CN112765400A - Weight updating method of interest tag, content recommendation method, device and equipment - Google Patents
Weight updating method of interest tag, content recommendation method, device and equipment Download PDFInfo
- Publication number
- CN112765400A CN112765400A CN202011619811.1A CN202011619811A CN112765400A CN 112765400 A CN112765400 A CN 112765400A CN 202011619811 A CN202011619811 A CN 202011619811A CN 112765400 A CN112765400 A CN 112765400A
- Authority
- CN
- China
- Prior art keywords
- target
- interest
- weight
- interest tag
- tag
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 75
- 230000003993 interaction Effects 0.000 claims abstract description 146
- 230000002452 interceptive effect Effects 0.000 claims abstract description 67
- 230000006399 behavior Effects 0.000 claims description 171
- 238000004891 communication Methods 0.000 claims description 19
- 238000012216 screening Methods 0.000 claims description 15
- 238000004590 computer program Methods 0.000 claims description 13
- 230000004927 fusion Effects 0.000 claims description 11
- 238000001514 detection method Methods 0.000 claims description 3
- 230000008569 process Effects 0.000 description 9
- 238000004364 calculation method Methods 0.000 description 6
- 230000006870 function Effects 0.000 description 6
- 230000009471 action Effects 0.000 description 4
- 230000001960 triggered effect Effects 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 230000008859 change Effects 0.000 description 2
- 230000002238 attenuated effect Effects 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 239000000835 fiber Substances 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 238000012163 sequencing technique Methods 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/70—Information retrieval; Database structures therefor; File system structures therefor of video data
- G06F16/73—Querying
- G06F16/735—Filtering based on additional data, e.g. user or group profiles
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/70—Information retrieval; Database structures therefor; File system structures therefor of video data
- G06F16/75—Clustering; Classification
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Multimedia (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computational Linguistics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The embodiment of the invention provides a weight updating method, a content recommendation method, a device and equipment of an interest tag, and belongs to the technical field of computers. The weight updating method comprises the following steps: determining a target interaction behavior corresponding to a target account; determining a target interest tag for the target interaction behavior from the interest tags associated with the target account; each interest tag is provided with a weight representing the interest matching degree of the interest tag and the target account; and updating the weight of the target interest tag based on the interaction attribute of the target interaction behavior. Therefore, in the scheme, after the target account generates the interactive behavior, the weight of each interest tag associated with the target account can be updated according to the interactive attribute of the interactive behavior, so that when content recommendation is performed on the target account based on the interest tags, the situation that the recommended content is not matched with the current interest point of the user can be avoided, and the user experience is improved.
Description
Technical Field
The invention relates to the technical field of computers, in particular to a weight updating method, a content recommendation device and electronic equipment of interest tags.
Background
With the development of information technology, more and more service scenes begin to adopt personalized recommendation. For example: when a user watches videos through a client, videos which may be of interest are recommended for the user. In the prior art, based on collected behavior information of a user, for example: and determining the interest tag of the user according to the behavior of clicking the video by the user, or the behavior of canceling the attention of a certain video author by the user, and recommending the user based on the interest tag.
However, the user may have different points of interest at different stages, and if the user is recommended only according to the determined interest tags during recommendation, the recommended content is easily not matched with the current points of interest of the user, which results in poor user experience during recommendation.
Disclosure of Invention
The embodiment of the invention aims to provide a weight updating method of an interest tag, a content recommending method, a content recommending device and electronic equipment, so as to solve the problem of poor user experience caused by mismatching of recommended content and the current interest point of a user. The specific technical scheme is as follows:
in a first aspect of the present invention, there is provided a method for updating a weight of an interest tag, including:
determining a target interaction behavior corresponding to a target account;
determining a target interest tag for the target interaction behavior from the interest tags associated with the target account; each interest tag is provided with a weight representing the interest matching degree of the interest tag and the target account;
and updating the weight of the target interest tag based on the interaction attribute of the target interaction behavior.
