer carrying capacity is not less than the customer carrying capacity threshold, the salesperson recommendation platform will count the customer carrying capacity score of the salesperson corresponding to the customer carrying capacity not less than the customer carrying capacity threshold as 0 points;

[0045] If there is a salesperson whose capacity is less than the customer carrying capacity threshold, the salesperson recommendation platform will use the salesperson whose capacity is less than the customer carrying capacity threshold as the sixth salesperson;

[0046] The salesperson recommendation platform obtains the estimated order placement time period for the sixth salesperson corresponding to the number of customers who have not placed an order;

[0047] The salesperson recommendation platform determines whether the business demand time period conflicts with the expected order time period;

[0048] The salesperson recommendation platform will count the customer carrying score of the salesperson whose business demand time period conflicts with the expected order time period as 0 points among the sixth salespersons;

[0049] The salesperson recommendation platform will assign a customer carrying score of 1 to the salesperson among the sixth salespersons whose business demand time period does not conflict with the expected order placement time period.

[0050] Furthermore, the salesperson recommendation platform calculates the comprehensive recommendation value of each salesperson based on the hybrid recommendation algorithm and the single score value, including:

[0051] The salesperson recommendation platform inputs the distance score, the previous relationship score, the business demand matching score, the sales professional score and the customer carrying score into the salesperson recommendation model based on the hybrid recommendation algorithm;

[0052] The expression of the salesperson recommendation model based on the hybrid recommendation algorithm is:

[0053] S n =W1S 1n +W2S 2n +W3S 3n +W4S 4n +W5S 5n , n represents the nth salesperson, n is a positive integer greater than 0, S 1n is the distance score of the nth salesperson, S 2n is the previous relationship score of the nth salesperson, S 3n S is the business demand matching score of the nth salesperson. 4n is the sales professional score of the nth salesperson, S 5n is the customer load score value of the nth salesperson, W1, W2, W3, W4 and W5 are preset weight values;

[0054] The salesperson recommendation model outputs the comprehensive recommendation value S of the nth salesperson n .

[0055] In a second aspect, an insurance salesperson recommendation system based on a hybrid recommendation algorithm is provided, comprising:

[0056] Salesperson recommendation platform, sales-end equipment, client equipment and insurance company business platform;

[0057] A data link is established between the salesperson recommendation platform and the sales-end equipment, the client equipment and the insurance company's business platform;

[0058] The client device is used to generate an insurance business request according to the insurance business demand when the target customer initiates an insurance business demand through the client device, and send the insurance business request to the salesperson recommendation platform;

[0059] The salesperson recommendation platform is used to obtain the salesperson information of all salespersons from the insurance company's business platform according to insurance business requests, and calculate the individual scoring value of each salesperson according to the insurance business request and the salesperson information. The individual scoring values ​​include distance score, previous relationship score, business demand matching score, sales professional score and customer carrying score;

[0060] The salesperson recommendation platform is also used to calculate the comprehensive recommendation value of each salesperson based on the hybrid recommendation algorithm and the single score value;

[0061] The salesperson recommendation platform is also used to select the salesperson with the highest comprehensive recommendation value as the best salesperson, and send business instructions to the sales-end device corresponding to the best salesperson, so that the best salesperson can handle insurance business for the target customer according to the business instructions.

[0062] The beneficial effects achieved by the present invention are:

[0063] When a target customer initiates an insurance business demand through a client device, the client device generates an insurance business request based on the insurance business demand, and sends the insurance business request to the salesperson recommendation platform; the salesperson recommendation platform obtains the salesperson information of all salespersons from the insurance company's business platform based on the insurance business request, and calculates a single score value for each salesperson based on the insurance business request and the salesperson information. The single score value includes a distance score value, a previous relationship score value, a business demand matching score value, a sales professional score value, and a customer carrying score value; the salesperson recommendation platform calculates a comprehensive recommendation value for each salesperson based on a hybrid recommendation algorithm and single score values; the salesperson recommendation platform selects the salesperson with the largest comprehensive recommendation value as the best salesperson, and sends a business instruction to the sales-end device corresponding to the best salesperson, so that the best salesperson handles insurance business for the target customer according to the business instruction. When matching customers with salespeople, the distance between the customer and the salesperson, their previous relationship and business needs, as well as the sales professionalism and customer capacity of the salesperson are taken into consideration. A comprehensive recommendation value is calculated based on a hybrid recommendation algorithm to match the customer with the best salesperson, thereby improving the quality of the insurance service process and customer satisfaction, which in turn helps to improve customer conversion rates and sales efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] Figure 1 A flowchart of the insurance salesperson recommendation method based on the hybrid recommendation algorithm of the present invention;

