A client industry determination method and apparatus, electronic device, and storage medium

Abstract

Embodiments of the present application disclose a customer industry determination method and device, electronic equipment and a storage medium; the industry prediction reference information in the customer data can be matched with an industry word library to determine a target word, the first candidate industry of the customer to which the customer data belongs can be determined from the preset industry according to the target word in the industry prediction reference information, the first industry identification information of the first candidate industry can be obtained based on the corresponding relationship between the preset industry and the first industry identification information, the first customer description information and the first industry identification information can be converted into a target industry description vector, and then the target second industry identification information can be determined according to the preset mapping relationship between the industry description vector and the second industry identification information, and the target second industry identification information is set as the industry identification information of the customer to which the customer data belongs. Therefore, the dependence on manual work in determining the customer industry can be reduced, the human resources can be saved, and the efficiency of determining the industry to which the customer belongs in the financial field can be improved on the basis of ensuring accuracy.

Description

Technical Field

[0001] This invention relates to the field of financial security, and more specifically to a method, apparatus, electronic device, and storage medium for determining a customer's industry. Background Technology

[0002] With the rapid development of the current economy, more and more customers are conducting financial transactions, and the volume of financial transactions has also increased significantly. In order to ensure the security and compliance of financial transactions, it is necessary to determine the industry to which the customers conducting financial transactions belong.

[0003] Currently, the main method for determining a client's industry is for auditors to determine the client's industry based on the information provided and then add the client's industry information during the audit process. However, this approach cannot guarantee that auditors will conduct a comprehensive audit based on client data, and it is highly dependent on human resources, resulting in a slow audit process and hindering efficiency in determining a client's industry. Summary of the Invention

[0004] This invention provides a method, apparatus, electronic device, and storage medium for determining a customer's industry, which can reduce reliance on manual labor, save human resources, and improve the efficiency of determining the customer's industry while ensuring accuracy.

[0005] This invention provides a method for determining a customer's industry, including:

[0006] The first customer description information in the customer data to be analyzed that matches the preset specific customer characteristics is determined, and the information in the customer data other than the first customer description information is determined as industry prediction reference information.

[0007] The industry prediction reference information is matched with industry thesaurus of at least two preset industries to determine the target words in the industry prediction reference information that match the industry-related words in the industry thesaurus.

[0008] Based on the target words corresponding to each preset industry in the industry prediction reference information, the first candidate industry of the customer to which the customer data belongs is determined from the preset industries. Based on the correspondence between the preset industries and the first industry identification information, the first industry identification information of the first candidate industry is obtained.

[0009] The first customer description information and the first industry identification information are converted into feature vectors, and the converted feature vectors are used as the target industry description vectors corresponding to the customer data.

[0010] Based on the preset mapping relationship between the industry description vector and the second industry identification information, and the target industry description vector, the target second industry identification information corresponding to the customer data is determined, and the target second industry identification information is set as the industry identification information of the customer to which the customer data belongs.

[0011] Accordingly, embodiments of the present invention also provide a customer industry determination device, which includes:

[0012] The information determination unit is used to determine the first customer description information in the customer data to be analyzed that matches the preset specific customer characteristics, and to determine the information in the customer data other than the first customer description information as industry prediction reference information.

[0013] The matching unit is used to match the industry prediction reference information with industry thesaurus of at least two preset industries to determine the target words in the industry prediction reference information that match the industry-related words in the industry thesaurus.

[0014] The first industry determination unit is used to determine the first candidate industry of the customer to which the customer data belongs from the preset industries based on the target words corresponding to each preset industry in the industry prediction reference information, and to obtain the first industry identification information of the first candidate industry based on the correspondence between the preset industries and the first industry identification information.

[0015] The vector conversion unit is used to convert the first customer description information and the first industry identification information into feature vectors, and use the converted feature vectors as the target industry description vectors corresponding to the customer data.

[0016] The second industry determination unit is used to determine the target second industry identification information corresponding to the customer data based on the preset mapping relationship between the industry description vector and the second industry identification information, and the target industry description vector, and set the target second industry identification information as the industry identification information of the customer to which the customer data belongs.

[0017] In an optional example, the first industry determination unit includes a first candidate industry determination unit, which is used to determine the correlation between the industry prediction reference information and each preset industry based on the industry correlation characterization information of the industry related words of each preset industry and the target words matched by the industry prediction reference information under each industry thesaurus. The industry correlation characterization information is used to characterize the correlation between the industry related words and the preset industry to which they belong.

[0018] Based on the correlation between the industry forecast reference information and each preset industry, the first candidate industry of the customer to which the customer data belongs is determined.

[0019] In an optional example, the first candidate industry determination unit can also be used to obtain the similarity between the target word and the industry-related words that match the target word;

[0020] Based on the industry relevance representation information of industry-related terms for each preset industry and the similarity, the relevance between the target term and each preset industry is determined;

[0021] Based on the target words in the industry forecast reference information and the correlation between the target words and each preset industry, the correlation between the industry forecast reference information and each preset industry is determined.

[0022] In an optional example, before the information determination unit, a mapping relationship establishment unit is further included, used to obtain the historical industry description vector of the historical customer data and the second industry identification information of the historical customer;

[0023] A mapping relationship is established between the second industry identification information and the historical industry description vector belonging to the same historical customer, resulting in a preset mapping relationship between the industry description vector and the second industry identification information.

[0024] Correspondingly, the second industry determination unit includes a second industry identification information determination unit, used to determine the relevance between the target industry description vector and the industry description vectors in the preset mapping relationship;

[0025] Based on the relevance corresponding to each industry description vector and the second industry identification information corresponding to each industry description vector, the target second industry identification information corresponding to the customer data is determined.

[0026] In one example, the second industry identification information determination unit includes a relevance determination subunit, which is used to calculate the distance between the target industry description vector and each of the industry description vectors, and use the distance as the relevance between the target industry description vector and each of the industry description vectors;

[0027] Correspondingly, the second industry identification information determination unit further includes a target second industry identification information determination unit, used to select a preset number of industry description vectors as reference industry description vectors from the industry description vectors based on the relevance.

[0028] Based on the second industry identification information corresponding to the reference industry description vector, the number of reference industry description vectors under each second industry identification information is counted.

[0029] Based on the number of reference industry description vectors under each second industry identifier, the target second industry identifier corresponding to the customer data is determined.

[0030] In one example, the target second industry identification information determination unit can also be used to calculate the ratio between the number of reference industry description vectors corresponding to each second industry identification information and the preset number, so as to obtain the vote difference corresponding to each second industry identification information.

[0031] If there is a second industry identifier information whose vote difference is greater than a preset difference, determine the second industry identifier information corresponding to the customer data from the second industry identifier information whose vote difference is greater than the preset difference, and obtain the target second industry identifier information;

[0032] If there is no second industry identification information where the difference in votes is greater than the preset difference, the customer data and the number of reference industry description vectors corresponding to the customer data under each second industry identification information are stored as anomaly prediction information in the anomaly prediction information set.

[0033] Send the abnormal prediction information in the abnormal prediction information set to the manual review platform;

[0034] The system receives the manual review results for the abnormal prediction information sent by the manual review platform. If the manual review results include target industry identification information set for the customer data, the target industry identification information is determined as the target second industry identification information corresponding to the customer data.

[0035] In one example, after the second industry determination unit, there is also a message generation unit, which is used to determine the target industry terminology corresponding to the target second industry identification information from the industry terminology of the at least two preset industries;

[0036] Based on the target industry thesaurus, determine the target industry related words in the target industry thesaurus that match the industry prediction reference information;

[0037] Obtain the customer analysis message template corresponding to the target second industry identification information. The customer analysis message template includes filling instruction information for the position to be filled. The filling instruction information is used to indicate the industry-related words that need to be filled at the position to be filled.

[0038] Based on the filling instruction information, the industry related words to be filled in the target industry related words are determined, and the industry related words to be filled in are filled into the corresponding filling positions in the customer analysis message template to obtain the customer analysis message.

[0039] Accordingly, embodiments of the present invention also provide an electronic device, including a memory and a processor; the memory stores an application program, and the processor is used to run the application program in the memory to perform the operations in any of the customer industry determination methods provided in embodiments of the present invention.

[0040] Furthermore, embodiments of the present invention also provide a storage medium storing a plurality of instructions adapted for loading by a processor to execute steps in any of the customer industry determination methods provided in embodiments of the present invention.

