A method, system, device, and medium for hierarchical classification of user information.
By preprocessing user information and vectorizing words to generate a tag vector space, and using a hierarchical classification matrix to determine similarity, the problems of low accuracy and high manual cost in traditional methods are solved, and efficient automatic hierarchical classification is achieved.
Patent Information
- Application Number
- CN202311297507.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2023-10-08
- Publication Date
- 2026-11-10
- Estimated Expiration
- 2043-10-08
AI Technical Summary
Traditional methods for classifying and categorizing financial and insurance data are not very accurate and require manual annotation, resulting in high labor costs.
By acquiring and preprocessing user information, a tag vector space is generated, and a pre-defined hierarchical classification matrix is used to determine similarity, thereby automatically classifying user information into hierarchical categories.
It improved the accuracy of user information classification and hierarchical classification, and reduced labor costs.
Smart Images

Figure CN117520539B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of natural language processing technology, and in particular to a method, system, device, and medium for hierarchical classification of user information. Background Technology
[0002] In recent years, with the continuous development of society, the amount of financial and insurance-related data has been growing at an increasingly rapid pace. Under these circumstances, in order to perform timely and reliable operations such as querying, deleting, and adding financial and insurance data, people have put forward the need for hierarchical classification of financial and insurance data (mainly user information).
[0003] Currently, traditional methods for classifying and categorizing financial and insurance data generally rely on simple classification algorithms. These algorithms are not very accurate when classifying user information, and they require manual labeling of data levels and categories, resulting in high labor costs.
[0004] Therefore, the problems existing in the current technology still need to be solved and optimized. Summary of the Invention
[0005] The purpose of this invention is to at least partially solve one of the technical problems existing in the related art.
[0006] Therefore, the first objective of this invention is to provide a hierarchical classification method for user information, which can effectively improve the accuracy of hierarchical classification of user information and reduce labor costs.
[0007] The second objective of this application is to provide a hierarchical classification system for user information.
[0008] To achieve the above-mentioned technical objectives, the technical solutions adopted in the embodiments of this application include:
[0009] In a first aspect, embodiments of this application provide a hierarchical classification method for user information, including:
[0010] Obtain user information to be classified and categorized;
[0011] The user information is preprocessed to obtain first information, which is used to characterize the keyword field and tag field of the user information;
[0012] The first information is processed into word vectors to generate a tag vector space;
[0013] Based on the preset hierarchical classification matrix, the similarity of the tag vector space is judged to obtain the hierarchical classification result of the user information.
[0014] In addition, the hierarchical classification method for user information according to the above embodiments of this application may also have the following additional technical features:
[0015] Furthermore, in one embodiment of this application, the preprocessing of the user information to be classified and categorized to obtain first information includes:
[0016] The user information is mapped and queried to obtain tag information;
[0017] The tag information and the user information are subjected to data cleaning, data transformation, and data standardization to obtain intermediate tag information;
[0018] The intermediate tag information and the user information are subjected to global word grouping processing to obtain the first information.
[0019] Furthermore, in one embodiment of this application, the step of performing word vectorization processing on the first information to generate a tag vector space includes:
[0020] The first information is transformed into a word vector space to obtain a tag matrix;
[0021] The label matrix is reduced in dimensionality to obtain the label vector space.
[0022] Furthermore, in one embodiment of this application, the step of performing a similarity judgment on the label vector space based on a preset hierarchical classification matrix to obtain a hierarchical classification result includes:
[0023] Obtain the preset similarity threshold;
[0024] A similarity score is obtained by performing a similarity judgment on the current vector element in the label vector space and the first sub-matrix element in the hierarchical classification matrix;
[0025] The similarity score is subjected to threshold filtering based on the similarity threshold to obtain a first similarity score;
[0026] The hierarchical classification result is determined based on the second sub-matrix element corresponding to the largest first similarity score and the current vector element;
[0027] The hierarchical classification matrix includes several matrix elements, each matrix element including a first sub-matrix element and a second sub-matrix element. The first sub-matrix element is used to represent the matrix indicator field, and the second sub-matrix element is used to represent the category and level corresponding to the matrix indicator field.
[0028] Furthermore, in one embodiment of this application, the hierarchical classification matrix is obtained through the following steps:
[0029] Obtain a standard training set, which includes standard keyword data, as well as level data and category data corresponding to the standard keyword data;
[0030] The standard training set is encoded and mapped to obtain a sequence dataset;
[0031] The sequence dataset is decoded and activated to obtain the hierarchical classification matrix.
