Material pushing method and device based on customer intention, equipment and medium

Abstract

The application discloses a material pushing method and device based on customer intention, equipment and medium. The material pushing method comprises the following steps: obtaining the reply content of the customer to the first round of material; identifying the customer intention based on the reply content; matching the strategy scene according to the customer intention; if the matching is successful, determining whether to send the next round of material to the customer according to the first round of material and the matched strategy scene. The application can identify the customer intention based on the reply content of the customer after the financial manager sends the material, match the strategy scene according to the customer intention, and if the matching is successful, determine whether to send the next round of material to the customer according to the sent material and the matched strategy scene, so as to realize the automatic sending of the next round of material to the customer and improve the material pushing efficiency. In addition, the application can not only identify the customer intention, but also match the strategy scene according to the customer intention, so as to improve the accuracy of the sent next round of material.

Description

Technical Field

[0001] This application relates to the field of financial materials, and more particularly to a method, apparatus, computer equipment, and readable storage medium for delivering materials based on customer intent. Background Technology

[0002] When it comes to customer management, the first thing that comes to mind is chatting with customers, sending them suitable materials, and guiding them to purchase products. Tencent's Enterprise WeChat app is a common customer outreach tool among chat applications, and it has also become an interactive platform between bank wealth management managers and customers. Wealth management managers select specific materials through the bank's tools and manually send them to customers. This process involves the wealth management manager manually identifying customer preferences and manually searching the material library for suitable materials to push. Since a wealth management manager may be managing several, dozens, or even hundreds of customers, this method is very inefficient for the wealth management manager and provides a poor customer experience. Summary of the Invention

[0003] Therefore, it is necessary to propose a material delivery method, device, computer equipment, and readable storage medium based on customer intentions to address the above problems, which can automatically push the next round of materials according to the strategy scenario that matches the customer intentions.

[0004] In a first aspect, this application provides a method for pushing materials based on customer intent, the method comprising:

[0005] Obtain the customer's response to the first round of materials;

[0006] Identify customer intent based on the content of the response;

[0007] Match the target strategy scenario based on the customer's intent;

[0008] If a match is successful, determine whether to send the next round of materials to the customer based on the materials from the first round and the matched strategy scenario.

[0009] In some embodiments, identifying customer intent based on the response content includes:

[0010] The response content is input into the first recognition model to obtain customer intent, which is random.

[0011] In some embodiments, identifying customer intent based on the response content includes:

[0012] Obtain the intention template corresponding to the strategy scenario;

[0013] The response content and the intention template are input into the second recognition model to obtain customer intentions, which are then matched with the intention template.

[0014] In some embodiments, identifying customer intent based on the response content includes:

[0015] Obtain the intention template corresponding to the strategy scenario;

[0016] The response content and the intention template are input into the second recognition model to obtain customer intentions, which are then matched with the intention template.

[0017] In some embodiments, the strategy scenario of matching hits based on the customer intent includes:

[0018] Obtain the policy scenarios issued by the policy system and create a mapping relationship between keywords and policy scenario IDs;

[0019] Match the target strategy scenario ID based on the customer's intended keywords and the mapping relationship.

[0020] In some embodiments, the method further includes:

[0021] Configure the validity period of the mapping relationship;

[0022] If no response from the customer regarding the first round of materials is received within the stated validity period, the process will end.

[0023] In some embodiments, if a match is successful, determining whether to send the next round of materials to the customer based on the first round of materials and the matched strategy scenario includes:

[0024] If a match is successful, obtain the customer ID and financial manager account based on the response content;

[0025] Based on the customer ID and the financial manager account, check if the matched strategy scenario ID exists in the cache.

[0026] If the matched policy scenario ID exists in the cache, then the matched policy scenario ID and the customer ID are sent to the policy system;

[0027] When the strategy system receives the matched strategy scenario ID and the customer ID, it issues the next round of materials.

[0028] In some embodiments, the validity period of the cache is consistent with the validity period of the mapping relationship.

[0029] Secondly, this application also provides a material delivery device based on customer intent, the device comprising:

[0030] The acquisition module is used to acquire the customer's response to the first round of materials;

[0031] The identification module is used to identify customer intent based on the content of the response;

[0032] The matching module is used to match the target strategy scenario based on the customer's intention;

[0033] The sending module is used to determine whether to send the next round of materials to the customer if a match is successful, based on the materials in the first round and the hit strategy scenario.

