Customer affiliation management method and system based on customer information reporting

Abstract

A customer affiliation management method based on customer information reporting comprises the following steps that reporting information uploaded by a sales client is received, and the reporting information comprises customer information and sales information; generating an identification symbol based on the customer information, and generating storage information of the block chain on the user based on the reporting information and the identification symbol; querying whether the number of the identifiers on the block chain is 0; if the number is 0, uploading the storage information, and attributing the customer of the report information to the sales in the sales information; according to the invention, the automatic division of the clients can be realized, and the corresponding storage information is generated through the report information and stored in the block chain, so that the authenticity of the client division basis is ensured, the client division basis is not changed, and a real data basis is provided for the automatic division of the clients.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a customer attribution management method and system based on customer information reporting. Background Art

[0002] With the continuous development of communication technology, the efficiency of information transmission is getting higher and higher. The improvement of information transmission efficiency has affected the sales methods of transmission. The company's sales management is not limited to using its own sales, but usually uses news channels to entrust third parties to sell on its behalf, relying on their channel capabilities to expand sales business.

[0003] In contrast, when a customer comes to the company to purchase goods, a certain commission or commission must be given to the salesperson or channel broker; however, in some cases, such as real estate sales, car sales, etc., it is difficult to distinguish whether the customer is contacted by the company's salesperson to purchase, or contacted by the channel broker on other platforms to purchase. This can easily lead to unclear customer attribution, resulting in conflicts in customer judgment, and thus customer interception. Summary of the invention

[0004] In view of the above-mentioned defects, the purpose of the present invention is to propose a customer attribution management method and system based on customer information reporting to solve the problem of unclear customer attribution and customer interception.

[0005] To achieve this purpose, 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: accepting the reporting information uploaded by the sales client, wherein the reporting information includes customer information and sales information;

[0007] Step S2: Generate an identification symbol based on the customer information, and generate the user's storage information on the blockchain based on the reporting information and the identification symbol;

[0008] Step S3: Query whether the number of the identification symbol on the blockchain is 0;

[0009] If the quantity is 0, the storage information will be uploaded to the chain, and the customer who reported the information this time will be attributed to the sales in the sales information;

[0010] If the quantity is greater than 0, 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, the customer of this reported information is attributed to the sales in the sales information;

[0011] If there is inconsistency, the customers reporting the information will be divided based on the time and / or region of uploading the different stored information.

[0012] Preferably, the specific steps of querying whether the number of the identification symbol on the blockchain is 1 in step S3 are: connecting to the blockchain using the QuikNode API interface;

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

[0014] Receive query data fed back by the service center.

[0015] Preferably, the method for dividing the customers who report the information this time based on the time and / or region of uploading different stored information in step S3 is as follows:

[0016] Get the sales region in the sales information of this storage information as the first region;

[0017] Obtain the sales region in the sales information stored previously as the second region;

[0018] Determine whether the same name appears in the first region and the second region. If not, the customer in the reported information is classified as the sales in the sales information;

[0019] If the same name appears, the upload 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 this reporting information is taken as the second time;

[0020] Determine whether the time difference between the first time and the second time is greater than the time threshold. If so, classify the customer who reported the information this time as a sales in the sales information;

[0021] If it is less than, communication data requests are sent to the sales client respectively, and the customers who report the information this time are divided based on the communication data requests.

[0022] Preferably, the steps of dividing the customers who report the information based on the communication data request are as follows:

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

[0024] Use natural language processing technology to perform semantic analysis on text content, divide text data based on the results of semantic analysis, and extract key information from it;

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

[0026] De-noising and filtering are performed on the speech content to obtain pre-processed speech, which is then input into the deep learning model. The extracted speech features are mapped to acoustic units of phonemes or subphonemes. According to the acoustic unit sequence and the natural language model, a vocabulary sequence is generated. The recognized vocabulary sequence is normalized, and semantic analysis of the normalized vocabulary sequence is performed using natural language processing technology to obtain key information.

