Label preference setting-based content recommendation method and device

Abstract

The invention provides a content recommendation method and device based on tag preference setting, and relates to the technical field of data processing. The tag preference setting-based content recommendation method comprises the following steps: in response to a search task established by a user, obtaining a target tag related to the search task and the number of times that the target tag is clicked by the user; determining effective interaction times according to the times and the cycle boundary value; determining preference state information according to the effective interaction times and the interaction times-preference state association relationship; and performing recommendation according to the preference state information to obtain a recommendation result. According to the technical scheme provided by the embodiment of the invention, the preference state information of the user for the target tag is determined through the effective interaction times of the target tag related to the search task of the user and the interaction times-preference state association relationship, and then recommendation is performed according to the preference state information. Recommendation is carried out on the basis of expression requirements of multi-file preferences of the user, and the recommendation result is high in accuracy.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data processing, in particular to a content recommendation method and device based on label preference setting. BACKGROUND

[0002] At present, with the rapid development of artificial intelligence technology, algorithm recommendation is applied more and more widely in the fields of e-commerce, content recommendation, social media, etc. The core of personalized recommendation algorithm lies in understanding user preferences. Traditional user preference setting methods mainly rely on implicit feedback and explicit feedback.

[0003] Implicit feedback infers user preferences by analyzing user historical behaviors (such as browsing, clicking, purchasing, etc.). This method can obtain a large amount of data reflecting the user's real behavior. However, implicit feedback is easily disturbed by accidental behavior (such as accidental clicks), and the obtained data contains noise and bias. It is necessary to extract effective information to distinguish positive and negative feedback, and it is relatively difficult to accurately interpret the user's real preferences. In contrast, explicit feedback expresses user interests through active preference setting, which is more intuitive than user's implicit behavior and can quickly obtain the user's core interest points.

[0004] Explicit feedback methods include label preference setting, form preference setting, slider bar preference setting, and rating evaluation methods. Among them, label preference setting, as a kind of explicit feedback method with small space occupation and simple operation, is widely used in various recommendation systems.

[0005] Label preference setting expresses user interest in specific content / goods through labels, has the advantages of simplicity, intuitiveness and easy understanding, and user operation is simple and space occupation is small. However, the traditional label preference setting method often only supports "yes / no" two states, which cannot meet the user's demand for expressing multi-grade preferences, resulting in inaccurate recommendation results. SUMMARY

[0006] Therefore, the present application provides a content recommendation method and device based on label preference setting, which can realize recommendation based on user's demand for expressing multi-grade preferences, and the recommendation result is accurate.

[0007] According to an aspect of the present application, a content recommendation method based on label preference setting is provided, comprising: in response to a search task established by a user, obtaining a target label related to the search task and the number of times the target label is clicked by the user; determining the effective interaction number of the target label according to the number of times and a pre-set cycle boundary value; determining the preference state information of the target label according to the effective interaction number and a pre-set interaction number-preference state association relationship; and recommending search content associated with the target label according to the preference state information to obtain a recommendation result.

[0008] According to some embodiments, the effective interaction number of the target label is determined according to the number and a preset loop boundary value, including: performing a modulo operation on the number and the preset loop boundary value to obtain an operation result as the effective interaction number of the target label.

[0009] According to some embodiments, the preference state information of the target label is determined according to the effective interaction number and a preset interaction number-preference state association, including: in the case that the effective interaction number is 0, the preference state information of the target label is determined according to the interaction number-preference state association to include a no explicit preference state; in the case that the effective interaction number is 1, the preference state information of the target label is determined according to the interaction number-preference state association to include a priority recommendation state; in the case that the effective interaction number is 2, the preference state information of the target label is determined according to the interaction number-preference state association to include a forced screening state.

[0010] According to some embodiments, the method further includes: determining a display style of the target label according to the effective interaction number of the target label; and updating the target label in real time according to the display style.

[0011] According to some embodiments, the display style of the target label is determined according to the effective interaction number of the target label, including: in the case that the effective interaction number is 0, the display style of the target label is determined to include a default style; in the case that the effective interaction number is 1, the display style of the target label is determined to include one of a blue style, a green style and an added check mark; in the case that the effective interaction number is 2, the display style of the target label is determined to include a red style or an added prohibition mark.