[0034] Thirdly, embodiments of this application also provide a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor causes the processor to perform the steps of the material delivery method based on customer intention as described in any of the first aspects.

[0035] Fourthly, embodiments of this application also provide a computer-readable storage medium storing a computer program, which, when executed by a processor, causes the processor to perform the steps of the material delivery method based on customer intention as described in any of the first aspects.

[0036] The embodiments of this application have at least the following beneficial effects:

[0037] This application embodiment obtains the customer's response to the first round of materials; identifies the customer's intention based on the response content; matches the customer's intention with a relevant strategy scenario; and if the match is successful, determines whether to send the next round of materials to the customer based on the first round of materials and the matched strategy scenario. This application embodiment can automatically send the next round of materials to the customer after the financial manager sends materials, based on the customer's response content, match the customer's intention with a relevant strategy scenario, and if the match is successful, determine whether to send the next round of materials to the customer based on the materials sent by the financial manager and the matched strategy scenario. This improves the efficiency of material delivery and thus enhances the customer experience. Furthermore, this application can not only identify customer intention but also match the matched strategy scenario, thereby improving the accuracy of the next round of materials sent, ensuring that the sent materials are the materials the customer needs. Attached Figure Description

[0038] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0039] in:

[0040] Figure 1This is a flowchart illustrating a material delivery method based on customer intent in one embodiment of this application;

[0041] Figure 2 This is a flowchart illustrating step S130 in one embodiment of this application;

[0042] Figure 3 This is a flowchart illustrating step S140 in one embodiment of this application;

[0043] Figure 4 This is a complete flowchart of a material delivery method based on customer intent in one embodiment of this application;

[0044] Figure 5 This is an internal structural diagram of a computer device in one embodiment of this application. Detailed Implementation

[0045] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0046] When it comes to customer management, the first thing that comes to mind is chatting with customers, sending them suitable materials, and guiding them to purchase products. Tencent's Enterprise WeChat app is a common customer outreach tool among chat applications, and it has also become an interactive platform between bank wealth management managers and customers. Before building an Enterprise WeChat middleware system, wealth management managers manually selected designated materials through bank staff tools and sent them to customers. This process involved wealth management managers manually identifying customer preferences and manually searching the material library for suitable materials to push. Since a wealth management manager may be responsible for several, dozens, or even hundreds of customers, this method is very inefficient for the wealth management manager and provides a poor customer experience. Based on the above problems, an Enterprise WeChat middleware system can be built. The system can create strategies, select customer groups, select distribution teams, select recommended materials, and then push them to the Enterprise WeChat app, and finally to customers.

[0047] However, this approach also presents a problem: how to automatically send the next round of materials after multiple rounds of dialogue. The financial manager sends multiple strategy scenarios to the client, extracts the client's responses, and then matches them using keywords. A match indicates the client has a certain interest. However, this method only indicates a possible interest; it doesn't reveal which specific strategy scenario the client's response corresponds to, thus preventing the sending of the next round of materials.

[0048] Strategy scenarios refer to the context in which specific materials are sent, with each scenario corresponding to a particular material. For example, in an event scenario, a financial manager might send a client materials related to an event, such as, "There's an upcoming birthday event. Please reply 'yes' if you wish to participate, otherwise reply 'no'." When a client replies "yes," the system automatically recognizes the client's intention and sends the next set of materials. Similarly, in an investment scenario, a financial manager might send a client materials related to an investment project, such as, "There are upcoming investment projects. Please reply '1' if you want to learn more, otherwise reply '0'." When a client replies "1," the system automatically recognizes the client's intention and sends the next set of materials.

[0049] When a financial advisor sends clients materials related to both promotional activities and investment strategies simultaneously, the advisor needs to match the client's response to determine which strategy scenario is being used. For example, if the client replies with "1," it matches the investment strategy scenario, and the advisor has already sent an investment strategy message. This successful match indicates the advisor sent the investment strategy message and the client responded, prompting the advisor to send the next set of materials. Therefore, the core issue remains identifying the appropriate strategy scenario. When an advisor sends multiple strategy scenarios to a client, they need to determine which scenario the client's response corresponds to before matching the next strategy scenario. This allows for automated customer journey routing.

