A customer information report-based customer attribution management method and system

Abstract

A customer attribution management method based on customer information reporting, comprising the following steps: accepting the uploaded reporting information of the sales client, wherein the reporting information comprises customer information and sales information; generating an identifier based on the customer information, and generating storage information on the blockchain based on the reporting information and the identifier; querying whether the number of the identifier on the blockchain is 0; if the number is 0, the storage information is chained, and the customer of this reporting information is attributed to the sales in the sales information; through the method, automatic customer division can be realized, corresponding storage information is generated through reporting information and saved in the blockchain, the authenticity of customer division basis is ensured, and tampering is avoided, thereby providing a real data basis for automatic customer division.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and in particular to a customer attribution management method and system based on customer information reporting. Background Technology

[0002] With the continuous development of communication technology, the efficiency of information transmission is becoming increasingly higher. This increased efficiency has impacted traditional sales methods. Companies are no longer limited to using their own sales force; they often leverage existing channels to outsource sales to third parties, relying on their channel capabilities to expand their sales business.

[0003] Conversely, when a customer comes to the company to purchase goods, a certain commission or bonus is required for the salesperson or channel agent. However, in some cases, such as real estate sales or car sales, it is difficult to distinguish whether the customer was contacted by the company's sales staff or by a channel agent contacting the customer through another platform. This can easily lead to unclear customer attribution, resulting in customer disputes and the phenomenon of customer interception. Summary of the Invention

[0004] To address the aforementioned shortcomings, the present invention aims to propose a customer attribution management method and system based on customer information reporting, thereby resolving the issue of unclear customer attribution and customer interception.

[0005] To achieve this objective, the present invention adopts the following technical solution: a customer attribution management method based on customer information reporting, comprising the following steps:

[0006] Step S1: Accept the reporting information uploaded by the sales client, wherein the reporting information includes customer information and sales information;

[0007] Step S2: Generate an identifier based on the customer information, and generate the user's blockchain storage information based on the reported information and the identifier.

[0008] Step S3: Check if the number of the identifier on the blockchain is 0;

[0009] If the quantity is 0, the storage information will be uploaded to the blockchain, and the customer whose information was reported this time will be included in the sales information.

[0010] If the quantity is greater than 0, then obtain the storage information with the same identifier on the blockchain, and determine whether the sales information in all the storage information is consistent. If they are consistent, then classify the customer of this report information into the sales information.

[0011] If there is a discrepancy, the customers for this reporting information will be categorized based on the on-chain time and / or region of the different storage information.

[0012] Preferably, the specific steps for querying whether the number of the identifier on the blockchain is 1 in step S3 are as follows: connect to the blockchain using the QuikNode API interface;

[0013] Based on the API service documentation, write an API interface activation request based on identifier lookup and send the activation request to the node service center;

[0014] Receive query data from the service center.

[0015] Preferably, the method for classifying customers of the reported information in step S3 based on the on-chain time and / or region of different storage information is as follows:

[0016] The sales location from the sales information stored this time is used as the first region;

[0017] Retrieve the sales region from the previously stored sales information and use it as the second region;

[0018] Determine if the first region and the second region have the same name. If no name is found, classify the customer in this report as a salesperson in the sales information.

[0019] If the same name appears, the on-chain time of the storage information corresponding to the second region with the same name is obtained as the first time, and the upload time of the information reported this time is used as the second time.

[0020] Determine if the time difference between the first and second times is greater than a time threshold. If it is, classify the customer whose information was reported this time as a salesperson in the sales information.

[0021] If the value is less than the threshold, a communication data request is sent to the sales client, and the customers whose information was reported are categorized based on the communication data request.

[0022] Preferably, the steps for classifying customers based on the communication data request for this report are as follows:

[0023] Generate the first program interface, and obtain the text content from the specified communication data through the first program;

[0024] Natural language processing technology is used to perform semantic analysis on text content. Based on the results of the semantic analysis, the text data is divided and key information is extracted from it.

[0025] Generate a second program interface, and use the second program to obtain the voice content in the specified communication data;

[0026] The speech content is denoised and filtered to obtain preprocessed speech. The preprocessed speech is input into a deep learning model, and the extracted speech features are mapped to acoustic units of phonemes or subphonemes. Based on the acoustic unit sequence and the natural language model, a word sequence is generated. The identified word sequence is normalized, and semantic analysis is performed on the normalized word sequence using natural language processing technology to obtain key information.