[0041] Using the scheme of this embodiment of the invention, a first customer description information matching a preset specific customer characteristic can be determined from the customer data to be analyzed. Information in the customer data other than the first customer description information is determined as industry prediction reference information. This industry prediction reference information is matched with industry thesauruses of at least two preset industries to determine target words in the industry prediction reference information that match industry-related words in the industry thesauruses. Based on the target words in the industry prediction reference information corresponding to each preset industry, a first candidate industry for the customer to which the customer data belongs is determined from the preset industries. Based on the correspondence between preset industries and first industry identification information, the first industry identification information of the first candidate industry is obtained. The first customer description information and the first industry identification information are then converted into a special... The feature vector, obtained through transformation, serves as the target industry description vector corresponding to the customer data. Based on the preset mapping relationship between the industry description vector and the second industry identifier information, and the target industry description vector, the target second industry identifier information corresponding to the customer data is determined. This target second industry identifier information is then set as the industry identifier information of the customer to which the customer data belongs. In this embodiment, industry-related word matching calculations are first performed on the industry prediction reference information to determine the first industry identifier information under the first candidate industry, making a preliminary prediction of the customer's industry. Then, combined with the first customer description information, calculations are performed according to the mapping relationship to finally obtain the industry identifier information of the customer to which the customer data belongs, further ensuring the accuracy of industry determination. Therefore, it can reduce the reliance on manual labor in determining the customer's industry, save human resources, and improve the efficiency of determining the customer's industry in the financial industry while ensuring accuracy. Attached Figure Description

[0042] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0043] Figure 1 This is a schematic diagram of a scenario illustrating the customer industry determination method provided in an embodiment of the present invention;

[0044] Figure 2 This is a flowchart of the customer industry determination method provided in an embodiment of the present invention;

[0045] Figure 3 This is another schematic diagram of the customer industry determination method provided in the embodiments of the present invention;

[0046] Figure 4 This is a schematic diagram of the customer data submission page provided in an embodiment of the present invention;

[0047] Figure 5 This is a schematic diagram of the manual review page provided in an embodiment of the present invention;

[0048] Figure 6 This is a schematic diagram of the page for generating customer analysis messages provided in an embodiment of the present invention;

[0049] Figure 7 This is a schematic diagram of the customer industry determination device provided in an embodiment of the present invention;

[0050] Figure 8 This is another structural schematic diagram of the customer industry determination device provided in an embodiment of the present invention;

[0051] Figure 9 This is a schematic diagram of the structure of the electronic device provided in an embodiment of the present invention. Detailed Implementation

[0052] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0053] This invention provides a customer industry determination method, apparatus, electronic device, and storage medium. Specifically, this invention provides a customer industry determination method applicable to a customer industry determination apparatus, which can be integrated into an electronic device.

[0054] The electronic device can be a terminal or other device, including but not limited to mobile terminals and fixed terminals. For example, mobile terminals include but are not limited to smartphones, smartwatches, tablets, laptops, smart vehicles, etc., while fixed terminals include but are not limited to desktop computers, smart TVs, etc.

[0055] The electronic device can also be a server or other device. The server can be an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms, but it is not limited to these.

[0056] The customer industry determination method of this invention can be implemented by a server, or by a terminal and a server together.

[0057] The following example illustrates the method of determining the customer's industry using a combination of terminal and server implementation.

[0058] like Figure 1 As shown, the customer industry determination system provided in this embodiment of the invention includes a terminal 10 and a server 20, etc. The terminal 10 and the server 20 are connected through a network, such as a wired or wireless network. The terminal 10 can serve as a terminal for users to send customer data to be analyzed to the server 20.

[0059] Terminal 10 can be a terminal for users to upload customer data to be analyzed, and is used to send the customer data to be analyzed to server 20.

[0060] Server 20 can be used to determine the first customer description information in the customer data to be analyzed that matches the preset specific customer characteristics, determine the information in the customer data other than the first customer description information as industry prediction reference information, match the industry prediction reference information with the industry thesaurus of at least two preset industries, determine the target words in the industry prediction reference information that match the industry related words in the industry thesaurus, determine the first candidate industry of the customer to which the customer data belongs from the preset industries based on the target words in the industry prediction reference information that correspond to each preset industry, obtain the first industry identification information of the first candidate industry based on the correspondence between the preset industries and the first industry identification information, convert the first customer description information and the first industry identification information into feature vectors, use the converted feature vectors as the target industry description vectors corresponding to the customer data, and determine the target second industry identification information corresponding to the customer data based on the preset mapping relationship between the industry description vectors and the second industry identification information, and the target industry description vectors.

[0061] Server 20 can send the target second industry identification information to terminal 10 so that terminal 10 can display the target second industry identification information.

[0062] Terminal 10 can display the target second industry identification information after receiving it from server 20. Terminal 10 can adjust the target second industry identification information based on the actions of the reviewer and send the adjusted result to server 20. Server 20 generates new target second industry identification information based on the received adjusted result.

[0063] The following sections provide detailed descriptions of each example. It should be noted that the order in which the embodiments are described is not intended to limit the preferred order of the embodiments.

[0064] The embodiments of the present invention will be described from the perspective of a customer industry determination device, which can be integrated into a server or terminal.

[0065] like Figure 2 As shown, the specific process of the customer industry determination method in this embodiment can be as follows:

[0066] 201. Determine the first customer description information in the customer data to be analyzed that matches the preset specific customer characteristics, and determine the information in the customer data other than the first customer description information as industry prediction reference information.

[0067] Among them, the preset specific customer characteristics can be pre-set based on the information that customers will provide related to themselves during actual application; the server can match the customer data to be analyzed with the preset specific customer characteristics to obtain the first customer description information belonging to the preset specific customer characteristics.

[0068] In practical applications, this can be achieved through the terminal, such as... Figure 4 On the page shown, customers fill in their information, providing some or all of the data that matches the preset specific customer characteristics. The terminal sends the data in the table and the preset specific customer characteristics to which the data belongs to the corresponding data to the server. The server can directly determine the first customer description information based on the data in the table and the preset specific customer characteristics to which the data belongs, further improving the efficiency of determining the first customer description information and industry forecast reference information.

[0069] It should be noted that the customer data involved in the specific embodiments of the present invention requires customer permission or consent when the embodiments provided by the present invention are applied to specific products or technologies, and the collection, use and processing of the relevant data must comply with the relevant laws, regulations and standards of the relevant countries and regions.

[0070] Understandably, to ensure the accuracy of the final target second industry identification information, before analyzing the customer data to obtain the target customer data, it is necessary to pre-establish a mapping relationship to be used in the customer industry determination method. Before step 201, the following is also included:

[0071] Obtain the historical industry description vector and the second industry identification information of historical customers from their historical customer data.

[0072] Establish a mapping relationship between the second industry identification information and the historical industry description vector belonging to the same historical customer, and obtain the preset mapping relationship between the industry description vector and the second industry identification information;

[0073] Historical customer data can be authentic customer data provided by customers within a certain period of time. For example, when establishing a mapping relationship, customer data from all customers of a bank over the past three years can be selected as historical customer data. Historical customer data can also be customer data compiled by developers based on actual application usage. Developers can design more complex information content and types in the compiled customer data to enhance the accuracy of the mapping relationship.

[0074] The second industry identifier information for historical customers can be generated manually after analyzing historical customer data to determine the industry to which the historical user belongs.

[0075] Optionally, in the mapping relationship, the second industry identification information of a historical customer can be mapped to only one historical industry description vector. The second industry identification information of historical customers mapped to different historical industry description vectors can be the same. Alternatively, the second industry identification information of a historical customer can be mapped to multiple historical industry description vectors.

[0076] 202. Match the industry forecast reference information with the industry terminology databases of at least two preset industries to determine the target words in the industry forecast reference information that match the industry-related words in the industry terminology databases.

[0077] The preset industries can be categories of industries determined based on industry classifications in daily life or in the financial field. In this embodiment of the invention, there are at least two preset industries.

[0078] In one example, each industry has its own industry-specific thesaurus, which can include proper nouns, commonly used industry terms, and other industry-specific vocabulary. Different industry thesauruses may contain some identical terms.

[0079] Specifically, when some industry terms in the industry terminology library match industry prediction reference information, the industry terms that match the industry prediction reference information are industry-related terms, and the terms in the industry prediction reference information that match the industry-related terms in the industry terminology library are target terms.

[0080] It is understandable that the industry terms corresponding to different preset industries may be partially or entirely the same, and the corresponding target terms may be partially or entirely the same.

[0081] Understandably, in order to ensure the relevance of target words to the industry, when determining target words in industry forecasting reference information, target words may include not only words that completely match industry-related words, but also modifiers of those words.

