User classification method, device, electronic device and readable storage medium

Abstract

The present invention relates to data processing, and discloses a user classification method, including: establishing hierarchical labels for each data set based on the description information of the data set, establishing a label tree based on the hierarchical labels, and taking the set of data sets corresponding to the same leaf node label in the label tree as a data set group; determining the target leaf node label based on the data set requirement text, taking the data set group corresponding to the target leaf node label as the target data set group, and calculating the comprehensive score of each data set in the target data set group based on historical modeling data; sorting the data sets in the target data set group based on the comprehensive score, determining the target data set based on the sorting result, and performing joint modeling based on the target data set to obtain a target user classification model; inputting user data into the target user classification model to obtain a user classification result. The present invention also provides a user classification device, an electronic device, and a readable storage medium. The present invention improves the accuracy of user classification.

Description

Technical Field

[0001] The present invention relates to the field of data processing, and in particular to a user classification method, device, electronic device and readable storage medium. Background Art

[0002] With the development of science and technology, models are being used more and more widely. For example, models are used to accurately classify users. Currently, user classification models are usually built based on the collected data sets. Due to data privacy and security requirements, the collected data sets are relatively limited, resulting in an incomplete data set and the classification accuracy of the model obtained by modeling is not high.

[0003] Nowadays, data sets can be improved through multi-party joint modeling. During the modeling process, federation members need to select appropriate data sets for joint modeling based on the description information of each party's data set. However, due to the relatively brief description information, it is impossible to accurately understand the details of each data set, resulting in improper selection of the target data set and low classification accuracy of the user classification model obtained by modeling. Therefore, a user classification method is urgently needed to improve the accuracy of user classification. Summary of the invention

[0004] In view of the above, it is necessary to provide a user classification method to improve the accuracy of user classification.

[0005] The user classification method provided by the present invention comprises:

[0006] Parse the joint modeling request for the user classification model issued by the first client, obtain the data set requirement text carried by the joint modeling request, obtain the description information of the data sets corresponding to each federation member from the first database, establish a hierarchical label for each data set based on the description information, establish a label tree based on the hierarchical label, and take the set of data sets corresponding to the same leaf node label in the label tree as a data set group;

[0007] Determine a target leaf node label corresponding to the joint modeling request based on the data set requirement text, use the data set group corresponding to the target leaf node label as the target data set group, obtain historical modeling data of each data set in the target data set group from a second database, and calculate a comprehensive score of each data set in the target data set group based on the historical modeling data;

[0008] sorting the data sets in the target data set group in descending order of comprehensive scores, determining the target data set based on the sorting result, and performing joint modeling based on the target data set to obtain a target user classification model;

[0009] The user classification request sent by the second client is parsed, user data carried in the user classification request is obtained, and the user data is input into the target user classification model to obtain a user classification result.

[0010] Optionally, determining a target leaf node label corresponding to the joint modeling request based on the data set requirement text includes:

[0011] Performing word segmentation processing on the data set demand text to obtain a word set;

[0012] Each word in the word set is matched with a tag library corresponding to each level tag of the tag tree, and a target leaf node tag is determined based on the matching result.

[0013] Optionally, the matching result includes a successfully matched tag name and a tag level, and determining the target leaf node tag based on the matching result includes:

[0014] Based on the tag name and the tag level, determine whether the successfully matched tags contain leaf node tags;

[0015] When it is determined that the successfully matched label contains a leaf node label, the successfully matched leaf node label is used as the target leaf node label.

[0016] Optionally, after determining whether the successfully matched tags contain leaf node tags based on the tag name and the tag level, the method further includes:

[0017] If it is determined that the successfully matched tags do not contain leaf node tags, the successfully matched tags will be sorted in descending order according to the tag hierarchy, and the sub-tag selection interfaces of each level of the last sorted tag will be displayed on the interface of the first client for the user of the first client to select the target leaf node tag.

[0018] Optionally, the performing joint modeling based on the target data set includes:

[0019] Perform data alignment processing and feature extraction processing on the target data set to obtain an aligned data set and a feature data set corresponding to the target data set, respectively, and perform joint modeling based on the target data set, the aligned data set and the feature data set.