[0032] Furthermore, in one embodiment of this application, the step of performing a similarity judgment on the current vector element of the label vector space and the first sub-matrix element in the hierarchical classification matrix to obtain a similarity score includes:
[0033] The distance between the current vector element in the label vector space and the first sub-matrix element in the hierarchical classification matrix is processed to obtain the first distance;
[0034] The similarity score is generated based on the first distance.
[0035] Furthermore, in one embodiment of this application, the hierarchical classification method for user information further includes:
[0036] Based on the classification results, obtain the updated dataset;
[0037] The current hierarchical classification matrix is updated based on the updated dataset to obtain the next hierarchical classification matrix.
[0038] Secondly, embodiments of this application provide a hierarchical classification system for user information, including:
[0039] The acquisition module is used to acquire user information to be classified and categorized.
[0040] The processing module is used to preprocess the user information to obtain first information, wherein the first information is used to characterize the keyword field and tag field of the user information;
[0041] The vector module is used to perform word vectorization processing on the first information to generate a tag vector space;
[0042] The judgment module is used to perform similarity judgment on the tag vector space according to the preset hierarchical classification matrix to obtain the hierarchical classification result of the user information.
[0043] Thirdly, embodiments of this application also provide a computer device, including:
[0044] At least one processor;
[0045] At least one memory for storing at least one program;
[0046] When the at least one program is executed by the at least one processor, the at least one processor implements the hierarchical classification method for user information described above.
[0047] Fourthly, embodiments of this application also provide a computer-readable storage medium storing a processor-executable program, which, when executed by the processor, is used to implement the hierarchical classification method for user information described above.
[0048] The advantages and beneficial effects of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this application:
[0049] This application discloses a hierarchical classification method for user information. The method involves: acquiring user information to be classified; preprocessing the user information to obtain first information, which represents the keyword and tag fields of the user information; performing word vectorization on the first information to generate a tag vector space; and performing similarity judgment on the tag vector space according to a preset hierarchical classification matrix to obtain the hierarchical classification result of the user information. This hierarchical classification method can effectively improve the accuracy of user information hierarchical classification and reduce manual costs. Attached Figure Description
[0050] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the following description is provided with accompanying drawings of the relevant technical solutions in the embodiments of this application or the prior art. It should be understood that the accompanying drawings described below are only for the purpose of clearly illustrating some embodiments of the technical solutions in this application. For those skilled in the art, other drawings can be obtained based on these drawings without any creative effort.
[0051] Figure 1 A flowchart illustrating a hierarchical classification method for user information provided in an embodiment of this application;
[0052] Figure 2 A schematic diagram of the structure of a hierarchical classification system for user information provided in this application embodiment;
[0053] Figure 3 This is a schematic diagram of the structure of a computer device provided in an embodiment of this application. Detailed Implementation
[0054] The embodiments of this application are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain this application, and should not be construed as limiting this application. The step numbers in the following embodiments are set only for ease of explanation, and there is no limitation on the order between the steps. The execution order of each step in the embodiments can be adaptively adjusted according to the understanding of those skilled in the art.
[0055] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.
[0056] It should be noted that in all specific embodiments of this application, when processing data related to user identity or characteristics, such as user information, user behavior data, user historical data, and user location information, user permission or consent is obtained first. Furthermore, the collection, use, and processing of this data comply with relevant laws, regulations, and standards. In addition, when embodiments of this application require access to sensitive personal information of users, separate permission or consent from the user is obtained through pop-ups or redirection to confirmation pages. Only after obtaining the user's separate permission or consent is the necessary user-related data required for the proper functioning of these embodiments acquired.
[0057] Currently, traditional methods for classifying and categorizing financial and insurance data generally rely on simple, single-iteration classification algorithms. These algorithms are not very accurate when classifying user information, and they require manual labeling of data levels and categories, resulting in high labor costs.
[0058] In view of this, embodiments of the present invention provide a hierarchical classification method for user information, which can effectively improve the accuracy of hierarchical classification of user information and reduce labor costs.
[0059] Reference Figure 1 In this application embodiment, a hierarchical classification method for user information includes:
[0060] Step 110: Obtain user information to be classified and categorized;
[0061] It is understood that, in the embodiments of this application, the user information to be classified and categorized may be the information entered by the user when registering an account, or the information generated by the user when using a financial or insurance account, etc.
