A label modification method based on user behavior, a server and a storage medium

Abstract

A label modification method based on user behavior, a server and a storage medium are used to quickly and effectively clean up the label content that is not interested by the user. The embodiment of the application comprises: adding a first main label to an initial label set to obtain a first label set; adding a plurality of first-level sub-labels of the first main label to the first label set of the user portrait to obtain a second label set; determining a first number of the plurality of first-level sub-labels corresponding to the labels of the not interested push content marked by the user within a first preset time period; and rolling back the user portrait to the initial label set. The label modification method based on user behavior provided by the embodiment of the application avoids the user from frequently receiving the push of not interested content through the rollback of the user portrait, and improves the experience of the user receiving the related push through the efficient cleaning of the user label.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of big data information, and in particular to a label modification method based on user behavior, a server and a storage medium. BACKGROUND

[0002] With the development of the Internet and mobile Internet, the acquisition and processing of user data have become increasingly important. For enterprises and institutions, understanding the interests, needs and behavior habits of users can help them better market, product design and service optimization. User label portrait system emerges as the times require, aiming to analyze and process user data through automated algorithms and machine learning techniques, extract the key features and behavior patterns of users, and form user label portraits.

[0003] Due to the large amount of content on the push platform, users are bound to be mislabeled due to misoperation, and then be pushed a large number of related content, most of which may not meet the user's interests and needs, and greatly affect the user experience. Related technologies use a feedback response module to delete the labels corresponding to the pushed content when the user clicks on the not interested option for the pushed content.

[0004] However, there is often a correlation between labels, and simply deleting labels cannot completely clean up related labels, and cannot guarantee that the user will not receive other related push content that is not interested. SUMMARY

[0005] The present application provides a label modification method based on user behavior, a server and a storage medium, which is used to rollback the user label portrait and avoid the user from frequently receiving not interested content push, and improve the user's experience of accepting related push through efficient cleaning of user labels.

[0006] In a first aspect, the present application provides a label modification method based on user behavior, the method comprising: in response to a user's retrieval operation on a first main label, adding the first main label to an initial label set of a user portrait to obtain a first label set; in response to the user's selection operation on the first main label related push content, adding a plurality of first-level sub-labels of the first main label to the first label set of the user portrait to obtain a second label set; the plurality of first-level sub-labels are determined from the first main label related push content; determining a first number of a first unsuitable label corresponding to the user marked not interested push content in a first preset time period in the plurality of first-level sub-labels; when the ratio of the first number to the total number of the plurality of first-level sub-labels is greater than a preset rollback threshold, rolling back the user portrait to the initial label set.

[0007] In the above embodiment, the number of labels marked as not interested by the user is determined, and when the proportion of the number reaches a certain value, the entire label is rolled back, making the label clearing more efficient and clean, avoiding the user receiving the corresponding push information after marking the label as not interested, and ensuring the user experience.

[0008] In some embodiments of the first aspect, after the first plurality of sub-level labels of the first main label are added to the first label set of the user portrait to obtain a second label set in response to the selection operation of the user on the first main label related push content, the method further comprises: determining a second number of second unsuitable labels in the first plurality of sub-level labels that are not viewed by the user for push content within a second preset time period; and rolling back the user portrait to the initial label set when a ratio of the second number to a total number of the first plurality of sub-level labels is greater than a preset rollback threshold.

[0009] In the above embodiment, the user's not viewing of the push content related to the label is recorded, and on the other hand, the label not interested by the user is detected, ensuring the accuracy of the corresponding label collection, avoiding the continuous push of the push content not interested by the user, and further improving the user experience.

[0010] In some embodiments of the first aspect, after the second number of second unsuitable labels in the first plurality of sub-level labels that are not viewed by the user for push content within a second preset time period is determined, the method further comprises: deleting the second unsuitable labels in the user portrait to obtain a third label set when a ratio of the second number to a total number of the first plurality of sub-level labels is not greater than a preset rollback threshold.

[0011] In the above embodiment, when the proportion of the labels marked as not interested by the user does not reach the specified threshold, the corresponding sub-level labels are deleted instead of rolling back the main label, avoiding the problem of losing part of the content interested by the user due to rollback, and ensuring that the user normally receives the related push.

[0012] In some embodiments of the first aspect, after the second number of second unsuitable labels in the first plurality of sub-level labels that are not viewed by the user for push content within a second preset time period is determined, the method further comprises: rolling back the user portrait to the initial label set when a ratio of the first number, the second number, and a total number of the first plurality of sub-level labels is greater than a preset rollback threshold.

[0013] In the above embodiment, the labels marked as not interested by the user and the labels not viewed by the user and inferred as not interested are recorded at the same time, further ensuring the accuracy of the collection of the degree of interest of the user, and the label is rolled back when the proportion of the label reaches a certain value, ensuring the timely cleaning of the content not interested by the user.

[0014] In some embodiments of the first aspect, in response to the user selecting the first main label related push content, the method further comprises: adding the first main label and the first number of first-level sub-labels to the first label set of the user portrait, to obtain a second label set; determining a second number of second-level sub-labels of the first-level sub-labels in the second label set; in response to the user selecting the first-level sub-label related push content, binding the second-level sub-labels of the first-level sub-label to the user portrait, to obtain a sixth label set; determining a third number of second-level unsuitable labels of the second-level sub-labels corresponding to the unsuitable push content marked by the user in a first preset time period; and deleting the first unsuitable label when a ratio of the third number to a total number of the second-level sub-labels is greater than a preset threshold.

[0015] In the above embodiments, the similarity of the sub-labels is calculated, and the sub-labels with low similarity are retained after the main label is rolled back, thereby avoiding the misdeletion of irrelevant content of the label marked by the user as not interesting, and ensuring the accuracy of the label cleaning.