[0082] For example, developers set up two preset industries: retail and wholesale. Both the retail and wholesale industry thesaurus contain the related term "pork." If the industry forecast information provided by client A includes "1000 jin of pork sold daily," then "1000 jin of pork" will be used as the target term.

[0083] In another example, if some words in the industry prediction reference information do not completely match the industry-related words in the industry thesaurus, but the similarity is greater than the preset value, then these words are also used as target words, and the similarity corresponding to each target word is recorded.

[0084] For example, the industry terminology database for the wholesale industry includes the related term "pork". If the industry forecast reference information provided by Customer A includes "1000 jin of braised pork sold daily", then when determining the target term, "1000 jin of braised pork" will be used as the target term, and the similarity between "braised pork" and "pork" will be recorded.

[0085] In practical applications, to ensure the accuracy and timeliness of industry terms in the industry thesaurus, the industry terms in the thesaurus can be adjusted based on historical customer data. Optionally, before the step "determining the first customer description information in the customer data to be analyzed that matches the preset specific customer characteristics", the following may also be included:

[0086] Obtain historical industry forecast reference information and corresponding historical target keywords;

[0087] Historical industry prediction reference information is matched with industry terminology databases to be trained for at least two preset industries to determine target words that match industry-related words in the industry terminology databases to be trained.

[0088] Based on the historical target words and the target words, the industry words in the industry thesaurus to be trained are adjusted to obtain the industry thesaurus.

[0089] When adjusting the industry terms in the industry terminology database to be trained, one can modify existing industry terms in the database, or add or delete industry terms, and so on.

[0090] 203. Based on the target words corresponding to each preset industry in the industry forecast reference information, determine the first candidate industry of the customer to which the customer data belongs from the preset industries, and obtain the first industry identification information of the first candidate industry based on the correspondence between the preset industries and the first industry identification information.

[0091] The different first industry identifiers can be used to represent different preset industries. These identifiers can be the industry name itself, or different numerical values ​​corresponding to different industries. For example, developers can set different numerical values ​​for different industries, such as a value of 1 for the retail industry and a value of 2 for the wholesale industry. It's understandable that the numerical values ​​for different industries only serve to represent those industries.

[0092] Specifically, the step "Based on the target terms corresponding to each preset industry in the industry forecast reference information, determine the first candidate industry of the customer to which the customer data belongs from the preset industries" can include:

[0093] Based on the industry relevance representation information of industry-related terms for each preset industry, and the target terms matched by the industry prediction reference information in each industry terminology database, the relevance between the industry prediction reference information and each preset industry is determined. The industry relevance representation information is used to represent the relevance between industry-related terms and their respective preset industries.

[0094] Based on industry forecast reference information and the correlation between various preset industries, the first candidate industry of the customer to which the customer data belongs is determined.

[0095] Among them, the industry relevance representation information of industry-related words for each preset industry can be used for measurement. Based on the industry-related words, the industry corresponding to the user corresponding to the industry prediction reference information is predicted as the influence degree of each preset industry. The industry relevance representation information of industry-related words for each preset industry can be expressed in the form of industry prediction probability or industry prediction score, such as 15% or 3 points.

[0096] In different preset industries, the industry relevance representation information of industry-related terms can be different.

[0097] For example, if the industry relevance information of industry-related terms for each preset industry is represented by industry prediction scores, in the retail industry, the industry prediction score corresponding to the industry-related term "large quantity" can be 1, and in the wholesale industry, the industry prediction score corresponding to the industry-related term "large quantity" can be 12.

[0098] In one example, when determining the correlation between industry prediction reference information and each preset industry, if the industry correlation representation information is expressed in the form of industry prediction probability, then the correlation of the industry prediction reference information in different preset industries can be comprehensively calculated by weighted calculation and other methods based on the industry correlation representation information of the industry related words of each preset industry.

[0099] In another example, when determining the correlation between industry forecast reference information and each preset industry, if the industry correlation representation information is expressed in the form of industry forecast scores, then the correlation of the industry forecast reference information in different preset industries can be comprehensively calculated by weighted calculation, cumulative summation, and other methods based on the industry correlation representation information of the industry related words of each preset industry.

[0100] In this process, developers set rules based on the correlation between industry forecast reference information and each preset industry to determine the first candidate industry of the customer to which the customer data belongs, according to actual usage. For example, the rule can be set to select only the industry with the highest correlation with the industry forecast reference information as the first candidate industry.

[0101] In another example, to avoid potential errors in selecting only one first candidate industry, the correlation between the industry prediction reference information and each preset industry can be sorted from highest to lowest. The top N industries corresponding to the correlation with the industry prediction reference information can then be selected as the first candidate industries. Here, N can be determined by the developers based on actual usage.

[0102] It is understood that different correlations are obtained under different preset industries. Therefore, after obtaining the industry prediction reference information and the correlations of each preset industry, the industry corresponding to different correlations can be directly determined, which improves the efficiency of the embodiments of the present invention.

[0103] Understandably, if some words in the industry prediction reference information do not completely match the industry words in the industry thesaurus, but these words have a certain degree of similarity to the industry words, then these words will also be used as target words, and the similarity between each target word and the corresponding industry related words will be recorded.

[0104] The step "Determine the correlation between the industry prediction reference information and each preset industry based on the industry relevance representation information of the industry-related terms for each preset industry, and the target terms matched by the industry prediction reference information in each industry thesaurus" can include:

[0105] Obtain the similarity between the target word and the industry-related words that match the target word;

[0106] Based on the industry relevance representation information and similarity of industry-related terms for each preset industry, the relevance between the target term and each preset industry is determined;

[0107] Based on the target words in the industry forecast reference information and the correlation between the target words and each preset industry, the correlation between the industry forecast reference information and each preset industry is determined.

[0108] In determining the relevance of a target word to each preset industry based on the industry relevance representation information and similarity of industry-related terms for each preset industry, the weight of the relevance between different target words and their respective preset industries can be determined first based on the similarity, and then a weighted calculation can be performed to obtain the relevance between the target word and each preset industry.

[0109] In practical applications, to improve the accuracy of the correlation between industry-related terms and their respective preset industries, and to minimize the impact of manually set industry prediction values ​​on the final target customer data, the correlation between industry-related terms and their respective preset industries can be adjusted based on historical customer data. Optionally, embodiments of the present invention may further include:

[0110] Obtain historical target keywords and their corresponding historical industry relevance to historical industry forecast reference information;

[0111] Based on the industry relevance representation information of the industry-related words to be trained for each preset industry, as well as the similarity between historical target words and industry-related words that match historical target words, the relevance of industry prediction reference information under each preset industry is determined.

[0112] Based on historical industry relevance and correlation, the industry relevance representation information of the industry-related words of each preset industry to be trained is adjusted to obtain the industry relevance representation information of the industry-related words of each preset industry after training.

[0113] In adjusting the industry relevance representation information of industry-related words in each preset industry to be trained, one can modify the industry relevance representation information of industry-related words in different industries, or modify the industry relevance representation information of industry-related words in different industries, etc.

[0114] 204. Convert the first customer description information and the first industry identification information into feature vectors, and use the converted feature vectors as the target industry description vectors corresponding to the customer data.

[0115] The target industry description vector can comprehensively describe the first customer description information and the first industry identification information. In one example, the server can establish a target vector space based on the industry description vector, and the dimension of the target vector space can be determined according to preset specific customer characteristics. If the first industry identification information is a numerical value corresponding to the industry, then the target industry description vector can represent the first customer description information and the first industry identification information in the target vector space in the form of position.

[0116] In another example, the server can perform multiple linear regression calculations on the first customer description information and the first industry identification information, with the second industry identification information as an unknown dependent variable and the first customer description information and the first industry identification information as variables, to calculate a set of regression coefficients and random error terms, and use this set of regression coefficients and random error terms as the target industry description vector.

[0117] 205. Based on the preset mapping relationship between the industry description vector and the second industry identification information, and the target industry description vector, determine the target second industry identification information corresponding to the customer data, and set the target second industry identification information as the industry identification information of the customer to which the customer data belongs.

[0118] The preset mapping relationship is the correspondence between industry description vectors and second industry identification information that is established in advance through manual annotation or machine learning.

[0119] Optionally, in the preset mapping relationship, a customer's second industry identification information may correspond to only one industry description vector, and the second industry identification information corresponding to different industry description vectors may be the same, or a customer's second industry identification information may be mapped to multiple industry description vectors.