[0027] Generate a third program interface, and obtain the picture content in the specified communication data through the third program;

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

[0029] Store key information in information groups;

[0030] Calling the artificial intelligence language model interface, outputting each information group to the artificial intelligence language model in sequence, and the artificial intelligence language model reorganizing the content of the information group through semantic analysis to obtain a reorganized sentence;

[0031] Combine the restructured statements and customer information sources to obtain complete business opportunity information;

[0032] Based on the number of business opportunity information, the customers who reported the information this time are assigned to the sales with the most business opportunity information.

[0033] Preferably, the following steps need to be performed before storing the key information in the information grouping:

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

[0035] And map the cleaned key information into the defined information groups.

[0036] A customer attribution management system based on customer information reporting, wherein the customer attribution management method based on customer information reporting comprises an uploading module, a processing module and an uplink 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 identification symbol based on the customer information, and to generate storage information of the user on the blockchain based on the reporting information and the identification symbol;

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

[0040] If the quantity is 0, the storage information will be uploaded to the chain, and the customer who reported the information this time will be attributed to the sales in the sales information;

[0041] If the quantity is greater than 0, 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, the customer of this reported information is attributed to the sales in the sales information;

[0042] If there is inconsistency, the customers reporting the information will be divided based on the time and / or region of uploading the different stored information.

[0043] Preferably, the uplink module includes a query submodule;

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

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

[0046] Receive query data fed back by the service center.

[0047] Preferably, the uplink module also includes a customer segmentation submodule;

[0048] The customer classification submodule is used to obtain the sales region in the sales information of the current stored information as the first region;

[0049] Obtain the sales region in the sales information stored previously as the second region;

[0050] Determine whether the same name appears in the first region and the second region. If not, the customer in the reported information is classified as the sales in the sales information;

[0051] If the same name appears, the upload 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 this reporting information is taken as the second time;

[0052] Determine whether the time difference between the first time and the second time is greater than the time threshold. If so, classify the customer who reported the information this time as a sales in the sales information;

[0053] If it is less than, communication data requests are sent to the sales client respectively, and the customers who report the information this time are divided based on the communication data requests.

[0054] Preferably, the customer segmentation submodule includes a text information processing unit, a voice information processing unit, a picture information processing unit, a reorganization 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] Use natural language processing technology to perform semantic analysis on text content, divide text data based on the results of semantic analysis, and extract key information from it;

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

[0058] De-noising and filtering are performed on the speech content to obtain pre-processed speech, which is then input into the deep learning model. The extracted speech features are mapped to acoustic units of phonemes or subphonemes. According to the acoustic unit sequence and the natural language model, a vocabulary sequence is generated. The recognized vocabulary sequence is normalized, and semantic analysis of the normalized vocabulary sequence is performed using natural language processing technology to obtain key information.

[0059] The picture information unit is used to generate a third program interface, and obtain the picture content in the specified communication data through the third program;

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

[0061] The reassembly unit is used to store key information in the information group;

[0062] Calling the artificial intelligence language model interface, outputting each information group to the artificial intelligence language model in sequence, and the artificial intelligence language model reorganizing the content of the information group through semantic analysis to obtain a reorganized sentence;

[0063] Combine the restructured statements and customer information sources to obtain complete business opportunity information;

[0064] The allocation unit is used to allocate the customers who reported the information this time to the sales with the most business opportunity information based on the amount of business opportunity information.

[0065] Preferably, the recombination unit further comprises a pretreatment subunit;

[0066] The pre-processing sub-unit is used to clean key information and remove redundant, erroneous or irrelevant data;

[0067] And map the cleaned key information into 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 division of customers, and the corresponding storage information is generated through the reporting information and saved in the blockchain, which ensures the authenticity of the basis for customer division and will not be tampered with, and provides a real data basis for the automatic division of customers. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] Figure 1 is a flow chart of an embodiment of the method of the present invention.