[0016] In some embodiments of the first aspect, in response to the user selecting the first main label related push content, the method further comprises: adding the first main label and the first number of first-level sub-labels to the first label set of the user portrait, to obtain a second label set; determining a second number of second-level sub-labels of the first-level sub-labels in the second label set; in response to the user selecting the first-level sub-label related push content, binding the second-level sub-labels of the first-level sub-label to the user portrait, to obtain a sixth label set; determining a third number of second-level unsuitable labels of the second-level sub-labels corresponding to the unsuitable push content marked by the user in a first preset time period; and deleting the first unsuitable label when a ratio of the third number to a total number of the second-level sub-labels is greater than a preset threshold.

[0017] In the above embodiments, the similar labels to the label marked by the user as not interesting among all the labels in the user portrait are calculated and deleted even if they are not under the same branch of the main label, thereby further ensuring the depth of the relevant content deletion, improving the accuracy of the push based on the user label, and ensuring the user experience.

[0018] In some embodiments of the first aspect, in response to the user selecting the first main label related push content, the method further comprises: adding the first main label and the first number of first-level sub-labels to the first label set of the user portrait, to obtain a second label set; determining a second number of second-level sub-labels of the first-level sub-labels in the second label set; in response to the user selecting the first-level sub-label related push content, binding the second-level sub-labels of the first-level sub-label to the user portrait, to obtain a sixth label set; determining a third number of second-level unsuitable labels of the second-level sub-labels corresponding to the unsuitable push content marked by the user in a first preset time period; and deleting the first unsuitable label when a ratio of the third number to a total number of the second-level sub-labels is greater than a preset threshold.

[0019] In the above embodiments, when analyzing each sub-label under the main label, the same method as the main label is adopted, and the sub-label is rolled back when the number of the sub-label marked by the user reaches a certain value, so that the method is more widely applicable, and the accurate implementation of the label deletion is ensured.

[0020] In a second aspect, an embodiment of the present application provides a server, comprising: a main label determination module, configured to add a first main label to an initial label set of a user portrait in response to a user's retrieval operation on the first main label, to obtain a first label set; a sub-label determination module, configured to add a plurality of first-level sub-labels of the first main label to the first label set of the user portrait in response to the user's selection operation on the first main label related push content, to obtain a second label set; an unsuitability determination module, configured to determine a first number of a first unsuitable label corresponding to the user-labeled uninterested push content in a first preset time period in the plurality of first-level sub-labels; and a portrait rollback module, configured to rollback the user portrait to the initial label set when a ratio of the first number to a total number of the plurality of first-level sub-labels is greater than a preset rollback threshold.

[0021] In a third aspect, an embodiment of the present application provides a server, comprising: one or more processors and a memory; the memory is coupled to the one or more processors, and the memory is configured to store computer program codes, the computer program codes comprising computer instructions, and the one or more processors are configured to invoke the computer instructions to enable the server to perform the method described in the first aspect and any possible implementation manner of the first aspect.

[0022] In a fourth aspect, an embodiment of the present application provides a computer program product comprising instructions, which, when executed on a server, cause the server to perform the method described in the first aspect and any possible implementation manner of the first aspect.

[0023] In a fifth aspect, an embodiment of the present application provides a computer-readable storage medium comprising instructions, which, when executed on a server, cause the server to perform the method described in the first aspect and any possible implementation manner of the first aspect.

[0024] It can be understood that the server provided in the second aspect and the third aspect, the computer program product provided in the fourth aspect, and the computer storage medium provided in the fifth aspect are all used to execute the method provided in the embodiments of the present application. Therefore, the beneficial effects that can be achieved are referred to the beneficial effects in the corresponding method, which will not be described here.

[0025] The one or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages:

[0026] 1. Since the method of automatically deleting the corresponding main label when the user-labeled uninterested label reaches a certain proportion is adopted, the removal of the corresponding label is more efficient and clean, and the problem that there are still related push contents after the user frequently labels the uninterested contents is effectively solved, thereby the user experience of the push based on the label is improved.

[0027] 2. The method of automatically determining the corresponding label as a label of no interest of the user if the user does not view the relevant push content, so that the label of no interest of the user is collected from multiple aspects, the problem that the related art cannot accurately determine the label of no interest of the user is effectively solved, and the accuracy of label rollback and deletion is realized.

[0028] 3. The method of judging the similarity of each label and the label of no interest marked by the user, so that the deletion of the label of interest of the user is avoided, and similar labels under other main labels are deleted, the problem of incomplete label deletion is effectively solved, the precision of label deletion is realized, and the related push is more beneficial. BRIEF DESCRIPTION OF DRAWINGS

[0029] Figure 1 is an application scenario diagram of a label modification based on user behavior in the embodiment of the present application;

[0030] Figure 2 is an exemplary scenario diagram of a label modification based on user behavior in the related art;

[0031] Figure 3 is an exemplary scenario diagram of a label modification based on user behavior provided by the embodiment of the present application;

[0032] Figure 4 is a flow diagram of a label modification based on user behavior provided by the embodiment of the present application;

[0033] Figure 5 is another exemplary scenario diagram of a label modification based on user behavior provided by the embodiment of the present application;

[0034] Figure 6 is another flow diagram of a label modification based on user behavior provided by the embodiment of the present application;

[0035] Figure 7 is a functional module structure diagram of a server provided by the embodiment of the present application;

[0036] Figure 8 is an entity device structure diagram of a server provided by the embodiment of the present application. DETAILED DESCRIPTION

[0037] The terminology used in the following embodiments of the present application is for the purpose of describing particular embodiments only and is not intended to be limiting of the present application. As used in the description of the embodiments of the application and the appended claims, the singular forms "a," "an" and "the" are intended to include both singular and plural forms, unless the context clearly indicates otherwise. It will be further understood that the terms "and / or," as used in the description of the embodiments of the application, refers to any one or more of the listed items, optionally including zero.