[0020] Optionally, the historical modeling data includes the number of successful modeling of the data set and the initial score of the data set by the modeling participants after each successful modeling. The calculation formula of the comprehensive score is:

[0021]

[0022] Among them, Y iis the comprehensive score of the i-th dataset in the target dataset group, M is the number of modeling successes of the dataset with the least number of modeling successes in the target dataset group, and N i is the number of successful modeling of the i-th dataset in the target dataset group, A is the average of the initial scores of all datasets in the target dataset group, S i is the average of the initial scores of the i-th dataset in the target dataset group.

[0023] Optionally, “determining a target leaf node label corresponding to the joint modeling request based on the data set requirement text” is replaced by:

[0024] The cascading label selection interface corresponding to the label tree is displayed to the interface of the first client, so that the user of the first client can select the target leaf node label based on the cascading label selection interface.

[0025] In order to solve the above problem, the present invention further provides a user classification device, the device comprising:

[0026] Establish a module, used to parse the joint modeling request for the user classification model issued by the first client, obtain the data set requirement text carried by the joint modeling request, obtain the description information of the data set corresponding to each federation member from the first database, establish a hierarchical label for each data set based on the description information, establish a label tree based on the hierarchical label, and take the set of data sets corresponding to the same leaf node label in the label tree as a data set group;

[0027] a calculation module, configured to determine a target leaf node label corresponding to the joint modeling request based on the data set requirement text, take a data set group corresponding to the target leaf node label as a target data set group, obtain historical modeling data of each data set in the target data set group from a second database, and calculate a comprehensive score of each data set in the target data set group based on the historical modeling data;

[0028] A modeling module, used to sort the data sets in the target data set group in descending order of comprehensive scores, determine the target data set based on the sorting result, and perform joint modeling based on the target data set to obtain a target user classification model;

[0029] The classification module is used to parse the user classification request sent by the second client, obtain the user data carried in the user classification request, input the user data into the target user classification model, and obtain the user classification result.

[0030] In order to solve the above problem, the present invention further provides an electronic device, the electronic device comprising:

[0031] at least one processor; and,

[0032] a memory communicatively connected to the at least one processor; wherein,

[0033] The memory stores a user classification program executable by the at least one processor, and the user classification program is executed by the at least one processor so that the at least one processor can perform the user classification method.

[0034] In order to solve the above problem, the present invention further provides a computer-readable storage medium, on which a user classification program is stored. The user classification program can be executed by one or more processors to implement the above user classification method.

[0035] Compared with the prior art, the present invention first establishes hierarchical labels for each data set based on the description information of the data set, establishes a label tree based on the hierarchical labels, and takes the set of data sets corresponding to the same leaf node label in the label tree as a data set group. This step performs labeling processing on the data set, which can facilitate the subsequent rapid and accurate finding of the desired data set group; then, the target leaf node label is determined based on the data set requirement text, and the data set group corresponding to the target leaf node label is taken as the target data set group. The comprehensive score of each data set in the target data set group is calculated based on the historical modeling data. This step makes the comprehensive score more objective and accurate; then, the data sets in the target data set group are sorted in descending order according to the comprehensive score, the target data set is determined based on the sorting result, and the target user classification model is obtained by joint modeling based on the target data set. This step makes the selection of the target data set faster and more accurate, and the classification accuracy of the target user classification model obtained by modeling is higher; finally, the user data is input into the target user classification model to obtain the user classification result. Therefore, the present invention improves the accuracy of user classification. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 A schematic diagram of a flow chart of a user classification method provided by an embodiment of the present invention;

[0037] Figure 2 A schematic diagram of a module of a user classification device provided by an embodiment of the present invention;

[0038] Figure 3 A schematic diagram of the structure of an electronic device for implementing a user classification method provided by an embodiment of the present invention;

[0039] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0040] In order to make the purpose, technical scheme and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0041] It should be noted that the descriptions of "first", "second", etc. in the present invention are only used for descriptive purposes and cannot be understood as indicating or implying their relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In addition, the technical solutions between the various embodiments can be combined with each other, but they must be based on the ability of ordinary technicians in the field to implement them. When the combination of technical solutions is contradictory or cannot be implemented, it should be deemed that such a combination of technical solutions does not exist and is not within the scope of protection required by the present invention.