[0027] Generate a third-party program interface to retrieve image content from specified communication data through the third program;

[0028] Call the OCR interface to extract text content from the image using OCR technology, and then perform semantic analysis on the text content using natural language processing technology to obtain key information;

[0029] Store key information within information groups;

[0030] The AI ​​language model interface is called to output each information group sequentially to the AI ​​language model. The AI ​​language model reorganizes the content of the information groups through semantic analysis to obtain the reorganized sentences.

[0031] By combining the reorganized statements with customer information sources, complete business opportunity information can be obtained;

[0032] Based on the number of business opportunity information, the customers who submitted the information this time were assigned to the salesperson with the most business opportunity information.

[0033] Preferably, the following steps should be performed before storing key information in information groups:

[0034] Used to clean critical information and remove redundant, erroneous, or irrelevant data;

[0035] The cleaned key information is then mapped to the defined information groups.

[0036] A customer attribution management system based on customer information reporting, and a customer attribution management method based on customer information reporting, including an uploading module, a processing module, and an on-chain module;

[0037] The upload module is used to accept the reporting information uploaded by the sales client, wherein the reporting information includes customer information and sales information;

[0038] The processing module is used to generate an identifier based on the customer information, and to generate the user's storage information on the blockchain based on the reporting information and the identifier.

[0039] The on-chain module is used to query whether the number of the identifier on the blockchain is 0;

[0040] If the quantity is 0, the storage information will be uploaded to the blockchain, and the customer whose information was reported this time will be included in the sales information.

[0041] If the quantity is greater than 0, then obtain the storage information with the same identifier on the blockchain, and determine whether the sales information in all the storage information is consistent. If they are consistent, then classify the customer of this report information into the sales information.

[0042] If there is a discrepancy, the customers for this reporting information will be categorized based on the on-chain time and / or region of the different storage information.

[0043] Preferably, the on-chain module includes a query sub-module;

[0044] The query submodule is used to connect to the blockchain using the QuikNode API interface;

[0045] Based on the API service documentation, write an API interface activation request based on identifier lookup and send the activation request to the node service center;

[0046] Receive query data from the service center.

[0047] Preferably, the on-chain module further includes a customer segmentation sub-module;

[0048] The customer segmentation submodule is used to obtain the sales location in the sales information of the stored information, and use it as the first region;

[0049] Retrieve the sales region from the previously stored sales information and use it as the second region;

[0050] Determine if the first region and the second region have the same name. If no name is found, classify the customer in this report as a salesperson in the sales information.

[0051] If the same name appears, the on-chain time of the storage information corresponding to the second region with the same name is obtained as the first time, and the upload time of the information reported this time is used as the second time.

[0052] Determine if the time difference between the first and second times is greater than a time threshold. If it is, classify the customer whose information was reported this time as a salesperson in the sales information.

[0053] If the value is less than the threshold, a communication data request is sent to the sales client, and the customers whose information was reported are categorized based on the communication data request.

[0054] Preferably, the customer segmentation submodule includes a text information processing unit, a voice information processing unit, an image information processing unit, a recombination unit, and an allocation unit;

[0055] The text information processing unit is used to generate a first program interface and obtain the text content in the specified communication data through the first program.

[0056] Natural language processing technology is used to perform semantic analysis on text content. Based on the results of the semantic analysis, the text data is divided and key information is extracted from it.

[0057] The voice information processing unit is used to generate a second program interface, and to obtain the voice content in the specified communication data through the second program.

[0058] The speech content is denoised and filtered to obtain preprocessed speech. The preprocessed speech is input into a deep learning model, and the extracted speech features are mapped to acoustic units of phonemes or subphonemes. Based on the acoustic unit sequence and the natural language model, a word sequence is generated. The identified word sequence is normalized, and semantic analysis is performed on the normalized word sequence using natural language processing technology to obtain key information.

[0059] The image information unit is used to generate a third program interface, through which the third program obtains the image content in the specified communication data;

[0060] Call the OCR interface to extract text content from the image using OCR technology, and then perform semantic analysis on the text content using natural language processing technology to obtain key information;

[0061] The reorganization unit is used to store key information within information groups;

[0062] The AI ​​language model interface is called to output each information group sequentially to the AI ​​language model. The AI ​​language model reorganizes the content of the information groups through semantic analysis to obtain the reorganized sentences.