[0038] Hereinafter, the terms "first", "second" are only for the purpose of description, and cannot be understood as implying or suggesting relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined with "first", "second" can explicitly or implicitly include one or more of the features, and in the description of the embodiments of the present application, the meaning of "multiple" is two or more, unless otherwise specified.

[0039] The technical solutions provided by the present application can be applied to the scenario of user behavior-based label modification of a server. Please refer to Figure 1 , which is a schematic diagram of an application scenario of a user behavior-based label modification in the embodiments of the present application.

[0040] Figure 1 A search box interface is shown in the figure, in which a user inputs a keyword and performs a search. The server collects the keyword information input by the user, generates a corresponding label, and then binds the label to the portrait model of the user. The user portrait is a model used to represent the user's habits and preferences. The labels on the portrait are used to represent the user's preference data. The server can classify users according to their portrait labels to facilitate information pushing and management. It should be noted that the user portrait can also be described as a portrait, a label portrait, a portrait model, etc. in the following text. Figure 1 In the figure, the user searches for CC, and the portrait of the user is bound with the label CC.

[0041] For ease of understanding, let's take another practical example. A certain enterprise user uses a certain scientific and technological platform to view patent-related information, which indicates the development direction of the enterprise. After the enterprise undergoes track transformation, its development strategy and direction will change, and the pushing of patent-related information also needs to be adjusted accordingly. In a certain period of time, the development direction of the enterprise is electronic display screen. He searches for the label content of electronic display screen on the scientific and technological platform, and his user portrait is bound with the label of electronic display screen. Correspondingly, the scientific and technological platform will subsequently match and push relevant patent information with the label of electronic display screen to him according to the label of electronic display screen on his portrait. Relatively speaking, after a certain competing enterprise that produces electronic display screens has relevant actions of patent application, the server of the platform will also push corresponding information to the enterprise user.

[0042] It should be noted that in the above or later described examples, the enterprise user is only an exemplary concept representing a user using the scientific and technological innovation platform for patent information retrieval, which can be a company, a department, or an individual, and is also described as a user or an enterprise in the following.

[0043] It should be noted that the determination of the user tag is not only the search operation described above, but also comes from the clicking and browsing of the user on the pushed content. The clicking and browsing of the user on the pushed content marked with the corresponding tag can better reflect the user's habits and preferences, and the user tag can be better determined. For the tag content frequently clicked and browsed by the user, the server will also strengthen the push weight. In actual life, the server will also collect chat record keywords, voice keywords, location, weather, and other information to determine the user tag, so as to perform more accurate information pushing.

[0044] The tag modification based on the user behavior can bring a better information acquisition experience to the user, but in the case that the user determines a tag that is not suitable for the user due to a mistaken clicking operation, an operation of another user, or a mistaken judgment of the server, or the user is not suitable for a tag later, the deletion of the corresponding tag and the reduction of the pushing of related content are also crucial.

[0045] Please refer to Figure 2 , which is an exemplary scenario diagram of a tag modification based on user behavior in the related art.

[0046] Figure 2 A user tag portrait is shown in . In the user portrait in the figure, three main tags A, B, and C are bound, and correspondingly, A1, A2, A3, B1, and C1, which are five first-level sub-tags respectively belonging to the three main tags A, B, and C, are bound, and A11, A21, and A22, which are three second-level sub-tags respectively belonging to the first-level sub-tags A1 and A2, are also bound. The first-level sub-tag is a branch tag of the main tag, which is a refinement and extension of the main tag content. Correspondingly, the second-level sub-tag is a branch tag of the first-level sub-tag. Correspondingly, there are also third-level sub-tags, fourth-level sub-tags, and the like.

[0047] After the user is not interested in some tags, the user will click the feedback control to select the not interested option. Correspondingly, the tags that the user is not interested in will be deleted, as shown in Figure 2 , A3, A11, and A21, which are selected by the dashed line frame, are not interested tags clicked and deleted by the user. However, simple deletion of the tag may not meet the user's needs, and some related information pushing cannot be completely prohibited.

[0048] For example, a certain enterprise user's user portrait is bound with three main tags of electronic display screen, external speaker and recording device, and the electronic display screen is provided with a desktop computer display screen and a television display screen as first-level sub-tags; the desktop computer display screen is branched to have small screen, large screen and super large screen as second-level sub-tags; and the television display screen is branched to have super clear screen and super thin screen as second-level sub-tags. In a certain time period, because the business on the product is not smooth, the enterprise user cancels the R&D investment on the electronic display screen, and the corresponding product focus is shifted to other products. He marks and deletes the small screen, super large screen, super thin screen and other information tags through the uninterested feedback of the patent push content in a short time. However, the feedback deletion operation of the user is not good, and the large screen and super clear screen are still pushed, and the user needs to continue to feed back.

[0049] The label modification method based on user behavior provided by the embodiment of the application can quickly and effectively clean up the user label through the collection of user feedback information, and ensure that the push content received by the user is clean and comfortable.

[0050] Please refer to Figure 3 for an exemplary scenario diagram of a label modification based on user behavior provided by the embodiment of the application.

[0051] and Figure 2 The user in Figure 3 deleted three labels A3, A11 and A21 in a short time. The scenario is different from Figure 2 , in which the server reverts the user portrait and unbinds the entire main label A after detecting that the user deletes the three labels A3, A11 and A21 in a short time.

