Customer matching method and device based on big data analysis

[0022]Example one

[0023]Such asfigure 1 As shown, the embodiment of the present application provides a customer matching method based on big data analysis, wherein the method further includes:

[0024]Step S100: Obtain the first course information;

[0025]Specifically, the first course information is the course information that the customer learns through the online education platform. The first course information has a wide variety of types, including various professional courses and various interest tutoring groups. No specific settings are made here. set.

[0026]Step S200: Obtain first course object information according to the first course information;

[0027]Specifically, the first course object information is an object for learning through online education, and the first course object information corresponds to the first course information one-to-one, which can be further understood as when the first course information is In a children's English training course, the first course object information is children and the like.

[0028]Step S300: Obtain first recommended client information according to the first course object information;

[0029]Specifically, the first recommended customer information is one of many objects of the first course object, that is, a customer is selected from the first course object information, and the first recommended customer information is the same as the The direction and content of the first course object information learning are the same.

[0030]Step S400: Obtain client IP information according to the first recommended client information;

[0031]Specifically, the client IP information is the network address information of the first recommended client information for surfing the Internet. The IP address is a unified address format provided by the IP protocol. It assigns a logical address to each network and each host on the Internet to shield the differences in physical addresses. Because of this unique address, This ensures that users can efficiently and conveniently select the objects they need from tens of thousands of computers when operating on a networked computer.

[0032]Step S500: Obtain first customer network data according to the customer IP information, where the first customer network data is the group chat information that the first customer participated in;

[0033]Specifically, if the customer IP information is known, the first customer network data can be further obtained. The first customer network data is the group chat information that the first customer participates in, that is, the first user learns The chat information of the learning exchange group added to the subject, the first customer can learn independently based on the chat information of the learning exchange group.

[0034]Step S600: Input the first course object information and the first customer network data into a first training model, where the first training model is obtained through training of multiple sets of training data, and each of the multiple sets of training data A set of training data includes: the first course object information, the first customer network data, and identification information used to identify the first correlation data;

[0035]Step S700: Obtain output information of the first training model, where the output information includes the first correlation data, and the first correlation data indicates that the first customer network data and the first course Relevance between object information;

[0036]Specifically, the first course object information and the first customer network data may be input into a first training model, and the first training model is a model for continuous self-training and learning based on the training data. Further, The training model is a neural network model. The neural network model is a neural network model in machine learning. Neural Networks (NN) are widely connected to each other by a large number of simple processing units (called neurons). The formed complex neural network system reflects many basic characteristics of human brain function and is a highly complex nonlinear dynamic learning system. The neural network model is described based on the mathematical model of neurons. Artificial Neural Networks (Artificial Neural Networks) is a description of the first-order characteristics of the human brain system.

[0037]Simply put, it is a mathematical model. Training based on a large amount of training data, where each set of training data in the training data includes: the first course object information, the first customer network data, and identification information used to identify the first relevance data, The neural network model continuously revises itself, and when the output information of the neural network model reaches a predetermined accuracy/convergence state, the supervised learning process ends. Through the data training of the neural network model, the neural network model is further made to process the input data more accurately, and the output first relevance data, that is, the first customer network data and the first A more accurate correlation between the information of course objects. Based on the characteristics of more accurate data processing after the training model is trained, the first course object information and the first customer network data are input into the first training model, and the first result is accurately obtained through the output information of the training model. Modifications have been made to improve the accuracy of customer data, thereby achieving the technical effect of making the promotion customer match higher and increasing the promotion success rate.

[0038]Step S800: Determine whether the first correlation data meets a first predetermined threshold;

[0039]Step S900: When it is satisfied, a first instruction is obtained. The first instruction indicates that the first recommended client successfully matches the first course information, and the first recommended client is saved in a client database.

[0040]Specifically, the first predetermined threshold is a preset compliance value of the first correlation data, and the first correlation data can be judged to determine whether the first correlation data meets the first A predetermined threshold. When the first relevance data meets the first predetermined threshold, that is, the first relevance data is up to the standard, then a first instruction is obtained. The first instruction is the The matching of the first course information is successful, that is, the learning direction and content of the first recommended client are consistent with the teaching direction and content of the first course information, and the first recommended client is stored in a customer database, The customer database can also be updated in real time, achieving the technical effect of matching a suitable communication group for customers and improving the success rate of promotion.