Optionally, the updating the weight of the target interest tag based on the interaction attribute of the target interaction behavior includes:
determining an adjustment amount of the weight of the target interest tag based on the interaction attribute of the target interaction behavior;
and updating the weight of the target interest tag according to the adjustment amount of the weight of the target interest tag.
Optionally, the determining, based on the interaction attribute of the target interaction behavior, an adjustment amount of the weight of the target interest tag includes:
determining an amplitude coefficient corresponding to the interaction attribute of the target interaction behavior from a preset incidence relation;
calculating the adjustment amount of the weight of the target interest tag according to the determined amplitude coefficient;
the preset association relationship is an association relationship between each interaction attribute and an amplitude coefficient, and the amplitude coefficient corresponding to any interaction attribute is as follows: coefficients for adjusting the weights of the interest tags when generating the interactive behavior with the interactive attribute.
Optionally, the interaction attribute of the target interaction behavior comprises a plurality of attributes; calculating an adjustment amount of the weight of the target interest tag according to the determined amplitude coefficient, including:
and fusing the amplitude coefficients corresponding to the determined interactive attributes according to a preset fusion mode to obtain the adjustment quantity of the weight of the target interest tag.
Optionally, the method further comprises:
when a preset weight adjustment condition is met, acquiring a target time difference of the target account, wherein the target time difference is a difference value between the current time and the last time of the target account for generating the interactive behavior;
and determining an adjustment parameter used when the weight of the interest tag is adjusted according to the target time difference for each interest tag associated with the target account, and performing predetermined operation on the determined adjustment parameter and the weight of the interest tag to obtain an updated weight of the interest tag.
In another aspect of the present invention, there is also provided a content recommendation method, including:
when detecting that a target account meets a content recommendation condition, determining each interest tag associated with the target account;
based on the weight of each interest tag, screening the interest tags required to be utilized by the content recommendation from each interest tag; wherein the weight of each interest tag is updated based on the updating method of any one of claims 1-5;
and recommending the content for the target account based on the screened interest tags.
In another aspect of the present invention, there is also provided an apparatus for updating a weight of an interest tag, including:
the data acquisition module is used for determining a target interaction behavior corresponding to the target account;
a tag determination module, configured to determine, from the interest tags associated with the target account, a target interest tag for the target interaction behavior; each interest tag is provided with a weight representing the interest matching degree of the interest tag and the target account;
and the weight updating module is used for updating the weight of the target interest tag based on the interaction attribute of the target interaction behavior.
In another aspect of the present invention, there is also provided a content recommendation apparatus, including:
the condition detection module is used for determining each interest tag associated with the target account when the target account is detected to meet the content recommendation condition;
the tag screening module is used for screening the interest tags required to be utilized by the content recommendation from the interest tags based on the weight of each interest tag; wherein the weight of each interest tag is updated based on the updating method of any one of claims 1-5;
and the content recommendation module is used for recommending the content for the target account based on the interest tag obtained by screening.
In another aspect of the present invention, there is also provided an electronic device, including a processor, a communication interface, a memory and a communication bus, where the processor, the communication interface, and the memory complete communication with each other through the communication bus;
a memory for storing a computer program;
and the processor is used for realizing the weight updating method of any interest label or the content recommendation method when executing the program stored in the memory.
In yet another aspect of the present invention, there is further provided a computer-readable storage medium, in which a computer program is stored, and the computer program, when executed by a processor, implements the method for updating the weight of any interest tag or the method for recommending content.
In yet another aspect of the present invention, there is also provided a computer program product containing instructions which, when run on a computer, causes the computer to perform any of the above-mentioned method for updating the weight of an interest tag or the above-mentioned method for recommending content.
The embodiment of the invention provides a scheme, and the target interaction behavior corresponding to the target account is determined; determining a target interest tag for the target interaction behavior from the interest tags associated with the target account; each interest tag is provided with a weight representing the interest matching degree of the interest tag and the target account; and updating the weight of the target interest tag based on the interaction attribute of the target interaction behavior. Therefore, in the scheme, after the target account generates the interactive behavior, the weight of each interest tag associated with the target account can be updated according to the interactive attribute of the interactive behavior, so that when content recommendation is performed on the target account based on the interest tags, the situation that the recommended content is not matched with the current interest point of the user can be avoided, and the user experience is improved.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below.