[0012] According to some embodiments, the search content associated with the target label is recommended according to the preference state information to obtain a recommendation result, including: in the case that the preference state information includes a no explicit preference state, the search content is recommended by using a default recommendation algorithm to obtain the recommendation result; in the case that the preference state information includes a priority recommendation state, a recommendation weight of the search content associated with the target label is increased, and the search content is recommended according to the recommendation weight to obtain the recommendation result; in the case that the preference state information includes a forced screening state, the search content irrelevant to the target label is excluded, and the search content remaining after the exclusion is recommended to obtain the recommendation result.

[0013] According to some embodiments, the search content associated with the target label is recommended according to the preference state information to obtain a recommendation result, including: in the case that the number of target labels related to the search task is greater than 1, the search content associated with multiple target labels is recommended according to a preset label weight to obtain the recommendation result.

[0014] According to some embodiments, in response to a search task established by a user, a target label related to the search task and a number of times the target label is clicked by the user are obtained, including: in response to the search task established by the user, performing semantic analysis on the search task to obtain an intention analysis result of the user; mapping the intention analysis result to a pre-set label to obtain the target label related to the search task; and obtaining the number of times the target label is clicked by the user.

[0015] According to some embodiments, obtaining the number of times the target label is clicked by the user includes: in a case where no label clicking operation performed by the user is detected at a current time step, obtaining the number of times the target label is clicked by the user in historical data; and in a case where a label clicking operation performed by the user is detected at the current time step, updating the number of times the corresponding label is clicked by the user in the historical data according to the label clicking operation, and obtaining the number of times the updated target label is clicked by the user.

[0016] According to an aspect of the present application, a content recommendation device based on a label preference setting includes: a number of times obtaining module configured to obtain, in response to a search task established by a user, a target label related to the search task and a number of times the target label is clicked by the user; an effective interaction module configured to determine an effective interaction number of times of the target label according to the number of times and a pre-set cycle boundary value; a preference state module configured to determine preference state information of the target label according to the effective interaction number of times and a pre-set interaction number of times-preference state association relationship; and a recommendation result module configured to recommend search content associated with the target label according to the preference state information to obtain a recommendation result.

[0017] According to an aspect of the present application, an electronic device is provided, which includes: one or more processors; a storage device configured to store one or more programs; and when the one or more programs are executed by the one or more processors, the one or more processors implement the method as described above.

[0018] According to an aspect of the present application, a computer readable medium having stored thereon a computer program or instructions, which, when executed by a processor, implement the method as described above.

[0019] According to the above embodiments provided by the present application, the effective interaction number of times of a target label related to a search task of a user is clicked by the user, and the interaction number of times-preference state association relationship is used to determine the preference state information of the target label by the user; the present application sets different preference states for different effective interaction numbers of times of the label by setting the interaction number of times-preference state association relationship, so that the user can complete the setting of the preference state by only a simple clicking operation on the label, which is convenient for operation and provides more fine-grained preference expression capability. Then, the recommendation is performed according to the preference state information, so that the recommendation is performed on the basis of the expression demand of the multi-grade preference of the user, and the recommendation result has high accuracy. Attached Figure Description

[0020] It should be understood that the above general description and the following detailed description are merely exemplary and do not limit this application.

[0021] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings, without exceeding the scope of protection claimed by this application.

[0022] Figure 1 A flowchart illustrating the content recommendation method based on tag preference settings provided in this application embodiment;

[0023] Figure 2 A flowchart illustrating how the preference state information of a target tag is determined based on the number of valid interactions and a pre-set relationship between the number of interactions and the preference state, as provided in this embodiment of the application.

[0024] Figure 3 A flowchart for determining the display style of a target tag based on the number of valid interactions with the target tag, provided for embodiments of this application;

[0025] Figure 4 A flowchart illustrating how to recommend search content associated with a target tag based on preference status information, as provided in this embodiment of the application, yields recommendation results.

[0026] Figure 5 A flowchart illustrating how, in response to a user-created search task, the target tags related to the search task and the number of times the target tags are clicked by the user are obtained, provided in an embodiment of this application.

[0027] Figure 6 A flowchart for obtaining the number of times a target tag is clicked by a user, provided as an embodiment of this application;

[0028] Figure 7 A block diagram of a content recommendation device based on tag preference settings provided in an embodiment of this application;

[0029] Figure 8 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0030] With reference to the drawings, the technical solutions in the embodiments of the present application will be clearly and completely described below, obviously, the described embodiments are some of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all the other embodiments obtained by those skilled in the art without creative work belong to the scope of protection of the present application.