[0041]When the first client network data contains a result, the embodiment of the present application further includes:

[0042]Step S1010: Determine whether the first correlation data meets a first predetermined threshold;

[0043]Step S1020: When it is not satisfied, a second instruction is obtained, and the second instruction is used to delete the first recommended client.

[0044]Specifically, when the first customer network data already contains a result, that is, the first customer has added a learning exchange group, then the added one exchange learning group is input into the first customer A training model is trained to obtain the corresponding first correlation data, and determine whether the first correlation data meets the first predetermined threshold. When the first correlation data does not meet the first predetermined threshold, the first correlation data is obtained. The second instruction, the second instruction is used to delete the first recommended customer, by ensuring that the correlation data learned by the customer in the customer database meets the first predetermined threshold, thereby ensuring that the customer’s learning direction and content are consistent with The teaching direction and content of the course information are consistent, which achieves the technical effect of making the promotion customer match higher and improving the promotion success rate.

[0045]When the first customer network data contains two or more results, the embodiment of the present application further includes:

[0046]Step S1110: Obtain the first relevance data according to the first result in the first course information and the first customer network data; according to the first course information and the first result in the first customer network data The second result is to obtain the second relevance data; and so on, until the Pth relevance data is obtained according to the Pth result in the first course information and the first customer network data, where P is greater than 1. Natural number;

[0047]Step S1120: separately determine whether the first relevance data, the second relevance data, and the Pth relevance data meet the first predetermined threshold;

[0048]Step S1130: When the judgment result has at least one satisfaction, a third instruction is obtained. The third instruction is that the first recommended client matches the first course information successfully, and the first recommended client is saved in the Customer database.

[0049]Specifically, when the first customer's network data contains two or more results, that is, the first customer has added more than one learning exchange group, and may add two or two including language, mathematics, etc. For the above learning exchange group, however, the first course information has only one subject, which can be any one of subjects such as Chinese, mathematics, foreign language, etc., which should be judged and matched separately. Then, the first relevance data can be obtained according to the first result in the first course information and the first customer network data, and so on, until according to the first course information, the first customer From the Pth result in the network data, the Pth relevance data is obtained, and it is determined whether the first relevance data, the second relevance data, and the Pth relevance data meet the first predetermined Threshold, and then different results are obtained. At least one of the different results should meet the first predetermined threshold, that is, at least one result matches the first course information, then when the judgment result When there is at least one satisfaction, a third instruction is obtained. The third instruction indicates that the first recommended customer is successfully matched with the first course information, and the first recommended customer is saved in the customer database. It is more accurate to match the first course information of the first recommended client to a suitable learning exchange group, so that the matching degree of the promotion client is higher, and the technical effect of improving the promotion success rate.

[0050]In order to match the corresponding learning exchange group with more customer groups, the embodiment of the application further includes:

[0051]Step S1210: Obtain first associated customer information according to the first recommended customer information and the first customer network data;

[0052]Step S1220: Obtain a first screening result according to the first associated customer information and the customer database;

[0053]Step S1230: When the first screening result is negative, a fourth instruction is obtained. The fourth instruction indicates that the first associated customer successfully matches the first course information, and the first associated customer is saved In the customer database.

[0054]Specifically, in order to match the corresponding learning exchange group with more customer groups, the first associated customer information can also be obtained according to the first recommended customer information and the first customer network data, and the first associated customer Information can be understood as customer information in the same learning exchange group as the first recommended customer, and then according to the first associated customer information and the customer database, the first screening result is obtained, that is, the customer database is performed Screening, judging whether the first associated customer information exists in the customer database, when the first screening result is no, that is, the first associated customer information is not in the customer database, and the fourth instruction is obtained, so The fourth instruction is that the first associated customer is successfully matched with the first course information, and the first associated customer is saved in the customer database, so as to achieve the corresponding learning exchange for matching with more customer groups Group, expand the technical effect of customer groups.