[0042] The present invention provides a user classification method. Figure 1 FIG. 1 is a flow chart of a user classification method provided by an embodiment of the present invention. The method can be executed by an electronic device, and the electronic device can be implemented by software and / or hardware.

[0043] In this embodiment, the user classification method includes:

[0044] S1. Parse the joint modeling request for the user classification model issued by the first client, obtain the data set requirement text carried by the joint modeling request, obtain the description information of the data set corresponding to each federation member from the first database, establish a hierarchical label for each data set based on the description information, establish a label tree based on the hierarchical label, and take the set of data sets corresponding to the same leaf node label in the label tree as a data set group.

[0045] In this embodiment, the first client is the client to which the initiator of the joint modeling request belongs. To ensure data security, each federation member stores its data set in a local storage space and only uploads the description information of the data set to the first database of the electronic device for other federation members to select appropriate data sets for joint modeling.

[0046] In this embodiment, hierarchical labels can be established for each data set based on the description information of the data set, and the description information includes the federation member information, data type, etc. of the data set. For example, based on the industry background of the company to which the federation member belongs, the federation members are first divided into labels such as finance, real estate, medical care, and e-commerce; different companies have different types of data, which can be further classified. Financial companies have loan data, overdue data, credit data, and user basic data; real estate companies have user basic data, house pre-purchase preferences, and housing asset information; user basic data can also be subdivided into interests and hobbies, behavioral characteristics, basic attributes, and purchasing power, and loan data can be subdivided into pre-loan data, mid-loan data, and post-loan data.

[0047] Tags at different levels are nested to form a tree structure, with the last tag being a leaf node tag. If a suitable tag cannot be found in the tag library, federation members can also customize the tag, which can be located at different levels of the tag tree to better classify and manage the data set.

[0048] After the label tree is established, the set of data sets corresponding to the same leaf node label in the label tree is regarded as a data set group.

[0049] S2. Determine the target leaf node label corresponding to the joint modeling request based on the data set requirement text, take the data set group corresponding to the target leaf node label as the target data set group, obtain the historical modeling data of each data set in the target data set group from the second database, and calculate the comprehensive score of each data set in the target data set group based on the historical modeling data.

[0050] The data set requirement text may be one sentence or multiple sentences.

[0051] The determining, based on the data set requirement text, a target leaf node label corresponding to the joint modeling request includes:

[0052] A11, performing word segmentation processing on the data set demand text to obtain a word set;

[0053] In this embodiment, a word list-based word segmentation algorithm (forward maximum matching algorithm, reverse maximum matching algorithm, bidirectional maximum matching algorithm) or a word segmentation method based on an N-gram language model may be used to segment the data set demand text.

[0054] A12. Match each word in the word set with a tag library corresponding to each level tag of the tag tree, and determine a target leaf node tag based on the matching result.

[0055] In this embodiment, each layer of labels in the label tree corresponds to a label library, and each word in the word set obtained by word segmentation is matched with each label library to determine the labels of each level of the required text of the data set.

[0056] The matching result includes the tag name and tag level of the successful match, and determining the target leaf node tag based on the matching result includes:

[0057] B11, judging whether the successfully matched tags contain leaf node tags based on the tag name and tag level;

[0058] B12. When it is determined that the successfully matched label contains a leaf node label, the successfully matched leaf node label is used as the target leaf node label.

[0059] After determining whether the successfully matched tags contain leaf node tags based on the tag name and the tag level, the method further includes:

[0060] If it is determined that the successfully matched tags do not contain leaf node tags, the successfully matched tags will be sorted in descending order according to the tag hierarchy, and the sub-tag selection interfaces of each level of the last sorted tag will be displayed on the interface of the first client for the user of the first client to select the target leaf node tag.

[0061] In this embodiment, the second database stores historical modeling data of each data set, and the historical modeling data includes the number of successful modeling of the data set and the initial score of the data set by the modeling participants after each successful modeling. The calculation formula of the comprehensive score is:

[0062]

[0063] Among them, Y i is the comprehensive score of the i-th dataset in the target dataset group, M is the number of modeling successes of the dataset with the least number of modeling successes in the target dataset group, and N i is the number of successful modeling of the i-th dataset in the target dataset group, A is the average of the initial scores of all datasets in the target dataset group, S i is the average of the initial scores of the i-th dataset in the target dataset group.