[0050] Based on this inventive concept, this application provides a method, apparatus, computer device, and readable storage medium for pushing materials based on customer intent. After a financial manager sends materials, it identifies the customer's intent based on the customer's response, matches the customer's intent with a relevant strategy scenario, and if a match is successful, determines whether to send the next round of materials to the customer based on the materials sent by the financial manager and the matched strategy scenario. This achieves automatic delivery of the next round of materials to the customer, improving material delivery efficiency and thus enhancing customer experience. Furthermore, this application not only identifies customer intent but also matches the matched strategy scenario, thereby improving the accuracy of the next round of materials sent—ensuring that the materials sent are those the customer needs.

[0051] The technical solutions shown in this application will now be described in detail through specific embodiments. It should be noted that the following embodiments may exist independently or in combination with each other; identical or similar content will not be repeated in different embodiments.

[0052] In a first aspect, embodiments of this application provide a method for pushing materials based on customer intentions. Figure 1 This is a flowchart illustrating a material delivery method based on customer intent in one embodiment of this application. Please refer to... Figure 1 In some embodiments, the customer intent-based material delivery method includes:

[0053] S110: Obtain the customer's response to the first round of materials.

[0054] In some embodiments, the material delivery method of this application is mainly implemented based on the Enterprise WeChat platform. As mentioned above, the financial manager and the customer communicate through the Enterprise WeChat platform and the Enterprise WeChat APP. After the financial manager sends the first round of materials to the customer, the Enterprise WeChat platform obtains the customer's response. The first round of materials can be understood as the first round of strategy scenarios, which includes multiple strategy scenarios.

[0055] In some embodiments, materials can be understood as specific wording, including but not limited to text, images, links, etc. For example, in a financial management scenario, the material could be, "There is currently a financial management activity; would you like to participate?"

[0056] S120: Identify customer intent based on response content.

[0057] In some embodiments, after obtaining the customer's reply content, the Enterprise WeChat platform identifies the customer's intention based on the customer's reply content, or sends the customer's reply content to a dedicated algorithm platform to identify the customer's intention.

[0058] In some embodiments, identifying customer intent based on response content can include the following two methods:

[0059] The response content is input into the first recognition model to obtain customer intent, which is random.

[0060] or,

[0061] Obtain the intention template corresponding to the strategy scenario, input the response content and intention template into the second recognition model to obtain customer intention, and match the customer intention with the intention template.

[0062] The first recognition model is typically a small model, such as the BERT model. Small models are robot models trained offline, using high-interaction dialogue content from production environments to train the model, and the result is an intention classification. Different small models can be used for different policy scenarios.

[0063] The customer's response is input into the first recognition model for identification, yielding the customer's intent. Different sub-models produce different customer intents. Therefore, the customer intent obtained using the first recognition model is random. Examples include: "I need to participate," "I don't want to participate," "111," "000," etc.

[0064] The second recognition model is generally a large model, and different strategy scenarios use the same large model, such as the GPT model or other deep learning models. Different strategy scenarios are configured with different intention templates, which contain intention categories. For example, the intention template for strategy scenario 1 contains three intentions: "positive / negative / other", the intention template for strategy scenario 2 contains two intentions: "participate / not participate", and the intention template for strategy scenario 3 contains two intentions: "1 / 0".

[0065] The customer's response and intent template are input into a second recognition model for identification, resulting in customer intent. This customer intent is then matched against the intent template. Specifically, customer intent can only be one of "positive / negative / other," or one of "participate / not participate," or one of "1 / 0"—it is not random. Using a large model for identification facilitates subsequent matching strategy scenarios.

[0066] Optionally, the first recognition model and the second recognition model can be selected through the configuration page to choose which model to use and the corresponding intention template.

[0067] S130: Match the target strategy scenario based on the customer's intention.

[0068] Figure 2 This is a flowchart illustrating step S130 in one embodiment of this application. Please refer to... Figure 2 In some embodiments, matching strategies based on customer intent include:

[0069] S131: Obtain the policy scenarios issued by the policy system and create a mapping relationship between keywords and policy scenario IDs.

[0070] S132: Match the target strategy scenario ID based on the keywords and mapping relationships of the customer's intent.

[0071] Strategy scenarios are typically created by the strategy system, which then distributes them to the Enterprise WeChat platform. Upon receiving the strategy scenarios, the Enterprise WeChat platform creates a mapping between keywords and the strategy scenarios. Specifically, this can involve creating a unique ID for each strategy scenario and then establishing a mapping between keywords and the strategy scenario ID. For example, the keywords "participate" and "do not participate" are mapped to activity scenario IDs, and the keywords "1" and "0" are mapped to financial management scenario IDs.