[0055]After determining whether the first correlation data meets a first predetermined threshold, step S800 further includes:

[0056]Step S810: When the first association data meets the first predetermined threshold, obtain a first online group chat topic according to the first customer network data;

[0057]Step S820: Obtain a first course theme according to the first course information;

[0058]Step S830: Obtain first similarity data according to the first online group chat topic and the first course topic;

[0059]Step S840: Determine whether the first similarity data meets a second predetermined threshold;

[0060]Step S850: When it is satisfied, a fifth instruction is obtained. The fifth instruction is used to list the first recommended customer as a candidate customer and save it in the candidate customer set.

[0061]Specifically, in order to perform customer group chat matching more humanely, when the first relevance data meets the first predetermined threshold, that is, the learning direction and content of the customer and the teaching direction and content of the course information are Consistently, according to the network data of the first customer, a first online group chat topic is obtained, and the first online group chat topic is a chat topic in the learning exchange group added by the first customer, which can be further understood as a parent For the English tutoring class reported to the child, the corresponding group chat theme should be English as the theme, the first course theme is the main content of the first course information, and it can be further understood that the child is learning primary school English , The first course theme corresponds to the learning content of elementary school English, rather than the super-level content such as junior high school and high school English, and then the first similarity is obtained according to the first online group chat theme and the first course theme The first similarity information can be understood as the similarity between the chat topic of the learning exchange group added by the parent and the content that the child actually needs to learn, and it is determined whether the first similarity data meets a second predetermined threshold , That is, to determine whether the first similarity data meets the basic requirements, when it is met, that is, the first similarity data meets the basic requirements, the chat topics of the learning exchange group added by the parents and the children actually need to learn If the similarity between the content is very close, then the fifth instruction is obtained. The fifth instruction is used to list the first recommended customer as a candidate customer and save it in the candidate customer set. Select customers, wait until the end of the learning and coaching class of the children's newspaper, and then recommend relevant learning and training courses to the children's parents, achieving a more humane technical effect of matching customer group chats and increasing the success rate of promotion.

[0062]After determining whether the first similarity data meets a second predetermined threshold, step S840 further includes:

[0063]Step S841: When the first similarity data does not meet a second predetermined threshold, obtain the first instruction.

[0064]Specifically, if the first similarity data does not meet the second predetermined threshold, it can be understood that the chat topic of the learning exchange group added by the parent does not match the content that the child actually needs to learn, which does not promote the child’s learning Function, the first instruction is obtained, and according to the first instruction, the first recommended client is successfully matched with the first course information, and the child is matched to a suitable learning exchange group, achieving a more reasonable and scientific The technical effect of matching customer group chats.

[0065]Furthermore, the embodiments of this application also include:

[0066]Step S1310: Use the first client IP information as first input data;

[0067]Step S1320: Use the first customer network data as second input data;

[0068]Step S1330: Input the first input data and the second input data into a second training model, where the second training model is obtained through training of multiple sets of training data, each of the multiple sets of training data The training data all include: the first input data, the second input data, and identification information used to identify the first activity information;

[0069]Step S1340: Obtain output information of the second training model, where the output information includes the first activity information, and the first activity information indicates that the first recommended customer is in the first customer network Activity in the data, where the first activity information includes activity time and activity degree;

[0070]Step S1350: Determine whether the first activity information meets a third predetermined threshold;

[0071]Step S1360: when it is not satisfied, obtain a first active interval time according to the first activity information;

[0072]Step S1370: Determine whether the first active interval time meets a fourth predetermined threshold;

[0073]Step S1380: When it is satisfied, a sixth instruction is obtained. The sixth instruction indicates that the first recommended customer and the first course information are successfully matched, and the first associated customer is removed from the backup customer set and stored in the store. In the customer database.