[0064] Through this calculation formula, a data set with fewer modeling times may also be calculated to obtain a higher comprehensive score, making the comprehensive score more objective.

[0065] For example, suppose there are three datasets in the target dataset group, namely dataset A, dataset B, and dataset C:

[0066] Dataset A was successfully modeled 6 times, with the corresponding initial scores being 6, 7, 8, 9, 9, and 9 respectively; Dataset B was successfully modeled 5 times, with the corresponding initial scores being 6, 7, 8, 9, and 9 respectively; Dataset C was successfully modeled once, with the corresponding initial score being 9.

[0067] Then we can calculate that A=(6+7+8+9+9+9+6+7+8+9+9+9) / 12=8, M=1, and the corresponding S1 of data set A is=(6+7+8+9+9+9) / 6=8, N1=6, and Y1=1 / (1+6)*8+6 / (1+6)*8; the corresponding S2 of data set B is=(6+7+8+9+9) / 5=7.8, N2=5, and Y2=1 / (1+5)*8.27+5 / (1+5)*7.8.

[0068] In another embodiment of the present invention, “determining the target leaf node label corresponding to the joint modeling request based on the data set requirement text” is replaced by:

[0069] The cascading label selection interface corresponding to the label tree is displayed to the interface of the first client, so that the user of the first client can select the target leaf node label based on the cascading label selection interface.

[0070] S3. Sort the data sets in the target data set group in descending order of comprehensive scores, determine the target data set based on the sorting result, perform joint modeling based on the target data set, and obtain a target user classification model.

[0071] In this embodiment, one or more data sets ranked higher in the target data set group may be used as the target data set.

[0072] The joint modeling based on the target data set includes:

[0073] Perform data alignment processing and feature extraction processing on the target data set to obtain an aligned data set and a feature data set corresponding to the target data set, respectively, and perform joint modeling based on the target data set, the aligned data set and the feature data set.

[0074] S4. Parse the user classification request sent by the second client, obtain user data carried in the user classification request, input the user data into the target user classification model, and obtain a user classification result.

[0075] It can be seen from the above embodiments that the user classification method proposed by the present invention, first, establishes hierarchical labels for each data set based on the description information of the data set, establishes a label tree based on the hierarchical labels, and takes the set of data sets corresponding to the same leaf node label in the label tree as a data set group. This step performs labeling processing on the data set, which can facilitate the subsequent rapid and accurate finding of the desired data set group; then, determines the target leaf node label based on the data set requirement text, takes the data set group corresponding to the target leaf node label as the target data set group, and calculates the comprehensive score of each data set in the target data set group based on the historical modeling data. This step makes the comprehensive score more objective and accurate; then, sorts the data sets in the target data set group in order from high to low according to the comprehensive score, determines the target data set based on the sorting result, and performs joint modeling based on the target data set to obtain the target user classification model. This step makes the selection of the target data set faster and more accurate, and the classification accuracy of the target user classification model obtained by modeling is higher; finally, the user data is input into the target user classification model to obtain the user classification result. Therefore, the present invention improves the accuracy of user classification.

[0076] like Figure 2 FIG. 1 is a schematic diagram of a module of a user classification device provided by an embodiment of the present invention.

[0077] The user classification device 100 of the present invention can be installed in an electronic device. According to the functions to be implemented, the user classification device 100 may include a building module 110, a calculation module 120, a modeling module 130 and a classification module 140. The module of the present invention may also be referred to as a unit, which refers to a series of computer program segments that can be executed by a processor of an electronic device and can complete fixed functions, and is stored in the memory of the electronic device.

[0078] In this embodiment, the functions of each module / unit are as follows:

[0079] Establish module 110, which is used to parse the joint modeling request for the user classification model issued by the first client, obtain the data set requirement text carried by the joint modeling request, obtain the description information of the data set corresponding to each federation member from the first database, establish a hierarchical label for each data set based on the description information, establish a label tree based on the hierarchical label, and treat the set of data sets corresponding to the same leaf node label in the label tree as a data set group.