[0063] By combining the reorganized statements with customer information sources, complete business opportunity information can be obtained;

[0064] The allocation unit is used to assign customers with the most reported business opportunities to the salesperson with the most business opportunities based on the number of business opportunities reported.

[0065] Preferably, the recombination unit further includes a preprocessing subunit;

[0066] The preprocessing subunit is used to clean key information and remove redundant, erroneous, or irrelevant data;

[0067] The cleaned key information is then mapped to the defined information groups.

[0068] One of the above technical solutions has the following advantages or beneficial effects: the present invention can realize the automatic classification of customers, and generate corresponding storage information through the reporting information and store it in the blockchain, which ensures the authenticity of the basis for customer classification and prevents tampering, thus providing a real data foundation for the automatic classification of customers. Attached Figure Description

[0069] Figure 1 This is a flowchart of one embodiment of the method of the present invention.

[0070] Figure 2 This is a schematic diagram of the structure of one embodiment of the system of the present invention. Detailed Implementation

[0071] Embodiments of the present invention are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and should not be construed as limiting the present invention.

[0072] In the description of embodiments of the present invention, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the stated features. In the description of embodiments of the present invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0073] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this invention, unless otherwise stated, "a plurality of" means two or more. Those skilled in the art will understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0074] like Figures 1-2 As shown, a customer attribution management method based on customer information reporting includes the following steps:

[0075] Step S1: Accept the reporting information uploaded by the sales client, wherein the reporting information includes customer information and sales information;

[0076] Step S2: Generate an identifier based on the customer information, and generate the user's blockchain storage information based on the reported information and the identifier.

[0077] Step S3: Check if the number of the identifier on the blockchain is 0;

[0078] If the quantity is 0, the storage information will be uploaded to the blockchain, and the customer whose information was reported this time will be included in the sales information.

[0079] If the quantity is greater than 0, then obtain the storage information with the same identifier on the blockchain, and determine whether the sales information in all the storage information is consistent. If they are consistent, then classify the customer of this report information into the sales information.

[0080] If there is a discrepancy, the customers for this reporting information will be categorized based on the on-chain time and / or region of the different storage information.

[0081] In order to better segment customers and ensure that each salesperson receives their commission fairly, this invention includes a sales client. After receiving a customer, the salesperson needs to report the customer information on the sales client and send the report information. The customer information includes the customer's name, phone number, and other customer-related information, while the sales information includes the salesperson's name, sales region, phone number, and other sales-related information.

[0082] After uploading the registration information, the system will convert it into corresponding storage information. This includes generating a unique identifier associated with the customer information, using methods such as binary encoding, hash encoding, and character encoding. The customer's social security code or ID card number can be incorporated during identifier encoding to ensure uniqueness. The generated storage information is not immediately uploaded to the blockchain; it is only uploaded after customer ownership is determined. In subsequent communication regarding customer commission disputes, the corresponding storage information can be retrieved from the blockchain to establish customer classification. Specifically, before uploading the storage information to the blockchain, the identifier is searched to determine if a matching identifier exists. If no matching identifier exists, it means no salesperson has previously contacted the customer, and the customer can be classified as the corresponding salesperson in this registration information. The storage information is then uploaded to the blockchain. Conversely, if a matching identifier exists, it means a salesperson has already contacted the customer. In this case, the customer in this registration information needs to be classified based on the upload time and / or region of the stored information. If it is necessary to reclassify customers, the sales information in the current storage information needs to be modified to match the previous sales information before uploading it to the blockchain.

[0083] The following is an example for explanation. In this instance, the sales client of Salesperson A uploads the reporting information. This reporting information includes Customer A's customer information and Salesperson A's sales information. An identifier A is generated from Customer A's customer information. The reporting information and identifier A together generate storage information A on the blockchain. Then, identifier A is used to search the blockchain. If no other storage information is found, storage information A is uploaded to the blockchain, and Customer A is assigned to Salesperson A. If other storage information B is found, it is determined whether the sales information in storage information B matches the sales information in storage information A. If they match, it means the same salesperson is still assigned to the same customer, and Customer A is still assigned to Salesperson A, and storage information A is uploaded to the blockchain again. If they do not match, the customer in this reporting information needs to be reassigned based on the upload time and / or region of the different storage information. It is worth noting that if Customer A is not assigned to Salesperson A in this reporting information upload, storage information A will not be uploaded to the blockchain, and a customer assignment failure message will be returned to the sales client. Salesperson A can use this message to contact the regional manager for appeals, etc.