[0052] Correspondingly, in an actual example, a user only needs to feed back the deletion of the small screen, super large screen and super thin screen information tags, and the entire electronic display screen label will be reverted and unbound, and the scientific and technological innovation platform will not push the display screen related patent information.

[0053] In combination with the above scenario, the label modification method based on user behavior in the embodiment of the application is described below. Please refer to Figure 4 for a flow diagram of a label modification based on user behavior provided by the embodiment of the application.

[0054] S401, in response to a retrieval operation of a first main tag by a user, adding the first main tag to an initial tag set of a user portrait to obtain a first tag set;

[0055] Specifically, after detecting that the user performs a keyword search, the server determines the keyword as a first main tag, and adds the first main tag to an initial tag set representing all tags of a user portrait, to obtain a first tag set. In addition to the first main tag, the first tag set also includes other main tags determined in the initial tag set.

[0056] For example, the user inputs the keyword "electronic display screen" in the search box. The server adds "electronic display screen" as a main tag to the initial tag set of the user portrait. At this time, the first tag set corresponding to the user portrait includes the tag "electronic display screen".

[0057] In some embodiments, after the user inputs a keyword of a lower level, such as "television screen", the server extracts the upper-level concept "electronic display screen" corresponding to the keyword as a main tag, and adds "television screen" as a branch tag of the "electronic display screen" tag. Finally, the two tags are added to the user portrait, and the first tag set of the user portrait at this time includes the two tags "electronic display screen" and "television screen".

[0058] S402, in response to the selection operation of the user on the first main tag related push content, adding a plurality of first-level sub-level tags of the first main tag to the first tag set of the user portrait to obtain a second tag set;

[0059] Specifically, after determining the first main tag, the server pushes the content corresponding to the first main tag to the user. These contents are the contents derived from the first main tag, and are associated with first-level sub-level tags. After the user selects the contents pushed by the server, the server determines a plurality of first-level sub-level tags selected by the user, and binds these first-level sub-level tags to the user portrait, that is, generates a second tag set in combination with the original first tag set. In addition to the first main tag and the plurality of first-level sub-level tags branched from the first main tag, the second tag set also includes other main tags and other branch tags determined in the past.

[0060] For example, after the user searches for "electronic display screen", the server pushes the content of "computer display screen", "television display screen", "mobile phone display screen", etc. to the user, and the user clicks to view the patent content of a plurality of computer display screens and television display screens. In response to the selection of the user, the server binds "computer display screen" and "television display screen" as first-level sub-level tags branched from "electronic display screen" to the user portrait. At this time, the user portrait corresponds to a second tag set, and the second tag set includes the main tag "electronic display screen" and the first-level sub-level tags "computer display screen" and "television display screen".

[0061] S403, determining a first number of first unsuitable tags in the plurality of first-level sub-level tags corresponding to the push content not interested by the user marked in a first preset time period;

[0062] Specifically, in the first preset time period, if it is detected that the user marks multiple push contents of no interest, the server determines that the marked contents correspond to a first unsuitable label. Further, the server counts the number of the first unsuitable labels and records the number as a first number.

[0063] In the above example, after the development focus of the enterprise deviates from the display screen field, the user of the enterprise clicks the no interest option for the push contents of multiple computer display screens and television display screens within one day. Then, the server records the computer display screen and the television display screen as the first unsuitable label, further counts the computer display screen and the television display screen label in the second label set, and records the first number as 2.

[0064] S404, when the ratio of the first number to the total number of the plurality of first sub-labels is greater than a preset rollback threshold, rollback the user portrait to the initial label set.

[0065] Specifically, after recording the number of the first unsuitable label, i.e., the first number, the server calculates the ratio of the first number to the total number of the first sub-labels corresponding to the first main label, i.e., calculates the proportion of the labels unsuitable for the user in the first sub-labels. The higher the proportion is, the less interested the user is in the content corresponding to the first main label. When the proportion exceeds the preset rollback threshold, the user portrait is rolled back to the initial label set stage, i.e., all information related to the first main label is removed. The preset rollback threshold is a value set in advance. The higher the value is, the more unsuitable labels are needed for rollback.

[0066] It should be noted that the rollback step only takes effect for the content corresponding to the first main label and should not affect other labels. After determining the first label set, the main consideration is the content corresponding to the first main label. Rolling back to the initial label set means emptying the label set after removing all labels corresponding to the first main label, and the content of other main labels is not affected, i.e., the rollback is regional and directional.

[0067] In the above example, after the user of the enterprise clicks the no interest option for the push contents of multiple computer display screens and television display screens, the server determines that the first number is 2, and calculates the ratio of the first number to the total number of the first sub-labels as 100%, which is greater than the preset 70%. Then, the rollback is performed immediately, and the first main label, i.e., the electronic display screen, on the user portrait is unbound. If only the first main label is considered, this step is equivalent to rolling back the user portrait to before the first main label, i.e., the electronic display screen, is bound.

[0068] It can be understood that, compared with the case that the related art cannot quickly and effectively clear the label of no interest of the user, the method provided in the embodiment of the present application can record the label marked by the user, and judge whether the user is not interested in the main label. After determining that the user is not interested, the label state is rolled back, so that the main label is quickly unbound, and the push of the related label content that is not cleaned up completely does not affect the user experience.

[0069] The above is one of the ideal scenarios and corresponding embodiments of the present application. In actual life, there can be other cases where the labels of other main label branches affect the user experience. The present application proposes some embodiments for these cases, which can be used to optimize the above problems and further improve the user experience.