[0120] The second industry identifier information can be generated manually after analyzing historical customer data and industry description vectors to determine the industry to which the customer described by the industry description vector belongs. The target second industry identifier information can represent the industry of the customer to which the customer data belongs. The target second industry identifier information can represent at least one industry. The server can send the target second industry identifier information representing at least two industries to the terminal for display, and the unique industry corresponding to the customer data is determined through manual review.

[0121] Specifically, the step "determine the target second industry identifier information corresponding to the customer data based on the preset mapping relationship between the industry description vector and the second industry identifier information, and the target industry description vector" may include:

[0122] Determine the target industry description vector and its relevance to the industry description vectors in the preset mapping relationship;

[0123] Based on the relevance of each industry description vector and the second industry identification information corresponding to each industry description vector, the target second industry identification information corresponding to the customer data is determined.

[0124] The relevance can be a numerical value, such as a number like 3, to represent the degree of relevance between the description vectors of various industries and the description vector of the target industry.

[0125] In one example, the step "determine the relevance between the target industry description vector and the industry description vectors in the preset mapping relationship" may include:

[0126] Calculate the distance between the target industry description vector and the description vectors of each industry, and use the distance as the correlation between the target industry description vector and the description vectors of each reference industry.

[0127] The distance can be calculated using the following formula:

[0128]

[0129] Where S represents the distance between the target industry description vector and the industry description vector, n represents the vector dimension of the industry description vector, and x ip Let x represent the vector component of the n-dimensional target industry description vector in the i-th dimension. iq Let i represent the vector component of the n-dimensional reference industry description vector in the i-th dimension.

[0130] In one example, to ensure the accuracy of the target second industry identification information while reducing the resource consumption of the customer industry determination method and improving the running speed, K-Nearest Neighbor can be used in practical applications. ， The KNN algorithm is used to calculate the target second industry identifier information. Optionally, the step "determine the target second industry identifier information corresponding to the customer data based on the relevance of each industry description vector and the industry identifier information" may include:

[0131] Based on relevance, a preset number of industry description vectors are selected as reference industry description vectors from the industry description vectors.

[0132] Based on the second industry identification information corresponding to the reference industry description vector, count the number of reference industry description vectors under each second industry identification information.

[0133] Based on the number of reference industry description vectors under each second industry identifier, the target second industry identifier corresponding to the customer data is determined.

[0134] The preset quantity is generally an odd number to avoid the situation where the number of reference industry description vectors in the two preset industries is the same when there are only two preset industries.

[0135] Optionally, if the target industry description vector represents a set of regression coefficients and random error terms, after obtaining the reference industry description vector, the regression coefficients and random error terms represented by the reference industry description vector can be weighted according to relevant factors to obtain a new set of regression coefficients and random error terms. The corresponding second industry identification information can then be determined as the target second industry identification information based on the new regression coefficients and random error terms.

[0136] In one example, when setting the target second industry identifier information as the industry identifier information of the customer to which the customer data belongs, if the customer data originally has industry identifier information, the original industry identifier information can be updated to the target second industry identifier information; if the customer data originally does not have industry identifier information, the target second industry identifier information can be written into the customer data.

[0137] To avoid the risk of errors in calculating the target second industry identifier information, the obtained target second industry identifier information can be further reviewed to improve the reliability of the embodiments of the present invention. Optionally, the step "determining the target second industry identifier information corresponding to the customer data based on the number of reference industry description vectors under each second industry identifier information" includes:

[0138] Calculate the ratio between the number of reference industry description vectors corresponding to each second industry identifier and the preset number to obtain the vote difference corresponding to each second industry identifier.

[0139] If there is a second industry identifier information whose vote difference is greater than the preset difference, determine the second industry identifier information corresponding to the customer data from the second industry identifier information whose vote difference is greater than the preset difference, and obtain the target second industry identifier information;

[0140] If there is no second industry identifier information where the difference in votes is greater than the preset difference, the customer data and the number of reference industry description vectors corresponding to the customer data under each second industry identifier information are stored as anomaly prediction information in the anomaly prediction information set.

[0141] Send the abnormal prediction information from the abnormal prediction information set to the manual review platform;

[0142] Receive the manual review results of the abnormal prediction information sent by the manual review platform. If the manual review results include the target industry identification information set for the customer data, determine the target industry identification information as the target second industry identification information corresponding to the customer data.

[0143] In one example, the vote difference corresponding to each second industry identifier can be calculated. If multiple vote differences greater than the preset difference are obtained, the vote differences greater than the preset difference and the corresponding second industry identifier are sent to the terminal for manual review. The target second industry identifier corresponding to the customer data is determined through manual review.

[0144] In another example, the vote difference corresponding to each second industry identifier can be calculated. If multiple vote differences greater than a preset difference are obtained, all second industry identifiers with vote differences greater than the preset difference are taken as target second industry identifiers. That is, target second industry identifiers can include multiple second industry identifiers. When setting the target second industry identifier as the industry identifier of the customer to which the customer data belongs, the target second industry identifier corresponding to the customer data with the highest vote difference can be set as the industry identifier of the customer to which the customer data belongs. At the same time, other target second industry identifiers and their corresponding vote differences are set in the customer data with supplementary explanations, remarks, etc.

[0145] In another example, only the vote difference corresponding to the number of reference industry description vectors with the largest value can be calculated. If the vote difference is greater than the preset difference, the second industry identification information corresponding to the vote difference can be directly used as the target second industry identification information corresponding to the customer data.

[0146] Among them, when storing abnormal prediction information into the abnormal prediction information set, the number of abnormal prediction information entries in the abnormal prediction information set can be counted in real time. When the number of stored abnormal prediction information entries exceeds the preset value, all abnormal prediction information in the abnormal prediction information set is sent to the manual review platform.

[0147] Optionally, to improve the speed and accuracy of customer industry identification in this embodiment of the invention, manually reviewed samples can be used as new historical customer data for training. Optionally, this embodiment of the invention may further include:

[0148] If the manual review results include target second industry identification information corresponding to the abnormal prediction information, the target second industry identification information and the corresponding customer data in the manual review results will be used as new historical customer data.

[0149] Understandably, to ensure the security of financial transactions and protect assets, if manual review reveals that the industry information of some clients cannot be determined, further review can be conducted on these clients whose industry information cannot be determined. One example also includes:

[0150] If the manual review results include suspicious user identifiers corresponding to the abnormal prediction information, the customer data corresponding to the suspicious user identifiers will be sent to the abnormal transaction review platform to trigger the abnormal transaction review platform to review the customer data for abnormal transactions.

[0151] Among them, the abnormal transaction review platform can determine whether a user's transaction poses a financial risk, or it can further review the customer's industry information and conduct a higher level of monitoring of the customer's transaction behavior.

[0152] To meet the reporting requirements of financial regulatory agencies, the required reporting messages can be automatically generated based on industry forecast reference information and target industry identification information. Optionally, embodiments of the present invention also include:

[0153] From at least two preset industry terminology databases, determine the target industry terminology database corresponding to the target second industry identifier information;

[0154] Based on the target industry thesaurus, identify the target industry related terms that match the industry prediction reference information in the target industry thesaurus;

[0155] Obtain the customer analysis message template corresponding to the target second industry identifier information. The customer analysis message template includes fill instruction information for the positions to be filled. The fill instruction information is used to indicate the industry-related words that need to be filled in the positions to be filled.

[0156] Based on the fill instruction information, determine the industry related words to be filled in the target industry related words, and fill the industry related words to be filled in the corresponding fill position in the customer analysis report template to obtain the customer analysis report.

[0157] The customer analysis message template is a formatted message template set by developers to meet the requirements of financial regulatory authorities, based on actual applications. Different secondary industry identification information can correspond to different customer analysis message templates.

[0158] In one example, the fill instruction information can also be used to indicate the first customer description information that needs to be filled in the position to be filled. When filling the customer analysis message template, the first customer description information can also be filled into the corresponding position to be filled in the customer analysis message template according to different preset specific customer characteristics, based on the fill instruction information and the first customer description information.

[0159] As can be seen from the above, the embodiments of the present invention can reduce reliance on manual labor, save human resources, and improve the efficiency of determining the industry to which a customer belongs in the financial field while ensuring accuracy.

[0160] Based on the methods described in the preceding embodiments, the following examples will provide further detailed explanations.

[0161] In this embodiment, the combination Figure 1 The system will be explained.

[0162] like Figure 3 As shown, the customer industry determination method in this embodiment can be described in the following specific process:

[0163] 301. The terminal receives customer data submitted by the user and sends the customer data to be analyzed to the server.

[0164] Among them, customers or auditors can use the terminal such as Figure 4 On the page shown, fill in the customer data to be analyzed.

