Tag completion method and tag completion apparatus

Abstract

The application discloses a label completion method and a label completion device, and relates to the technical field of computer multimedia. The method comprises the following steps: determining the user similarity between a first user and each second user according to the personal label set of the first user and the personal label set of each second user; determining at least one similar user corresponding to the first user in each second user according to the user similarity between the first user and each second user; determining at least one completion label of the first user according to the personal label set of the first user and the personal label set of the at least one similar user, and updating the personal label set of the first user according to the at least one completion label. The method can realize label transmission from similar users to the first user according to the similarity between the similar users, so that the label of the first user can be completed by the labels of the similar users.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of computer multimedia, in particular to a label completion method and a label completion device. BACKGROUND

[0002] At present, by establishing a systematic multi-dimensional label, the characteristics of users, user preferences, personalized recommendation content, product categories, communication feedback and other attribute labels are labeled, and intelligent and automated personalized recommendation processes are used to provide personalized recommendation experience for user groups. However, the current label system still has many problems. Generally, the personal labels of users are selected by the users themselves or generated by an algorithm marking method. The labels generated by the two label generation methods are not comprehensive, so label completion is needed. SUMMARY

[0003] Therefore, the present application provides a label completion method and a label completion device, which can rely on the similarity between similar users to realize label transmission from similar users to a first user, so that the labels of a first user can be completed by the labels of similar users.

[0004] To achieve the above purpose, in a first aspect, the present application provides a label completion method, which comprises:

[0005] According to the personal label set of the first user and the personal label set of each second user, the user similarity between the first user and each second user is determined.

[0006] According to the user similarity between the first user and each second user, at least one similar user corresponding to the first user is determined from each second user.

[0007] According to the personal label set of the first user and the personal label set of the at least one similar user, at least one completed label of the first user is determined, and the personal label set of the first user is updated according to the at least one completed label.

[0008] In some examples, the determination of the user similarity between the first user and each second user according to the personal label set of the first user and the personal label set of each second user comprises:

[0009] For each second user, the similar label set between the first user and the second user is determined according to the personal label set of the first user and the personal label set of the second user.

[0010] According to the similar label set between the first user and each second user, the user similarity between the first user and each second user is determined.

[0011] In some examples, the determining, according to the set of similar tags of the first user and each second user, a user similarity between the first user and each second user comprises:

[0012] respectively for each set of similar tags, determining a semantic similarity of the set of similar tags according to a tag similarity between each similar tag included in the set of similar tags;

[0013] determining a user similarity algorithm corresponding to a preset interval to which the semantic similarity belongs, and determining the user similarity between the first user and the second user corresponding to the set of similar tags according to the user similarity algorithm.

[0014] In some examples, the determining, respectively for each set of similar tags, a semantic similarity of the set of similar tags according to a tag similarity between each similar tag included in the set of similar tags comprises:

[0015] respectively for each set of similar tags, obtaining a word vector of each similar tag in the set of similar tags, and calculating a tag similarity between each two similar tags according to the word vectors of the two similar tags;

[0016] determining the semantic similarity of the set of similar tags according to the tag similarity between each two similar tags.

[0017] In some examples, the preset interval comprises a first preset interval, a second preset interval and a third preset interval, and the semantic similarities corresponding to the first preset interval, the second preset interval and the third preset interval increase in turn; and the determining a user similarity algorithm corresponding to a preset interval to which the semantic similarity belongs, and determining the user similarity between the first user and the second user corresponding to the set of similar tags according to the user similarity algorithm comprises:

[0018] for each semantic similarity of the set of similar tags, determining a preset interval to which the semantic similarity belongs;

[0019] if the semantic similarity of the set of similar tags belongs to the first preset interval, determining the user similarity W between the first user and the second user corresponding to the set of similar tags according to the following user similarity algorithm: 1,2

[0020]

[0021] if the semantic similarity of the set of similar tags belongs to the second preset interval, determining the user similarity W between the first user and the second user corresponding to the set of similar tags according to the following user similarity algorithm: 1,2

[0022] ​​

[0023]

[0024] If the semantic similarity of the similar label set belongs to a third preset interval, the user similarity W 1,2 = 1;

[0025] Wherein, labels represents the calculated similar label set, N is the number of each similar label in the similar label set, S ij represents the similarity of any two similar labels in the similar label set.

[0026] In some examples, the determining at least one similar user corresponding to the first user from each second user according to the user similarity between the first user and each second user comprises:

[0027] For each user similarity, comparing the user similarity with a preset user similarity;

[0028] If the user similarity is greater than or equal to the preset user similarity, the second user corresponding to the user similarity is the similar user corresponding to the first user.