[0080] In this embodiment, the first client is the client to which the initiator of the joint modeling request belongs. To ensure data security, each federation member stores its data set in a local storage space and only uploads the description information of the data set to the first database of the electronic device for other federation members to select appropriate data sets for joint modeling.

[0081] In this embodiment, hierarchical labels can be established for each data set based on the description information of the data set, and the description information includes the federation member information, data type, etc. of the data set. For example, based on the industry background of the company to which the federation member belongs, the federation members are first divided into labels such as finance, real estate, medical care, and e-commerce; different companies have different types of data, which can be further classified. Financial companies have loan data, overdue data, credit data, and user basic data; real estate companies have user basic data, house pre-purchase preferences, and housing asset information; user basic data can also be subdivided into interests and hobbies, behavioral characteristics, basic attributes, and purchasing power, and loan data can be subdivided into pre-loan data, mid-loan data, and post-loan data.

[0082] Tags at different levels are nested to form a tree structure, with the last tag being a leaf node tag. If a suitable tag cannot be found in the tag library, federation members can also customize the tag, which can be located at different levels of the tag tree to better classify and manage the data set.

[0083] After the label tree is established, the set of data sets corresponding to the same leaf node label in the label tree is regarded as a data set group.

[0084] The computing module 120 is used to determine the target leaf node label corresponding to the joint modeling request based on the data set requirement text, take the data set group corresponding to the target leaf node label as the target data set group, obtain the historical modeling data of each data set in the target data set group from the second database, and calculate the comprehensive score of each data set in the target data set group based on the historical modeling data.

[0085] The data set requirement text may be one sentence or multiple sentences.

[0086] The determining, based on the data set requirement text, a target leaf node label corresponding to the joint modeling request includes:

[0087] A21, performing word segmentation processing on the data set demand text to obtain a word set;

[0088] In this embodiment, a word list-based word segmentation algorithm (forward maximum matching algorithm, reverse maximum matching algorithm, bidirectional maximum matching algorithm) or a word segmentation method based on an N-gram language model may be used to segment the data set demand text.

[0089] A22. Match each word in the word set with a tag library corresponding to each level tag of the tag tree, and determine a target leaf node tag based on the matching result.

[0090] In this embodiment, each layer of labels in the label tree corresponds to a label library, and each word in the word set obtained by word segmentation is matched with each label library to determine the labels of each level of the required text of the data set.

[0091] The matching result includes the tag name and tag level of the successful match, and determining the target leaf node tag based on the matching result includes:

[0092] B21, based on the tag name and tag level, determining whether the successfully matched tags contain a leaf node tag;

[0093] B22. When it is determined that the successfully matched label contains a leaf node label, the successfully matched leaf node label is used as the target leaf node label.

[0094] After determining whether the successfully matched tags contain leaf node tags based on the tag name and the tag level, the calculation module 120 is further used to:

[0095] If it is determined that the successfully matched tags do not contain leaf node tags, the successfully matched tags will be sorted in descending order according to the tag hierarchy, and the sub-tag selection interfaces of each level of the last sorted tag will be displayed on the interface of the first client for the user of the first client to select the target leaf node tag.

[0096] In this embodiment, the second database stores historical modeling data of each data set, and the historical modeling data includes the number of successful modeling of the data set and the initial score of the data set by the modeling participants after each successful modeling. The calculation formula of the comprehensive score is:

[0097]

[0098] Among them, Y i is the comprehensive score of the i-th dataset in the target dataset group, M is the number of modeling successes of the dataset with the least number of modeling successes in the target dataset group, and N i is the number of successful modeling of the i-th dataset in the target dataset group, A is the average of the initial scores of all datasets in the target dataset group, S i is the average of the initial scores of the i-th dataset in the target dataset group.

[0099] Through this calculation formula, a data set with fewer modeling times may also be calculated to obtain a higher comprehensive score, making the comprehensive score more objective.

[0100] For example, suppose there are three datasets in the target dataset group, namely dataset A, dataset B, and dataset C:

[0101] Dataset A was successfully modeled 6 times, with the corresponding initial scores being 6, 7, 8, 9, 9, and 9 respectively; Dataset B was successfully modeled 5 times, with the corresponding initial scores being 6, 7, 8, 9, and 9 respectively; Dataset C was successfully modeled once, with the corresponding initial score being 9.