[0065] Figure 2 A flow chart for calculating a single item scoring value for the present invention;

[0066] Figure 3 This is a structural diagram of the insurance salesperson recommendation system based on the hybrid recommendation algorithm of the present invention. DETAILED DESCRIPTION

[0067] The present invention will be further described below in conjunction with the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, and cannot be used to limit the protection scope of the present invention.

[0068] like Figure 1 As shown, an embodiment of the present invention provides an insurance salesperson recommendation method based on a hybrid recommendation algorithm, comprising:

[0069] 101, when the target customer initiates an insurance business demand through the client device, the client device generates an insurance business request according to the insurance business demand, and sends the insurance business request to the salesperson recommendation platform;

[0070] In this embodiment, the insurance salesperson recommendation system based on the hybrid recommendation algorithm is composed of a salesperson recommendation platform, a sales-end device, a client device and an insurance company business platform. The insurance company business platform is used for data processing and data storage for insurance-related business; the sales-end device is a terminal electronic device of the insurance salesperson, such as a personal computer, a mobile phone, etc.; the client device is a terminal electronic device used by the customer; the salesperson recommendation platform establishes a data transmission link with the sales-end device, the client device and the insurance company business platform;

[0071] When the target customer has insurance business needs, an input operation is performed on the client device. The client device generates an insurance business request based on the insurance business needs input by the target customer, and sends the insurance business request to the salesperson recommendation platform. The insurance business request includes expected insurance type information, target customer location information and target customer unique identification. The expected insurance type information is the expected type of insurance that the target customer wants to purchase, and the target customer location information is generated by the GPS or Beidou positioning system of the client device. The target customer unique identification can be the device unique ID of the client device used by the target customer, or it can be the target customer's identity ID.

[0072] 102. The salesperson recommendation platform obtains the salesperson information of all salespersons from the insurance company business platform according to the insurance business request, and calculates the individual scoring value of each salesperson according to the insurance business request and the salesperson information. The individual scoring value includes the distance score, the previous relationship score, the business demand matching score, the sales professional score and the customer carrying score;

[0073] In this embodiment, after the salesperson recommendation platform receives the insurance business request sent by the client device, it extracts the salesperson information of all salespersons from the database of the insurance company's business platform, and then calculates the individual score value of each salesperson based on the insurance business request and the salesperson information. The specific calculation process is as follows: Figure 2 Shown are:

[0074] 201, the salesperson recommendation platform analyzes the insurance business request to obtain the expected insurance type information, the target customer location information and the target customer unique identifier;

[0075] 202. The salesperson recommendation platform obtains salesperson information of all salespersons from the insurance company business platform; the salesperson information includes the salesperson's unique identification, salesperson's location information, insurance types that can be undertaken, sales performance information, and the number of customers who have not yet issued a policy;

[0076] 203, the salesperson recommendation platform calculates a distance score value based on the target customer location information and the salesperson location information;

[0077] The salesperson recommendation platform uses the preset map navigation software to obtain the physical distance value between the target customer location information and the salesperson location information; since salespeople usually meet offline to sign contracts for customers when handling insurance business, the physical distance needs to be taken into consideration when recommending salespeople. For example, the preset distance value is 100km, indicating that salespeople within 100 kilometers are selected as candidate salespeople. The salesperson recommendation platform selects salespeople whose physical distance value is not greater than the preset distance value as candidate salespeople, forming a set of candidate salespeople;

[0078] The salesperson recommendation platform normalizes the physical distance values ​​corresponding to all candidate salespersons to obtain a distance score value for each candidate salesperson. The distance score value ranges from [0,1]. Within this interval, a larger value indicates a longer distance, and a smaller value indicates a longer distance.