[0070] Next, another scenario related to these embodiments will be introduced. Please refer to Figure 5 , which is another example scenario of label modification based on user behavior provided by the embodiment of the present application.

[0071] And Figure 3 , the scenario is similar to Figure 5 , the user in Figure 3 deleted the A11 and A21 labels in a short time. And the difference between the scenario and

[0072] is that the server deletes the related A1, A2, B11 labels after detecting that the user deletes the A11 and A21 labels in a short time. The labels A1, A2, and B11 are labels associated with the labels A11 and A21.

[0073] Next, the method steps provided by the present application will be further described in combination with the above scenario. Please refer to Figure 6 , which is another flowchart of label modification based on user behavior provided by the embodiment of the present application.

[0074] S601, in response to the user's retrieval operation on the first main label, adding the first main label to the initial label set of the user portrait to obtain a first label set;

[0075] Referring to step S401, after detecting that the user performs a keyword retrieval, the server determines the keyword as the first main label, and adds the first main label to the initial label set representing all labels of the user portrait to obtain a first label set. In addition to the first main label, the first label set also has other main labels determined in the previous initial label set.

[0076] In some embodiments, after the user inputs a lower-level keyword such as a computer display screen, the server extracts the corresponding upper-level concept electronic display screen as the main tag, and then binds the computer display screen as a branch tag of the electronic display screen tag. Finally, the two tags are added to the user portrait together, and the first tag set of the user portrait at this time includes the two tags of electronic display screen and computer display screen.

[0077] In some embodiments, after the user inputs a lower-level keyword such as a computer display screen, the server extracts the corresponding upper-level concept electronic display screen, and after detecting that the electronic display screen is the main tag in the initial tag set of the user portrait, the server binds the computer display screen as a branch tag of the electronic display screen.

[0078] S602, in response to the selection operation of the user on the first main tag related push content, adding a plurality of first level sub-level tags of the first main tag to the first tag set of the user portrait to obtain a second tag set;

[0079] Referring to S402, after determining the first main tag, the server pushes the content corresponding to the first main tag to the user. These contents are the contents derived from the first main tag, and will have first level sub-level tags. After the user selects the server pushed content, the server determines the plurality of first level sub-level tags selected by the user, and binds these first level sub-level tags to the user portrait, that is, generates a second tag set combined with the original first tag set.

[0080] In some embodiments, in response to the user's click and gradual deepening of the push content, the second tag set will not only include the first main tag and the corresponding first level sub-level tag, but also include the corresponding second level sub-level tag, third level sub-level tag, etc. These tags are branch tags that are gradually refined. For example, the electronic display screen branches the computer display screen, and the computer display screen branches the large screen, the small screen, the curved screen, etc.

[0081] In some embodiments, in response to the user's click and gradual deepening of the push content corresponding to the first level sub-level tag, the server binds a plurality of second level sub-level tags corresponding to the first level sub-level tag to the user portrait to obtain a sixth tag set. In comparison, similar to steps S403 and S404, the server determines a plurality of second level sub-level tags corresponding to the push content not interested by the user in the first preset time period through the not interested marking of the user, and records the second level sub-level tags as second level unsuitable tags, and then determines the total number of the second level unsuitable tags, and records the total number as a third number. Finally, when the ratio of the third number to the total number of second level sub-level tags exceeds a specified threshold, the corresponding first level sub-level tag is determined and deleted. That is, the user can delete the entire computer display screen tag by deleting the small screen and super large screen tags.

[0082] S603, determining a first number of first unsuitable tags in the plurality of first-level sub-level tags corresponding to the uninterested push contents marked by the user in the first preset time period;

[0083] Referring to S403, in the first preset time period, if it is detected that the user marks a plurality of uninterested push contents, the server determines that the marking tags corresponding to these marked contents are first unsuitable tags. Further, the server counts the number of these first unsuitable tags, and records the number as a first number.

[0084] S604, determining a second number of second unsuitable tags in the plurality of first-level sub-level tags that are not viewed by the user in the second preset time period;

[0085] Corresponding to S603, the server determines the push contents that are not viewed by the user in the second preset time period, and determines that the marking tags corresponding to these ignored contents are second unsuitable tags. Further, the server counts the number of these second unsuitable tags, and records the number as a second number.

[0086] S605, when a ratio of the first number, the second number, and a total number of the plurality of first-level sub-level tags is greater than a preset rollback threshold, rolling back the user portrait to the initial tag set;

[0087] After determining the first number and the second number, the server sums the first number and the second number to obtain a total of unsuitable tags for the user. Further, the server calculates a ratio of the total of unsuitable tags to a total number of first-level sub-level tags corresponding to the first main tag, that is, calculates a total proportion of the first-level sub-level tags that are unsuitable for the user, and the higher the proportion, the more uninterested the user is in the content corresponding to the first main tag.

[0088] When the proportion exceeds the preset rollback threshold, the server rolls back the user portrait to the initial tag set stage, that is, removes all information related to the first main tag. The preset rollback threshold is a value set in advance, and the higher the value, the more unsuitable tags are required for rollback.

[0089] S606, determining whether a ratio of the first number and a total number of the plurality of first-level sub-level tags is greater than a preset rollback threshold;

[0090] Referring to S404, after recording the number of first unsuitable tags, that is, the first number, the server calculates a ratio of the first number to a total number of first-level sub-level tags corresponding to the first main tag, that is, calculates a proportion of the first-level sub-level tags that are unsuitable for the user, and the higher the proportion, the more uninterested the user is in the content corresponding to the first main tag. Further, the server determines whether the proportion is greater than a preset rollback threshold, and performs subsequent tag processing according to the determination result.