[0102] Then we can calculate that A=(6+7+8+9+9+9+6+7+8+9+9+9) / 12=8, M=1, and the corresponding S1 of data set A is=(6+7+8+9+9+9) / 6=8, N1=6, and Y1=1 / (1+6)*8+6 / (1+6)*8; the corresponding S2 of data set B is=(6+7+8+9+9) / 5=7.8, N2=5, and Y2=1 / (1+5)*8.27+5 / (1+5)*7.8.

[0103] In another embodiment of the present invention, “determining the target leaf node label corresponding to the joint modeling request based on the data set requirement text” is replaced by:

[0104] The cascading label selection interface corresponding to the label tree is displayed to the interface of the first client, so that the user of the first client can select the target leaf node label based on the cascading label selection interface.

[0105] The modeling module 130 is used to sort the data sets in the target data set group in descending order of comprehensive scores, determine the target data set based on the sorting result, and perform joint modeling based on the target data set to obtain a target user classification model.

[0106] In this embodiment, one or more data sets ranked higher in the target data set group may be used as the target data set.

[0107] The joint modeling based on the target data set includes:

[0108] Perform data alignment processing and feature extraction processing on the target data set to obtain an aligned data set and a feature data set corresponding to the target data set, respectively, and perform joint modeling based on the target data set, the aligned data set and the feature data set.

[0109] The classification module 140 is used to parse the user classification request sent by the second client, obtain the user data carried in the user classification request, input the user data into the target user classification model, and obtain the user classification result.

[0110] like Figure 3 FIG. 1 is a schematic diagram of the structure of an electronic device for implementing a user classification method provided by an embodiment of the present invention.

[0111] The electronic device 1 is a device that can automatically perform numerical calculations and / or information processing according to pre-set or stored instructions. The electronic device 1 can be a computer, a single network server, a server group consisting of multiple network servers, or a cloud based on cloud computing consisting of a large number of hosts or network servers, wherein cloud computing is a type of distributed computing, a super virtual computer composed of a group of loosely coupled computer sets.

[0112] In this embodiment, the electronic device 1 includes, but is not limited to, a memory 11, a processor 12, and a network interface 13 that can be interconnected and communicated via a system bus. The memory 11 stores a user classification program 10 that can be executed by the processor 12. Figure 3 Only the electronic device 1 having components 11-13 and the user classification program 10 is shown, and those skilled in the art can understand that Figure 3 The structure shown does not constitute a limitation on the electronic device 1 , and may include fewer or more components than shown in the figure, or combine certain components, or arrange the components differently.

[0113] The memory 11 includes a memory and at least one type of readable storage medium. The memory provides a cache for the operation of the electronic device 1; the readable storage medium may be a non-volatile storage medium such as a flash memory, a hard disk, a multimedia card, a card-type memory (e.g., SD or DX memory, etc.), a random access memory (RAM), a static random access memory (SRAM), a read-only memory (ROM), an electrically erasable programmable read-only memory (EEPROM), a programmable read-only memory (PROM), a magnetic memory, a disk, an optical disk, etc. In some embodiments, the readable storage medium may be an internal storage unit of the electronic device 1, such as a hard disk of the electronic device 1; in other embodiments, the non-volatile storage medium may also be an external storage device of the electronic device 1, such as a plug-in hard disk equipped on the electronic device 1, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card, etc. In this embodiment, the readable storage medium of the memory 11 is generally used to store an operating system and various application software installed on the electronic device 1, such as storing the code of the user classification program 10 in one embodiment of the present invention. In addition, the memory 11 may also be used to temporarily store various types of data that have been output or are to be output.

[0114] In some embodiments, the processor 12 may be a central processing unit (CPU), a controller, a microcontroller, a microprocessor, or other data processing chips. The processor 12 is generally used to control the overall operation of the electronic device 1, such as performing control and processing related to data interaction or communication with other devices. In this embodiment, the processor 12 is used to run the program code stored in the memory 11 or process data, such as running the user classification program 10.

[0115] The network interface 13 may include a wireless network interface or a wired network interface, and the network interface 13 is used to establish a communication connection between the electronic device 1 and a client (not shown in the figure).