[0072] Extracting keywords from customer intent. Specifically, if customer intent is identified through the first recognition model, since the identified customer intent is random, it is necessary to extract keywords from it. For example, the keywords extracted from "I need to participate," "I don't want to participate," "111," and "000" would be "participate," "not participate," "1," and "0." If customer intent is identified through the second recognition model, it is generally not necessary to extract keywords; the customer intent output by the second recognition model can be directly used as the keywords.

[0073] Based on the extracted keyword query mapping relationship, the strategy scenario IDs matching the keywords are determined. The ID for the activity scenario is set as X1Y2Z3, and the ID for the financial management scenario is set as O1P2Q3. If the extracted keyword is "participate" or "do not participate," the matched strategy scenario ID is the activity scenario ID X1Y2Z3, resulting in a successful match. If the extracted keyword is "1" or "0," the matched strategy scenario ID is the financial management scenario ID O1P2Q3, also resulting in a successful match. If the extracted keyword is an ambiguous word such as "okay" or "learn more," the match fails.

[0074] S140: If the match is successful, determine whether to send the next round of materials to the customer based on the materials in the first round and the matching strategy scenario.

[0075] If the match is successful in step S130, then it is determined whether to send the next round of materials to the customer based on the materials in the first round and the matching strategy scenario. If the match fails in step S130, no action is taken.

[0076] Furthermore, when sending the next round of materials to customers, appropriate materials can be sent based on the identified customer intentions. For example:

[0077] If the extracted keyword is "participate", the next round of materials sent can be activity details, including the activity address, time, precautions, etc.

[0078] If the extracted keyword is "not participating", what is the reason for sending the next round of materials if it is "not participating"?

[0079] If the extracted keyword is "1", the next round of materials sent can be the details of the financial project, including a detailed introduction of the financial project.

[0080] If the extracted keyword is "0", then the next round of materials sent can be those that do not need to be understood. What is the reason for this?

[0081] In summary, the embodiments of this application can identify customer intentions based on customer responses after the financial manager sends materials, match the target strategy scenario according to the customer's intentions, and if the match is successful, determine whether to send the next round of materials to the customer based on the materials sent by the financial manager and the target strategy scenario, thereby realizing automatic sending of the next round of materials to the customer, improving material push efficiency, and thus improving customer experience.

[0082] In addition, this application can not only identify customer intentions, but also match which strategy scenario is hit based on the customer's intentions, thereby improving the accuracy of the next round of materials sent, that is, the materials sent are the materials that the customer needs.

[0083] In some embodiments, after creating the mapping relationship between keywords and strategy scenario IDs, the validity period of this mapping relationship can also be configured. If no response from the customer regarding the first round of materials is received within the validity period, the process ends, the mapping relationship becomes invalid, and the mapping relationship is deleted to save space. Even if the customer responds after the validity period expires, the next process will not be triggered.

[0084] Figure 3 This is a flowchart illustrating step S140 in one embodiment of this application. Please refer to... Figure 3 In some embodiments, if a match is successful, it is determined whether to send the next round of materials to the customer based on the materials from the first round and the matching strategy scenario, including:

[0085] S141: If a match is successful, obtain the customer ID and financial manager account based on the reply content.

[0086] Specifically, the WeChat Work platform has a listener that listens to the first round of materials sent by the financial manager to the client. At this time, the financial manager's account, client ID, and the strategy scenario ID corresponding to the sent materials are stored in the cache. The storage format can be a phrase composed of [financial manager account, client ID, strategy scenario ID].

[0087] When the WeChat Work platform obtains a customer's reply, it can simultaneously obtain the customer's ID and the account of the financial manager who sent the strategy scenario to the customer, based on the dialogue relationship.

[0088] S142: Check if the matching strategy scenario ID exists in the cache based on the customer ID and financial manager account.

[0089] The enterprise platform queries the cached phrases based on the obtained customer ID and financial manager account, and checks whether the phrases contain a matching strategy scenario ID.

[0090] S143: If the policy scenario ID that is matched exists in the cache, then send the matched policy scenario ID and customer ID to the policy system.

[0091] When a matching strategy scenario ID is found within a query phrase (meaning the strategy scenario ID in the phrase matches the matching strategy ID), it indicates that the financial manager sent the strategy scenario to the client and the client responded accordingly. The WeChat platform then sends the matching strategy scenario ID and the client ID to the strategy system.