FIG. 1 is a flowchart illustrating a method for updating weights of interest tags according to an embodiment of the present invention;
FIG. 2 is another flowchart of a method for updating the weight of an interest tag according to an embodiment of the present invention;
FIG. 3 is a flowchart of a content recommendation method according to an embodiment of the present invention;
FIG. 4 is a schematic structural diagram of an interest tag weight updating apparatus according to an embodiment of the present invention;
FIG. 5 is a schematic structural diagram of a content recommendation apparatus according to an embodiment of the present invention;
fig. 6 is a schematic structural diagram of an electronic device in an embodiment of the invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be described below with reference to the drawings in the embodiments of the present invention.
In order to solve the problem of poor user experience caused by mismatching of recommended content and the current interest point of a user, the embodiment of the invention provides a weight updating method of an interest tag, a content recommending method, a content recommending device and electronic equipment.
It should be noted that the method for updating the weight of the interest tag provided in the embodiment of the present invention is applied to an electronic device, and the electronic device may be any terminal device having a network connection function and running a multimedia client, or may be a server of the multimedia client. In practical applications, the terminal device may be: smart phones, tablets, laptops, and the like.
Specifically, the execution subject of the method for updating the weight of the interest tag provided by the embodiment of the present invention may be a weight updating apparatus for an interest tag running in an electronic device. For example, if the electronic device is a terminal device, the weight updating device of the interest tag may be a multimedia client running in the electronic device or a plug-in the multimedia client. The multimedia client may be an APP or a web client. The content recommended by the embodiment of the present invention may be multimedia data, such as video data, and accordingly, the multimedia client may be a video client, which is not limited to this.
The method for updating the weight of the interest tag provided by the embodiment of the invention can comprise the following steps:
determining a target interaction behavior corresponding to a target account;
determining a target interest tag for the target interaction behavior from the interest tags associated with the target account; each interest tag is provided with a weight representing the interest matching degree of the interest tag and the target account;
and updating the weight of the target interest tag based on the interaction attribute of the target interaction behavior.
Therefore, in the scheme, after the interactive behavior is generated in the target account, the weight of each interest tag associated with the target account can be updated according to the behavior interactive attribute of the interactive behavior, so that when content recommendation is performed on the target account based on the interest tags, the situation that the recommended content is not matched with the current interest point of the user can be avoided, and the user experience is improved.
The following describes a method for updating the weights of interest tags according to an embodiment of the present invention with reference to the accompanying drawings.
As shown in fig. 1, a method for updating weights of interest tags according to an embodiment of the present invention may include:
s101, determining a target interaction behavior corresponding to a target account;
the target interaction behavior may be a behavior generated when a user using the target account interacts with the multimedia client. The target account is an account of content to be recommended, and the target account may be any account which can access multimedia content in the multimedia client after logging in the multimedia client. In addition, the account information of the target account may include an account identifier that can identify the target account, such as: user avatars, user nicknames, and the like. And, the target interactive behavior may be the latest interactive behavior of the target account, or a plurality of interactive behaviors of the target account within a predetermined time.
The target account may perform a plurality of types of interactive behaviors with the client, for example: the behavior types may be: the types of behaviors used to characterize the positive interest of the target account, such as: utilizing the target account to approve and collect any video in the multimedia client, or paying attention to an uploading author of any video, and the like; the following steps can be also included: the types of behaviors used to characterize the negative interest of the target account, such as: canceling the approval and collection of a video in the multimedia client by using the target account, or canceling the uploading author paying attention to the video, and the like.