[0029] In some examples, the determining at least one complementary label of the first user according to the personal label set of the first user and the personal label set of the at least one similar user comprises:

[0030] Obtaining the personal label set of the first user and the personal label set of the at least one similar user, and generating an associated label set of the first user;

[0031] Determining the association probability of each associated label in the associated label set;

[0032] According to the association probability of each associated label, determining at least one complementary label of the first user.

[0033] In some examples, the determining the association probability of each associated label in the associated label set comprises:

[0034] For each associated label, determining a first initial association probability of the first user corresponding to the associated label;

[0035] Determining a second initial association probability of at least one similar user to which the associated label belongs;

[0036] According to the first initial association probability, the second initial association probability, and the user similarity between the first user corresponding to the associated label and at least one similar user, determining the association probability of the associated label.

[0037] In some examples, determining the association probability of the association tag based on the first initial association probability, the second initial association probability, and the user similarity between the first user corresponding to the association tag and at least one of the similar users includes:

[0038] The association probability P of each associated tag is calculated according to the following formula: i,a :

[0039]

[0040] Among them, P1 i,a is the first initial association probability, P2 j,a is the second initial association probability, j is a similar user, V is the set of similar users of the first user, w ij is the similarity between the first user and similar users.

[0041] In some examples, determining at least one completion tag for the first user based on the association probabilities of the associated tags includes:

[0042] comparing the association probability of each of the association tags with a preset association probability;

[0043] If the association probability of the association tag is greater than or equal to the preset association probability, the association tag is determined to be a completion tag.

[0044] In some examples, determining at least one completion tag for the first user based on the association probabilities of the associated tags includes:

[0045] sorting the association probabilities of the associated tags from large to small;

[0046] A preset number of associated tags with the largest associated probabilities are selected as the completion tags.

[0047] In a second aspect, the present invention provides a label completion device, comprising:

[0048] A user similarity determination module, configured to determine a user similarity between the first user and each second user based on the first user's personal tag set and each second user's personal tag set;

[0049] a similar user determining module, configured to determine at least one similar user corresponding to the first user among each second user based on the user similarity between the first user and each second user;

[0050] a tag completion module configured to determine at least one completion tag of the first user according to the personal tag set of the first user and the personal tag set of the at least one similar user, and update the personal tag set of the first user according to the at least one completion tag.

[0051] In a third aspect, the present application provides an electronic device, comprising:

[0052] at least one processor; and

[0053] a memory connected with the at least one processor; wherein,

[0054] the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method described above.

[0055] In a fourth aspect, the present application provides a non-transitory computer readable storage medium storing computer instructions, wherein the computer instructions are used to enable the computer to perform the method described above.

[0056] The present application has at least the following advantages:

[0057] The tag completion method and the tag completion device provided by the present application can determine the similar users of the first user, and then determine the completion tags of the first user according to the personal tag set of the first user and the personal tag set of the similar users, and update the personal tag set of the first user according to the completion tags, so that the tag transmission from the similar users to the first user can be realized according to the similarity between the similar users and the first user, and the tags of the similar users can be used to complete the tag completion of the first user. BRIEF DESCRIPTION OF DRAWINGS

[0058] The accompanying drawings are included to provide a further understanding of the present application, and constitute a part of the specification, and are used together with the following detailed description to explain the present application, but do not constitute a limitation of the present application.

[0059] Figure 1 A flowchart of an embodiment of the tag completion method provided by the present application;

[0060] Figure 2 A flowchart of another embodiment of the tag completion method provided by the present application;

[0061] Figure 3 A structural diagram of an embodiment of the tag completion device provided by the present application;

[0062] Figure 4 A structural diagram of an embodiment of the electronic device provided by the present application. DETAILED DESCRIPTION

[0063] In order to better understand the technical solutions of the present application, the present application will be further described in detail below in conjunction with the drawings and specific embodiments.

[0064] It can be understood that the specific embodiments and drawings described herein are merely used to explain the present application, but not to limit the present application.

[0065] It can be understood that the embodiments in the present application and the features in the embodiments can be combined with each other without conflict.

[0066] It can be understood that, for the convenience of description, only the parts related to the present application are shown in the drawings of the present application, and the parts unrelated to the present application are not shown in the drawings.

[0067] It can be understood that each unit and module involved in the embodiments of the present application can correspond to only one entity structure, or can be composed of multiple entity structures, or multiple units and modules can be integrated into one entity structure.

[0068] It can be understood that the functions and steps marked in the flowcharts and block diagrams of the present application can occur in an order different from that marked in the drawings without conflict.