[0031] In addition, the described features, structures or characteristics can be combined in any suitable way in one or more embodiments. In the following description, numerous specific details are provided to give a sufficient understanding of the embodiments of the present application. However, those skilled in the art will realize that the technical solutions of the present application can be practiced without one or more of the specific details, or can employ other methods, components, devices, steps, etc. In other cases, well-known methods, devices, implementations or operations are not shown or described in detail to avoid obscuring the aspects of the present application.

[0032] The block diagrams shown in the drawings are only functional entities, which do not necessarily correspond to physically independent entities. That is, these functional entities can be implemented in the form of software, or in one or more hardware modules or integrated circuits, or in different networks and / or processor devices and / or microcontroller devices.

[0033] The flowcharts shown in the drawings are only exemplary illustrations, which do not necessarily include all the contents and operations / steps, and are not necessarily executed in the order described. For example, some operations / steps can be further decomposed, and some operations / steps can be combined or partially combined, so that the actual execution order can be changed according to the actual situation.

[0034] It should be understood that although the terms first, second, third, etc. can be used herein to describe various components, these components should not be limited by these terms. These terms are used to distinguish one component from another component. Therefore, the first component discussed below can be called the second component without departing from the teachings of the present application concepts. As used herein, the term "and / or" includes any one and all combinations of the associated listed items.

[0035] The specific implementation can refer to the following embodiments.

[0036] Figure 1 The flowchart of the content recommendation method based on tag preference setting provided by the embodiments of the present application is shown in FIG. 1. As shown in the figure, the method comprises steps S110-S140. Figure 1

[0037] In step S110, in response to the search task established by the user, the target tag related to the search task and the number of times the target tag is clicked by the user are obtained. ​

[0038] The construction of the search task includes user inputting at least one search word, etc., which is not limited in the present application.

[0039] The label referred to in the present application includes a label common to all search tasks and a label only for a certain type of search task, and a keyword used to describe the content / characteristics of the search task, such as "sports", "technology", "food", etc.

[0040] According to an example embodiment, the label includes a commodity preference category, a merchant preference category, a service preference category, etc.

[0041] Further, in a specific embodiment, the commodity preference category label includes new products / designs, hot trends, bestsellers, low prices, brand alternatives, brand authorizations, etc. The merchant preference category label includes super factories, powerful merchants, deeply certified factories, places of production / industrial belts, etc. The service preference category label includes 24-hour delivery, 48-hour delivery, 24-hour late pickup guaranteed, 48-hour late pickup guaranteed, return shipping insurance, free shipping, one-piece delivery, small order customization, and encrypted face sheet, etc.

[0042] The present application does not limit the number of gears of the label. Different gears of the label correspond to different preference states and meanings, and the user can interact with the label by clicking.

[0043] The existing label preference setting method has deficiencies in space occupation, interaction complexity, and multi-gear preference expression capability, making it difficult for users to accurately express their personal preferences and affecting the recommendation effect. The present application aims to establish a correlation between the number of interactions between the user and the label and the preference state, simplify user operations, and reduce decision-making costs, so as to make recommendations according to the corresponding preference state, improve the accuracy of algorithm recommendations, improve the efficiency of users discovering content / merchandise, enhance user satisfaction and platform activity, and ultimately bring about an increase in business value.

[0044] When the user establishes a search task, the label related to the search task is obtained, denoted as a target label, and the number of times the target label is clicked by the user is also obtained.

[0045] In step S120, the effective interaction number of the target label is determined according to the number of times and a pre-set cycle boundary value.

[0046] The cycle boundary value is set for the label in advance. The significance of the cycle boundary value is to determine the cycle mode of the interaction between the user and the label. For example, in some embodiments, the cycle boundary value is set to 3, and by clicking the label multiple times, the multi-gear preference state of "state 1-state 2-state 3" is switched.

[0047] On the basis of the set cycle boundary value, the actual effective interaction times of the user and the target label can be determined according to the number of times that the target label is clicked by the user.

[0048] In step S130, the preference state information of the target label is determined according to the effective interaction times and the pre-set interaction times-preference state association.