[0062] Step 120: Preprocess the user information to obtain first information, which is used to characterize the keyword field and tag field of the user information;
[0063] Step 120 involves preprocessing the user information to obtain first information, which is used to characterize the keyword and tag fields of the user information, including:
[0064] Step 121: Perform mapping query processing on the user information to obtain tag information;
[0065] Step 122: Perform data cleaning, data transformation, and data standardization on the label information to obtain the intermediate label information;
[0066] Step 123: Perform global word grouping processing on the intermediate tag information and the user information to obtain the first information.
[0067] It is understandable that, in practical applications, user information and tag information are stored in different databases, and a mapping table is used to maintain the relationship between them. Therefore, in this embodiment, the corresponding tag information can be retrieved from the database based on the user information. Then, data cleaning, data transformation, and data standardization can be performed using regular expressions to remove irregular characters, irrelevant characters, and extract key characters from the tag information, thus obtaining intermediate tag information. It should be noted that some tag information is stored in a question-and-answer format (e.g., Question: What do you mainly use your income for?; Answer: Investment and financial management). This tag information needs to undergo data transformation to remove the question portion and retain only the answer portion. Furthermore, global word grouping can involve sequentially grouping user information with each intermediate tag information to form multi-sequence word groups, and using these multi-sequence word groups as the first piece of information.
[0068] It is worth noting that the user's tag information mainly includes age range, education level, occupation category, income, location, insurance purchase intention index, preferred insurance types (medical insurance, critical illness insurance, pension insurance, accident insurance), whether they have purchased insurance, their knowledge of insurance, duration of attention to insurance information, health status, risk avoidance methods, annual insurance expenditure, medical information, hospitalization history, family situation, insured amount, type of insurance product, insured party, marital status, lifestyle habits, emergency response measures, past insurance service experience, IP address location, mobile phone location, credit card information, investment and financial management information, activity trajectory, nature of the enterprise or institution they belong to, job position, monthly consumption purpose, etc., which will not be elaborated further in this application.
[0069] Step 130: Perform word vectorization on the first information to generate a tag vector space;
[0070] Step 130, performing word vectorization on the first information to generate a tag vector space, includes:
[0071] Step 131: Perform word vector space transformation on the first information to obtain the tag matrix;
[0072] Step 132: Perform dimensionality reduction on the label matrix to obtain the label vector space.
[0073] It is understandable that tag information can be transformed into a word vector space using the word2vec word vector model or the ELMO model. Each keyword field in the tag information is mapped into a word vector space matrix, and a tag matrix is generated based on the word vector space matrix of each keyword field. Then, the dimension of the tag matrix is reduced to handle the case where the matrix element is 0 during the encoding process, thereby obtaining the tag vector space.
[0074] Step 140: Perform similarity judgment on the preset hierarchical classification matrix based on the tag vector to obtain the hierarchical classification result of the user information.
[0075] Step 140, which involves performing a similarity judgment on a preset hierarchical classification matrix based on the tag vector to obtain the hierarchical classification result of the user information, includes:
[0076] Step 141: Obtain the preset similarity threshold;
[0077] Step 142: Perform similarity judgment on the current vector element in the label vector space and the first sub-matrix element in the hierarchical classification matrix to obtain a similarity score;
[0078] Step 142, which involves performing a similarity judgment on the current vector element of the label vector space and the first sub-matrix element in the hierarchical classification matrix to obtain a similarity score, includes:
[0079] Step 1421: Perform distance processing on the current vector element in the label vector space and the first sub-matrix element in the hierarchical classification matrix to obtain the first distance;
[0080] Step 1422: Generate the similarity score based on the first distance.
[0081] It is understandable that the matrix keyword field can include the matrix keyword field and the matrix label field. The first sub-matrix element is the vector representation of the matrix keyword field and the matrix label field. The current vector element can be distanced with the current first sub-matrix element in the hierarchical classification matrix to obtain the vector distance (first distance) between the current vector element and all first sub-matrix elements. Then, the corresponding similarity score is generated based on each first distance.
[0082] It is worth mentioning that there are often multiple first sub-matrix elements in the hierarchical classification matrix. After the current vector element and the current first sub-matrix element have completed the similarity judgment, the next first sub-matrix element is taken as the current first sub-matrix element, and the process returns to step 142 until the current vector element and the first sub-matrix elements in the hierarchical classification matrix have all undergone similarity judgment.