[0079] 204, the salesperson recommendation platform calculates the past relationship score value according to the salesperson unique identifier and the target customer unique identifier;

[0080] The salesperson recommendation platform searches the insurance company's business platform database based on the target customer's unique identifier to determine whether the target customer has an association with a salesperson. It is pre-specified that after a salesperson and a customer have a business transaction or communication, the salesperson will upload and save the business data or chat records to the insurance company's business platform database in a timely manner.

[0081] If there is no relationship with any salesperson, the previous relationship score of all the salespersons to be selected will be 0 points;

[0082] If there is an association relationship with the first salesperson, the salesperson recommendation platform determines whether the first salesperson is in the set of salespersons to be selected based on the salesperson unique identifier of the first salesperson;

[0083] If the target customer is in the set of salespeople to be selected, the salesperson recommendation platform determines whether the relationship between the target customer and the first salesperson is a communication relationship or a business relationship; the communication relationship may be that the first salesperson and the target customer have had telephone communication or communication via chat software; the business relationship may be that the first salesperson and the target customer have signed an insurance business contract; if it is a communication relationship, the previous relationship score of the first salesperson is calculated as 0.5 points; if it is a business relationship, the previous relationship score of the first salesperson is calculated as 1 point; the previous relationship scores of other salespeople in the set of salespeople to be selected, except the first salesperson, are calculated as 0 points;

[0084] If the candidate salesperson is not in the set of candidate salespersons, the salesperson recommendation platform will count the past relationship scores of all candidate salespersons in the set as 0, and add the first salesperson as a candidate salesperson to the set of candidate salespersons; since the candidate salesperson is not in the set of candidate salespersons, the distance between the candidate salesperson and the target customer must exceed the preset distance value, and the distance score value corresponding to the candidate salesperson is assigned 0 at this time.

[0085] 205. The salesperson recommendation platform calculates the demand matching score value based on the information of the insurance types that can be undertaken and the information of the expected insurance types;

[0086] After calculating the previous relationship score values ​​above, a new set of salespeople to be selected may be formed, or the set of salespeople to be selected may remain unchanged;

[0087] The salesperson recommendation platform determines whether the insurance types that all salespersons in the candidate salesperson set can undertake include the expected insurance types. Assuming that the expected insurance type information is accident insurance, each salesperson can undertake different types of insurance, for example, a salesperson who specializes in auto insurance.

[0088] If the insurance types that the second salesperson can undertake include the expected insurance types, the salesperson recommendation platform will count the demand matching score of the second salesperson as 1 point;

[0089] If the insurance types that the third salesperson can undertake do not include the expected insurance types, the salesperson recommendation platform will extract insurance type keywords related to the expected insurance types from the sales performance information of the third salesperson. For example, the expected insurance type is family accident insurance, but the insurance types that the third salesperson can undertake do not include family accident insurance. However, the insurance type keyword extracted from "family accident insurance" is "accident insurance". Assuming that the sales performance information of the third salesperson includes "personal accident insurance", the third salesperson is capable of connecting with the target customer to complete the business of "family accident insurance" and can undertake the business as the fourth salesperson. Assuming that the sales performance information of the third salesperson does not include the insurance type keyword "accident insurance", the third salesperson is not capable of undertaking the business of "family accident insurance" and can serve as the fifth salesperson.

[0090] If the fourth salesperson is able to take the job, the salesperson recommendation platform will assign the fourth salesperson's corresponding demand matching score to 0.5 points;

[0091] If the fifth salesperson is unable to take the job, the salesperson recommendation platform will count the demand matching score of the fifth salesperson as 0.

[0092] 206, the salesperson recommendation platform calculates the sales professional score value according to the sales performance information;

[0093] The salesperson recommendation platform analyzes sales performance information to obtain job grade information, cumulative sales data, claim amounts, and abnormal operation behaviors, and obtains sales professional scores of all salespeople in the set of candidate salespeople based on preset sales professional ability scoring rules; the sales professional score is a normalized value, and the sales professional score range is [0,1]. The larger the value, the stronger the professional ability, and the smaller the value, the lower the professional ability.

[0094] 207. The salesperson recommendation platform calculates the customer carrying score value according to the number of customers who have not placed an order.