[0049] The interaction times-preference state association is pre-set, for example, in some embodiments, the cycle boundary value is set to 3, the interaction times of 0 corresponds to the initial state, the interaction times of 1 corresponds to state 1, and the interaction times of 2 corresponds to state 2.

[0050] According to the effective interaction times and the pre-set interaction times-preference state association, the preference state information of the target label can be determined. For example, based on the above embodiment, assuming that the effective interaction times is 2, the preference state information includes state 2.

[0051] The multiple states correspond to different degrees of preference of the user to different labels, for example, the initial state can correspond to no interest, state 1 can correspond to general interest, and state 2 can correspond to very interested, etc.

[0052] In step S140, the search content associated with the target label is recommended according to the preference state information, and a recommendation result is obtained.

[0053] The preference state and the recommendation algorithm are associated, the search content associated with the target label is recommended according to the recommendation algorithm corresponding to the preference state information, and the recommended content is obtained as the recommendation result.

[0054] The application determines the preference state information of the user to the target label through the effective interaction times of the target label clicked by the user in the search task of the user and the interaction times-preference state association. The application sets different preference states for different effective interaction times of the label by setting the interaction times-preference state association. The user only needs to perform a simple click operation on the label to complete the setting of the preference state, which is convenient to operate and provides more fine-grained preference expression ability. Then, the recommendation is performed according to the preference state information, so as to realize the recommendation on the basis of the expression demand of the multiple preferences of the user, and the recommendation result has high accuracy.

[0055] According to some embodiments, in step S120, the effective interaction times of the target label is determined according to the number of times and the pre-set cycle boundary value, which can be implemented through step S121.

[0056] In step S121, the number of times and the pre-set cycle boundary value are used for remainder operation, and the operation result is obtained as the effective interaction times of the target label.

[0057] The effective interaction number is determined by using a remainder operation. Specifically, by using the remainder operation, the click number that leads to an integer period cycle is removed, and the remaining click number is reserved as the effective interaction number.

[0058] According to some embodiments, with reference to Figure 2 In step S130, the preference state information of the target label is determined according to the effective interaction number and the pre-set interaction number-preference state association relationship. Specifically, it can be implemented by steps S210-S230.

[0059] In step S210, in the case that the effective interaction number is 0, the preference state information of the target label is determined according to the interaction number-preference state association relationship, which includes no explicit preference state.

[0060] In this embodiment, the cycle boundary value is set to 3, and the interaction number-preference state association relationship is set as follows: the interaction number 0 corresponds to the initial state, the interaction number 1 corresponds to state 1, and the interaction number 2 corresponds to state 2. The initial state is set as no explicit preference state, state 1 is set as a preferred recommendation state, and state 2 is set as a forced screening state.

[0061] In the case that the effective interaction number is 0, it corresponds to no explicit preference state, and therefore the preference state information of the target label includes no explicit preference state, which can be understood as that the user's preference state for the target label is no explicit preference state.

[0062] In step S220, in the case that the effective interaction number is 1, the preference state information of the target label is determined according to the interaction number-preference state association relationship, which includes a preferred recommendation state.

[0063] In the case that the effective interaction number is 1, it corresponds to a preferred recommendation state, and therefore the preference state information of the target label includes a preferred recommendation state, which can be understood as that the user's preference state for the target label is a preferred recommendation state.

[0064] In step S230, in the case that the effective interaction number is 2, the preference state information of the target label is determined according to the interaction number-preference state association relationship, which includes a forced screening state.

[0065] In the case that the effective interaction number is 2, it corresponds to a forced screening state, and therefore the preference state information of the target label includes a forced screening state, which can be understood as that the user's preference state for the target label is a forced screening state.

[0066] According to some embodiments, the method further includes steps S150-S160.

[0067] In step S150, the display style of the target label is determined according to the effective interaction number of the target label.

[0068] In order to help the user to distinguish different preference states corresponding to different click numbers of the label, different display styles are set for the label corresponding to different effective interaction numbers.

[0069] For example, based on the above embodiment, the cycle boundary value is set to 3, and the interaction number-preference state association relationship is set as follows: the interaction number 0 corresponds to the initial state, the interaction number 1 corresponds to state 1, and the interaction number 2 corresponds to state 2.

[0070] On this basis, the style of the label in the initial state is set to style 1, the style of the label in state 1 is set to style 2, and the style of the label in state 2 is set to style 3.