[0092] S144: When the strategy system receives the matched strategy scenario ID and customer ID, it issues the next round of materials.

[0093] Upon receiving the strategy scenario ID and customer ID, the strategy system sends the next round of materials to the Enterprise WeChat platform. The Enterprise WeChat platform then sends these materials to the customer's Enterprise WeChat app.

[0094] In some embodiments, the cache expiration period is consistent with the expiration period of the mapping relationship between keywords and policy scenario IDs. That is, the cache expiration time is set according to the expiration period configured in the mapping relationship, and the cache is automatically cleared when the expiration time expires to prevent the cache space from becoming too full and causing the storage of new phrases to fail.

[0095] The technical solution of this application will be described below with a relatively complete embodiment. Figure 4 This is a schematic diagram illustrating the complete process of a material delivery method based on customer intent in one embodiment of this application. Please refer to... Figure 4 The interaction process between the WeChat Work App, WeChat Work middleware platform, algorithm platform, and strategy system includes the following steps:

[0096] Step 1: The strategy system creates a strategy scenario and distributes the strategy scenario to the Enterprise WeChat middle platform.

[0097] Step 2: Configure the mapping relationship between strategy scenario IDs and keywords in the Enterprise WeChat middleware platform. Optionally, the validity period of this mapping relationship can also be configured.

[0098] Step 3: The financial manager sends multiple first-round materials (i.e., multiple first-strategy scenarios) to the client's WeChat APP through the WeChat platform.

[0099] Step 4: The WeChat Work platform obtains the customer's response to the first round of materials and sends the response to the algorithm platform for identification to determine the customer's intent. Simultaneously, the WeChat Work platform stores the customer ID, financial manager account, and the sent strategy scenario ID in a cache.

[0100] Step 5: The algorithm platform returns the keywords of customer intent to the Enterprise WeChat platform.

[0101] Step 6: The Enterprise WeChat platform queries the mapping relationship based on keywords and matches the hit strategy scenario ID.

[0102] Step 7: The enterprise WeChat platform checks the cache for the strategy scenario ID matched in Step 6 based on the customer ID and financial manager account. Optionally, the cache's validity period can be configured to match the validity period of the mapping relationship.

[0103] Step 8: If the policy scenario ID matched in step 6 is found in the cache, the Enterprise WeChat platform will send the retrieved policy scenario ID and customer ID back to the policy system.

[0104] Step 9: The strategy system sends the next round of materials (i.e., the next round of strategy scenarios) to the Enterprise WeChat platform.

[0105] Step 10: The financial manager sends the next round of materials to the client's WeChat Work APP through the WeChat Work platform.

[0106] It should be noted that the implementation process of each of the above steps has been described in detail in the above embodiments, and will not be repeated here.

[0107] In some embodiments, the algorithm platform can be a standalone algorithm platform or it can be integrated into the enterprise micro-platform or strategy system.

[0108] In some embodiments, the Enterprise WeChat platform, algorithm platform, and strategy system can be integrated into a single system that can communicate with the customer's Enterprise WeChat APP to achieve the aforementioned steps.

[0109] Through practice, it has been found that the customer intent-based material push method proposed in this application, by calling the algorithm platform through the conversation content for intent recognition, queries the mapping relationship of the obtained intent keywords, matches the hit strategy scenario ID, and checks whether the above-mentioned hit strategy scenario ID exists in the cache through the customer ID and the sender account (i.e., the financial manager account). If it exists, it is sent back to the strategy system. The use of intelligent recognition of customer intent and intelligent parsing of strategy information has played a significant role in accurate multi-turn dialogue and automatic strategy delivery, improving the efficiency of customer outreach and management, and improving the efficiency of financial management operations, which is in line with the goal of digital operation.

[0110] Secondly, embodiments of this application also provide a material delivery device based on customer intent, comprising:

[0111] The acquisition module is used to acquire the customer's response to the first round of materials;

[0112] The identification module is used to identify customer intent based on the content of the response;

[0113] The matching module is used to match relevant strategy scenarios based on customer intent.

[0114] The sending module is used to determine whether to send the next round of materials to the customer if a match is successful, based on the materials in the first round and the matching strategy scenario.

[0115] Thirdly, embodiments of this application also provide a computer device, including a memory and a processor. The memory stores a computer program, and when the computer program is executed by the processor, the processor performs the steps of the material delivery method based on customer intention as described in any of the first aspects.