[0083] Step 143: Perform threshold filtering on the similarity score according to the similarity threshold to obtain the first similarity score;
[0084] Step 144: Determine the hierarchical classification result based on the second sub-matrix element corresponding to the largest first similarity score and the current vector element;
[0085] The hierarchical classification matrix includes several matrix elements, each matrix element including a first sub-matrix element and a second sub-matrix element. The first sub-matrix element is used to represent the matrix indicator field, and the second sub-matrix element is used to represent the category and level corresponding to the matrix indicator field.
[0086] In this embodiment of the application, all similarity scores can be initially screened using a similarity threshold to exclude similarity scores that are less than the similarity threshold, resulting in several first similarity scores to reduce the computational load of comparison. Then, the first similarity score with the largest value among the several first similarity scores is selected, and a hierarchical classification result is generated using the second sub-matrix element corresponding to the first similarity score and the current vector element.
[0087] It is understood that, since the current vector element in this application is used to represent the user's current keyword field and tag field, and the second sub-matrix element is used to represent the category and level corresponding to the matrix keyword field, in the process of determining the hierarchical classification result based on the second sub-matrix element and the current vector element corresponding to the largest first similarity score, the current vector element and the second sub-matrix element can be simply mapped to output the user's current keyword field and tag field, as well as the corresponding category and level; or it can be output in the form of user information, corresponding level and category, which will not be elaborated further in this application.
[0088] For example, steps 141 to 143 can be implemented using the Deep Structured Semantic Models (DSSM). In this embodiment, the label vector space is represented as [1,0,0,1], [1,1,0,0], [1,1,1,0], with the current vector element being [1,0,0,1]. The hierarchical classification matrix is represented as [1,0,1,1]-[2,1], [1,1,2,0]-[3,2], [1,2,0,0]-[3,1]. The first matrix element is [1,0,1,1]-[2,1], the first sub-matrix element is [1,0,1,1], the second sub-matrix element is [2,1], and the remaining vector elements or other matrix elements can be deduced by analogy. This application will not elaborate further here. After similarity judgment, the hierarchical classification result obtained from the label vector space can be represented in vector form as [1,0,0,1]-[2,1].
[0089] In some embodiments, the hierarchical classification matrix is obtained through the following steps:
[0090] Step 150: Obtain a standard training set, which includes standard keyword data, as well as level data and category data corresponding to the standard keyword data;
[0091] Step 160: Perform encoding mapping processing on the standard training set to obtain a sequence dataset;
[0092] Step 170: Decode and activate the sequence dataset to obtain the hierarchical classification matrix.
[0093] In this embodiment, the standard training set is used as the standard for user classification. For categories, it can be mainly divided into three categories: user data, business data, and system operation and security data. For levels, it can be mainly divided into three levels: user privacy information related, second-level business data related, and first-level behavioral data and log data. Specifically, the standard keyword data in the standard training set can be obtained by statistically analyzing keywords widely used in the financial and insurance industry. Then, the level data and category data can be obtained by manually or machine labeling the corresponding levels or categories, which will not be elaborated further here. For example, the standard training set can be encoded and mapped to the semantic space using the Encoder module in the Encoder-Decoder model, resulting in several fixed-dimensional vectors, which are then used as a sequence dataset. Then, the sequence dataset is decoded and activated by the Decoder module in the Decoder model to output a classification matrix.
[0094] In some embodiments, the hierarchical classification method for user information further includes:
[0095] Step 180: Obtain the updated dataset based on the classification results.
[0096] Step 190: Update the current hierarchical classification matrix according to the updated dataset to obtain the next hierarchical classification matrix.
[0097] In this embodiment, the updated dataset can be the hierarchical classification result corresponding to a non-maximum first similarity score. Specifically, in practice, there are often multiple similarity scores greater than the similarity threshold, resulting in multiple first similarity scores. In this embodiment, after generating the output hierarchical classification result based on the maximum first similarity score, the hierarchical classification results corresponding to the remaining first similarity scores can be added to the standard training set to achieve adaptive updating of the standard training set. By performing encoding mapping and decoding activation processing on the updated standard training set, the next hierarchical classification matrix is obtained, thereby improving the accuracy of user information hierarchical classification.