[0095] The insurance business request also includes the business demand time period.

[0096] The salesperson recommendation platform determines whether the number of unordered customers of all salespersons in the candidate salesperson set is less than the customer capacity threshold. Because each salesperson has limited business processing capabilities, assuming that the customer capacity threshold is 20, and salesperson A still has 30 unordered customers, then salesperson A is saturated with workload and cannot connect with target customers. In this case, the customer capacity score is 0.

[0097] If there is a salesperson whose customer carrying capacity is not less than the customer carrying capacity threshold, the salesperson recommendation platform will count the customer carrying capacity score of the salesperson corresponding to the customer carrying capacity not less than the customer carrying capacity threshold as 0 points;

[0098] Assume that the customer capacity threshold is 20, and salesperson B still has 5 customers who have not placed orders. Salesperson B is less than the customer capacity threshold, so the salesperson recommendation platform will select salesperson B as the sixth salesperson;

[0099] The salesperson recommendation platform obtains the estimated order-issuing time periods for the number of customers who have not placed an order corresponding to the sixth salesperson; assuming that the estimated order-issuing time periods for the five customers who have not placed an order for the sixth salesperson C are 8:00-12:00 on September 1, 10:00-12:00 on September 2, 12:00-15:00 on September 4, 16:00-18:00 on September 5, and 10:00-14:00 on September 5;

[0100] The salesperson recommendation platform determines whether the business demand time period conflicts with the expected order time period; for example, the target customer's business demand time period is around 10:00 on September 1, which conflicts with the sixth salesperson C's time period of 8:00-12:00 on September 1, so the sixth salesperson C cannot connect with the target customer; the sixth salesperson C's customer carrying score is 0 points;

[0101] The salesperson recommendation platform will assign a customer carrying score of 1 to the salesperson among the sixth salespersons whose business demand time period does not conflict with the expected order placement time period.

[0102] 103. The salesperson recommendation platform calculates the comprehensive recommendation value of each salesperson based on the hybrid recommendation algorithm and the single score value;

[0103] Among them, the salesperson recommendation platform inputs the distance score, previous relationship score, business demand matching score, sales professional score and customer carrying score into the salesperson recommendation model based on the hybrid recommendation algorithm;

[0104] The expression of the salesperson recommendation model based on the hybrid recommendation algorithm is:

[0105] S n =W1S 1n +W2S 2n +W3S 3n +W4S 4n +W5S 5n , n represents the nth salesperson, n is a positive integer greater than 0, S 1n is the distance score of the nth salesperson, S 2n is the previous relationship score of the nth salesperson, S 3n S is the business demand matching score of the nth salesperson. 4n is the sales professional score of the nth salesperson, S 5n The customer load score value of the nth salesperson, W1, W2, W3, W4 and W5 are preset weight values, and the values ​​of W1, W2, W3, W4 and W5 can be equal, or can be set according to which single score value is preset;

[0106] The salesperson recommendation model outputs the comprehensive recommendation value S of the nth salesperson n .

[0107] 104. The salesperson recommendation platform selects the salesperson with the largest comprehensive recommendation value as the best salesperson, and sends a business instruction to the sales terminal device corresponding to the best salesperson, so that the best salesperson can handle insurance business for the target customer according to the business instruction.

[0108] The best salesperson with the largest comprehensive recommendation value is selected from multiple salespeople, and a business instruction is sent to the sales-end device corresponding to the best salesperson. The sales-end device displays the target customer to the best salesperson according to the business instruction, so that the best salesperson can handle insurance business for the target customer.

[0109] The implementation principle of the embodiment of the present invention is:

[0110] When a target customer initiates an insurance business demand through a client device, the client device generates an insurance business request based on the insurance business demand, and sends the insurance business request to the salesperson recommendation platform; the salesperson recommendation platform obtains the salesperson information of all salespersons from the insurance company's business platform based on the insurance business request, and calculates a single score value for each salesperson based on the insurance business request and the salesperson information. The single score value includes a distance score value, a previous relationship score value, a business demand matching score value, a sales professional score value, and a customer carrying score value; the salesperson recommendation platform calculates a comprehensive recommendation value for each salesperson based on a hybrid recommendation algorithm and single score values; the salesperson recommendation platform selects the salesperson with the largest comprehensive recommendation value as the best salesperson, and sends a business instruction to the sales-end device corresponding to the best salesperson, so that the best salesperson handles insurance business for the target customer according to the business instruction. When matching customers with salespeople, the distance between the customer and the salesperson, their previous relationship and business needs, as well as the sales professionalism and customer capacity of the salesperson are taken into consideration. A comprehensive recommendation value is calculated based on a hybrid recommendation algorithm to match the customer with the best salesperson, thereby improving the quality of the insurance service process and customer satisfaction, which in turn helps to improve customer conversion rates and sales efficiency.

[0111] In combination with the insurance salesperson recommendation method based on the hybrid recommendation algorithm described in the above embodiments, the insurance salesperson recommendation system based on the hybrid recommendation algorithm is described below through an embodiment.

[0112] like Figure 3 As shown, an embodiment of the present invention provides an insurance salesperson recommendation system based on a hybrid recommendation algorithm, including:

[0113] Salesperson recommendation platform 301, sales terminal device 302, client device 303 and insurance company business platform 304;

[0114] A data link is established between the salesperson recommendation platform 301 and the sales terminal device 302, the client device 303 and the insurance company business platform 304;

[0115] The client device 303 is used to generate an insurance business request according to the insurance business demand when the target customer initiates an insurance business demand through the client device 303, and send the insurance business request to the salesperson recommendation platform 301;

[0116] The salesperson recommendation platform 301 is used to obtain the salesperson information of all salespersons from the insurance company business platform 304 according to the insurance business request, and calculate the individual scoring value of each salesperson according to the insurance business request and the salesperson information. The individual scoring values ​​include the distance score, the previous relationship score, the business demand matching score, the sales professional score and the customer carrying score;

[0117] The salesperson recommendation platform 301 is also used to calculate the comprehensive recommendation value of each salesperson based on the hybrid recommendation algorithm and the single scoring value;

[0118] The salesperson recommendation platform 301 is also used to select the salesperson with the largest comprehensive recommendation value as the best salesperson, and send business instructions to the sales-end device 302 corresponding to the best salesperson, so that the best salesperson can handle insurance business for the target customer according to the business instructions.

[0119] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0120] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0121] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0122] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0123] The above are merely embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention are included in the scope of the claims of the present invention to be approved.

Claims

1. A method for recommending insurance salespersons based on a hybrid recommendation algorithm, characterized in that: Applied to an insurance salesperson recommendation system based on a hybrid recommendation algorithm, the insurance salesperson recommendation system based on a hybrid recommendation algorithm includes a salesperson recommendation platform, a sales end device, a client device and an insurance company business platform, and the method includes: When the target customer initiates an insurance business demand through the client device, the client device generates an insurance business request according to the insurance business demand, and sends the insurance business request to the salesperson recommendation platform; The salesperson recommendation platform obtains the salesperson information of all salespersons from the insurance company business platform according to the insurance business request, and calculates the individual scoring value of each salesperson according to the insurance business request and the salesperson information, wherein the individual scoring value includes the distance score value, the previous relationship score value, the business demand matching score value, the sales professional score value and the customer carrying score value; The salesperson recommendation platform calculates a comprehensive recommendation value for each salesperson based on a hybrid recommendation algorithm and the individual scoring values; The salesperson recommendation platform selects the salesperson with the largest comprehensive recommendation value as the best salesperson, and sends a business instruction to the sales-end device corresponding to the best salesperson, so that the best salesperson handles insurance business for the target customer according to the business instruction.