[0116] Figure 5 This is an internal structural diagram of a computer device according to one embodiment of this application. Please refer to... Figure 5 In some embodiments, the computer device includes a processor, memory, and a network interface connected via a system bus. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system and may also store a computer program that, when executed by the processor, causes the processor to perform the steps of any of the above-mentioned material delivery methods based on customer intent. The internal memory may also store a computer program that, when executed by the processor, causes the processor to perform the steps of any of the above-mentioned material delivery methods based on customer intent. Those skilled in the art will understand that... Figure 5 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0117] Fourthly, embodiments of this application also provide a computer-readable storage medium storing a computer program, which, when executed by a processor, causes the processor to perform the steps of the material delivery method based on customer intention as described in any of the first aspects.

[0118] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, the program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and RAMbus dynamic RAM (RDRAM), etc.

[0119] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0120] The above embodiments merely illustrate several implementation methods of this application, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this patent application should be determined by the appended claims.

Claims

1. A method for pushing materials based on customer intent, characterized in that, The method includes: Obtain customer responses to the first round of materials, which include multiple strategy scenarios; Identify customer intent based on the content of the response; Match the target strategy scenario based on the customer's intent; If a match is successful, determine whether to send the next round of materials to the customer based on the materials from the first round and the matched strategy scenario. The strategy scenario of matching based on customer intent includes: Obtain the policy scenarios issued by the policy system and create a mapping relationship between keywords and policy scenario IDs; Match the target strategy scenario ID based on the customer's intended keywords and the mapping relationship; If a match is successful, determine whether to send the next round of materials to the customer based on the materials from the first round and the matched strategy scenario, including: If a match is successful, obtain the customer ID and the financial manager account that sent the strategy scenario to the customer based on the reply content; Based on the customer ID and the financial manager account, check if the matched strategy scenario ID exists in the cache. If the matched policy scenario ID exists in the cache, then the matched policy scenario ID and the customer ID are sent to the policy system; When the strategy system receives the matched strategy scenario ID and the customer ID, it issues the next round of materials.

2. The material delivery method based on customer intent according to claim 1, characterized in that, The process of identifying customer intent based on the response content includes: The response content is input into the first recognition model to obtain customer intent, which is random.

3. The material delivery method based on customer intent according to claim 1, characterized in that, The process of identifying customer intent based on the response content includes: Obtain the intention template corresponding to the strategy scenario; The response content and the intention template are input into the second recognition model to obtain customer intentions, which are then matched with the intention template.

4. The material delivery method based on customer intent according to claim 1, characterized in that, The method further includes: Configure the validity period of the mapping relationship; If no response from the customer regarding the first round of materials is received within the stated validity period, the process will end.

5. The material delivery method based on customer intent according to claim 1, characterized in that, The validity period of the cache is consistent with the validity period of the mapping relationship.

6. A material delivery device based on customer intent, characterized in that, The device includes: The acquisition module is used to acquire the customer's response to the first round of materials, which includes multiple strategy scenarios; The identification module is used to identify customer intent based on the content of the response; The matching module is used to match the target strategy scenario based on the customer's intention; The sending module is used to determine whether to send the next round of materials to the customer if a match is successful, based on the first round of materials and the hit strategy scenario. The strategy scenario of matching based on customer intent includes: Obtain the policy scenarios issued by the policy system and create a mapping relationship between keywords and policy scenario IDs; Match the target strategy scenario ID based on the customer's intended keywords and the mapping relationship; If a match is successful, determine whether to send the next round of materials to the customer based on the materials from the first round and the matched strategy scenario, including: If a match is successful, obtain the customer ID and the financial manager account that sent the strategy scenario to the customer based on the reply content; Based on the customer ID and the financial manager account, check if the matched strategy scenario ID exists in the cache. If the matched policy scenario ID exists in the cache, then the matched policy scenario ID and the customer ID are sent to the policy system; When the strategy system receives the matched strategy scenario ID and the customer ID, it issues the next round of materials.

7. A computer device, characterized in that, It includes a memory and a processor, the memory storing a computer program that, when executed by the processor, causes the processor to perform the steps of the customer intention-based material delivery method as described in any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that, The device contains a computer program that, when executed by a processor, causes the processor to perform the steps of the material delivery method based on customer intent as described in any one of claims 1 to 5.

Images