[0071] Based on this, the display style of the target label is determined according to the effective interaction number of the target label.

[0072] For example, in some embodiments, the effective interaction number of the target label is 2, corresponding to state 2, and the display style of the target label is determined to be style 3.

[0073] In step S160, the target label is updated in real time according to the display style.

[0074] According to the determined display style, the style of the target label is updated in real time, which is displayed to the user for intuitive distinction of the label state.

[0075] The present application clearly shows the user the state of the current label through the change of the label style, avoiding the misunderstanding of the user.

[0076] According to some embodiments, with reference to Figure 3 In step S150, the display style of the target label is determined according to the effective interaction number of the target label, which can be implemented through steps S310-S330.

[0077] In step S310, in the case that the effective interaction number is 0, the display style of the target label includes a default style.

[0078] In this embodiment, the cycle boundary value is set to 3, and the interaction number-preference state association relationship is set as follows: the interaction number 0 corresponds to the initial state, the interaction number 1 corresponds to state 1, and the interaction number 2 corresponds to state 2. The style of the label in the initial state is set to the default style, the style of the label in state 1 is set to one of the blue style, the green style, and the added check mark, and the style of the label in state 2 is set to the red style or the added prohibition mark.

[0079] In the case of 0 valid interaction times, corresponding to the initial state, the display style of the target label includes the default style.

[0080] In step S320, in the case of 1 valid interaction times, it is determined that the display style of the target label includes one of the blue style, the green style and the addition of the selected mark.

[0081] In the case of 1 valid interaction times, corresponding to state 1, the display style of the target label includes one of the blue style, the green style and the addition of the selected mark.

[0082] In step S330, in the case of 2 valid interaction times, it is determined that the display style of the target label includes the red style or the addition of the prohibited mark.

[0083] In the case of 2 valid interaction times, corresponding to state 1, the display style of the target label includes the red style or the addition of the prohibited mark.

[0084] According to some embodiments, with reference to Figure 4 In step S140, the search content associated with the target label is recommended according to the preference state information to obtain a recommendation result, which can be implemented through steps S410-S430.

[0085] In step S410, in the case of no explicit preference state in the preference state information, a default recommendation algorithm is used for recommendation to obtain a recommendation result.

[0086] In this embodiment, the cycle boundary value is set to 3, and the interaction times-preference state association relationship is set as follows: 0 interaction times correspond to the initial state, 1 interaction times correspond to state 1, and 2 interaction times correspond to state 2. The initial state is set as no explicit preference state, state 1 is set as a preferred recommendation state, and state 2 is set as a forced screening state.

[0087] The recommendation algorithm for no explicit preference state is set as the default recommendation algorithm, which is the recommendation algorithm of the platform itself in the e-commerce, content recommendation, social media and other scenarios, meaning that the algorithm normally recommends in the current state without special consideration of the target label.

[0088] In the actual implementation process, the search content associated with the target label is recommended by using the default recommendation algorithm, and the recommended content is obtained as the recommendation result.

[0089] In step S420, in the case of the preferred recommendation state in the preference state information, the recommendation weight of the search content associated with the target label is increased, and the recommendation is performed according to the recommendation weight to obtain a recommendation result.

[0090] The recommendation algorithm of the priority recommendation state is set as an algorithm recommendation weighting, the algorithm recommendation weighting recommendation algorithm increases the recommendation weight of the search content associated with the target tag, and recommendation is performed based on the weight, which means that in the current state, the algorithm preferentially recommends the content / merchandise related to the tag.

[0091] In actual implementation, the recommendation weight of the search content associated with the target tag is increased, and recommendation is performed according to the recommendation weight, to obtain the recommended content as the recommendation result.

[0092] In step S430, in the case where the preference state information includes a forced screening state, search content irrelevant to the target tag is excluded, and recommendation is performed according to the remaining search content, to obtain a recommendation result.

[0093] The recommendation algorithm of the forced screening state is set as forced screening filtering, the forced screening filtering recommendation algorithm forcibly filters out search content irrelevant to the target tag, and recommendation is performed based on the remaining content, which means that in the current state, the algorithm forcibly excludes content / merchandise related to the tag.

[0094] In actual implementation, the search content irrelevant to the target tag is excluded, and recommendation is performed according to the remaining search content, to obtain the recommended content as the recommendation result.