[0069] It can be understood that in the flowcharts and block diagrams of the present application, the architecture, functions and operations of the possible implementations of the system, device, equipment and method according to the embodiments of the present application are shown. Each block in the flowchart or block diagram can represent a unit, module, program segment, code, which contains executable instructions for realizing the specified functions. Moreover, each block or combination of blocks in the block diagram and flowchart can be realized by a hardware-based system for realizing the specified functions, or by a combination of hardware and computer instructions.

[0070] It can be understood that the units and modules involved in the embodiments of the present application can be realized in the form of software or hardware, for example, the units and modules can be located in a processor.

[0071] Embodiment one,

[0072] Referring to Figure 1 , Figure 1 A flowchart of a label completion method provided by one embodiment of the present application is shown. The present embodiment provides a label completion method, which comprises the following steps:

[0073] S1, determining the user similarity between the first user and each second user according to the personal label set of the first user and the personal label set of each second user.

[0074] Each user (including the first user and the second user) has an initial personal tag set, and the tags in the personal tag set can be generated by the user himself or by the system according to the user information, which is not limited here.

[0075] For each second user, the user similarity between the first user and the second user can be calculated according to the tags included in the personal tag set of the first user and the tags included in the personal tag set of the second user. Specifically, similar tags in the personal tag set of the first user and the personal tag set of the second user can be filtered out to generate a similar tag set, and the user similarity between the first user and the second user can be determined according to the semantic similarity between each similar tag in the similar tag set. Wherein, the similar tags can be tags in the personal tag set of the first user and the personal tag set of the second user with a semantic similarity higher than a preset threshold, and the similar tags can also be the same tags in the personal tag set of the first user and the personal tag set of the second user, which is not limited here.

[0076] It should be noted that in the user cluster, the first user is the user to be generated a complete tag, and the second user is the user in the user cluster other than the first user. Each second user becomes the first user when performing tag completion.

[0077] S2, determining at least one similar user corresponding to the first user from each second user according to the user similarity between the first user and each second user.

[0078] Specifically, after calculating the user similarity between the first user and each second user based on the previous step, each user similarity can be compared with a preset user similarity in turn, and whether the second user corresponding to each user similarity is a similar user of the first user can be determined according to the comparison result.

[0079] S3, determining at least one complete tag of the first user according to the personal tag set of the first user and the personal tag set of at least one similar user, and updating the personal tag set of the first user according to the at least one complete tag.

[0080] Specifically, after filtering out the similar users of the first user from each second user based on the previous step, since the first user and the similar users have a high similarity, the tags in the personal tag set of the similar users but not in the personal tag set of the first user are most likely to be the complete tags of the first user. Based on the above, after determining the similar users of the first user, the relevance of the tags in the personal tag set of the first user and the tags in the personal tag set of at least one similar user can be calculated in a certain way, and the tags with high relevance can be filtered out as the complete tags of the first user, so that the personal tag set of the first user can be updated with the complete tags.

[0081] In the label completion method provided by the embodiment, the similar user of the first user is determined, and then the complete label of the first user is determined according to the personal label set of the first user and the personal label set of the similar user, and the personal label set of the first user is updated according to the complete label, so that the label transmission from the similar user to the first user can be realized according to the similarity between the similar user and the first user, and the label of the first user can be completed by the label of the similar user of the first user.

[0082] Embodiment two,

[0083] Referring to Figure 2 , Figure 2 A flowchart of a label completion method provided by another embodiment of the application is shown. The label completion method provided by the embodiment includes:

[0084] S01, for each second user, the similar label set of the first user and the second user is determined according to the personal label set of the first user and the personal label set of the second user.

[0085] Each user (including the first user and the second user) has an initial personal label set, and the labels in the personal label set can be generated by the user himself or generated by the system according to the user information, which is not limited here.

[0086] For each second user, the similar label set of the first user and the second user can be generated according to the labels included in the personal label set of the first user and the labels included in the personal label set of the second user. Specifically, the similar labels in the personal label set of the first user and the personal label set of the second user can be filtered to generate a similar label set, and then the user similarity between the first user and the second user can be determined according to the semantic similarity of the similar labels in the similar label set.

[0087] Among them, the similar label can be a label with a semantic similarity higher than a preset threshold in the personal label set of the first user and the personal label set of the second user, and the similar label can also be a same label in the personal label set of the first user and the personal label set of the second user, which is not limited here. Taking the similar label as an example, the personal label set of the first user includes {film and television, makeup, decoration}, and the personal label set of a second user includes {makeup, decoration, photography}, and the similar label set of the first user and the second user is {makeup, decoration}.

[0088] S02, the user similarity between the first user and each second user is determined according to the similar label set of the first user and each second user.

[0089] After generating the similar tag set of the first user and each second user, the semantic similarity of the similar tags in the similar tag set can be calculated, and then the user similarity between the first user and the second user can be calculated according to the semantic similarity; or the user similarity between the first user and the second user can be directly determined according to the number of similar tags in the similar tag set, which is not limited here.