[0091] S607, determining whether the ratio of the second number to the total number of the plurality of first sub-level labels is greater than a preset fallback threshold value;

[0092] Referring to the S404 part, after recording the second number of second unsuitable labels, the server calculates the ratio of the second number to the total number of the first sub-level labels corresponding to the first main label, that is, calculates the proportion of the second unsuitable labels in the first sub-level labels that are not suitable for the user. The higher the proportion, the less interested the user is in the content corresponding to the first main label. Further, the server determines whether the proportion is greater than a preset fallback threshold value, and performs subsequent label processing according to the determination result.

[0093] S608, deleting the first and second unsuitable labels in the user portrait respectively to obtain a third label set;

[0094] When it is determined that the proportion of the first unsuitable label of the user is not greater than the preset fallback threshold value, the server determines that the user is not interested in all the content of the first main label at this time, and directly deletes the corresponding first unsuitable label. The user portrait after deleting the corresponding unsuitable label is the third label set at this time.

[0095] When it is determined that the proportion of the second unsuitable label of the user is not greater than the preset fallback threshold value, the server determines that the user is not interested in all the content of the second main label at this time, and directly deletes the corresponding second unsuitable label. The user portrait after deleting the corresponding unsuitable label is the third label set at this time.

[0096] S609, falling back the user portrait to the initial label set;

[0097] When it is determined that the proportion of the first unsuitable label of the user is greater than the preset fallback threshold value or it is determined that the proportion of the second unsuitable label of the user is greater than the preset fallback threshold value, the user portrait is fallen back to the initial label set stage, that is, all information related to the first main label is removed. The preset fallback threshold value is a value set in advance, and the higher the value, the more unsuitable labels are required for fallback.

[0098] S610, determining the similarity of the plurality of first sub-level labels and the first unsuitable label;

[0099] After determining the plurality of first unsuitable labels and the number thereof, the server determines the similarity of the first main label branch and the first unsuitable labels. The higher the similarity, the more likely the sub-level label is to become the next object of the user to mark and delete; the lower the similarity, the less clear the user's attitude towards the label, and the suggestion is to keep the label.

[0100] S611, when the similarity is lower than a preset proximity threshold value, keeping the corresponding first sub-level label in the user portrait to obtain a fourth label set;

[0101] Correspondingly, the similarity of other sub-level labels is determined, and it is determined whether the similarity is lower than a preset similarity threshold. For the labels whose similarity is lower than the preset similarity threshold, the labels are reserved in the process of backtracking of the first main label, and the user portrait of the sub-level labels is reserved. At this time, it is the fourth label set. In some embodiments, only the first sub-level label is reserved.

[0102] S612, determining a similar label in the user portrait which has a similarity higher than a preset similarity threshold with the first unsuitable label;

[0103] After determining the first unsuitable labels and the number thereof, the server determines the similarity of branch labels of other main labels with the first unsuitable labels. The higher the similarity is, the more likely the user is not interested in. For the labels on other main labels whose similarity is higher than the preset similarity threshold, the server determines that the labels are similar labels.

[0104] S613, deleting the similar label in the user portrait to obtain a fifth label set.

[0105] After obtaining the similar labels on other main labels, the server deletes the similar labels to ensure that the related push content is deleted completely. In the above scenario, this step is embodied as that the user deletes the small screen, the super large screen and the super thin screen, and the server deletes the large screen and the super clear screen.

[0106] The above examples are only used for understanding the scheme, and in actual situations, the number of label branches is much larger than the above examples, and is more complex. The deletion and backtracking operation of the labels is more responsible. The effect of the method in actual situations is better and can be predicted.

[0107] It can be understood that the implementation of the above embodiments can infer the degree of user interest in related content according to the operation of the user deleting related labels, and then perform different label processing operations. When the user deletes some labels, the corresponding related labels are deleted, so that the push content becomes more suitable for the user, improves the label deletion efficiency, and ensures the user experience.

[0108] The server in the embodiments of the present application will be introduced from the perspective of modules. Please refer to Figure 7 A functional module structure diagram of a server provided by the embodiments of the present application.

[0109] The server comprises:

[0110] The main label determination module 701 is configured to add the first main label to an initial label set of a user portrait in response to a retrieval operation of the user on the first main label, to obtain a first label set.

[0111] The sub-label determination module 702 is configured to add the first main label and a plurality of first-level sub-labels of the first main label to the first label set of the user portrait to obtain a second label set in response to the selection operation of the user on the first main label related push content.

[0112] The discomfort determination module 703 is configured to determine a first number of first discomfort labels in the plurality of first-level sub-labels corresponding to the user-labeled uninterested push content in a first preset time period.

[0113] The portrait rollback module 704 is configured to rollback the user portrait to the initial label set when a ratio of the first number to a total number of the plurality of first-level sub-labels is greater than a preset rollback threshold.

[0114] In some embodiments, the discomfort determination module 703 of the server is further configured to determine a second number of second discomfort labels in the plurality of first-level sub-labels that are not viewed by the user in the push content in a second preset time period; and the portrait rollback module 704 is further configured to rollback the user portrait to the initial label set when a ratio of the second number to the total number of the plurality of first-level sub-labels is greater than the preset rollback threshold.

[0115] In some embodiments, the server further comprises:

[0116] The label deletion module 705 is configured to delete the second discomfort label in the user portrait to obtain a third label set when the ratio of the second number to the total number of the plurality of first-level sub-labels is not greater than the preset rollback threshold.

[0117] In some embodiments, the portrait rollback module 704 of the server is further configured to rollback the user portrait to the initial label set when the ratio of the first number, the second number, and the total number of the plurality of first-level sub-labels is greater than the preset rollback threshold.