[0098] Reference Figure 2 This application also provides a hierarchical classification system for user information, including:
[0099] Module 101 is used to obtain user information to be classified and categorized.
[0100] Processing module 102 is used to preprocess the user information to obtain first information, wherein the first information is used to characterize the keyword field and tag field of the user information;
[0101] Vector module 103 is used to perform word vectorization processing on the first information to generate a tag vector space;
[0102] The judgment module 104 is used to perform similarity judgment on the tag vector space according to the preset hierarchical classification matrix to obtain the hierarchical classification result of the user information.
[0103] It is understood that the content of the above method embodiments is applicable to this system embodiment. The specific functions implemented in this system embodiment are the same as those in the above method embodiments, and the beneficial effects achieved are also the same as those achieved in the above method embodiments.
[0104] Reference Figure 3 This application also provides a computer device, including:
[0105] At least one processor 201;
[0106] At least one memory 202 is used to store at least one program;
[0107] When the at least one program is executed by the at least one processor 201, the at least one processor 201 implements the above-described method embodiments.
[0108] Similarly, it can be understood that the content of the above method embodiments is applicable to this device embodiment. The specific functions implemented by this device embodiment are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.
[0109] This application also provides a computer-readable storage medium storing a program executable by a processor 201, which, when executed by the processor 201, is used to implement the above-described method embodiments.
[0110] Similarly, the content of the above method embodiments is applicable to the present computer-readable storage medium embodiments. The specific functions implemented by the present computer-readable storage medium embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.
[0111] In some alternative embodiments, the functions / operations mentioned in the block diagrams may not occur in the order shown in the operation diagrams. For example, depending on the functions / operations involved, two consecutively shown blocks may actually be executed substantially simultaneously, or the blocks may sometimes be executed in reverse order. Furthermore, the embodiments presented and described in the flowcharts of this application are provided by way of example to provide a more comprehensive understanding of the technology. The disclosed methods are not limited to the operations and logic flows presented herein. Alternative embodiments are contemplated in which the order of various operations is changed and sub-operations described as part of a larger operation are executed independently.
[0112] Furthermore, although this application is described in the context of functional modules, it should be understood that, unless otherwise stated to the contrary, one or more of the functions and / or features may be integrated into a single physical device and / or software module, or one or more functions and / or features may be implemented in a separate physical device or software module. It is also understood that a detailed discussion of the actual implementation of each module is unnecessary for understanding this application. Rather, given the properties, functions, and internal relationships of the various functional modules in the apparatus disclosed herein, the actual implementation of the module will be understood within the scope of conventional technology for an engineer. Therefore, those skilled in the art can implement the application set forth in the claims using ordinary techniques without excessive experimentation. It is also understood that the specific concepts disclosed are merely illustrative and not intended to limit the scope of this application, which is determined by the full scope of the appended claims and their equivalents.
[0113] If a function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.
[0114] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device.
[0115] More specific examples of computer-readable media (a non-exhaustive list) include: electrical connections (electronic devices) having one or more wires, portable computer disk drives (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, computer-readable media can even be paper or other suitable media on which programs can be printed, because programs can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in computer memory.
[0116] It should be understood that various parts of this application can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented using software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.
[0117] In the foregoing description of this specification, the references to terms such as "one embodiment," "another embodiment," or "some embodiments," etc., indicate that a specific feature, structure, material, or characteristic described in connection with an embodiment or example is included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.
[0118] Although embodiments of this application have been shown and described, those skilled in the art will understand that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of this application, the scope of which is defined by the claims and their equivalents.
[0119] The above is a detailed description of the preferred embodiments of this application, but this application is not limited to the embodiments. Those skilled in the art can make various equivalent modifications or substitutions without departing from the spirit of this application, and these equivalent modifications or substitutions are all included within the scope defined by the claims of this application.