[0116] Optionally, the electronic device 1 may further include a user interface, which may include a display (Display), an input unit such as a keyboard (Keyboard), and the optional user interface may also include a standard wired interface and a wireless interface. Optionally, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, and an OLED (Organic Light-Emitting Diode) touch device, etc. Among them, the display may also be appropriately referred to as a display screen or a display unit, which is used to display information processed in the electronic device 1 and to display a visual user interface.

[0117] It should be understood that the embodiment is for illustration only and the scope of the patent application is not limited to this structure.

[0118] The user classification program 10 stored in the memory 11 of the electronic device 1 is a combination of multiple instructions. When running in the processor 12, it can achieve:

[0119] Parse the joint modeling request for the user classification model issued by the first client, obtain the data set requirement text carried by the joint modeling request, obtain the description information of the data sets corresponding to each federation member from the first database, establish a hierarchical label for each data set based on the description information, establish a label tree based on the hierarchical label, and take the set of data sets corresponding to the same leaf node label in the label tree as a data set group;

[0120] Determine a target leaf node label corresponding to the joint modeling request based on the data set requirement text, use the data set group corresponding to the target leaf node label as the target data set group, obtain historical modeling data of each data set in the target data set group from a second database, and calculate a comprehensive score of each data set in the target data set group based on the historical modeling data;

[0121] sorting the data sets in the target data set group in descending order of comprehensive scores, determining the target data set based on the sorting result, and performing joint modeling based on the target data set to obtain a target user classification model;

[0122] The user classification request sent by the second client is parsed, user data carried in the user classification request is obtained, and the user data is input into the target user classification model to obtain a user classification result.

[0123] Specifically, the specific implementation method of the processor 12 for the user classification program 10 can refer to Figure 1 The description of the relevant steps in the corresponding embodiments will not be repeated here.

[0124] Furthermore, if the module / unit integrated in the electronic device 1 is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. The computer-readable medium can be non-volatile or non-volatile. The computer-readable medium can include: any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a mobile hard disk, a magnetic disk, an optical disk, a computer memory, and a read-only memory (ROM).

[0125] The computer-readable storage medium stores a user classification program 10, which can be executed by one or more processors. The specific implementation of the computer-readable storage medium of the present invention is basically the same as the above-mentioned user classification method embodiments, and will not be described in detail here.

[0126] In the several embodiments provided by the present invention, it should be understood that the disclosed devices, apparatuses and methods can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of the modules is only a logical function division, and there may be other division methods in actual implementation.

[0127] The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0128] In addition, each functional module in each embodiment of the present invention may be integrated into one processing unit, each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of hardware plus software functional modules.

[0129] It is obvious to those skilled in the art that the present invention is not limited to the details of the above exemplary embodiments, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0130] Therefore, no matter from which point of view, the embodiments should be regarded as illustrative and non-restrictive, and the scope of the present invention is limited by the appended claims rather than the above description, so it is intended that all changes falling within the meaning and scope of the equivalent elements of the claims are included in the present invention. Any attached figure mark in the claims should not be regarded as limiting the claims involved.

[0131] In addition, it is clear that the word "comprising" does not exclude other units or steps, and the singular does not exclude the plural. Multiple units or devices stated in the system claim can also be implemented by one unit or device through software or hardware. The second and other words are used to indicate names, but not to indicate any particular order.

[0132] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solution of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solution of the present invention.

Claims

1. A user classification method, characterized in that: The method comprises: Parse the joint modeling request for the user classification model issued by the first client, obtain the data set requirement text carried by the joint modeling request, obtain the description information of the data sets corresponding to each federation member from the first database, establish a hierarchical label for each data set based on the description information, establish a label tree based on the hierarchical label, and take the set of data sets corresponding to the same leaf node label in the label tree as a data set group; Based on the matching results of the words obtained by word segmentation processing of the data set demand text and the labels of each level of the label tree, determine the target leaf node label corresponding to the joint modeling request, take the data set group corresponding to the target leaf node label as the target data set group, obtain the number of modeling successes of each data set in the target data set group and the initial score of the data set by the modeling participants after each modeling success from the second database, and calculate the comprehensive score of each data set in the target data set group based on the number of modeling successes and the initial score of the data set by the modeling participants after each modeling success; sorting the data sets in the target data set group in descending order of comprehensive scores, determining the target data set based on the sorting result, and performing joint modeling based on the target data set to obtain a target user classification model; The user classification request sent by the second client is parsed, user data carried in the user classification request is obtained, and the user data is input into the target user classification model to obtain a user classification result.