[0090] In some examples, S02 includes:

[0091] S021, for each similar tag set respectively, the semantic similarity of the similar tag set is determined according to the tag similarity between each similar tag in the similar tag set.

[0092] Specifically, the tag similarity between each similar tag in each similar tag set can be calculated first, and then the semantic similarity of the entire similar tag set can be determined based on the similarities.

[0093] In some examples, S021 includes a plurality of sub-steps:

[0094] Sub-step one, for each similar tag set respectively, the word vector of each similar tag in the similar tag set is obtained, and the tag similarity between each two similar tags is calculated according to the word vectors of the two similar tags.

[0095] Specifically, for the similar tag set of the first user and each second user respectively, the word vector of each similar tag in the similar tag set is obtained, and then the cosine similarity algorithm is used to calculate the tag similarity between the two similar tags according to the word vectors of the two similar tags. The word vector of each similar tag includes a plurality of dimensional vectors, and the cosine similarity algorithm calculates the cosine value between the two similar tags according to the vector of each dimension in the word vectors of the two similar tags, that is, the included angle between the two similar tags can be obtained. The smaller the included angle, the higher the tag similarity between the two similar tags, and vice versa.

[0096] Sub-step two, the semantic similarity of the similar tag set is determined according to the tag similarity between each two similar tags.

[0097] Specifically, for the similar tag set of the first user and each second user respectively, after calculating the tag similarity between each two similar tags in the similar tag set, the similarity of the entire similar tag set is evaluated by synthesizing the tag similarities, and the evaluation result is the semantic similarity of the similar tag set. In some examples, the average of a plurality of tag similarities can be taken as the semantic similarity of the similar tag set, and of course other methods can also be used, which are not limited here.

[0098] S022、determine a user similarity algorithm corresponding to the preset interval to which the semantic similarity belongs, and determine the user similarity between the first user and the second user corresponding to the similar tag set according to the user similarity algorithm.

[0099] Specifically, different user similarity algorithms are selected according to whether the semantic similarity of each similar tag set is high or low, and the user similarity between the first user and the second user corresponding to each similar tag set is calculated.

[0100] In some examples, the preset interval includes a first preset interval, a second preset interval and a third preset interval, and the semantic similarity corresponding to the first preset interval, the second preset interval and the third preset interval increases in turn. S022 includes a plurality of sub-steps:

[0101] Sub-step one, for the semantic similarity of each similar tag set, determine the preset interval to which the semantic similarity belongs.

[0102] Based on the above calculation of the semantic similarity of each similar tag set, according to the value of the semantic similarity of each similar tag set, determine the preset interval to which the value of the semantic similarity belongs, for example, the first preset interval, the second preset interval or the third preset interval, and determine the user similarity algorithm corresponding to the preset interval according to the value of the semantic similarity.

[0103] Sub-step two (1), if the semantic similarity of the similar tag set belongs to the first preset interval, the user similarity W between the first user and the second user corresponding to the similar tag set is determined according to the following user similarity algorithm 1,2 :

[0104]

[0105] Wherein, N is the number of similar tags in the similar tag set. In the case of weak semantic similarity of the similar tag set, the relevance and reference of the similar tags are low, so the tag similarity between the similar tags can be ignored, and the user similarity W between the first user and the second user corresponding to the similar tag set is directly determined by the number of similar tags in the similar tag set 1,2 .

[0106] It should be noted that if the similar tag set is empty, i.e. if the second user corresponding to the similar tag set has no similar tags, then

[0107] Sub-step two (2), if the semantic similarity of the similar tag set belongs to the second preset interval, the user similarity W between the first user and the second user corresponding to the similar tag set is determined according to the following user similarity algorithm 1,2 :

[0108]

[0109]

[0110] wherein, labels represents the calculated similar label set, N is the number of similar labels contained in the similar label set, S ij The similarity between any two similar labels (label i or label j) in the similar label set is represented by the similarity between the two labels. The similarity can be calculated using covariance.

[0111] Sub-step two (3), if the semantic similarity of the similar label set belongs to the third preset interval, the user similarity W 1,2 = 1. Specifically, the third preset interval can be an interval in which the semantic similarity of the similar label set is 1, which means that the similar label set only has one identical label, so W 1,2 = 1.

[0112] It should be noted that steps S01 and S02 are specific steps of one implementation of step S1, and step S1 can also be implemented in other ways, which are not limited here.

[0113] S03, for each user similarity, compare the user similarity with the preset user similarity.