[0070] Figure 2 It is a schematic diagram of the structure of an embodiment of the system of the present invention. DETAILED DESCRIPTION

[0071] The embodiments of the present invention are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals represent the same or similar elements or elements having the same or similar functions from beginning to end. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and cannot be understood as limiting the present invention.

[0072] In the description of the embodiments of the present invention, the terms "first" and "second" are used for descriptive purposes only and are not to be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the embodiments of the present invention, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0073] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, unless otherwise specified, the meaning of "plurality" is two or more. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to 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: accepting the reporting information uploaded by the sales client, wherein the reporting information includes customer information and sales information;

[0076] Step S2: Generate an identification symbol based on the customer information, and generate the user's storage information on the blockchain based on the reporting information and the identification symbol;

[0077] Step S3: Query whether the number of the identification symbol on the blockchain is 0;

[0078] If the quantity is 0, the storage information will be uploaded to the chain, and the customer who reported the information this time will be attributed to the sales in the sales information;

[0079] If the quantity is greater than 0, 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, the customer of this reported information is attributed to the sales in the sales information;

[0080] If there is inconsistency, the customers reporting the information will be divided based on the time and / or region of uploading the different stored information.

[0081] In order to better divide customers and enable each salesperson to fairly obtain relevant commissions, a sales client is provided in the present invention. After receiving a customer, the salesperson needs to report on the sales client and send reporting information, wherein the customer information includes customer-related information such as customer name and customer phone number, and the sales information includes salesperson's name, sales area, phone number and other sales-related information.

[0082] After uploading the filing information, the system will convert the filing information to generate corresponding storage information, in which a unique representation symbol associated with the customer information will be generated according to the customer information. The method for generating the identification symbol can be: binary encoding, hash encoding, character encoding, etc. When encoding the identification symbol, the customer's social unified code or provincial certificate can be added to ensure the uniqueness of the identification symbol. After the storage information is generated, it will not be immediately stored on the chain. Only after the customer's ownership is determined, the storage information will be put on the chain. In the subsequent communication of customer commission disputes, the corresponding storage information can be found from the above-mentioned blockchain to realize the basis for customer division. Specifically, before the storage information is put on the chain, it is necessary to use the identification symbol to search in the blockchain to determine whether there is the same identification symbol in the blockchain. If there is no same identification symbol, it means that no sales have connected with the customer before. At this time, the customer can be divided into the corresponding sales in this filing information. Then the storage information is stored on the chain. When there is the same identification symbol, it means that the sales have already connected with the customer. At this time, it is necessary to divide the customers of this filing information according to the time and / or region of the chain of different storage information. If customers need to be reclassified, the sales information in the stored information needs to be modified, and the sales information needs to be modified to the previous sales information before being uploaded to the chain.

[0083] The following is an example to explain that the sales client of sales A uploads the filing information. At this time, the filing information includes the customer information of customer A and the sales information of sales A. The identification symbol A is generated through the customer information of customer A, and the filing information and the identification symbol A together generate the storage information A of the user on the blockchain. Then the identification symbol A will be used to search on the blockchain. If no other storage information is retrieved, the storage information A will be uploaded to the chain. At this time, customer A will be divided into sales A. If other storage information B is retrieved, it is determined whether the sales information in storage information B is consistent with the sales information of storage information A. If it is consistent, it means that the same sales are connected to the same customer. At this time, customer A will still be divided into sales A, and storage information A will continue to be uploaded to the chain. If it is inconsistent, it is necessary to divide the customers of this filing information based on the time and / or region of different storage information. It is worth mentioning that if the customer A this time is not divided into the sales A uploaded by this filing information, the storage information A will not be uploaded to the blockchain, and the message of customer division failure will be returned to the sales client. Sales A can use this message to find the manager of the region to appeal and perform other operations.

[0084] The present invention can realize automatic division of customers, and generate corresponding storage information through reporting information and save it in the blockchain, ensuring the authenticity of the basis for customer division and preventing tampering, thus providing a real data basis for the automatic division of customers.