[0084] This invention enables automatic customer segmentation and generates corresponding storage information through reporting information, which is then stored in the blockchain. This ensures the authenticity of the customer segmentation basis and prevents tampering, providing a genuine data foundation for automatic customer segmentation.

[0085] Preferably, the specific steps for querying whether the number of the identifier on the blockchain is 1 in step S3 are as follows: connect to the blockchain using the QuikNode API interface;

[0086] Based on the API service documentation, write an API interface activation request based on identifier lookup and send the activation request to the node service center;

[0087] Receive query data from the service center.

[0088] For large companies, their business locations are numerous and distributed across different regions. QuikNode supports multiple geographical areas, which helps optimize network call times and improve the speed and performance of DApps. Users can choose the appropriate region to deploy nodes according to their needs. This allows sales personnel from different regions to upload reporting information, improving the ease of on-chain processing.

[0089] Preferably, the method for classifying customers of the reported information in step S3 based on the on-chain time and / or region of different storage information is as follows:

[0090] The sales location from the sales information stored this time is used as the first region;

[0091] Retrieve the sales region from the previously stored sales information and use it as the second region;

[0092] Determine if the first region and the second region have the same name. If no name is found, classify the customer in this report as a salesperson in the sales information.

[0093] If the same name appears, the on-chain time of the storage information corresponding to the second region with the same name is obtained as the first time, and the upload time of the information reported this time is used as the second time.

[0094] Determine if the time difference between the first and second times is greater than a time threshold. If it is, classify the customer whose information was reported this time as a salesperson in the sales information.

[0095] If the value is less than the threshold, a communication data request is sent to the sales client, and the customers whose information was reported are categorized based on the communication data request.

[0096] Because the blockchain is connected via the QuikNode API interface through the client, regional information can be uploaded during the connection, or the sales region information can be added to the sales information. This allows for the rapid acquisition of data from the first and second regions. Since customers of the same company type may make purchases in different sales regions across the country, different sales representatives will handle these transactions in different regions. To address this, we check if the first and second regions have the same name. If no name is found, it means the customer is appearing in the first region for the first time, and the customer can be assigned to the sales representative whose information was uploaded in this report. If the first and second regions have the same name, it means the customer is not appearing in that region for the first time. Therefore, we need to check if the time difference between the upload time of this report and the first report meets a time threshold greater than six months. If the time difference is greater than the time threshold, it means the customer has not contacted the previous sales representative for a long time, and therefore the customer can be assigned to the sales representative whose information was uploaded in this report, and the corresponding storage information is uploaded to the blockchain. If the time difference is less than the time threshold, it means that the customer has been communicating with the previous salesperson in the short term. At this time, it is impossible to directly determine the customer's affiliation. Therefore, a communication data request will be sent to the sales client. After receiving the request, the sales client can select the corresponding customer's communication data to upload. The communication data can be used to determine which salesperson communicated with the customer to what extent, and the customer affiliation can be determined based on the depth of communication.

[0097] Preferably, the steps for classifying customers based on the communication data request for this report are as follows:

[0098] Generate the first program interface, and obtain the text content from the specified communication data through the first program;

[0099] Natural language processing technology is used to perform semantic analysis on text content. Based on the results of the semantic analysis, the text data is divided and key information is extracted from it.

[0100] For text content recognition, existing and mature Natural Language Processing (NLP) techniques can be used to perform semantic analysis on the text content. The analysis results can determine customer transaction intentions, transaction details, etc., thereby segmenting the text data and extracting corresponding key information. For example, if the goal is to extract transaction time, quantity, product type, etc., the text data will be segmented according to these requirements, resulting in different sets of key information.

[0101] Generate a second program interface, and use the second program to obtain the voice content in the specified communication data;

[0102] The speech content is denoised and filtered to obtain preprocessed speech. The preprocessed speech is input into a deep learning model, and the extracted speech features are mapped to acoustic units of phonemes or subphonemes. Based on the acoustic unit sequence and the natural language model, a word sequence is generated. The identified word sequence is normalized, and semantic analysis is performed on the normalized word sequence using natural language processing technology to obtain key information.