[0114] Specifically, the preset user similarity is set according to the need. If the number of labels to be completed is large and the range is wide, a smaller preset user similarity can be set. If the accuracy of the labels to be completed is high and the number is moderate, a larger preset user similarity can be set. For the user similarity of the first user and each second user, each user similarity is compared with the preset user similarity in turn, and then it is determined according to the comparison result whether the second user corresponding to the user similarity is a similar user of the first user.

[0115] S04, if the user similarity is greater than or equal to the preset user similarity, the second user corresponding to the user similarity is the similar user corresponding to the first user.

[0116] Specifically, for the user similarity of the first user and each second user, the user similarity is compared with the preset user similarity. If the user similarity is greater than or equal to the preset user similarity, the second user corresponding to the user similarity is the similar user of the first user. If the user similarity is less than the preset user similarity, the second user corresponding to the user similarity is a non-similar user. By comparing each user similarity with the preset user similarity in turn, the similar user of the first user can be screened from each second user.

[0117] It should be noted that steps S03 and S04 are specific steps of an implementation of step S2, and step S2 can also be implemented in other manners, which are not limited herein.

[0118] Based on the above steps, each user in the user cluster is sequentially calculated to obtain the similar users of each user, and then a user-tag graph is generated. Each node in the graph is a user (a first user or a second user), and a first user is connected to each of its similar users. The edge between the first user and the similar user is the similarity W between the two nodes. 1,2 .

[0119] S05, obtaining the personal tag set of the first user and the personal tag set of at least one similar user, and generating the associated tag set of the first user.

[0120] Based on the above user-tag graph, the personal tag set of the first user is obtained, and the personal tag set of each adjacent node (similar user) of the first user is obtained. The tags in these personal tag sets are collected to generate the associated tag set of the first user. In other words, the associated tag set of the first user is the union of the personal tag set of the first user and the personal tag set of each similar user of the first user.

[0121] S06, determining the association probability of each associated tag in the associated tag set.

[0122] After generating the associated tag set based on the personal tag set of each similar user and the personal tag set of the first user, the association probability of each associated tag in the associated tag set needs to be calculated sequentially. The association probability is used to indicate the association of the associated tag with the first user.

[0123] In some examples, S06 determines the association probability of each associated tag in the associated tag set includes a plurality of sub-steps:

[0124] Sub-step one, for each associated tag, determining the first initial association probability of the first user corresponding to the associated tag.

[0125] Specifically, an associated tag can belong to the personal tag set of the first user and / or at least one similar user. For each associated tag, if the associated tag exists in the personal tag set of the first user, the first initial association probability P1 of the first user is taken as 1, and if the associated tag does not exist in the personal tag set of the first user, i.e., the corresponding result of the associated tag and the first user is empty, the first initial association probability P1 of the first user is taken as 0. i,a i,a .

[0126] Sub-step two, determining the second initial association probability of at least one similar user to which the associated tag belongs.​

[0127] Specifically, one association tag can belong to the personal tag set of the first user and / or at least one similar user. For the personal tag set of each similar user, if one association tag belongs to the personal tag set, the second initial association probability P2 of the association tag to the similar user is 1. j,a If one association tag does not belong to the personal tag set, the second initial association probability P2 of the association tag to the similar user is 0. j,a Specifically, one association tag can belong to the personal tag set of the first user and / or at least one similar user. For the personal tag set of each similar user, if one association tag belongs to the personal tag set, the second initial association probability P2 of the association tag to the similar user is 1.

[0128] Sub-step three, determining the association probability of the association tag according to the first initial association probability P1 i,a , the second initial association probability P2 j,a , and the user similarity w of the first user corresponding to the association tag and the at least one similar user. ij

[0129] Specifically, sub-step three specifically includes:

[0130] The association probability P of each association tag is calculated according to the following formula: i,a

[0131]

[0132] wherein P1 i,a is the first initial association probability, P2 j,a is the second initial association probability, j is the similar user, V is the set of similar users of the first user, and w ij is the similarity of the first user (i represents the first user) and the similar user (j represents the similar user) currently calculated.

[0133] S07, determining at least one completion tag of the first user according to the association probability of each association tag.

[0134] Specifically, the association probability is used to indicate the association of the association tag with the first user, and the association tag with higher association is selected as the completion tag of the first user according to the size of the association probability, and the personal tag set of the first user is updated according to the completion tag.

[0135] In some examples, S07 includes multiple manners. For example, manner one and manner two.

[0136] Manner one:

[0137] Step one, comparing the association probability of each association tag with a preset association probability.

[0138] ​​Based on the above calculation, the association probability of each association label in the association label set of the first user is calculated, for each association label, the association probability of the association label is compared with the preset association probability, and whether the association label is a complete label of the first user is determined according to the comparison result.

[0139] Step two, if the association probability of the association label is greater than or equal to the preset association probability, the association label is determined as a complete label.