[0085] Preferably, the specific steps of querying whether the number of the identification symbol on the blockchain is 1 in step S3 are: connecting to the blockchain using the QuikNode API interface;

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

[0087] Receive query data fed back by the service center.

[0088] For large companies, there are multiple business locations in different regions, and QuikNode supports multiple different geographical regions, which helps optimize network call time and improve the speed and performance of DApp. Users can choose the appropriate region to deploy nodes according to their needs. Sales at business locations in different regions can upload reporting information to improve the convenience of chaining.

[0089] Preferably, the method for dividing the customers who report the information this time based on the time and / or region of uploading different stored information in step S3 is as follows:

[0090] Get the sales region in the sales information of this storage information as the first region;

[0091] Obtain the sales region in the sales information stored previously as the second region;

[0092] Determine whether the same name appears in the first region and the second region. If not, the customer in the reported information is classified as the sales in the sales information;

[0093] If the same name appears, the upload 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 this reporting information is taken as the second time;

[0094] Determine whether the time difference between the first time and the second time is greater than the time threshold. If so, classify the customer who reported the information this time as a sales in the sales information;

[0095] If it is less than, communication data requests are sent to the sales client respectively, and the customers who report the information this time are divided based on the communication data requests.

[0096] Since the client connects to the blockchain through the QuikNode API interface, it can upload its regional information when connecting, or add the sales region information to the sales information, so that the data of the first region and the second region can be quickly obtained. Since customers of the same company type may purchase in different sales areas across the country. There will be different sales to receive customers in different regions. In this case, it is determined whether the same name appears in the first and second regions. If the same name does not appear, it means that the customer appears in the first region for the first time, and the customer can be divided into the sales uploaded in this reporting information. If the same name appears in the first and second regions, it means that the customer is not the first time to appear in this area, so at this time it is necessary to determine whether the time difference between the upload time of this reporting information and the first time is for the time threshold, where the time threshold is greater than half a year. When the time difference is greater than the time threshold, it means that the customer has not contacted the previous sales for a long time, so the customer can be divided into the sales uploaded in this reporting information, and the corresponding storage information can be uploaded to the chain. If the time difference is less than the time threshold, it means that the customer has been communicating with the previous sales in a short period of time. At this time, the customer's affiliation cannot be directly determined, so 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 for upload. The communication data can be used to determine the depth of communication between the salesperson and the customer, and the customer affiliation can be divided according to the depth of communication.

[0097] Preferably, the steps of dividing the customers who report the information based on the communication data request are as follows:

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

[0099] Use natural language processing technology to perform semantic analysis on text content, divide text data based on the results of semantic analysis, and extract key information from it;

[0100] For text content recognition, the existing and mature natural language processing (NLP) technology can be used to perform semantic analysis on the text content. The results of the analysis can be used to determine the customer's transaction intention, the content of the transaction, etc., so as to divide the text data and extract the corresponding key information. For example, if the transaction time, quantity, product type, etc. need to be extracted, the text data will be divided according to the above needs to extract the transaction time, quantity, and product, and obtain different key information for each group.

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

[0102] De-noising and filtering are performed on the speech content to obtain pre-processed speech, which is then input into the deep learning model. The extracted speech features are mapped to acoustic units of phonemes or subphonemes. According to the acoustic unit sequence and the natural language model, a vocabulary sequence is generated. The recognized vocabulary sequence is normalized, and semantic analysis of the normalized vocabulary sequence is performed using natural language processing technology to obtain key information.

[0103] Since the speech content may be disturbed by the environment during recognition, it needs to be denoised and filtered before being converted into text content to reduce the impact of external factors on the speech content during the recognition process. When performing speech recognition, it can be recognized through the existing deep learning model. The deep learning model maps the extracted speech features to the acoustic units of phonemes or subphonemes, and then generates the corresponding vocabulary sequence based on the acoustic unit sequence and the natural language model. The recognized vocabulary sequence is normalized, such as adding punctuation marks according to the pause position, and finally a complete text sentence (normalized vocabulary sequence) is obtained. Then semantic analysis can be used to obtain key information.