[0103] Because speech content may be affected by environmental interference during recognition, it needs to undergo denoising and filtering before being converted into text content. This reduces the impact of external factors on the speech content recognition process. Speech recognition can be performed using existing deep learning models. These models extract speech features and map them to acoustic units of phonemes or subphonemes. Then, based on the acoustic unit sequence and a natural language model, a corresponding word sequence is generated. The recognized word sequence is then normalized, for example, by adding punctuation marks based on pauses, ultimately resulting in a complete text sentence (the normalized word sequence). Semantic analysis can then be used to obtain key information.

[0104] Generate a third-party program interface to retrieve image content from specified communication data through the third program;

[0105] Call the OCR interface to extract text content from the image using OCR technology, and then perform semantic analysis on the text content using natural language processing technology to obtain key information;

[0106] Store key information within information groups;

[0107] The AI ​​language model interface is called to output each information group sequentially to the AI ​​language model. The AI ​​language model reorganizes the content of the information groups through semantic analysis to obtain the reorganized sentences.

[0108] By combining the reorganized statements with customer information sources, complete business opportunity information can be obtained;

[0109] Existing artificial intelligence language models can be used to compile multiple independent words within a group of information into a complete reconstructed sentence. This reconstructed sentence is concise but includes all the information that is helpful for the transaction.

[0110] Furthermore, because the reconstructed statements are synthesized using an AI language model and are not manually reviewed, they may sometimes not meet usability standards. Therefore, customer information sources are included in the business opportunity information. When salespeople find the reconstructed statements unsatisfactory, they can re-examine the customer's information based on the source data.

[0111] Based on the number of business opportunity information, the customers who submitted the information this time were assigned to the salesperson with the most business opportunity information.

[0112] Since business opportunity information reflects the amount of communication between sales staff and customers, more communication indicates greater effort invested by the salesperson in communicating with that customer. Theoretically, the customer listed in this report should be classified as the salesperson with the most business opportunity information. Of course, if sales staff disagree with the system's classification, they can appeal offline.

[0113] Preferably, the following steps should be performed before storing key information in information groups:

[0114] Used to clean critical information and remove redundant, erroneous, or irrelevant data;

[0115] The cleaned key information is then mapped to the defined information groups.

[0116] By removing redundant information and ensuring its cleanliness and accuracy, the content generated in subsequent sentences will be more realistic and fluent.

[0117] A customer attribution management system based on customer information reporting, and a customer attribution management method based on customer information reporting, including an uploading module, a processing module, and an on-chain module;

[0118] The upload module is used to accept the reporting information uploaded by the sales client, wherein the reporting information includes customer information and sales information;

[0119] The processing module is used to generate an identifier based on the customer information, and to generate the user's storage information on the blockchain based on the reporting information and the identifier.

[0120] The on-chain module is used to query whether the number of the identifier on the blockchain is 0;

[0121] If the quantity is 0, the storage information will be uploaded to the blockchain, and the customer whose information was reported this time will be included in the sales information.

[0122] If the quantity is greater than 0, then obtain the storage information with the same identifier on the blockchain, and determine whether the sales information in all the storage information is consistent. If they are consistent, then classify the customer of this report information into the sales information.

[0123] If there is a discrepancy, the customers for this reporting information will be categorized based on the on-chain time and / or region of the different storage information.

[0124] Preferably, the on-chain module includes a query sub-module;

[0125] The query submodule is used to connect to the blockchain using the QuikNode API interface;

[0126] Based on the API service documentation, write an API interface activation request based on identifier lookup and send the activation request to the node service center;

[0127] Receive query data from the service center.

[0128] Preferably, the on-chain module further includes a customer segmentation sub-module;

[0129] The customer segmentation submodule is used to obtain the sales location in the sales information of the stored information, and use it as the first region;

[0130] Retrieve the sales region from the previously stored sales information and use it as the second region;

[0131] Determine if the first region and the second region have the same name. If no name is found, classify the customer in this report as a salesperson in the sales information.

[0132] If the same name appears, the on-chain time of the storage information corresponding to the second region with the same name is obtained as the first time, and the upload time of the information reported this time is used as the second time.

[0133] Determine if the time difference between the first and second times is greater than a time threshold. If it is, classify the customer whose information was reported this time as a salesperson in the sales information.

[0134] If the value is less than the threshold, a communication data request is sent to the sales client, and the customers whose information was reported are categorized based on the communication data request.