[0140] The preset association probability can be set in various ways, for example: the association probability of each label contained in the personal label set of the first user is calculated; the minimum value of the association probability of each label is selected as the preset association probability.

[0141] For each association label, the association probability of the association label is compared with the preset association probability, if the association probability of the association label is greater than or equal to the preset association probability, the association label is determined as a complete label; if the association probability of the association label is less than the preset association probability, the association label is a non-complete label.

[0142] Method two:

[0143] Step one, the association probability of each association label is sorted from large to small.

[0144] Based on the above calculation, the association probability of each association label in the association label set of the first user is calculated, and then each association label is sorted according to the association probability from large to small.

[0145] Step two, select the preset number of association labels with the largest association probability as complete labels.

[0146] Based on the above sorting result, the preset number is determined according to the number of complete labels required, and the preset number of association labels with the largest association probability is selected as complete labels according to the sorting result.

[0147] After determining the complete labels according to method one or method two, the personal label set of the first user is updated according to the complete labels, so as to complete the labels of the first user.

[0148] It should be noted that steps S05, S06 and S07 are specific steps of one implementation manner of step S3, and step S3 can also be implemented by other manners, which is not limited here.

[0149] The label completion method provided in the embodiment determines the similar users of the first user, and then determines the associated label set of the first user according to the personal label set of the first user and the personal label set of the similar users, and then determines the completion label of the first user according to the association probability of each associated label in the associated label set, and updates the personal label set of the first user according to the completion label, so that the label transmission from the similar users to the first user can be realized by relying on the similarity between the similar users and the first user, and the label of one first user can be completed by using the label of the similar user.

[0150] Embodiment three,

[0151] Referring to Figure 3 , Figure 3 The structure schematic diagram of the label completion device provided in one embodiment of the application is shown. The label completion device provided in the embodiment comprises:

[0152] The user similarity determination module 101 is configured to determine the user similarity between the first user and each second user according to the personal label set of the first user and the personal label set of each second user.

[0153] The similar user determination module 102 is configured to determine at least one similar user corresponding to the first user from each second user according to the user similarity between the first user and each second user.

[0154] The label completion module 103 is configured to determine at least one completion label of the first user according to the personal label set of the first user and the personal label set of the at least one similar user, and update the personal label set of the first user according to the at least one completion label.

[0155] The label completion device provided in the embodiment determines the similar users of the first user, and then determines the completion label of the first user according to the personal label set of the first user and the personal label set of the similar users, and updates the personal label set of the first user according to the completion label, so that the label transmission from the similar users to the first user can be realized by relying on the similarity between the similar users and the first user, and the label of one first user can be completed by using the label of the similar user.

[0156] Embodiment four,

[0157] The application provides an electronic device, comprising:

[0158] at least one processor, and

[0159] a memory connected to the at least one processor in communication. Wherein,

[0160] The memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the short message verification described above.

[0161] In the technical solutions of the present disclosure, the acquisition, storage and application of user personal information are in line with relevant laws and regulations and do not violate public order and good customs.

[0162] Referring to Figure 4 , Figure 4 A schematic block diagram of an example electronic device 800 that can be used to implement embodiments of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptops, desktops, tablets, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. The electronic device can also represent various forms of mobile devices, such as personal digital processors, cellular telephones, smart phones, wearable devices, and other similar computing devices. The components shown here, their connections and relationships, and their functions, are meant to be examples only, and are not meant to limit implementations of the present disclosure described and / or claimed in this document.

[0163] As shown in Figure 4 , the device 800 includes a computing unit 801 that can perform various appropriate actions and processes in accordance with a computer program stored in a read-only memory (ROM) 802 or a computer program loaded from a storage unit 808 into a random access memory (RAM) 803. Various programs and data required for the operation of the device 800 can also be stored in the RAM 803. The computing unit 801, the ROM 802, and the RAM 803 are connected to each other through a bus 804. An input / output (I / O) interface 805 is also connected to the bus 804.

[0164] Various components in the device 800 are connected to the I / O interface 805, including: an input unit 806, such as a keyboard, a mouse, etc. An output unit 807, such as various types of displays, speakers, etc. A storage unit 808, such as a magnetic disk, an optical disk, etc. And a communication unit 809, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 809 allows the device 800 to exchange information / data with other devices through a computer network, such as the Internet, and / or various telecommunications networks.