[0104] Generate a third program interface, and obtain the picture content in the specified communication data through the third program;

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

[0106] Store key information in information groups;

[0107] Calling the artificial intelligence language model interface, outputting each information group to the artificial intelligence language model in sequence, and the artificial intelligence language model reorganizing the content of the information group through semantic analysis to obtain a reorganized sentence;

[0108] Combine the restructured statements and customer information sources to obtain complete business opportunity information;

[0109] Through the existing artificial intelligence language model, multiple independent words in an information group can be compiled into a complete reorganized sentence. The reorganized sentence is short but includes all the information that is helpful for the transaction.

[0110] In addition, since the reorganized sentences are synthesized by the artificial intelligence language model and have not been manually reviewed, sometimes the constructed reorganized sentences may not meet the usage standards. Therefore, the customer information source will be attached to the business opportunity information. When the salesperson feels that the reorganized sentence is not useful, he can re-understand the customer's information based on the customer information source.

[0111] Based on the number of business opportunity information, the customers who reported the information this time are assigned to the sales with the most business opportunity information.

[0112] Since business opportunity information reflects the amount of business communication between sales and customers, the more communication, the more cost the salesperson has to invest in communicating with the customer. In theory, the customers in this reported information should be classified as the salesperson with the most business opportunity information. Of course, if salespeople disagree with the results of the system classification, they can make an offline appeal.

[0113] Preferably, the following steps need to be performed before storing the key information in the information grouping:

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

[0115] And map the cleaned key information into the defined information groups.

[0116] Clear redundant information and ensure the cleanliness and accuracy of the information, so that the content will be more realistic and fluent in the subsequent generation of corresponding sentences.

[0117] A customer attribution management system based on customer information reporting, wherein the customer attribution management method based on customer information reporting comprises an uploading module, a processing module and an uplink 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 identification symbol based on the customer information, and to generate storage information of the user on the blockchain based on the reporting information and the identification symbol;

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

[0121] If the quantity is 0, the storage information will be uploaded to the chain, and the customer who reported the information this time will be attributed to the sales in the sales information;

[0122] If the quantity is greater than 0, 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, the customer of this reported information is attributed to the sales in the sales information;

[0123] If there is inconsistency, the customers reporting the information will be divided based on the time and / or region of uploading the different stored information.

[0124] Preferably, the uplink module includes a query submodule;

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

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

[0127] Receive query data fed back by the service center.

[0128] Preferably, the uplink module also includes a customer segmentation submodule;

[0129] The customer classification submodule is used to obtain the sales region in the sales information of the current stored information as the first region;

[0130] Obtain the sales region in the sales information stored previously as the second region;

[0131] Determine whether the same name appears in the first region and the second region. If not, the customer in the reported information is classified as the sales in the sales information;

[0132] If the same name appears, the upload 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 this reporting information is taken as the second time;

[0133] Determine whether the time difference between the first time and the second time is greater than the time threshold. If so, classify the customer who reported the information this time as a sales in the sales information;

[0134] If it is less than, communication data requests are sent to the sales client respectively, and the customers who report the information this time are divided based on the communication data requests.

[0135] Preferably, the customer segmentation submodule includes a text information processing unit, a voice information processing unit, a picture information processing unit, a reorganization 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] Use natural language processing technology to perform semantic analysis on text content, divide text data based on the results of semantic analysis, and extract key information from it;

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

[0139] De-noising and filtering are performed on the speech content to obtain pre-processed speech, which is then input into the deep learning model. The extracted speech features are mapped to acoustic units of phonemes or subphonemes. According to the acoustic unit sequence and the natural language model, a vocabulary sequence is generated. The recognized vocabulary sequence is normalized, and semantic analysis of the normalized vocabulary sequence is performed using natural language processing technology to obtain key information.