It is understood that when the target account generates the target interactive behavior in the multimedia client, the multimedia client may record the target interactive behavior. In addition, in order to reduce the calculation amount of the multimedia client and reduce the content occupation of the multimedia client, the multimedia client may upload the behavior data of the target interaction behavior of the target account to a server corresponding to the multimedia client after recording the target interaction behavior of the target account. Correspondingly, when the method for updating the weight of the interest tag is applied to the multimedia client, determining the target interaction behavior corresponding to the target account may include: determining a target interaction behavior corresponding to the target account based on the locally recorded behavior; or requesting the behavior data of the target interactive behavior from the server, so as to determine the target interactive behavior corresponding to the target account. When the method for updating the weight of the interest tag is applied to the server, determining the target interaction behavior corresponding to the target account may include: and determining the target interaction behavior corresponding to the target account based on the behavior data of the interaction behavior corresponding to the target account reported by the multimedia.
S102, determining a target interest tag for a target interaction behavior from the interest tags associated with the target account;
each interest label is provided with a weight which represents the interest matching degree of the interest label and the target account.
It is understood that each interactive behavior may correspond to an interest tag, and each interest tag may be provided with a weight representing a degree of interest matching the interest tag with the target account. In order to facilitate adjustment of the weight of the interest tag associated with the target account, after the target interaction behavior is determined, the target interest tag targeted by the target interaction behavior is determined from the interest tags associated with the target account, and then the weight of the target interest tag is adjusted subsequently.
For example, in one implementation, determining a target interest tag for the target interaction behavior from among the interest tags associated with the target account may include: and searching for an interest tag which is the same as the interest tag corresponding to the target interactive behavior from the interest tags associated with the target account, and taking the searched interest tag as the target interest tag for the target interactive behavior.
There may be a plurality of ways to determine the interest tags associated with the target account. For example, in one implementation, when the weight update method is applied to the multimedia client, the determination of each interest tag associated with the target account may include: sending an interest acquisition request carrying an account identifier of a target account to a target server so that the target server sends each interest tag associated with the target account to a multimedia client according to the interest acquisition request; the target server is a server corresponding to the multimedia client.
For example, in another implementation manner, when the weight updating method is applied to the server, the determining manner of each interest tag associated with the target account may include: and searching each interest tag associated with the target account from the preset association relation between the account identification and the interest tag.
In the embodiment of the present invention, the interaction behavior of the target account is analyzed to determine a manner of the interest tag corresponding to the interaction behavior, which is not specifically limited. The following are exemplary: and determining a target object acted by the interactive behavior of the target account, acquiring attribute information of the target object, and determining the interest tag corresponding to the interactive behavior of the target account according to the attribute information. For example: when the interaction behavior of the target account includes: when watching the video a, the video upload author a paying attention to the video a, then the interest tag corresponding to the interaction behavior of the target account may include: types of video a, such as: type of food, and/or video upload author a, etc.
S103, updating the weight of the target interest tag based on the interaction attribute of the target interaction behavior.
The interaction attribute of the target interaction behavior can reflect the interest degree of the user of the target account in the target interest tag, so that the weight of the target interest tag can be updated based on the interaction attribute of the target interaction behavior. The number of the interaction attributes may be one or more, and if the number of the interaction attributes is one, the interaction attributes may be a behavior type of the target interaction behavior, a tag type of an interest tag corresponding to the target interaction behavior, or a behavior time period of the target interaction behavior, which is not limited to this; if there are multiple interaction attributes, the interaction attributes may be at least two of the behavior type of the target interaction behavior, the interest tag corresponding to the target interaction behavior, and the behavior time period of the target interaction behavior, but are not limited thereto. Optionally, in an implementation manner, updating the weight of the target interest tag based on the interaction attribute of the target interaction behavior may include: determining the adjustment amount of the weight of the target interest tag based on the interaction attribute of the target interaction behavior; and updating the weight of the target interest label according to the adjustment amount of the weight of the target interest label.