[0165] The computing unit 801 can be various general and / or special purpose processing components with processing and computing capabilities. Some examples of the computing unit 801 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various specialized artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 801 performs various methods and processes described above, such as the label completion method. For example, in some embodiments, the label completion method can be implemented as a computer software program tangibly embodied in a machine-readable medium, such as the storage unit 808. In some embodiments, part or all of the computer program can be loaded and / or installed onto the device 800 via the ROM 802 and / or the communication unit 809. When the computer program is loaded onto the RAM 803 and executed by the computing unit 801, one or more steps of the label completion method described above can be performed. Alternatively, in other embodiments, the computing unit 801 can be configured to perform the label completion method by any other suitable means, such as by means of firmware.

[0166] Various implementations of the systems and techniques described above can be realized in digital electronic circuitry, integrated circuitry, a field programmable gate array (FPGA), an application specific integrated circuit (ASIC), a system on a chip (SOC), a programmable logic device (PLD), a computer hardware, firmware, software, and / or combinations thereof. These various implementations can include implementation in one or more computer programs that are executable and / or interpretable on a programmable system including at least one programmable processor, which can be special or general purpose, coupled to receive data and instructions from, and to transmit data and instructions to, a storage system, at least one input device, and at least one output device.

[0167] Program code for carrying out methods of the present disclosure can be written in any combination of one or more programming languages. The program code can be provided to a processor or controller of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the program code, when executed by the processor or controller, produces a function / operation specified in the flowchart and / or block diagram. The program code can be executed entirely on a machine, partially on a machine, partially on a machine as a stand-alone software package, and partially on a remote machine or a server.

[0168] Embodiment Five,

[0169] The present application provides a non-transitory computer readable storage medium storing computer instructions, wherein the computer instructions are used to make a computer execute the label completion method according to the above.

[0170] In the context of this disclosure, a machine-readable medium can be a tangible medium that contains or stores a program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. Machine-readable storage media can include, without limitation, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media can include, but are not limited to, an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0171] To provide for interaction with a user, the systems and techniques described here can be implemented on a computer having a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user and a keyboard and a pointing device (e.g., a mouse or a trackball) by which the user can provide input to the computer. Other kinds of devices can be used to provide for interaction with a user as well; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form, including acoustic, speech, or tactile input.

[0172] The systems and techniques described here can be implemented in a computing system that includes a back end component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front end component (e.g., a user computer having a graphical user interface or a Web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such back end, middleware, or front end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network (LAN), a wide area network (WAN), and the Internet.

[0173] The computer system can include clients and servers. A client and server are generally remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. The server can be a cloud server, a server of a distributed system, or a server combined with a blockchain.

[0174] It should be understood that the various forms of flow shown above can be re-ordered, added to, or have steps deleted, using the steps shown above. For example, the steps recited in the present disclosure can be performed in parallel, in series, or in a different order, as long as the desired results of the technology disclosed in the present disclosure are achieved, which are not limited herein.

[0175] It can be understood that the above embodiments are only exemplary embodiments adopted for illustrating the principles of the present application, and the present application is not limited thereto. Various modifications and improvements can be made by those of ordinary skill in the art without departing from the spirit and essence of the present application, and these modifications and improvements are also considered to be within the scope of protection of the present application.

Claims

1. A tag completion method, characterized in that: The method comprises: For each second user, respectively, determine a similar tag set between the first user and the second user based on the personal tag set of the first user and the personal tag set of the second user; for each similar tag set, determine the semantic similarity of the similar tag set based on the tag similarity between each similar tag included in the similar tag set; determine a user similarity algorithm corresponding to a preset interval to which the semantic similarity belongs, and determine the user similarity between the first user and the second user corresponding to the similar tag set based on the user similarity algorithm; determining, based on the user similarities between the first user and each second user, at least one similar user corresponding to the first user among each second user; Determining at least one complementary tag of the first user based on the personal tag set of the first user and the personal tag set of the at least one similar user, and updating the personal tag set of the first user based on the at least one complementary tag; The preset intervals include: a first preset interval, a second preset interval, and a third preset interval, and the semantic similarities corresponding to the first preset interval, the second preset interval, and the third preset interval increase in sequence; determining the user similarity algorithm corresponding to the preset interval to which the semantic similarity belongs, and determining the user similarity between the first user and the second user corresponding to the similar tag set according to the user similarity algorithm include: determining, for each of the semantic similarities of the similar tag set, the preset interval to which the semantic similarity belongs; If the semantic similarity of the similar tag set belongs to the first preset interval, the user similarity W between the first user and the second user corresponding to the similar tag set is determined according to the following user similarity algorithm: 1，2 : If the semantic similarity of the similar tag set belongs to the second preset interval, the user similarity W between the first user and the second user corresponding to the similar tag set is determined according to the following user similarity algorithm: 1，2 : If the semantic similarity of the similar tag set belongs to the third preset interval, then the user similarity W 1,2 =1; Among them, labels represents the calculated similarity label set, N is the number of similar labels in the similarity label set, S ij Represents the similarity between any two similar tags in the similar tag set.