[0074]Specifically, after judging whether the first relevance data meets the first predetermined threshold, it can also be judged whether the reported learning tutoring class is about to expire and end through the speaking activity in the learning exchange group and the group time. The child again matches the appropriate learning exchange group. It is possible to continuously train the input data by inputting the first customer IP information and the first customer network data into a second training model, the second training model is the same as the first training model, and the input data is not detailed here. In detail, until the output information of the neural network model reaches a predetermined accuracy/convergence state, the supervised learning process ends. Through the data training of the neural network model, the neural network model can process the input data more accurately, so that the output first activity information, that is, the first recommended customer is in the first The activity level in the customer network data, wherein the first activity level information includes active time and activity level, and then it is determined whether the first activity level information meets a third predetermined threshold, that is, the first customer is in the learning exchange group Whether the information such as the activeness of the speech and the speaking time meets the most basic activeness, when not satisfied, the first active interval time is obtained according to the first activeness information, and the first active interval time is the first The time interval for the customer to speak in the group, the activity level of the first customer can be further determined according to the first active interval time. When the first active interval time is too long, in summary, the first customer If the learning period of is about to expire, a sixth instruction is obtained. The sixth instruction is that the first recommended client matches the first course information successfully, and then the first associated client is removed from the backup client set and saved In the customer database, by judging whether the learning period is about to expire according to the user’s actual group chat activity, and recommending a suitable learning exchange group to the first customer in a timely manner, a more humane and more reasonable The matching learning exchange group makes the promotion customer match higher and improves the technical effect of promotion success rate.

[0075]In order to effectively record and save customer database information, the embodiments of this application further include:

[0076]Generating a first verification code according to the first customer data information, where the first verification code corresponds to the first customer data information one-to-one;

[0077]Generate a second verification code according to the second customer data information, the second verification code corresponds to the second customer data information one-to-one, and so on, the Nth customer data information can be obtained, and according to the Nth customer data Information generates the Nth verification code, where N is a natural number greater than 1;

[0078]Copy and save all customer data information and verification codes in M ​​devices, where M is a natural number greater than 1.

[0079]Specifically, in order to ensure that the customer database information can be effectively recorded and stored, it can be encrypted based on the blockchain to ensure that the data is not tampered with. Blockchain technology is a universal underlying technical architecture. It generates and synchronizes data on distributed nodes through a consensus mechanism, and uses programmable scripts to achieve automatic execution of contract terms and data operations. Blockchain is defined as a data structure in which data blocks are organized in chronological order, and different blocks form a chain-like connection in order. This data structure is used to build a digital ledger.

[0080]Generate a first verification code based on the first customer data information, the first verification code corresponds to the first customer data information one-to-one; generate a second verification code based on the second customer data information, the second verification code is The second customer data information is one-to-one correspondence, and so on, the Nth customer data information can be obtained, and the Nth verification code is generated according to the Nth customer data information, where N is a natural number greater than 1, and all customers The data information and the verification code are respectively copied and stored in M ​​devices, where M is a natural number greater than 1. Encrypted storage of the first customer data information, where each device corresponds to a node, and all nodes are combined to form a block chain. Such a block chain constitutes a convenient verification (as long as the last block is verified) The hash value of is equivalent to verifying the entire version), unchangeable (any change in transaction information will change the hash value of all subsequent blocks, so that it cannot be passed during verification).

[0081]The blockchain system adopts a distributed data form, so that each participating node can obtain a complete database backup. Unless 51% of the nodes in the entire system can be controlled at the same time, the modification of the database by a single node is invalid and impossible. Affect the data content on other nodes. Therefore, the more nodes participating in the system, the stronger the computing power, and the higher the data security in the system. The encryption processing of the customer database information based on the blockchain effectively ensures the storage security of the customer database information, and achieves the technical effect of safely recording and storing the customer database information.

[0082]In summary, the method and device for customer matching based on big data analysis provided by the embodiments of the present application have the following technical effects:

[0083]1. By obtaining the first course object information and the first customer network data information, and inputting the first course object information and the first customer network data information into the first neural network model for continuous training, The output result is more accurate, and the first relevance data output, that is, the relevance between the first customer network data and the first course object information is more accurate, and the matching degree of promotion customers is more accurate. High, improve the technical effect of promotion success rate.

[0084]2. By obtaining the activity level information of the first customer in the learning exchange group, it can be determined whether the learning period of the first customer is about to expire, and then matching other suitable learning exchange groups for the first customer, A more user-friendly technical effect of customer learning group chat matching is achieved.