[0135] Preferably, the customer segmentation submodule includes a text information processing unit, a voice information processing unit, an image information processing unit, a recombination unit, and an allocation unit;

[0136] The text information processing unit is used to generate a first program interface and obtain the text content in the specified communication data through the first program.

[0137] Natural language processing technology is used to perform semantic analysis on text content. Based on the results of the semantic analysis, the text data is divided and key information is extracted from it.

[0138] The voice information processing unit is used to generate a second program interface, and to obtain the voice content in the specified communication data through the second program.

[0139] The speech content is denoised and filtered to obtain preprocessed speech. The preprocessed speech is input into a deep learning model, and the extracted speech features are mapped to acoustic units of phonemes or subphonemes. Based on the acoustic unit sequence and the natural language model, a word sequence is generated. The identified word sequence is normalized, and semantic analysis is performed on the normalized word sequence using natural language processing technology to obtain key information.

[0140] The image information unit is used to generate a third program interface, through which the third program obtains the image content in the specified communication data;

[0141] Call the OCR interface to extract text content from the image using OCR technology, and then perform semantic analysis on the text content using natural language processing technology to obtain key information;

[0142] The reorganization unit is used to store key information within information groups;

[0143] The AI ​​language model interface is called to output each information group sequentially to the AI ​​language model. The AI ​​language model reorganizes the content of the information groups through semantic analysis to obtain the reorganized sentences.

[0144] By combining the reorganized statements with customer information sources, complete business opportunity information can be obtained;

[0145] The allocation unit is used to assign customers with the most reported business opportunities to the salesperson with the most business opportunities based on the number of business opportunities reported.

[0146] Preferably, the recombination unit further includes a preprocessing subunit;

[0147] The preprocessing subunit is used to clean key information and remove redundant, erroneous, or irrelevant data;

[0148] The cleaned key information is then mapped to the defined information groups.

[0149] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "illustrative embodiment," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0150] Although embodiments of the invention have been shown and described, those skilled in the art will understand that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the claims and their equivalents.

Claims

1. A customer attribution management method based on customer information reporting, characterized in that, Includes the following steps: Step S1: Accept the reporting information uploaded by the sales client, wherein the reporting information includes customer information and sales information; Step S2: Generate an identifier based on the customer information, and generate the user's blockchain storage information based on the reported information and the identifier. Step S3: Check if the number of the identifier on the blockchain is 0; If the quantity is 0, the storage information will be uploaded to the blockchain, and the customer whose information was reported this time will be included in the sales information. If the quantity is greater than 0, then obtain the storage information with the same identifier on the blockchain, and determine whether the sales information in all the storage information is consistent. If they are consistent, then classify the customer of this report information into the sales information. If there is a discrepancy, the sales region in the sales information stored this time will be used as the first region; The process involves: retrieving the sales region from previously stored sales information as the second region; determining if the first and second regions have the same name; if not, classifying the customer in this report as a sales representative within the sales information; if the same name exists, retrieving the upload time of the stored information corresponding to the second region with the same name as the first time, and using the upload time of this report as the second time; determining if the time difference between the first and second times is greater than a time threshold; if greater, classifying the customer in this report as a sales representative within the sales information; if less, sending communication data requests to the sales client and classifying the customer in this report based on the communication data requests. Specifically, this includes the following steps: generating a first program interface; retrieving text content from specified communication data through the first program; performing semantic analysis on the text content using natural language processing technology; classifying the text data based on the semantic analysis results; and extracting key information from the data. Generate a second program interface, and use the second program to obtain the voice content in the specified communication data; The speech content is denoised and filtered to obtain preprocessed speech. The preprocessed speech is input into a deep learning model, and the extracted speech features are mapped to acoustic units of phonemes or subphonemes. Based on the acoustic unit sequence and the natural language model, a word sequence is generated. The identified word sequence is normalized, and semantic analysis is performed on the normalized word sequence using natural language processing technology to obtain key information. Generate a third-party program interface to retrieve image content from specified communication data through the third program; Call the OCR interface to extract text content from the image using OCR technology, and then perform semantic analysis on the text content using natural language processing technology to obtain key information; Store key information within information groups; The AI ​​language model interface is invoked to sequentially output each information group to the AI ​​language model. The AI ​​language model reorganizes the content of the information group through semantic analysis to obtain reorganized sentences. The reorganized sentences are combined with the customer information sources to obtain complete business opportunity information. Based on the number of business opportunity information, the customers with the most reported information are assigned to the sales staff with the most business opportunity information.