[0118] In some embodiments, the server further comprises:

[0119] The similarity determination module 706 is configured to determine a similarity between the plurality of first-level sub-labels and the first discomfort label.

[0120] The label retention module 707 is configured to retain the corresponding first-level sub-label in the user portrait to obtain a fourth label set when the similarity is lower than a preset similarity threshold.

[0121] In some embodiments, the server further comprises:

[0122] The similarity search module 708 is configured to determine a similar label in the user portrait having a similarity higher than a preset similarity threshold with the first discomfort label.

[0123] The tag deletion module 705 of the server can also be configured to delete the similar tags in the user portrait to obtain a fifth tag set.

[0124] In some embodiments, the server further comprises:

[0125] a secondary determination module 709 configured to bind a plurality of secondary sub-level tags of the primary sub-level tag to the user portrait to obtain a sixth tag set in response to a selection operation of the user on the primary sub-level tag related push content.

[0126] The unsuitability determination module 703 of the server can also be configured to determine a third number of secondary unsuitable tags of the plurality of secondary sub-level tags corresponding to the user marked uninterested push content in a first preset time period.

[0127] The tag deletion module 705 of the server can also be configured to delete the primary unsuitable tag when a ratio of the third number to a total number of the plurality of secondary sub-level tags is greater than a preset threshold.

[0128] The server in the embodiments of the present application is described above from the perspective of a modular functional entity. The server in the embodiments of the present application is described below from the perspective of hardware processing. Please refer to Figure 8 FIG. 1 shows an entity device structure diagram of a server provided in an embodiment of the present application.

[0129] It should be noted that, Figure 8 The structure of the server shown is only an example and should not impose any limitation on the functions and use range of the embodiments of the present application.

[0130] As Figure 8 shown, the server includes a central processing unit (CPU) 801, which can perform various appropriate actions and processes according to programs stored in a read-only memory (ROM) 802 or loaded from a storage portion 808 to a random access memory (RAM) 803, such as performing the methods described in the above embodiments. Various programs and data required for system operation are also stored in the RAM 803. The CPU 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.

[0131] The following components are connected to the I / O interface 805: an input section 806 including a mouse, a keyboard, a microphone, a camera, and the like; an output section 807 including a Liquid Crystal Display (LCD), a speaker, and the like; a storage section 808 including a hard disk, and the like; and a communication section 809 including a network interface card such as a LAN (Local Area Network) card, a modem, and the like. The communication section 809 performs communication processing via a network such as the Internet. A drive 810 is also connected to the I / O interface 805 as necessary. A removable medium 811 such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, and the like is attached to the drive 810 as necessary, so that a computer program read out therefrom is installed in the storage section 808 as necessary.

[0132] In particular, the processes described above with reference to the flowcharts can be implemented as a computer software program according to embodiments of the present application. For example, embodiments of the present application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing a computer program for executing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network by the communication section 809, and / or installed from the removable medium 811. When the computer program is executed by the central processing unit (CPU) 801, various functions defined in the present application are executed.

[0133] It should be noted that the computer readable medium shown in the embodiments of the present application can be a computer readable signal medium or a computer readable storage medium or any combination of the two. The computer readable storage medium may, for example, but is not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or component, or any combination of the above. More specific examples of computer readable storage media can include, but are not limited to, an electrical connection having 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), a flash memory, an optical fiber, a portable compact disk read-only memory (Compact Disc Read-Only Memory, CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present application, the computer readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, device or component. In the present application, the computer readable signal medium can include a data signal carrying computer readable computer programs in a baseband or as part of a carrier wave. Such a propagated data signal can take various forms, including but not limited to an electromagnetic signal, an optical signal or any suitable combination thereof.

[0134] The flowcharts and block diagrams in the drawings illustrate the possible implementation architectures, functions and operations of the systems, methods and computer program products according to various embodiments of the present application. In the flowcharts or block diagrams, each block can represent a module, a program segment or a part of code containing one or more executable instructions for implementing the specified logic function. It should also be noted that in some alternative implementations, the functions noted in the blocks can occur in different order than that shown in the drawings. For example, two blocks indicated in succession can actually be executed substantially in parallel, and sometimes in reverse order, depending on the functions involved. It should also be noted that each block in the block diagram or flowchart, and the combination of blocks in the block diagram or flowchart, can be implemented by a dedicated hardware-based system that performs the specified functions or operations, or can be implemented by a combination of dedicated hardware and computer instructions.

[0135] Specifically, the server of the embodiment includes a processor and a memory, and the memory stores a computer program. When the computer program is executed by the processor, the spectrum information measurement method provided by the above embodiment is implemented.

[0136] As another aspect, the present application also provides a computer readable storage medium, which can be included in the server described in the above embodiments, or can exist separately and not be assembled into the server. The storage medium carries one or more computer programs, which, when executed by a processor of the server, enable the server to implement the method provided in the above embodiments.

[0137] The above-described embodiments are merely used to illustrate the technical solutions of the present application, but not for limiting the present application; even though the present application has been described in detail with reference to the foregoing embodiments, those ordinarily skilled in the art should understand that: they can still make modifications to the technical solutions recorded in the foregoing embodiments, or make equivalent replacements to some of the technical features; and the modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present application.

[0138] In the above embodiments, according to the context, the term “when” can be interpreted as meaning “if” or “after” or “in response to determining” or “in response to detecting”. Similarly, according to the context, the phrase “upon determining” or “if detecting (the stated condition or event)” can be interpreted as meaning “if determining” or “in response to determining” or “upon detecting (the stated condition or event)” or “in response to detecting (the stated condition or event)”.