[0165] In one example, the customer or auditor clicks as shown in the image. Figure 4 The control named "Submit" triggers the terminal to generate customer data to be analyzed based on the information on the current page, and the terminal sends the customer data to be analyzed to the server.

[0166] 302. After receiving the customer data to be analyzed, the server determines the first customer description information in the customer data that matches the preset specific customer characteristics, and determines the information in the customer data other than the first customer description information as industry prediction reference information.

[0167] Optionally, the step "determine the first customer description information that belongs to the preset specific customer characteristics in the customer data to be analyzed" can also be completed directly on the terminal, saving computing resources on the server.

[0168] In one example, once the server receives the customer data to be analyzed, it can process the received customer data immediately.

[0169] In another example, a storage area for the information to be analyzed can be pre-configured in the server. The customer data to be analyzed is stored in the storage area, and a fixed processing time interval is set. After each fixed processing time interval, the customer data in the storage area is analyzed and processed, thus saving computing resources on the server.

[0170] The processing time interval can be 24 hours or a specific processing time, such as starting analysis and processing at 8:00 AM every workday.

[0171] 303. The server matches the industry prediction reference information with the industry terminology databases of at least two preset industries to determine the target words in the industry prediction reference information that match the industry-related words in the industry terminology databases.

[0172] For example, the server determines the industry prediction reference information as "selling vegetables in Market S, selling 100 jin of vegetables and 30 jin of braised pork daily, employing 1 person, with a warehouse occupying 30 square meters and a stall occupying 10 square meters." The industry terminology for the retail industry includes "vegetables," "pork," "stall," and "employment," while the terminology for the wholesale industry includes "vegetables," "pork," "warehouse," and "employment." After matching calculation, the target words in the industry prediction reference information corresponding to the retail industry are determined to be: vegetables, braised pork, stall, and employment; and the target words corresponding to the wholesale industry are: vegetables, braised pork, warehouse, and employment. Among these, the similarity between "braised pork" and "pork" is judged to be 60%.

[0173] 304. The server determines the relevance between the industry prediction reference information and each preset industry based on the industry relevance representation information of the industry related words of each preset industry and the target words matched by the industry prediction reference information in each industry thesaurus.

[0174] The relevance between industry-related terms and their respective preset industries can be a specific numerical value, such as 1 or 3. The correspondence between industry-related terms and their respective preset industries can include the industry prediction score for each industry-related term in different preset industries. For example, in the retail industry, the industry prediction score for the industry-related term "hire" is 2, and in the wholesale industry, the industry prediction score for the industry-related term "hire" is 5.

[0175] Specifically, when determining the correlation between industry forecast reference information and each preset industry, the industry forecast scores corresponding to each industry-related term in different industries can be directly added together to obtain the correlation between industry forecast reference information and each preset industry.

[0176] In one example, if the target term and industry-related terms do not perfectly match but have a certain degree of similarity, the industry prediction scores corresponding to the target term in different industries can be weighted based on the similarity to obtain the correlation between the industry prediction reference information and each preset industry. The specific calculation method can be designed by the developers according to the actual situation and is not limited here.

[0177] For example, if the target words in the industry forecast reference information corresponding to the retail industry are: vegetables, braised pork, stall, employment, and the target words corresponding to the wholesale industry are: vegetables, braised pork, warehouse, employment, then the similarity between braised pork and pork is judged to be 60%.

[0178] Specifically, in the retail industry, the industry prediction score can be set as follows: vegetables = 5, pork = 5, stalls = 6, and employment = 1. In the wholesale industry, the industry prediction score can be set as follows: vegetables = 5, pork = 5, warehouses = 6, and employment = 6.

[0179] The correlation between industry forecast reference information and the retail industry can be calculated as 5 + 5 * 60% + 6 + 1 = 15, and the correlation with the wholesale industry is 5 + 5 * 60% + 6 + 6 = 20.

[0180] 305. Based on the correlation between industry prediction reference information and each preset industry, the server determines the first candidate industry of the customer to which the customer data belongs, and obtains the first industry identification information of the first candidate industry based on the correspondence between the preset industry and the first industry identification information.

[0181] Developers can set the first industry identifier to a specific number. For example, the first industry identifier for the retail industry is 1, and the first industry identifier for the wholesale industry is 2.

[0182] For example, if the industry forecast reference information has the highest relevance to the wholesale industry, then the wholesale industry will be selected as the first candidate industry, and the first industry identifier information will be 2.

[0183] 306. The server converts the first customer description information and the first industry identification information into feature vectors, and uses the converted feature vectors as the target industry description vectors corresponding to the customer data.

[0184] In one example, the first customer description information includes: monthly transaction frequency of 3 and first industry identifier of 2. The generated target industry description vector can be represented as (3, 2).

[0185] 307. The server calculates the distance between the target industry description vector and the description vectors of each industry, and uses the distance as the correlation between the target industry description vector and the description vectors of each industry.

[0186] The mapping relationship can include a multi-dimensional vector space determined based on preset specific customer characteristics and industry categories, as well as an industry description vector in the vector space determined based on historical customer data.

[0187] For example, the target industry description vector can be represented as (3, 2), and the mapping relationship also provides industry description vectors A (2, 2) and B (3, 2). The server can calculate the distance between the target industry description vector and industry description vectors A and B respectively, and use the obtained distance as the relevance between the target industry description vector and each industry description vector.

[0188] 308. The server selects a preset number of industry description vectors from the industry description vectors based on relevance as reference industry description vectors.

[0189] There can be multiple industry description vectors, with a preset number not exceeding the total number of industry description vectors.

[0190] When selecting reference industry description vectors, you can first sort the relevance, then select the relevance with the highest relevance according to the reference vector selection parameters, and use the industry description vector corresponding to the selected relevance as the reference industry description vector.

[0191] For example, given industry description vectors A, B, C, and D, and the distances between the target industry description vector and A, B, C, and D being 5, 3, 4, and 1 respectively, with a preset quantity of 3, sorting them according to their distances to the target industry description vectors, the order of distances from largest to smallest is 5, 4, 3, and 1. The three industry description vectors with the smallest distances can be selected as reference industry description vectors according to the preset quantity; that is, the selected reference industry description vectors are D, B, and C.

[0192] 309. Based on the second industry identification information of the reference industry description vector, the server counts the number of reference industry description vectors corresponding to the target industry description vector under each second industry identification information, and determines the target second industry identification information corresponding to the customer data based on the number of reference industry description vectors.

[0193] For example, if the second industry identification information corresponding to the three selected reference industry description vectors are 1, 1, and 2 respectively, then the number of reference industry description vectors for customer data in the retail industry with second industry identification information of 1 is 2, and the number of reference industry description vectors in the wholesale industry with second industry identification information of 2 is 1. Therefore, it can be determined that the second industry identification information corresponding to customer data is 1, and the target second industry identification information is 1.

[0194] In one example, based on the reference industry description vector and the corresponding second industry identifier information, the number of reference industry description vectors corresponding to customer data under different second industry identifier information can be obtained. The second industry identifier information corresponding to the highest number of reference industry description vectors is taken as the target second industry identifier information corresponding to customer data.

[0195] 310. The server calculates the ratio between the number of reference industry description vectors corresponding to each second industry identifier and the preset number, and obtains the vote difference value corresponding to each second industry identifier. If the vote difference value of the second industry identifier is greater than the preset difference value, step 313 is executed.

[0196] In one example, the ratio between each number of identifiers and the reference vector selection parameter can be calculated separately, and each ratio can be compared with a preset ratio.

[0197] In another example, the ratio between the highest number of identifiers and the reference vector selection parameter can be directly calculated and compared with a preset ratio.

[0198] 311. If the difference in votes for the second industry identifier information is not greater than the preset difference, the server will send the customer data and the number of reference industry description vectors corresponding to the customer data under each second industry identifier information as anomaly prediction information to the manual review platform for manual review.

[0199] If, after calculating the ratio between each number of identifiers and the reference vector selection parameter and comparing each ratio with the preset ratio, no ratio is greater than the preset ratio, or the ratio between the highest number of identifiers and the reference vector selection parameter is not greater than the preset ratio, then manual review is required.

[0200] 312. The terminal receives the abnormal prediction information sent by the server and displays it through the manual review platform so that the user can select the corresponding industry based on the abnormal prediction information. The terminal generates target industry identification information based on the user's selection and sends it to the server.

[0201] In one example, the terminal can transmit anomaly prediction information as follows: Figure 5 As shown in 501, the terminal will prioritize displaying the anomaly prediction information. Figure 5 In the 501 form, the "Number of Tickets" column represents the ratio between the highest number of identification information for each customer and the reference vector selection parameter; the "Predicted Industry" column prioritizes displaying the industry name corresponding to the highest number of identification information; the "Customer Information" column can display the customer's unique ID, customer's name, or customer data, etc.