2. The method according to claim 1, characterized in that The step of determining the semantic similarity of each similar tag set according to the tag similarities between the similar tags included in the similar tag set comprises: For each similar tag set, obtain the word vector of each similar tag in the similar tag set, and calculate the tag similarity between each two similar tags based on the word vectors of each two similar tags; The semantic similarity of the similar tag set is determined according to the tag similarity between every two similar tags.

3. The method according to claim 1, characterized in that The determining, based on the user similarities between the first user and each second user, at least one similar user corresponding to the first user among each second user includes: For each user similarity, compare the user similarity with the preset user similarity; If the user similarity is greater than or equal to the preset user similarity, the second user corresponding to the user similarity is the similar user corresponding to the first user.

4. The method according to claim 1, wherein The determining, based on the personal tag set of the first user and the personal tag set of the at least one similar user, at least one completion tag of the first user comprises: Acquire the personal tag set of the first user and the personal tag set of the at least one similar user to generate an associated tag set of the first user; Determining an association probability of each associated tag in the associated tag set; At least one completion tag of the first user is determined according to the association probabilities of the respective association tags.

5. The method according to claim 4, characterized in that Determining the association probability of each association tag in the association tag set includes: For each of the association tags, determining a first initial association probability that the association tag corresponds to the first user; Determining a second initial association probability of at least one of the similar users to which the association tag belongs; The association probability of the association tag is determined according to the first initial association probability, the second initial association probability, and the user similarity between the first user corresponding to the association tag and at least one of the similar users.

6. The method according to claim 5, characterized in that The determining the association probability of the association tag according to the first initial association probability, the second initial association probability, and the user similarity between the first user corresponding to the association tag and at least one of the similar users includes: The association probability P of each associated tag is calculated according to the following formula: i，a : Among them, P1 i，a is the first initial association probability, P2 j，a is the second initial association probability, j is a similar user, V is the set of similar users of the first user, w ij is the similarity between the first user and similar users.

7. The method according to any one of claims 4 to 6, characterized in that: The determining, based on the association probabilities of the associated tags, at least one completion tag for the first user includes: comparing the association probability of each of the association tags with a preset association probability; If the association probability of the association tag is greater than or equal to the preset association probability, the association tag is determined to be a completion tag.

8. The method according to any one of claims 4 to 6, characterized in that: The determining, based on the association probabilities of the associated tags, at least one completion tag for the first user includes: sorting the association probabilities of the associated tags from large to small; A preset number of associated tags with the largest associated probabilities are selected as the completion tags.

9. A label completion device, characterized in that: include: A user similarity determination module is configured to determine, for each second user, a similar tag set between the first user and the second user based on the first user's personal tag set and the second user's personal tag set; determine, for each similar tag set, a semantic similarity of the similar tag set based on the tag similarity between the respective similar tags contained in the similar tag set; determine a user similarity algorithm corresponding to a preset interval to which the semantic similarity belongs, and determine, based on the user similarity algorithm, a user similarity between the first user and the second user corresponding to the similar tag set; a similar user determining module, configured to determine at least one similar user corresponding to the first user among each second user based on the user similarity between the first user and each second user; a tag completion module, configured to determine at least one completion tag for the first user based on the personal tag set of the first user and the personal tag set of the at least one similar user, and update the personal tag set of the first user based on the at least one completion tag; Wherein, the preset intervals include: a first preset interval, a second preset interval and a third preset interval, and the semantic similarities corresponding to the first preset interval, the second preset interval and the third preset interval increase in sequence; the user similarity algorithm corresponding to the preset interval to which the semantic similarity belongs is determined, when the user similarity between the first user and the second user corresponding to the similar tag set is determined according to the user similarity algorithm, the user similarity determination module is used to: determine the preset interval to which the semantic similarity belongs for the semantic similarity of each similar tag set; If the semantic similarity of the similar tag set belongs to the first preset interval, the user similarity W between the first user and the second user corresponding to the similar tag set is determined according to the following user similarity algorithm: 1,2 : If the semantic similarity of the similar tag set belongs to the second preset interval, the user similarity W between the first user and the second user corresponding to the similar tag set is determined according to the following user similarity algorithm: 1,2 : If the semantic similarity of the similar tag set belongs to the third preset interval, then the user similarity W 1,2 =1; Among them, labels represents the calculated similarity label set, N is the number of similar labels in the similarity label set, S ij Represents the similarity between any two similar tags in the similar tag set.

10. An electronic device comprising: at least one processor; as well as a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 8.

11. A non-transitory computer-readable storage medium storing computer instructions, wherein: The computer instructions are used to cause the computer to execute the method according to any one of claims 1-8.

Images