[0085]Example two

[0086]Based on the same inventive concept as the customer matching method based on big data analysis in the foregoing embodiment, the present invention also provides a customer matching device based on big data analysis, such asfigure 2 As shown, the device includes:

[0087]First obtaining unit 11: The first obtaining unit 11 is used to obtain first course information;

[0088]Second obtaining unit 12: The second obtaining unit 12 is configured to obtain first course object information according to the first course information;

[0089]Third obtaining unit 13: The third obtaining unit 13 is configured to obtain first recommended client information according to the first course object information;

[0090]Fourth obtaining unit 14: The fourth obtaining unit 14 is configured to obtain client IP information according to the first recommended client information;

[0091]Fifth obtaining unit 15: The fifth obtaining unit 15 is configured to obtain first customer network data according to the customer IP information, where the first customer network data is group chat information that the first customer participates in;

[0092]First input unit 16: The first input unit 16 is used to input the first course object information and the first customer network data into a first training model, wherein the first training model uses multiple sets of training data Obtained by training, each of the multiple sets of training data includes: the first course object information, the first customer network data, and identification information used to identify the first relevance data;

[0093]Sixth obtaining unit 17: The sixth obtaining unit 17 is configured to obtain output information of the first training model, where the output information includes the first relevance data, and the first relevance data represents The association between the first customer network data and the first course object information;

[0094]First judging unit 18: The first judging unit 18 is used to judge whether the first correlation data meets a first predetermined threshold;

[0095]Seventh obtaining unit 19: The seventh obtaining unit 19 is configured to obtain a first instruction when it is satisfied. The first instruction indicates that the first recommended client matches the first course information successfully, and the first A recommended customer is stored in the customer database.

[0096]Further, the device further includes:

[0097]Second judgment unit: The second judgment unit is used to judge whether the first correlation data meets a first predetermined threshold;

[0098]Eighth obtaining unit: The eighth obtaining unit is used to obtain a second instruction when it is not satisfied, and the second instruction is used to delete the first recommended client.

[0099]Further, the device further includes:

[0100]Ninth obtaining unit: The ninth obtaining unit is configured to obtain the first correlation data according to the first result in the first course information and the first customer network data; according to the first course information , The second result in the first customer network data obtains the second relevance data; and so on, until the P th result is obtained according to the first course information and the P th result in the first customer network data Relevance data, where P is a natural number greater than 1;

[0101]Third judging unit: The third judging unit is configured to respectively judge whether the first relevance data, the second relevance data, and the Pth relevance data meet the first predetermined threshold;

[0102]Tenth obtaining unit: The tenth obtaining unit is used to obtain a third instruction when there is at least one satisfaction in the judgment result, and the third instruction indicates that the first recommended client matches the first course information successfully, and The first recommended customer is stored in the customer database.

[0103]Further, the device further includes:

[0104]Eleventh obtaining unit: The eleventh obtaining unit is configured to obtain first associated customer information according to the first recommended customer information and the first customer network data;

[0105]Twelfth obtaining unit: The twelfth obtaining unit is configured to obtain the first screening result according to the first associated customer information and the customer database;

[0106]Thirteenth obtaining unit: The thirteenth obtaining unit is used to obtain a fourth instruction when the first screening result is negative, and the fourth instruction is the first associated customer and the first course If the information is successfully matched, the first associated customer is stored in the customer database.

[0107]Further, the device further includes:

[0108]Fourteenth obtaining unit: The fourteenth obtaining unit is configured to obtain a first online group chat topic according to the first customer network data when the first association data meets the first predetermined threshold;

[0109]Fifteenth obtaining unit: The fifteenth obtaining unit is used to obtain the first course theme according to the first course information;

[0110]Sixteenth obtaining unit: The sixteenth obtaining unit is configured to obtain first similarity data according to the first online group chat topic and the first course topic;

[0111]Fourth judgment unit: The fourth judgment unit is used to judge whether the first similarity data meets a second predetermined threshold;

[0112]Seventeenth obtaining unit: The seventeenth obtaining unit is used to obtain a fifth instruction when satisfied, and the fifth instruction is used to list the first recommended customer as a candidate customer and save it in the candidate customer set .