[0202] If the "Customer Information" section only displays the customer's unique ID, such as "Customer No. 1", the auditor clicks... Figure 5 The "Customer No. 1" section in 501 displays the customer data corresponding to Customer No. 1.

[0203] In one example, if the reviewer, after analyzing customer data, believes that the industry prediction is inaccurate, they can click as shown. Figure 5 The controls shown in 501 are named "Retail" and "Wholesale," and drop-down menus display other industry options for selection.

[0204] Understandably, to ensure the accuracy of clients' industry information, auditors can also review normal forecast information by clicking on it. Figure 5 The control named "Vote Count (n>set)" shown in 501 allows the terminal to... Figure 5 As shown in 502, normal prediction information is displayed first.

[0205] Once the reviewers confirm that there are no issues with the industry assessment, they can click on the following: Figure 5 The button labeled "Submit" shown in 501 or 502 allows the terminal to generate target industry representation information based on the predicted industry and send it to the server.

[0206] 313. The server sets the target second industry identifier information to the industry identifier information of the customer to which the customer data belongs.

[0207] For example, when setting the target second industry identifier information to the industry identifier information of the customer to which the customer data belongs, the customer data originally contained industry identifier information, and the original industry identifier information was updated to the target second industry identifier information.

[0208] 314. The server generates customer analysis messages based on industry forecast reference information and customer analysis message templates.

[0209] Understandably, when generating transaction information in a preset format, the server can also fill in the customer analysis message template based on the first customer description information.

[0210] 315. The terminal retrieves and displays the generated customer analysis message from the server. After the user confirms that there are no problems with the customer analysis message through the terminal, the user can submit the customer analysis message through the terminal.

[0211] like Figure 6 As shown, the reviewer clicked... Figure 6 The 601 error message control named "Automatically Generate Message" allows the terminal to retrieve customer analysis messages corresponding to the "Retail" industry from the server based on the click operation, resulting in messages such as... Figure 6 The message shown is 602.

[0212] In one example, if the auditor believes that the customer's analysis message is incorrect, they can click as shown below. Figure 6 The control shown in 602, named "Clear Edit Box", clears the customer analysis message on the page, or allows you to directly modify the customer analysis message in the edit box.

[0213] In another example, if the auditor believes that the industry corresponding to the customer data is incorrect, they can click as shown. Figure 6 The control shown, named "Retail," allows you to reselect the industry corresponding to the customer.

[0214] Optionally, if the auditor confirms that there are no issues with the client's analysis report, they can directly click "as shown". Figure 6 The control shown, named "Submit," submits the client's analysis message.

[0215] As can be seen from the above, the embodiments of the present invention can reduce the reliance on manual labor when determining the industry of a customer, save human resources, and improve the efficiency of determining the industry of a customer in the financial industry while ensuring accuracy.

[0216] To better implement the above methods, the present invention also provides a customer industry determination device.

[0217] refer to Figure 7 The device includes:

[0218] The information determination unit 701 is used to determine the first customer description information in the customer data to be analyzed that matches the preset specific customer characteristics, and to determine the information in the customer data other than the first customer description information as industry prediction reference information.

[0219] The matching unit 702 is used to match the industry prediction reference information with the industry thesaurus of at least two preset industries to determine the target words in the industry prediction reference information that match the industry-related words in the industry thesaurus.

[0220] The first industry determination unit 703 is used to determine the first candidate industry of the customer to which the customer data belongs from the preset industries based on the target words corresponding to each preset industry in the industry prediction reference information, and to obtain the first industry identification information of the first candidate industry based on the correspondence between the preset industries and the first industry identification information.

[0221] The vector conversion unit 704 is used to convert the first customer description information and the first industry identification information into feature vectors, and use the converted feature vectors as the target industry description vectors corresponding to the customer data.

[0222] The second industry determination unit 705 is used to determine the target second industry identification information corresponding to the customer data based on the preset mapping relationship between the industry description vector and the second industry identification information, as well as the target industry description vector, and set the target second industry identification information as the industry identification information of the customer to which the customer data belongs.

[0223] In an optional example, the first industry determination unit 703 includes a first candidate industry determination unit 704, which is used to determine the correlation between the industry prediction reference information and each preset industry based on the industry correlation characterization information of the industry related words of each preset industry and the target words matched by the industry prediction reference information under each industry thesaurus. The industry correlation characterization information is used to characterize the correlation between the industry related words and the preset industry to which they belong.

[0224] Based on industry forecast reference information and the correlation between various preset industries, the first candidate industry of the customer to which the customer data belongs is determined.

[0225] In an optional example, the first candidate industry determination unit 704 can also be used to obtain the similarity between the target word and the industry-related words that match the target word;

[0226] Based on the industry relevance representation information and similarity of industry-related terms for each preset industry, the relevance between the target term and each preset industry is determined;

[0227] Based on the target words in the industry forecast reference information and the correlation between the target words and each preset industry, the correlation between the industry forecast reference information and each preset industry is determined.

[0228] In an optional example, such as Figure 8 As shown, before the information determination unit 701, there is also a mapping relationship establishment unit 706, which is used to obtain the historical industry description vector of historical customer data and the second industry identification information of historical customers;

[0229] A mapping relationship is established between the second industry identification information and the historical industry description vector belonging to the same historical customer, resulting in a preset mapping relationship between the industry description vector and the second industry identification information.

[0230] Correspondingly, the second industry determination unit 705 includes a second industry identification information determination unit 707, which is used to determine the target industry description vector and the correlation between it and the description vectors of each industry in the preset mapping relationship;

[0231] Based on the relevance of each industry description vector and the second industry identification information corresponding to each industry description vector, the target second industry identification information corresponding to the customer data is determined.

[0232] In one example, the second industry identification information determination unit 707 includes a relevance determination subunit 708, which is used to calculate the distance between the target industry description vector and each industry description vector, and use the distance as the relevance between the target industry description vector and each industry description vector.

[0233] Correspondingly, the second industry identification information determination unit 707 also includes a target second industry identification information determination unit 709, which is used to select a preset number of industry description vectors as reference industry description vectors from the industry description vectors based on relevance.

[0234] Based on the second industry identification information corresponding to the reference industry description vector, count the number of reference industry description vectors under each second industry identification information.

[0235] Based on the number of reference industry description vectors under each second industry identifier, the target second industry identifier corresponding to the customer data is determined.

[0236] In one example, the target second industry identification information determination unit 709 can also be used to calculate the ratio between the number of reference industry description vectors corresponding to each second industry identification information and the preset number, so as to obtain the vote difference corresponding to each second industry identification information.

[0237] If there is a second industry identifier information whose vote difference is greater than the preset difference, determine the second industry identifier information corresponding to the customer data from the second industry identifier information whose vote difference is greater than the preset difference, and obtain the target second industry identifier information;

[0238] If there is no second industry identification information where the difference in votes is greater than the preset difference, the customer data and the number of reference industry description vectors corresponding to the customer data under each second industry identification information are stored as anomaly prediction information in the anomaly prediction information set.

[0239] Send the abnormal prediction information from the abnormal prediction information set to the manual review platform;

[0240] Receive the manual review results for abnormal prediction information sent by the manual review platform. If the manual review results include target industry identification information set for customer data, determine the target industry identification information as the target second industry identification information corresponding to the customer data.

[0241] In one example, after the second industry determination unit 705, there is also a message generation unit 710, which is used to determine the target industry terminology corresponding to the target second industry identification information from at least two preset industry terminology databases.

[0242] Based on the target industry thesaurus, identify the target industry related terms that match the industry prediction reference information in the target industry thesaurus;

[0243] Obtain the customer analysis message template corresponding to the target second industry identifier information. The customer analysis message template includes fill instruction information for the positions to be filled. The fill instruction information is used to indicate the industry-related words that need to be filled in the positions to be filled.

[0244] Based on the fill instruction information, determine the industry related words to be filled in the target industry related words, and fill the industry related words to be filled in the corresponding fill position in the customer analysis report template to obtain the customer analysis report.

[0245] As can be seen from the above, the customer industry determination device can reduce the reliance on manual labor in determining the customer industry, save human resources, and improve the efficiency of determining the customer's industry in the financial industry while ensuring accuracy.