[0139] In the above embodiments, all or some of the embodiments can be implemented by software, hardware, firmware or any combination thereof. When implemented by software, all or some of the embodiments can be implemented in the form of a computer program product. The computer program product includes one or more computer instructions. When loaded and executed by a computer, the computer program instructions generate all or some of the processes or functions described in the embodiments of the present application. The computer can be a general purpose computer, a special purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer readable storage medium, or transmitted from one computer readable storage medium to another computer readable storage medium, for example, the computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center through wired (such as coaxial cable, optical fiber, digital subscriber line) or wireless (such as infrared, wireless, microwave, etc.) mode. The computer readable storage medium can be any available medium accessible by a computer or a data storage device such as a server, data center, etc. integrated with one or more available media. The available medium can be a magnetic medium (for example, floppy disk, hard disk, magnetic tape), an optical medium (for example, DVD), or a semiconductor medium (for example, solid state disk) and the like.

[0140] Those skilled in the art can understand that all or part of the processes in the above-mentioned method embodiments can be implemented by a computer program instructing relevant hardware to complete, the program can be stored in a computer readable storage medium, and the program can include the processes of the above-mentioned method embodiments when executed. The aforementioned storage medium includes ROM or random storage memory RAM, magnetic disc or optical disc and various storage code medium.

Claims

1. A method for modifying tags based on user behavior, applied to a server, characterized in that, The method includes: In response to the user's search operation on the first main tag, the first main tag is added to the initial tag set of the user profile to obtain the first tag set; In response to the user's selection of push content related to the first main tag, multiple first-level sub-tags of the first main tag are added to the first tag set of the user profile to obtain a second tag set; the multiple first-level sub-tags are determined from the push content related to the first main tag; Determine the first number of first uninteresting tags among the plurality of first-level sub-tags that correspond to the user-marked uninterested push content within a first preset time period; When the ratio of the first number to the total number of the plurality of first-level sub-labels is greater than a preset rollback threshold, the user profile is rolled back to the initial label set.

2. The method according to claim 1, characterized in that, After responding to the user's selection operation on the push content related to the first main tag, adding multiple first-level sub-tags of the first main tag to the first tag set of the user profile to obtain a second tag set, the method further includes: Determine the second number of second inappropriate tags among the plurality of first-level sub-tags that have not been viewed by users within a second preset time period; When the ratio of the second number to the total number of the plurality of first-level sub-labels is greater than a preset rollback threshold, the user profile is rolled back to the initial label set.

3. The method according to claim 2, characterized in that, After determining the second number of second inappropriate tags among the plurality of first-level sub-tags that have not been viewed by users within a second preset time period, the method further includes: When the ratio of the second number to the total number of the plurality of first-level sub-labels is not greater than a preset fallback threshold, the second inappropriate label is deleted from the user profile to obtain a third label set.

4. The method according to claim 2, characterized in that, After determining the second number of second inappropriate tags among the plurality of first-level sub-tags that have not been viewed by users within a second preset time period, the method further includes: When the ratio of the sum of the first number and the second number to the total number of the plurality of first-level sub-labels is greater than a preset rollback threshold, the user profile is rolled back to the initial label set.

5. The method according to claim 1, characterized in that, When the ratio of the first number to the total number of the plurality of first-level sub-labels is greater than a preset rollback threshold, after rolling back the user profile to the initial label set, the method further includes: Determine the similarity between the plurality of first-level sub-labels and the first inappropriate label; When the similarity is lower than a preset similarity threshold, the corresponding first-level sub-label is retained in the user profile to obtain a fourth label set.

6. The method according to claim 1, characterized in that, When the ratio of the first number to the total number of the plurality of first-level sub-labels is greater than a preset rollback threshold, after rolling back the user profile to the initial label set, the method further includes: Identify similar tags in the user profile that have a similarity higher than a preset threshold to the first inappropriate tag; The similar tags are deleted from the user profile to obtain the fifth tag set.

7. The method according to claim 1, characterized in that, After responding to the user's selection operation on the push content related to the first main tag, adding multiple first-level sub-tags of the first main tag to the first tag set of the user profile to obtain a second tag set, the method further includes: In response to the user's selection of push content related to the first-level sub-tag, multiple second-level sub-tags of the first-level sub-tag are bound to the user profile to obtain a sixth tag set; Determine the third number of secondary uninteresting tags corresponding to the multiple secondary sub-tags of the push content marked as uninteresting by the user within the first preset time period; When the ratio of the third number to the total number of the plurality of second-level sub-labels is greater than a preset threshold, the first-level sub-label is deleted.

8. A server, characterized in that, include: The primary tag determination module is used to respond to the user's search operation for the first primary tag, add the first primary tag to the initial tag set of the user profile, and obtain the first tag set; The sub-label determination module is used to respond to the user's selection operation on the push content related to the first main label, and add multiple first-level sub-labels of the first main label to the first label set of the user profile to obtain the second label set; The discomfort determination module is used to determine the first number of first discomfort tags among the multiple first-level sub-tags that correspond to the user-marked uninterested push content within a first preset time period. The profile rollback module is used to roll back the user profile to the initial tag set when the ratio of the first number to the total number of the plurality of first-level sub-tags is greater than a preset rollback threshold.

9. A server, characterized in that, include: One or more processors and memory; The memory is coupled to the one or more processors, the memory being used to store computer program code, the computer program code including computer instructions, the one or more processors invoking the computer instructions to cause the server to perform the method as described in any one of claims 1-7.

10. A computer-readable storage medium comprising instructions, characterized in that, When the instructions are executed on the server, the server causes the server to perform the method as described in any one of claims 1-7.

Images