2. The insurance salesperson recommendation method based on the hybrid recommendation algorithm according to claim 1, characterized in that: The insurance business request includes the expected insurance type information, the target customer location information and the target customer unique identification. The salesperson recommendation platform obtains the salesperson information of all salespersons from the insurance company business platform according to the insurance business request, and calculates the individual scoring value of each salesperson based on the hybrid recommendation algorithm according to the insurance business request and the salesperson information, including: The salesperson recommendation platform analyzes the insurance business request to obtain expected insurance type information, target customer location information and target customer unique identification; The salesperson recommendation platform obtains the salesperson information of all salespersons from the insurance company business platform; the salesperson information includes the salesperson's unique identification, salesperson's location information, insurance types that can be undertaken, sales performance information, and the number of customers who have not yet issued a policy; The salesperson recommendation platform calculates a distance score value according to the target customer location information and the salesperson location information; The salesperson recommendation platform calculates the past relationship score value according to the salesperson unique identifier and the target customer unique identifier; The salesperson recommendation platform calculates the demand matching score value according to the information of the insurance types that can be undertaken and the information of the expected insurance types; The salesperson recommendation platform calculates the sales professional score value based on the sales performance information; The salesperson recommendation platform calculates the customer carrying score value according to the number of customers who have not placed an order.

3. The insurance salesperson recommendation method based on the hybrid recommendation algorithm according to claim 2, characterized in that: The salesperson recommendation platform calculates a distance score value according to the target customer location information and the salesperson location information, including: The salesperson recommendation platform uses a preset map navigation software to obtain a physical distance value between the target customer location information and the salesperson location information; The salesperson recommendation platform selects the salesperson whose physical distance value is not greater than the preset distance value as a candidate salesperson to be selected, forming a set of candidate salespersons; The salesperson recommendation platform normalizes the physical distance values ​​corresponding to all candidate salespersons to obtain a distance score value for each candidate salesperson. The distance score value ranges from [0,1], where a larger value indicates a longer distance, and a smaller value indicates a longer distance.

4. The insurance salesperson recommendation method based on the hybrid recommendation algorithm according to claim 3 is characterized in that: The salesperson recommendation platform calculates the past relationship score value according to the salesperson unique identifier and the target customer unique identifier, including: The salesperson recommendation platform searches the database of the insurance company business platform according to the target customer unique identifier to determine whether the target customer has an association relationship with a salesperson; If there is no relationship with any salesperson, the previous relationship score of all the salespersons to be selected will be 0 points; If there is an association relationship with the first salesperson, the salesperson recommendation platform determines whether the first salesperson is in the set of salespeople to be selected according to the salesperson unique identifier of the first salesperson; If the target customer is in the set of salespeople to be selected, the salesperson recommendation platform determines whether the relationship between the target customer and the first salesperson is a communication relationship or a business relationship; if it is a communication relationship, the previous relationship score of the first salesperson is calculated as 0.5 points; if it is a business relationship, the previous relationship score of the first salesperson is calculated as 1 point; the previous relationship scores of other salespeople in the set of salespeople to be selected other than the first salesperson are calculated as 0 points; If the first salesperson is not in the set of salespeople to be selected, the salesperson recommendation platform will count the past relationship scores of all salespeople to be selected in the set as 0, and add the first salesperson as a candidate salesperson to the set of salespeople to be selected.

5. The insurance salesperson recommendation method based on the hybrid recommendation algorithm according to claim 4 is characterized in that: After adding the first salesperson as a candidate salesperson to the set of salespersons to be selected, the method further includes: The distance score value corresponding to the candidate salesperson is assigned a value of 0.

6. The insurance salesperson recommendation method based on the hybrid recommendation algorithm according to claim 5, characterized in that: The salesperson recommendation platform calculates the demand matching score value according to the information of the insurance types that can be undertaken and the information of the expected insurance types, including: The salesperson recommendation platform determines whether the insurance type information that can be undertaken by all salespersons in the set of salespersons to be selected includes the expected insurance type information; If the insurance types that can be undertaken by the second salesperson include the expected insurance types, the salesperson recommendation platform will count the demand matching score corresponding to the second salesperson as 1 point; If the insurance types that the third salesperson can undertake do not include the expected insurance types, the salesperson recommendation platform extracts insurance type keywords related to the expected insurance types from the sales performance information of the third salesperson, and analyzes whether the salesperson has the ability to undertake the expected insurance types corresponding to the expected insurance types; If the fourth salesperson is able to take the job, the salesperson recommendation platform will assign the demand matching score of the fourth salesperson to 0.5 points; If the fifth salesperson is unable to take the job, the salesperson recommendation platform will count the demand matching score corresponding to the fifth salesperson as 0.