[0246] Furthermore, embodiments of the present invention also provide an electronic device, which may be a terminal or a server, such as... Figure 9 As shown, it illustrates a structural schematic diagram of the electronic device involved in an embodiment of the present invention, specifically:

[0247] The electronic device may include a radio frequency (RF) circuit 901, a memory 902 including one or more computer-readable storage media, an input unit 903, a display unit 904, a sensor 905, an audio circuit 906, a wireless fidelity (WiFi) module 907, a processor 908 including one or more processing cores, and a power supply 909, etc. Those skilled in the art will understand that... Figure 6 The terminal structure shown does not constitute a limitation on the terminal and may include more or fewer components than shown, or combine certain components, or have different component arrangements. Wherein:

[0248] RF circuit 901 can be used for receiving and transmitting signals during information transmission or calls. Specifically, it receives downlink information from the base station and hands it over to one or more processors 908 for processing; additionally, it transmits uplink data to the base station. Typically, RF circuit 901 includes, but is not limited to, an antenna, at least one amplifier, a tuner, one or more oscillators, a Subscriber Identity Module (SIM) card, a transceiver, a coupler, a low-noise amplifier (LNA), a duplexer, etc. Furthermore, RF circuit 901 can also communicate wirelessly with networks and other devices. Wireless communication can use any communication standard or protocol, including but not limited to GSM, GPRS, CDMA, WCDMA, LTE, email, and SMS.

[0249] The memory 902 can be used to store software programs and modules. The processor 908 executes various functional applications and data processing by running the software programs and modules stored in the memory 902. The memory 902 may mainly include a program storage area and a data storage area. The program storage area may store the operating system, application programs required for at least one function (such as sound playback function, image playback function, etc.), etc.; the data storage area may store data created according to the use of the terminal (such as audio data, phone book, etc.). In addition, the memory 902 may include high-speed random access memory, and may also include non-volatile memory, such as at least one disk storage device, flash memory device, or other volatile solid-state storage device. Accordingly, the memory 902 may also include a memory controller to provide access to the memory 902 for the processor 908 and the input unit 903.

[0250] Input unit 903 can be used to receive input digital or character information, and to generate keyboard, mouse, joystick, optical, or trackball signal inputs related to user settings and function control. Specifically, in one embodiment, input unit 903 may include a touch-sensitive surface and other input devices. A touch-sensitive surface, also known as a touch display or touchpad, can collect user touch operations on or near it (e.g., user operations using fingers, styluses, or any suitable object or accessory on or near the touch-sensitive surface) and drive corresponding connection devices according to a pre-set program. Optionally, the touch-sensitive surface may include a touch detection device and a touch controller. The touch detection device detects the user's touch position and the signal generated by the touch operation, transmitting the signal to the touch controller; the touch controller receives touch information from the touch detection device, converts it into touch point coordinates, sends it to the processor 908, and can receive and execute commands from the processor 908. Furthermore, various types of touch-sensitive surfaces, such as resistive, capacitive, infrared, and surface acoustic wave, can be used. In addition to the touch-sensitive surface, input unit 903 may also include other input devices. Specifically, other input devices may include, but are not limited to, one or more of the following: physical keyboard, function keys (such as volume control buttons, power buttons, etc.), trackball, mouse, joystick, etc.

[0251] Display unit 904 can be used to display information input by the user or information provided to the user, as well as various graphical user interfaces of the terminal. These graphical user interfaces can be composed of graphics, text, icons, video, and any combination thereof. Display unit 904 may include a display panel, optionally configured as a liquid crystal display (LCD), organic light-emitting diode (OLED), or similar form. Furthermore, a touch-sensitive surface may cover the display panel. When the touch-sensitive surface detects a touch operation on or near it, it transmits the information to processor 908 to determine the type of touch event. Subsequently, processor 908 provides corresponding visual output on the display panel according to the type of touch event. Although in Figure 6 In this context, the touch-sensitive surface and the display panel are two separate components for implementing input and output functions. However, in some embodiments, the touch-sensitive surface and the display panel can be integrated to achieve both input and output functions.

[0252] The terminal may also include at least one sensor 905, such as a light sensor, a motion sensor, and other sensors. Specifically, the light sensor may include an ambient light sensor and a proximity sensor. The ambient light sensor can adjust the brightness of the display panel according to the ambient light level, and the proximity sensor can turn off the display panel and / or backlight when the terminal is moved to the ear. As a type of motion sensor, a gravity acceleration sensor can detect the magnitude of acceleration in various directions (generally three axes). When stationary, it can detect the magnitude and direction of gravity and can be used for applications that recognize the phone's posture (such as landscape / portrait switching, related games, magnetometer posture calibration), vibration recognition-related functions (such as pedometer, tapping), etc. Other sensors that the terminal may also be equipped with, such as gyroscopes, barometers, hygrometers, thermometers, and infrared sensors, will not be described in detail here.

[0253] Audio circuitry 906, a speaker, and a microphone provide an audio interface between the user and the terminal. Audio circuitry 906 converts received audio data into electrical signals, transmits them to the speaker, and the speaker converts them into sound signals for output. Conversely, the microphone converts collected sound signals into electrical signals, which are then received by audio circuitry 906, converted back into audio data, and processed by processor 908. The processed data is then transmitted via RF circuitry 901 to, for example, another terminal, or output to memory 902 for further processing. Audio circuitry 906 may also include an earphone jack to facilitate communication between a peripheral headset and the terminal.

[0254] WiFi is a short-range wireless transmission technology. Terminals using the WiFi module 907 can help users send and receive emails, browse web pages, and access streaming media, providing wireless broadband internet access. Although Figure 6 WiFi module 907 is shown, but it is understood that it is not a necessary component of the terminal and can be omitted as needed without changing the essence of the invention.

[0255] The processor 908 is the control center of the terminal, connecting various parts of the phone via various interfaces and lines. It executes various terminal functions and processes data by running or executing software programs and / or modules stored in the memory 902, and by calling data stored in the memory 902. Optionally, the processor 908 may include one or more processing cores; preferably, the processor 908 may integrate an application processor and a modem processor, wherein the application processor mainly handles the operating system, user interface, and applications, and the modem processor mainly handles wireless communication. It is understood that the modem processor may also not be integrated into the processor 908.

[0256] The terminal also includes a power supply 909 (such as a battery) that supplies power to various components. Preferably, the power supply can be logically connected to the processor 908 through a power management system, thereby enabling functions such as charging, discharging, and power consumption management through the power management system. The power supply 909 may also include one or more DC or AC power supplies, recharging systems, power fault detection circuits, power converters or inverters, power status indicators, and other arbitrary components.

[0257] Although not shown, the terminal may also include a camera, Bluetooth module, etc., which will not be described in detail here. Specifically, in this embodiment, the processor 908 in the terminal loads the executable files corresponding to the processes of one or more applications into the memory 902 according to the following instructions, and the processor 908 runs the applications stored in the memory 902 to realize various functions, as follows:

[0258] Identify the first customer description information in the customer data to be analyzed that matches the preset specific customer characteristics, and identify the information in the customer data other than the first customer description information as industry prediction reference information.

[0259] The industry forecast reference information is matched with the industry thesaurus of at least two preset industries to determine the target words in the industry forecast reference information that match the industry-related words in the industry thesaurus.

[0260] Based on the target words corresponding to each preset industry in the industry forecast reference information, determine the first candidate industry of the customer to which the customer data belongs from the preset industries, and obtain the first industry identification information of the first candidate industry based on the correspondence between the preset industries and the first industry identification information.

[0261] The first customer description information and the first industry identification information are converted into feature vectors, and the converted feature vectors are used as the target industry description vectors corresponding to the customer data.

[0262] Based on the preset mapping relationship between the industry description vector and the second industry identification information, and the target industry description vector, the target second industry identification information corresponding to the customer data is determined, and the target second industry identification information is set as the industry identification information of the customer to which the customer data belongs.

[0263] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be performed by instructions, or by instructions controlling related hardware. These instructions can be stored in a computer-readable storage medium and loaded and executed by a processor.

[0264] To this end, embodiments of the present invention provide a storage medium storing a plurality of instructions that can be loaded by a processor to execute steps in any of the customer industry determination methods provided in embodiments of the present invention. For example, the instructions can execute the following steps:

[0265] Identify the first customer description information in the customer data to be analyzed that matches the preset specific customer characteristics, and identify the information in the customer data other than the first customer description information as industry prediction reference information.

[0266] The industry forecast reference information is matched with the industry thesaurus of at least two preset industries to determine the target words in the industry forecast reference information that match the industry-related words in the industry thesaurus.