[0140] The picture information unit is used to generate a third program interface, and obtain the picture content in the specified communication data through the third program;

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

[0142] The reassembly unit is used to store key information in the information group;

[0143] Calling the artificial intelligence language model interface, outputting each information group to the artificial intelligence language model in sequence, and the artificial intelligence language model reorganizing the content of the information group through semantic analysis to obtain a reorganized sentence;

[0144] Combine the restructured statements and customer information sources to obtain complete business opportunity information;

[0145] The allocation unit is used to allocate the customers who reported the information this time to the sales with the most business opportunity information based on the amount of business opportunity information.

[0146] Preferably, the recombination unit further comprises a pretreatment subunit;

[0147] The pre-processing sub-unit is used to clean key information and remove redundant, erroneous or irrelevant data;

[0148] And map the cleaned key information into the defined information groups.

[0149] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "illustrative embodiments", "examples", "specific examples", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

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

Claims

1. A customer attribution management method based on customer information reporting, characterized in that: The steps include: Step S1: accepting the reporting information uploaded by the sales client, wherein the reporting information includes customer information and sales information; Step S2: Generate an identification symbol based on the customer information, and generate the user's storage information on the blockchain based on the reporting information and the identification symbol; Step S3: Query whether the number of the identification symbol on the blockchain is 0; If the quantity is 0, the storage information will be uploaded to the chain, and the customer who reported the information this time will be attributed to the sales in the sales information; If the quantity is greater than 0, 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, the customer of this reported information is attributed to the sales in the sales information; If there is inconsistency, the customers reporting the information will be divided based on the time and / or region of uploading the different stored information.

2. A customer attribution management method based on customer information reporting according to claim 1, characterized in that: The specific steps of querying whether the number of the identification symbol on the blockchain is 1 in step S3 are: connecting to the blockchain using the QuikNode API interface; According to the API service documentation, write an API interface activation request based on the identifier query and send the activation request to the node service center; Receive query data fed back by the service center.

3. A method for customer attribution management based on customer information reporting according to claim 2, characterized in that: The method for dividing the customers who report the information this time based on the time and / or region of uploading different stored information in step S3 is as follows: Get the sales region in the sales information of this storage information as the first region; Obtain the sales region in the sales information stored previously as the second region; Determine whether the same name appears in the first region and the second region. If not, the customer in the reported information is classified as the sales in the sales information; If the same name appears, the upload 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 this reporting information is taken as the second time; Determine whether the time difference between the first time and the second time is greater than the time threshold. If so, classify the customer who reported the information this time as a sales in the sales information; If it is less than, communication data requests are sent to the sales client respectively, and the customers who report the information this time are divided based on the communication data requests.

4. A method for customer attribution management based on customer information reporting according to claim 3, characterized in that: The specific steps of dividing the customers reporting information based on the communication data request are as follows: Generate a first program interface, and obtain text content in the specified communication data through the first program; Use natural language processing technology to perform semantic analysis on text content, divide text data based on the results of semantic analysis, and extract key information from it; Generate a second program interface, and obtain the voice content in the specified communication data through the second program; De-noising and filtering are performed on the speech content to obtain pre-processed speech, which is then input into the deep learning model. The extracted speech features are mapped to acoustic units of phonemes or subphonemes. According to the acoustic unit sequence and the natural language model, a vocabulary sequence is generated. The recognized vocabulary sequence is normalized, and semantic analysis of the normalized vocabulary sequence is performed using natural language processing technology to obtain key information. Generate a third program interface, and obtain the picture content in the specified communication data through the third program; Call the OCR interface, extract the text content in the image through OCR technology, and perform semantic analysis on the text content through natural language processing technology to obtain key information; Store key information in information groups; Calling the artificial intelligence language model interface, outputting each information group to the artificial intelligence language model in sequence, and the artificial intelligence language model reorganizing the content of the information group through semantic analysis to obtain a reorganized sentence; Combine the restructured statements and customer information sources to obtain complete business opportunity information; Based on the number of business opportunity information, the customers who reported the information this time are assigned to the sales with the most business opportunity information.