7. The insurance salesperson recommendation method based on the hybrid recommendation algorithm according to claim 5, characterized in that: The salesperson recommendation platform calculates the sales professional score value according to the sales performance information, including: The salesperson recommendation platform parses the sales performance information to obtain job grade information, cumulative sales data, claim amount and abnormal operation behavior, and obtains the sales professional score values ​​of all salespeople in the set of candidate salespeople according to the preset sales professional ability scoring rules; the sales professional score value is a normalized value, and the value range of the sales professional score value is [0,1]. The larger the value, the stronger the professional ability, and the smaller the value, the lower the professional ability.

8. The insurance salesperson recommendation method based on a hybrid recommendation algorithm according to claim 5, characterized in that: The insurance service request also includes a service requirement time period. The salesperson recommendation platform calculates the customer carrying score value according to the number of customers who have not placed an order, including: The salesperson recommendation platform determines whether the number of customers who have not placed an order for all salespersons in the salesperson set to be selected is less than a customer carrying capacity threshold; If there is a salesperson whose customer carrying capacity is not less than the customer carrying capacity threshold, the salesperson recommendation platform will count the customer carrying capacity score of the salesperson corresponding to the customer carrying capacity not less than the customer carrying capacity threshold as 0 points; If there is a salesperson whose capacity is less than the customer carrying capacity threshold, the salesperson recommendation platform will use the salesperson whose capacity is less than the customer carrying capacity threshold as the sixth salesperson; The salesperson recommendation platform obtains the estimated order placement time period of the sixth salesperson corresponding to the number of customers who have not placed an order; The salesperson recommendation platform determines whether the business demand time period conflicts with the expected order placement time period; The salesperson recommendation platform counts the customer carrying score of the salesperson whose business demand time period conflicts with the expected order issuance time period among the sixth salesperson as 0 points; The salesperson recommendation platform counts the customer carrying score of the sixth salesperson whose business demand time period does not conflict with the expected order issuance time period as 1 point.

9. The insurance salesperson recommendation method based on a hybrid recommendation algorithm according to any one of claims 1 to 8, characterized in that: The salesperson recommendation platform calculates the comprehensive recommendation value of each salesperson based on the hybrid recommendation algorithm and the single scoring value, including: The salesperson recommendation platform inputs the distance score, the past relationship score, the business demand matching score, the sales professional score and the customer carrying score into a salesperson recommendation model based on a hybrid recommendation algorithm; The expression of the salesperson recommendation model based on the hybrid recommendation algorithm is: S n =W1S 1n +W2S 2n +W3S 3n +W4S 4n +W5S 5n , where n represents the nth salesperson, n is a positive integer greater than 0, and S 1n is the distance score value of the nth salesperson, the S 2n is the previous relationship score of the nth salesperson, the S 3n is the business demand matching score of the nth salesperson, where S 4n is the sales professional score of the nth salesperson, the S 5n is the customer load score value of the nth salesperson, wherein W1, W2, W3, W4 and W5 are preset weight values; The salesperson recommendation model outputs the comprehensive recommendation value S of the nth salesperson n .

10. An insurance salesperson recommendation system based on a hybrid recommendation algorithm, characterized in that: include: Salesperson recommendation platform, sales-end equipment, client equipment and insurance company business platform; A data link is established between the salesperson recommendation platform and the sales terminal device, the client device and the insurance company business platform; The client device is used to generate an insurance business request according to the insurance business demand when the target customer initiates an insurance business demand through the client device, and send the insurance business request to the salesperson recommendation platform; The salesperson recommendation platform is used to obtain the salesperson information of all salespersons from the insurance company business platform according to the insurance business request, and calculate the individual scoring value of each salesperson according to the insurance business request and the salesperson information, wherein the individual scoring value includes the distance score value, the previous relationship score value, the business demand matching score value, the sales professional score value and the customer carrying score value; The salesperson recommendation platform is further used to calculate the comprehensive recommendation value of each salesperson based on the hybrid recommendation algorithm and the single scoring value; The salesperson recommendation platform is also used to select the salesperson with the largest comprehensive recommendation value as the best salesperson, and send a business instruction to the sales-end device corresponding to the best salesperson, so that the best salesperson can handle insurance business for the target customer according to the business instruction.

Images