[0095] According to some embodiments, in step S140, the search content associated with the target tag is recommended according to the preference state information, to obtain a recommendation result, which can be implemented through step S141.

[0096] In step S141, in the case where the number of target tags related to the search task is greater than 1, the search content associated with multiple target tags is recommended according to the pre-set tag weight, to obtain a recommendation result.

[0097] The user can set a custom weight value (denoted as a tag weight) for each tag, indicating the preference degree. In the recommendation process, the search content associated with multiple target tags is recommended by using the pre-set tag weight, to obtain a recommendation result.

[0098] According to some embodiments, with reference to Figure 5 In step S110, in response to the search task established by the user, the target tag related to the search task and the number of times that the target tag is clicked by the user are obtained, which can be implemented through steps S510-S530.

[0099] In step S510, in response to the search task established by the user, semantic analysis is performed on the search task, to obtain an intention analysis result of the user.

[0100] In the process of establishing the search task, the user can describe his / her preference using natural language, obtain a search task described by natural language, and then parse the user's intention by an algorithm, i.e., perform semantic analysis-based intention parsing on the search task according to a semantic analysis algorithm to obtain an intention parsing result.

[0101] In step S520, the intention parsing result is mapped to a preset label to obtain a target label related to the search task.

[0102] The intention parsing result of the user is mapped to an existing label to obtain a target label related to the search task.

[0103] In step S530, the number of times the target label is clicked by the user is obtained.

[0104] According to some embodiments, with reference to Figure 6 In step S530, the number of times the target label is clicked by the user is obtained, which can be achieved by steps S610-S620.

[0105] In step S610, in the case where no user performs a label clicking operation at the current time step, the number of times the target label is clicked by the user in the historical data is obtained.

[0106] If the current search user does not perform manual adjustment, the last setting is adopted, i.e., the number of times the target label is clicked by the user in the historical data is adopted.

[0107] Further, the algorithm can also be used to deposit the personal long-term preference of the user to affect the default setting of the filtering option.

[0108] In step S620, in the case where the user performs a label clicking operation at the current time step, the number of times the updated target label is clicked by the user is obtained according to the label clicking operation and by updating the number of times the corresponding label is clicked by the user in the historical data.

[0109] If the current search user performs manual adjustment, the adjusted setting is adopted to obtain the number of times the adjusted target label is clicked by the user. Meanwhile, the related content in the historical data is updated.

[0110] To further illustrate the content recommendation method based on label preference setting provided in the present application, a specific embodiment is given.

[0111] In this embodiment, the cycle boundary value is set to 3, and the interaction frequency-preference state association relationship is set as follows: the interaction frequency of 0 corresponds to the initial state, the interaction frequency of 1 corresponds to state 1, and the interaction frequency of 2 corresponds to state 2. The initial state is set as no explicit preference state, state 1 is set as a preferred recommendation state, and state 2 is set as a forced filtering state.

[0112] In the initial state, the label is displayed in a default style, indicating that the user has no explicit preference for the content / merchandise corresponding to the label. For example, the label color is gray or white.

[0113] If the user performs the first click, i.e., the effective interaction number is 1, after the label is clicked once, the label enters the "priority recommendation" state. The label style changes to prompt the user that the label has been selected. For example, the label color changes to blue or green, or a selected mark is added.

[0114] If the user performs the second click, i.e., the effective interaction number is 2, after the label is clicked again, the label enters the "forced filtering" state. The label style further changes to prompt the user that the label has been set to forced filtering. For example, the label color changes to red, or a prohibited mark is added.

[0115] If the user performs the third click, i.e., the effective interaction number is 3, after the label is clicked again, the label returns to the initial state (default state).

[0116] Then, using the algorithm associated with the specified state, the recommendation is made according to the different states of the label, and the recommendation result is obtained.

[0117] The content recommendation method based on label preference setting provided in the application solves the problem that the "yes / no" label preference setting cannot express multiple preferences, and through clicking, the switching of multiple preference states is realized, and more fine-grained preference expression capability is provided.

[0118] The application solves the problems of large space occupation and complicated operation of forms and sliders, and the preference setting can be completed only by simple clicking operation on the label, without the need to fill in forms or drag sliders, greatly simplifying the user operation, reducing the decision cost, and at the same time, realizing multiple preference settings in the smallest space size, saving screen space.