5. A method for customer attribution management based on customer information reporting according to claim 4, characterized in that: The following steps need to be performed before storing key information in information groups: Used to clean key information and remove redundant, erroneous or irrelevant data; And map the cleaned key information into the defined information groups.

6. A customer attribution management system based on customer information reporting, characterized in that: A customer attribution management method based on customer information reporting as described in any one of claims 1 to 5, comprising an uploading module, a processing module and an uplink 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 identification symbol based on the customer information, and to generate storage information of the user on the blockchain based on the reporting information and the identification symbol; The on-chain module is used to query whether the number of the identification symbol on the blockchain is 0; If the quantity is 0, the storage information will be uploaded to the chain, and the customer who reported the information this time will be attributed to the sales in the sales information; If the quantity is greater than 0, 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, the customer of this reported information is attributed to the sales in the sales information; If there is inconsistency, the customers reporting the information will be divided based on the time and / or region of uploading the different stored information.

7. A customer attribution management system based on customer information reporting according to claim 6, characterized in that: The uplink module includes a query submodule; The query submodule is used to connect to the blockchain using the QuikNode API interface; According to the API service documentation, write an API interface activation request based on the identifier query and send the activation request to the node service center; Receive query data fed back by the service center.

8. A customer attribution management system based on customer information reporting according to claim 7, characterized in that: The uplink module also includes a customer division submodule; The customer classification submodule is used to obtain the sales region in the sales information of the current stored information as the first region; Obtain the sales region in the sales information stored previously as the second region; Determine whether the same name appears in the first region and the second region. If not, the customer in the reported information is classified as the sales in the sales information; If the same name appears, the upload 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 this reporting information is taken as the second time; Determine whether the time difference between the first time and the second time is greater than the time threshold. If so, classify the customer who reported the information this time as a sales in the sales information; If it is less than, communication data requests are sent to the sales client respectively, and the customers who report the information this time are divided based on the communication data requests.

9. A customer attribution management system based on customer information reporting according to claim 8, characterized in that: The customer division submodule includes a text information processing unit, a voice information processing unit, a picture information processing unit, a reorganization 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; Use natural language processing technology to perform semantic analysis on text content, divide text data based on the results of semantic analysis, and extract key information from it; The voice information processing unit is used to generate a second program interface, and obtain the voice content in the specified communication data through the second program; De-noising and filtering are performed on the speech content to obtain pre-processed speech, which is then input into the deep learning model. The extracted speech features are mapped to acoustic units of phonemes or subphonemes. According to the acoustic unit sequence and the natural language model, a vocabulary sequence is generated. The recognized vocabulary sequence is normalized, and semantic analysis of the normalized vocabulary sequence is performed using natural language processing technology to obtain key information. The picture information unit is used to generate a third program interface, and obtain the picture content in the specified communication data through the third program; Call the OCR interface, extract the text content in the image through OCR technology, and perform semantic analysis on the text content through natural language processing technology to obtain key information; The reassembly unit is used to store key information in the information group; Calling the artificial intelligence language model interface, outputting each information group to the artificial intelligence language model in sequence, and the artificial intelligence language model reorganizing the content of the information group through semantic analysis to obtain a reorganized sentence; Combine the restructured statements and customer information sources to obtain complete business opportunity information; The allocation unit is used to allocate the customers who reported the information this time to the sales with the most business opportunity information based on the amount of business opportunity information.

10. A customer attribution management system based on customer information reporting according to claim 8, characterized in that: The recombination unit also includes a pretreatment subunit; The pre-processing sub-unit is used to clean key information and remove redundant, erroneous or irrelevant data; And map the cleaned key information into the defined information groups.

Images