[0113]Further, the device further includes:

[0114]Eighteenth obtaining unit: The eighteenth obtaining unit is configured to obtain the first instruction when the first similarity data does not meet a second predetermined threshold.

[0115]Further, the device further includes:

[0116]Second input unit: The second input unit is used to input the first input data and the second input data into a second training model, wherein the second training model is obtained through training of multiple sets of training data, so Each of the multiple sets of training data includes: the first input data, the second input data, and identification information used to identify first activity information;

[0117]Nineteenth obtaining unit: The nineteenth obtaining unit is configured to obtain output information of the second training model, wherein the output information includes the first activity information, and the first activity information represents all The activity degree of the first recommended client in the network data of the first client, wherein the first activity degree information includes an activity time and an activity degree;

[0118]Fifth judging unit: The fifth judging unit is used to judge whether the first activity information meets a third predetermined threshold;

[0119]Twentieth obtaining unit: the twentieth obtaining unit is configured to obtain a first active interval time according to the first activity information when it is not satisfied;

[0120]Sixth judging unit: The sixth judging unit belongs to judging whether the first active interval time meets a fourth predetermined threshold;

[0121]Twenty-first obtaining unit: The twenty-first obtaining unit is used to obtain a sixth instruction when it is satisfied. The sixth instruction indicates that the first recommended customer matches the first course information successfully, and the The first associated customer is collectively removed from the standby customer and stored in the customer database.

[0122]Aforementionedfigure 1 The various changes and specific examples of the customer matching method based on big data analysis in the first embodiment are also applicable to the customer matching device based on big data analysis of this embodiment. For the detailed description of the customer matching method, those skilled in the art can clearly know the implementation method of the customer matching device based on big data analysis in this embodiment, so for the sake of brevity of the description, it will not be detailed again.

[0123]Example three

[0124]Reference belowimage 3 To describe the electronic device of the embodiment of the present application.

[0125]image 3 It illustrates a schematic structural diagram of an electronic device according to an embodiment of the present application.

[0126]Based on the inventive concept of a customer matching method based on big data analysis in the foregoing embodiment, the present invention also provides a customer matching device based on big data analysis, on which a computer program is stored, which is realized when the program is executed by a processor The steps of any method of the customer matching method based on big data analysis described above.

[0127]Among them, inimage 3 In the bus architecture (represented by the bus 300), the bus 300 can include any number of interconnected buses and bridges. The bus 300 will include one or more processors represented by the processor 302 and various memories represented by the memory 304. The circuits are linked together. The bus 300 may also link various other circuits such as peripheral devices, voltage regulators, power management circuits, etc., which are all known in the art, and therefore, no further descriptions thereof are provided herein. The bus interface 305 provides an interface between the bus 300 and the receiver 301 and transmitter 303. The receiver 301 and the transmitter 303 may be the same element, namely a transceiver, which provides a unit for communicating with various other devices on the transmission medium. The processor 302 is responsible for managing the bus 300 and general processing, and the memory 304 may be used to store data used by the processor 302 when performing operations.

[0128]The embodiment of the application provides a customer matching method based on big data analysis, wherein the method further includes: obtaining first course information; obtaining first course object information according to the first course information; One course object information, obtain the first recommended customer information; according to the first recommended customer information, obtain the customer IP information; according to the customer IP information, obtain the first customer network data, the first customer network data is the Group chat information that the first customer participated in; input the first course object information and the first customer network data into a first training model, where the first training model is obtained by training with multiple sets of training data, and the multiple Each set of training data in the set of training data includes: the first course object information, the first customer network data, and identification information used to identify the first relevance data; obtaining the output of the first training model Information, wherein the output information includes the first relevance data, and the first relevance data represents the relevance between the first customer network data and the first course object information; judging the first Whether the relevance data meets the first predetermined threshold; when it is satisfied, a first instruction is obtained. The first instruction indicates that the first recommended customer matches the first course information successfully, and the first recommended customer is saved in Customer database.

[0129]Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a system, or a computer program product. Therefore, the present invention may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.