Claims
1. A method for hierarchical classification of user information, characterized in that, include: Obtain user information to be classified and categorized; The user information is preprocessed to obtain first information, which is used to characterize the keyword field and tag field of the user information; The first information is processed into word vectors to generate a tag vector space; the tag vector space is the vector space corresponding to the keyword field and the tag field. Based on the preset hierarchical classification matrix, the similarity of the tag vector space is judged to obtain the hierarchical classification result of the user information; The preprocessing of the user information to obtain the first information includes: The user information is mapped and queried to obtain tag information; The label information is subjected to data cleaning, data transformation, and data standardization to obtain intermediate label information; Global word grouping is performed on the intermediate tag information and the user information to obtain the first information; The step of performing similarity judgment on the label vector space according to the preset hierarchical classification matrix to obtain hierarchical classification results includes: Obtain the preset similarity threshold; A similarity score is obtained by performing a similarity judgment on the current vector element in the label vector space and all first sub-matrix elements in the hierarchical classification matrix; Based on the similarity threshold, all the similarity scores are subjected to threshold filtering to obtain several first similarity scores; The hierarchical classification result is determined based on the second sub-matrix element corresponding to the largest first similarity score and the current vector element; The hierarchical classification matrix includes several matrix elements, each matrix element including a first sub-matrix element and a second sub-matrix element. The first sub-matrix element is used to represent the matrix indicator field, and the second sub-matrix element is used to represent the category and level corresponding to the matrix indicator field. The matrix indicator field is the keyword field and the tag field.
2. The hierarchical classification method for user information according to claim 1, characterized in that, The step of performing word vectorization processing on the first information to generate a tag vector space includes: The first information is transformed into a word vector space to obtain a tag matrix; The label matrix is reduced in dimensionality to obtain the label vector space.
3. The hierarchical classification method for user information according to claim 1, characterized in that, The hierarchical classification matrix is obtained through the following steps: Obtain a standard training set, which includes standard keyword data, as well as level data and category data corresponding to the standard keyword data; The standard training set is encoded and mapped to obtain a sequence dataset; The sequence dataset is decoded and activated to obtain the hierarchical classification matrix.
4. The hierarchical classification method for user information according to claim 1, characterized in that, The step of performing a similarity judgment on the current vector element of the label vector space and the first sub-matrix element in the hierarchical classification matrix to obtain a similarity score includes: The distance between the current vector element in the label vector space and the first sub-matrix element in the hierarchical classification matrix is processed to obtain the first distance; The similarity score is generated based on the first distance.
5. The hierarchical classification method for user information according to claim 3, characterized in that, The hierarchical classification method for user information also includes: Based on the classification results, obtain the updated dataset; The current hierarchical classification matrix is updated based on the updated dataset to obtain the next hierarchical classification matrix.
6. A hierarchical classification system for user information, characterized in that, include: The acquisition module is used to acquire user information to be classified and categorized. The processing module is used to preprocess the user information to obtain first information, wherein the first information is used to characterize the keyword field and tag field of the user information; The vector module is used to perform word vectorization processing on the first information to generate a tag vector space; the tag vector space is a vector space corresponding to the keyword field and the tag field. The judgment module is used to perform similarity judgment on the tag vector space according to the preset hierarchical classification matrix to obtain the hierarchical classification result of the user information; The preprocessing of the user information to obtain the first information includes: The user information is mapped and queried to obtain tag information; The label information is subjected to data cleaning, data transformation, and data standardization to obtain intermediate label information; Global word grouping is performed on the intermediate tag information and the user information to obtain the first information; The step of performing similarity judgment on the label vector space according to the preset hierarchical classification matrix to obtain hierarchical classification results includes: Obtain the preset similarity threshold; A similarity score is obtained by performing a similarity judgment on the current vector element in the label vector space and all first sub-matrix elements in the hierarchical classification matrix; Based on the similarity threshold, all the similarity scores are subjected to threshold filtering to obtain several first similarity scores; The hierarchical classification result is determined based on the second sub-matrix element corresponding to the largest first similarity score and the current vector element; The hierarchical classification matrix includes several matrix elements, each matrix element including a first sub-matrix element and a second sub-matrix element. The first sub-matrix element is used to represent the matrix indicator field, and the second sub-matrix element is used to represent the category and level corresponding to the matrix indicator field. The matrix indicator field is the keyword field and the tag field.
7. A computer device, characterized in that, include: At least one processor; At least one memory for storing at least one program; When the at least one program is executed by the at least one processor, the at least one processor implements the hierarchical classification method for user information as described in any one of claims 1-5.
8. A computer-readable storage medium storing a processor-executable program, characterized in that, The processor-executable program, when executed by the processor, is used to implement the hierarchical classification method for user information as described in any one of claims 1-5.
Citation Information
Patent Citations
Method and device for determining illegal user, electronic equipment and storage medium
CN115203501A