[0267] Based on the target words corresponding to each preset industry in the industry forecast reference information, determine the first candidate industry of the customer to which the customer data belongs from the preset industries, and obtain the first industry identification information of the first candidate industry based on the correspondence between the preset industries and the first industry identification information.

[0268] The first customer description information and the first industry identification information are converted into feature vectors, and the converted feature vectors are used as the target industry description vectors corresponding to the customer data.

[0269] Based on the preset mapping relationship between the industry description vector and the second industry identification information, and the target industry description vector, the target second industry identification information corresponding to the customer data is determined, and the target second industry identification information is set as the industry identification information of the customer to which the customer data belongs.

[0270] For details on the implementation of each of the above operations, please refer to the previous examples, which will not be repeated here.

[0271] The storage medium may include: read-only memory (ROM), random access memory (RAM), disk or optical disk, etc.

[0272] Since the instructions stored in the storage medium can execute the steps in any of the customer industry determination methods provided in the embodiments of the present invention, the beneficial effects that any of the customer industry determination methods provided in the embodiments of the present invention can achieve can be realized, as detailed in the preceding embodiments, and will not be repeated here.

[0273] According to one aspect of this application, a computer program product or computer program is also provided, comprising computer instructions stored in a computer-readable storage medium. A processor of an electronic device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the electronic device to perform the methods provided in the various optional implementations of the above embodiments.

[0274] The foregoing has provided a detailed description of a customer industry determination method, apparatus, electronic device, and storage medium provided by embodiments of the present invention. Specific examples have been used to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of the present invention. At the same time, for those skilled in the art, there will be changes in specific implementation methods and application scope based on the ideas of the present invention. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A method for determining a customer's industry, characterized in that, include: The first customer description information in the customer data to be analyzed that matches the preset specific customer characteristics is determined, and the information in the customer data other than the first customer description information is determined as industry prediction reference information. The industry prediction reference information is matched with industry thesaurus of at least two preset industries to determine the target words in the industry prediction reference information that match the industry-related words in the industry thesaurus. Based on the target words corresponding to each preset industry in the industry prediction reference information, the first candidate industry of the customer to which the customer data belongs is determined from the preset industries. Based on the correspondence between the preset industries and the first industry identification information, the first industry identification information of the first candidate industry is obtained. The first customer description information and the first industry identification information are converted into feature vectors, and the converted feature vectors are used as the target industry description vectors corresponding to the customer data. Based on the preset mapping relationship between the industry description vector and the second industry identification information, and the target industry description vector, the target second industry identification information corresponding to the customer data is determined, and the target second industry identification information is set as the industry identification information of the customer to which the customer data belongs.

2. The customer industry determination method according to claim 1, characterized in that, The step of determining the first candidate industry of the customer to which the customer data belongs from the preset industries based on the target words corresponding to each preset industry in the industry prediction reference information includes: Based on the industry relevance representation information of industry-related terms for each preset industry, and the target terms matched by the industry prediction reference information in each industry terminology library, the relevance between the industry prediction reference information and each preset industry is determined. The industry relevance representation information is used to represent the relevance between the industry-related terms and their respective preset industries. Based on the correlation between the industry forecast reference information and each preset industry, the first candidate industry of the customer to which the customer data belongs is determined.

3. The customer industry determination method according to claim 2, characterized in that, The step of determining the relevance between the industry prediction reference information and each preset industry based on the industry relevance representation information of industry-related terms for each preset industry, and the target terms matched by the industry prediction reference information in each industry thesaurus, includes: Obtain the similarity between the target word and the industry-related words that match the target word; Based on the industry relevance representation information of industry-related terms for each preset industry and the similarity, the relevance between the target term and each preset industry is determined; Based on the target words in the industry forecast reference information and the correlation between the target words and each preset industry, the correlation between the industry forecast reference information and each preset industry is determined.

4. The customer industry determination method according to claim 1, characterized in that, Before determining the first customer description information in the customer data to be analyzed that matches a preset specific customer characteristic, the method further includes: Obtain the historical industry description vector of the historical customer data and the second industry identification information of the historical customer; Establish a mapping relationship between the second industry identification information and the historical industry description vector belonging to the same historical customer, and obtain the preset mapping relationship between the industry description vector and the second industry identification information; The step of determining the target second industry identifier information corresponding to the customer data based on the preset mapping relationship between the industry description vector and the second industry identifier information, and the target industry description vector, includes: Determine the correlation between the target industry description vector and the industry description vectors in the preset mapping relationship; Based on the relevance corresponding to each industry description vector and the second industry identification information corresponding to each industry description vector, the target second industry identification information corresponding to the customer data is determined.

5. The customer industry determination method according to claim 4, characterized in that, Determining the relevance between the target industry description vector and the industry description vectors in the preset mapping relationship includes: Calculate the distance between the target industry description vector and each of the industry description vectors, and use the distance as the correlation between the target industry description vector and each of the industry description vectors; The step of determining the target second industry identifier information corresponding to the customer data based on the relevance of each industry description vector and the second industry identifier information corresponding to each industry description vector includes: Based on the relevance, a preset number of industry description vectors are selected from the industry description vectors as reference industry description vectors; Based on the second industry identification information corresponding to the reference industry description vector, the number of reference industry description vectors under each second industry identification information is counted. Based on the number of reference industry description vectors under each second industry identifier, the target second industry identifier corresponding to the customer data is determined.

6. The customer industry determination method according to claim 5, characterized in that, The determination of the target second industry identification information corresponding to the customer data based on the number of reference industry description vectors under each second industry identification information includes: Calculate the ratio between the number of reference industry description vectors corresponding to each second industry identifier and the preset number to obtain the vote difference corresponding to each second industry identifier. If there is a second industry identifier information whose vote difference is greater than a preset difference, determine the second industry identifier information corresponding to the customer data from the second industry identifier information whose vote difference is greater than the preset difference, and obtain the target second industry identifier information; If there is no second industry identification information where the difference in votes is greater than the preset difference, the customer data and the number of reference industry description vectors corresponding to the customer data under each second industry identification information are stored as anomaly prediction information in the anomaly prediction information set. Send the abnormal prediction information in the abnormal prediction information set to the manual review platform; The system receives the manual review results for the abnormal prediction information sent by the manual review platform. If the manual review results include target industry identification information set for the customer data, the target industry identification information is determined as the target second industry identification information corresponding to the customer data.

7. The method for determining the customer industry according to any one of claims 1-6, characterized in that, Also includes: From the industry terminology databases of the at least two preset industries, determine the target industry terminology database corresponding to the target second industry identifier information; Based on the target industry thesaurus, determine the target industry related words in the target industry thesaurus that match the industry prediction reference information; Obtain the customer analysis message template corresponding to the target second industry identification information. The customer analysis message template includes filling instruction information for the position to be filled. The filling instruction information is used to indicate the industry-related words that need to be filled at the position to be filled. Based on the filling instruction information, the industry related words to be filled in the target industry related words are determined, and the industry related words to be filled in are filled into the corresponding filling positions in the customer analysis message template to obtain the customer analysis message.

8. A customer industry determination device, characterized in that, include: The information determination unit is used to determine the first customer description information in the customer data to be analyzed that matches the preset specific customer characteristics, and to determine the information in the customer data other than the first customer description information as industry prediction reference information. The matching unit is used to match the industry prediction reference information with industry thesaurus of at least two preset industries to determine the target words in the industry prediction reference information that match the industry-related words in the industry thesaurus. The first industry determination unit is used to determine the first candidate industry of the customer to which the customer data belongs from the preset industries based on the target words corresponding to each preset industry in the industry prediction reference information, and to obtain the first industry identification information of the first candidate industry based on the correspondence between the preset industries and the first industry identification information. The vector conversion unit is used to convert the first customer description information and the first industry identification information into feature vectors, and use the converted feature vectors as the target industry description vectors corresponding to the customer data. The second industry determination unit is used to determine the target second industry identification information corresponding to the customer data based on the preset mapping relationship between the industry description vector and the second industry identification information, and the target industry description vector, and set the target second industry identification information as the industry identification information of the customer to which the customer data belongs.

9. An electronic device, characterized in that, It includes a memory and a processor; the memory stores an application program, and the processor is used to run the application program within the memory to perform the operations in the customer industry determination method according to any one of claims 1 to 7.

10. A storage medium, characterized in that, The storage medium stores a plurality of instructions adapted for loading by a processor to execute the steps of the customer industry determination method according to any one of claims 1 to 7.

11. A computer program product, characterized in that, The computer program product includes computer instructions stored in a computer-readable storage medium; the processor of the electronic device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the electronic device to perform the steps in the customer industry determination method according to any one of claims 1 to 7.

Images