[0119] The application solves the problem of non-uniform evaluation standards, and uses explicit states (such as priority recommendation and forced filtering) instead of fuzzy level evaluation, so that the preference expression is more accurate. At the same time, the application is simpler and more intuitive, and the user does not need to perform complex numerical setting or language description, and is easier to use.

[0120] The device embodiment of the application is described below, which can be used to execute the method embodiment of the application. For details not disclosed in the device embodiment of the application, reference can be made to the method embodiment of the application.

[0121] Figure 7 A block diagram of a content recommendation device based on label preference setting according to an example embodiment is shown.

[0122] Figure 7The apparatus shown can perform the content recommendation method based on the label preference setting according to the embodiments of the present application.

[0123] As shown in Figure 7 The content recommendation apparatus based on the label preference setting can include:

[0124] Referring to Figure 7 According to the foregoing description, the times obtaining module 710 is configured to obtain a target label related to a search task established by a user and a number of times the target label is clicked by the user in response to the search task.

[0125] The effective interaction module 720 is configured to determine an effective interaction number of times of the target label according to the number of times and a preset cycle boundary value.

[0126] The preference state module 730 is configured to determine preference state information of the target label according to the effective interaction number of times and a preset interaction number of times-preference state correlation.

[0127] The recommendation result module 740 is configured to recommend search content associated with the target label according to the preference state information to obtain a recommendation result.

[0128] The apparatus performs similar functions to the method provided above, and other functions can be referred to the foregoing description, which will not be described here.

[0129] The embodiments of the present application disclose an electronic device, comprising: a processor; a memory, storing a computer program, when the computer program is executed by the processor, the processor executes the instruction generation method.

[0130] For example, referring to Figure 8 , Figure 8 The electronic device 800 shown includes a processor 801 and a memory 803. Wherein, the processor 801 and the memory 803 are connected, such as through the bus 802. Optionally, the electronic device 800 can also include a transceiver 804. It should be noted that the transceiver 804 is not limited to one in actual application, and the structure of the electronic device 800 does not constitute a limitation on the embodiments of the present application.

[0131] The processor 801 can be a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array) or other programmable logic device, transistor logic device, hardware component, or any combination thereof. It can implement or execute various exemplary logical blocks, modules and circuits described in the present disclosure. The processor 801 can also be a combination of computing functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, and the like.

[0132] The bus 802 can include a path for transmitting information between the above-mentioned components. The bus 802 can be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, or the like. The bus 802 can be divided into an address bus, a data bus, a control bus, and the like. For convenience of representation, Figure 8 In the figure, only one thick line is used to represent the bus, but it does not mean that there is only one bus or only one type of bus.

[0133] The memory 803 can be a ROM (Read Only Memory) or other type of static storage device that can store static information and instructions, a RAM (Random Access Memory) or other type of dynamic storage device that can store information and instructions, an EEPROM (Electrically Erasable Programmable Read Only Memory), a CD-ROM (Compact Disc Read Only Memory) or other optical disk storage, a magnetic disk storage medium or other magnetic storage device, or any other storage medium that can be used to carry or store desired program code in the form of instructions or data structures and that can be accessed by a computer, but is not limited thereto.

[0134] The memory 803 is configured to store application program codes for implementing the solutions of the present application, and the processor 801 is configured to execute the application program codes stored in the memory 803.

[0135] Figure 8 The electronic device shown is merely an example and should not limit the functions and use range of the embodiments of the present application.

[0136] The embodiments of the present application disclose a computer readable storage medium, which stores a computer program, and when the computer program is executed by a processor, the processor executes an instruction generation method.

[0137] It should be understood that, although each step in the flowchart of the accompanying drawings is shown in sequence according to the direction of the arrow, these steps are not necessarily executed in sequence according to the direction of the arrow. Unless otherwise specified herein, the execution of these steps is not strictly limited in sequence, and they can be executed in other sequences. Moreover, at least part of the steps in the flowchart of the accompanying drawings can include multiple sub-steps or multiple stages, which are not necessarily executed at the same time, but can be executed at different times, and the execution sequence is not necessarily sequential, but can be executed in rotation or alternation with at least part of other steps or sub-steps or stages of other steps.