2. The user classification method according to claim 1, characterized in that: The matching results of the words obtained based on the word segmentation processing of the data set demand text and the labels of each level of the label tree include: Performing word segmentation processing on the data set demand text to obtain a word set; Each word in the word set is matched with a tag library corresponding to each level tag of the tag tree, and a target leaf node tag is determined based on the matching result.

3. The user classification method according to claim 2, characterized in that: The matching result includes the tag name and tag level of the successful match, and determining the target leaf node tag based on the matching result includes: Based on the tag name and the tag level, determine whether the successfully matched tags contain leaf node tags; When it is determined that the successfully matched label contains a leaf node label, the successfully matched leaf node label is used as the target leaf node label.

4. The user classification method according to claim 3, characterized in that: After determining whether the successfully matched tags contain leaf node tags based on the tag name and the tag level, the method further includes: If it is determined that the successfully matched tags do not contain leaf node tags, the successfully matched tags will be sorted in descending order according to the tag hierarchy, and the sub-tag selection interfaces of each level of the last sorted tag will be displayed on the interface of the first client for the user of the first client to select the target leaf node tag.

5. The user classification method according to claim 1, characterized in that: The joint modeling based on the target data set includes: Perform data alignment processing and feature extraction processing on the target data set to obtain an aligned data set and a feature data set corresponding to the target data set, respectively, and perform joint modeling based on the target data set, the aligned data set and the feature data set.

6. The user classification method according to any one of claims 1 to 5, characterized in that: The calculation formula of the comprehensive score is: Among them, Y i is the comprehensive score of the i-th dataset in the target dataset group, M is the number of modeling successes of the dataset with the least number of modeling successes in the target dataset group, and N i is the number of successful modeling of the i-th dataset in the target dataset group, A is the average of the initial scores of all datasets in the target dataset group, S i is the average of the initial scores of the i-th dataset in the target dataset group.

7. The user classification method according to claim 1, characterized in that: "Determine the target leaf node label corresponding to the joint modeling request based on the data set requirement text" is replaced by: The cascading label selection interface corresponding to the label tree is displayed to the interface of the first client, so that the user of the first client can select the target leaf node label based on the cascading label selection interface.

8. A user classification device, characterized in that: The device comprises: Establish a module, used to parse the joint modeling request for the user classification model issued by the first client, obtain the data set requirement text carried by the joint modeling request, obtain the description information of the data set corresponding to each federation member from the first database, establish a hierarchical label for each data set based on the description information, establish a label tree based on the hierarchical label, and take the set of data sets corresponding to the same leaf node label in the label tree as a data set group; A calculation module, configured to determine the target leaf node label corresponding to the joint modeling request based on the matching results of the words obtained by word segmentation processing of the data set demand text and the labels of each level of the label tree, take the data set group corresponding to the target leaf node label as the target data set group, obtain the number of modeling successes of each data set in the target data set group and the initial score of the data set by the modeling participants after each modeling success from the second database, and calculate the comprehensive score of each data set in the target data set group based on the number of modeling successes and the initial score of the data set by the modeling participants after each modeling success; A modeling module, used to sort the data sets in the target data set group in descending order of comprehensive scores, determine the target data set based on the sorting result, and perform joint modeling based on the target data set to obtain a target user classification model; The classification module is used to parse the user classification request sent by the second client, obtain the user data carried in the user classification request, input the user data into the target user classification model, and obtain the user classification result.

9. An electronic device, characterized in that: The electronic device comprises: at least one processor; and, a memory communicatively connected to the at least one processor; wherein, The memory stores a user classification program executable by the at least one processor, and the user classification program is executed by the at least one processor so that the at least one processor can perform the user classification method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a user classification program, which can be executed by one or more processors to implement the user classification method according to any one of claims 1 to 7.

Images