2. The customer attribution management method based on customer information reporting according to claim 1, characterized in that, The specific steps for querying whether the number of the identifier on the blockchain is 1 in step S3 are as follows: connect to the blockchain using the QuikNode API interface; Based on the API service documentation, write an API interface activation request based on identifier lookup and send the activation request to the node service center; Receive query data from the service center.

3. The customer attribution management method based on customer information reporting according to claim 1, characterized in that, The following steps need to be performed before storing key information in information groups: Used to clean critical information and remove redundant, erroneous, or irrelevant data; The cleaned key information is then mapped to the defined information groups.

4. A customer attribution management system based on customer information reporting, characterized in that, The customer attribution management method based on customer information reporting as described in any one of claims 1 to 3 includes an uploading module, a processing module, and an on-chain module; The upload module is used to accept the reporting information uploaded by the sales client, wherein the reporting information includes customer information and sales information; The processing module is used to generate an identifier based on the customer information, and to generate the user's storage information on the blockchain based on the reporting information and the identifier. The on-chain module is used to query whether the number of the identifier on the blockchain is 0; If the quantity is 0, the storage information will be uploaded to the blockchain, and the customer whose information was reported this time will be included in the sales information. If the quantity is greater than 0, then obtain the storage information with the same identifier on the blockchain, and determine whether the sales information in all the storage information is consistent. If they are consistent, then classify the customer of this report information into the sales information. If there is a discrepancy, the customers for this reporting information will be categorized based on the on-chain time and / or region of the different storage information. The on-chain module also includes a customer segmentation sub-module; The customer segmentation submodule is used to obtain the sales location in the sales information of the stored information, and use it as the first region; Retrieve the sales region from the previously stored sales information and use it as the second region; Determine if the first region and the second region have the same name. If no name is found, classify the customer in this report as a salesperson in the sales information. If the same name appears, the on-chain time of the storage information corresponding to the second region with the same name is obtained as the first time, and the upload time of the information reported this time is used as the second time. Determine if the time difference between the first and second times is greater than a time threshold. If it is, classify the customer whose information was reported this time as a salesperson in the sales information. If the value is less than the target value, a communication data request is sent to the sales client respectively, and the customers whose information is reported are divided based on the communication data request. The customer segmentation submodule includes a text information processing unit, a voice information processing unit, an image information processing unit, a recombination unit, and an allocation unit; The text information processing unit is used to generate a first program interface and obtain the text content in the specified communication data through the first program. Natural language processing technology is used to perform semantic analysis on text content. Based on the results of the semantic analysis, the text data is divided and key information is extracted from it. The voice information processing unit is used to generate a second program interface, and to obtain the voice content in the specified communication data through the second program. The speech content is denoised and filtered to obtain preprocessed speech. The preprocessed speech is input into a deep learning model, and the extracted speech features are mapped to acoustic units of phonemes or subphonemes. Based on the acoustic unit sequence and the natural language model, a word sequence is generated. The identified word sequence is normalized, and semantic analysis is performed on the normalized word sequence using natural language processing technology to obtain key information. The image information unit is used to generate a third program interface, through which the third program obtains the image content in the specified communication data; Call the OCR interface to extract text content from the image using OCR technology, and then perform semantic analysis on the text content using natural language processing technology to obtain key information; The reorganization unit is used to store key information within information groups; The AI ​​language model interface is called to output each information group sequentially to the AI ​​language model. The AI ​​language model reorganizes the content of the information groups through semantic analysis to obtain the reorganized sentence. By combining the reorganized statements with customer information sources, complete business opportunity information can be obtained; The allocation unit is used to assign customers with the most reported business opportunities to the salesperson with the most business opportunities based on the number of business opportunities.

5. A customer attribution management system based on customer information reporting according to claim 4, characterized in that, The on-chain module includes a query sub-module; The query submodule is used to connect to the blockchain using the QuikNode API interface; Based on the API service documentation, write an API interface activation request based on identifier lookup and send the activation request to the node service center; Receive query data from the service center.

6. A customer attribution management system based on customer information reporting according to claim 4, characterized in that, The recombination unit also includes a preprocessing subunit; The preprocessing subunit is used to clean key information and remove redundant, erroneous, or irrelevant data; The cleaned key information is then mapped to the defined information groups.

Images