[0138] The above is only some of the embodiments of the present application, and it should be pointed out that, for those skilled in the art, without departing from the principles of the present application, a number of improvements and refinements can be made, and these improvements and refinements should also be considered as the protection scope of the present application.

Claims

1. A content recommendation method based on tag preference settings, characterized by, The method comprises the following steps: obtaining a target label related to a search task established by a user and a number of times the target label is clicked by the user in response to the search task; determining an effective interaction number of the target label according to the number of times and a preset cycle boundary value; determining preference state information of the target label according to the effective interaction number and a preset interaction number-preference state association relationship; recommending search content associated with the target label according to the preference state information to obtain a recommendation result.

2. The method of claim 1, wherein, The method further comprises the following steps: determining the effective interaction number of the target label by performing a modulo operation on the number of times and the preset cycle boundary value.

3. The method of claim 1, wherein, The method further comprises the following steps: determining the preference state information of the target label according to the effective interaction number and the preset interaction number-preference state association relationship, wherein in a case where the effective interaction number is 0, the preference state information of the target label is determined to include no explicit preference state according to the interaction number-preference state association relationship; in a case where the effective interaction number is 1, the preference state information of the target label is determined to include a priority recommendation state according to the interaction number-preference state association relationship; 4. The method of claim 1, wherein, in a case where the effective interaction number is 2, the preference state information of the target label is determined to include a forced screening state according to the interaction number-preference state association relationship. The method further comprises the following steps: determining a display style of the target label according to the effective interaction number of the target label; 5. The method of claim 4, wherein, updating the target label in real time according to the display style. The method further comprises the following steps: determining the display style of the target label according to the effective interaction number of the target label, wherein in a case where the effective interaction number is 0, the display style of the target label is determined to include a default style; 6. The method of claim 3, wherein, in a case where the effective interaction number is 1, the display style of the target label is determined to include one of a blue style, a green style and an added check mark; in a case where the effective interaction number is 2, the display style of the target label is determined to include a red style or an added prohibition mark. The method further comprises the following steps: recommending the search content associated with the target label according to the preference state information to obtain a recommendation result, wherein 7. The method of claim 3, wherein, in a case where the preference state information includes no explicit preference state, a default recommendation algorithm is used for recommendation to obtain the recommendation result; in a case where the preference state information includes a priority recommendation state, a recommendation weight of the search content associated with the target label is increased, and the search content is recommended according to the recommendation weight to obtain the recommendation result; in a case where the preference state information includes a forced screening state, search content irrelevant to the target label is excluded, and the remaining search content is recommended to obtain the recommendation result. The method further comprises the following steps: In the case that the number of target labels related to the search task is greater than 1, the search content associated with the plurality of target labels is recommended according to a preset label weight, and a recommendation result is obtained.

8. The method of claim 1, wherein, In response to a search task established by a user, a target label related to the search task and a number of times that the target label is clicked by the user are obtained, including: In response to a search task established by a user, the search task is subjected to semantic analysis, and an intention analysis result of the user is obtained. The intention analysis result is mapped to a preset label to obtain the target label related to the search task. The number of times that the target label is clicked by the user is obtained.

9. The method of claim 8, wherein, The number of times that the target label is clicked by the user is obtained, including: In the case that no label clicking operation of the user is detected at a current time step, the number of times that the target label is clicked by the user in historical data is obtained. In the case that the label clicking operation of the user is detected at the current time step, the number of times that the corresponding label is clicked by the user in the historical data is updated according to the label clicking operation, and the number of times that the target label is clicked by the user after the update is obtained.

10. A content recommendation device based on tag preference settings, characterized in that, Including: A number of times obtaining module is configured to obtain, in response to a search task established by a user, a target label related to the search task and a number of times that the target label is clicked by the user. An effective interaction module is configured to determine an effective interaction number of times of the target label according to the number of times and a preset cycle boundary value. A preference state module is configured to determine preference state information of the target label according to the effective interaction number of times and a preset interaction number of times-preference state association relationship. A recommendation result module is configured to recommend search content associated with the target label according to the preference state information, and obtain a recommendation result.

11. An electronic device, comprising: Including: One or more processors; A storage device configured to store one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1-9.

12. A computer readable storage medium having stored thereon a computer program or instructions, characterized in that, The computer program or instructions implement the method according to any one of claims 1-9 when executed by a processor.

Images