In order to update the weight of the target interest tag conveniently, an adjustment amount for updating the weight of the target interest tag may be determined based on the interaction attribute of the target interaction behavior, and the weight of the target interest tag is updated according to the adjustment amount of the weight of the target interest tag. For example, a corresponding threshold or threshold range may be associated with the interaction attribute of each target interaction behavior for updating the weight of the target interest tag. For example, in one implementation, determining an adjustment amount of the weight of the target interest tag based on the interaction attribute of the target interaction behavior may include:
determining an amplitude coefficient corresponding to the interaction attribute of the target interaction behavior from a preset incidence relation;
calculating the adjustment quantity of the weight of the target interest tag according to the determined amplitude coefficient;
the preset association relationship is the association relationship between each interaction attribute and an amplitude coefficient, and the amplitude coefficient corresponding to any interaction attribute is as follows: coefficients for adjusting the weights of the interest tags when generating the interactive behavior with the interactive attribute. It is understood that for any interactive behavior that characterizes positive interest of the target account, the amplitude coefficient corresponding to the interactive attribute of the interactive behavior may be a positive value, and for any interactive behavior that characterizes negative interest of the target account, the amplitude coefficient corresponding to the interactive attribute of the interactive behavior may be a negative value, or a value less than 1. This is so that: when any interactive behavior representing the positive interest of the target account is taken as a target interactive behavior, the weight of the determined target interest tag is increased; and when any interactive behavior representing the negative interest of the target account is taken as the target interactive behavior, the weight of the determined target interest tag is reduced. Considering that the interaction attribute of any target interaction behavior may be one or may include a plurality of interaction attributes, the interaction attribute of the target interaction behavior may correspond to one or more amplitude coefficients. In this way, there may be various implementations of calculating the adjustment amount of the weight of the target interest tag according to the determined amplitude coefficient.
When the interaction attribute of the target interaction behavior is one, calculating an adjustment amount of the weight of the target interest tag according to the determined amplitude coefficient, which may include: and scaling the determined amplitude coefficient according to a preset multiple, and taking the scaled amplitude coefficient as an adjustment quantity of the weight of the target interest tag.
When the interaction attribute of the target interaction behavior comprises a plurality of attributes, calculating an adjustment amount of the weight of the target interest tag according to the determined amplitude coefficient, which may include: and fusing the amplitude coefficients corresponding to the determined interactive attributes according to a preset fusion mode to obtain the adjustment quantity of the weight of the target interest tag. It will be appreciated that there may be a variety of such predetermined fusion patterns, for example: a fusion mode of adding the amplitude coefficients corresponding to the determined behavior attributes, a fusion mode of multiplying the amplitude coefficients corresponding to the determined behavior attributes, a fusion mode of adding the amplitude coefficients corresponding to the determined behavior attributes and scaling by a predetermined multiple, and the like, but the invention is not limited thereto.
If the plurality of interaction attributes comprise: when the behavior type of the target interaction behavior and the tag type of the interest tag corresponding to the target interaction behavior are determined, according to a predetermined fusion manner, fusing the amplitude coefficients corresponding to the determined interaction attributes to obtain an adjustment amount of the weight of the target interest tag, which may include:
fusing the amplitude coefficient corresponding to the determined behavior type and the amplitude coefficient corresponding to the tag type by using a preset fusion formula to obtain the adjustment quantity of the weight of the target interest tag;
wherein the predetermined fusion formula comprises:
Score1＝action_score*entity_score
wherein Score1 is the adjustment amount of the weight of the target interest tag, action _ Score is the amplitude coefficient corresponding to the determined behavior type, and entry _ Score is the amplitude coefficient corresponding to the determined tag type.
Considering that the behavior type of each interactive behavior generating the target interactive behavior may be different, the amplitude coefficients corresponding to different behavior types may also be different in specific applications; in addition, the tag type of the interest tag for each target interaction behavior may be different, and correspondingly, the amplitude systems corresponding to different tag types may also be different. In addition, different determination manners may exist for the specific values of the amplitude coefficients corresponding to different behavior types and tag types, for example: an empirical value set by a human, a value calculated by an execution subject in a preset calculation manner, and the like, and are not particularly limited herein.
Assume that the target interaction behavior is: if the currently displayed target video is not clicked, the behavior type of the target interaction behavior may be: the type of the non-click, the tag type of the interest tag corresponding to the target interaction behavior may be: a video identification of the target video, such as a video ID, and/or a video type of the target video, such as: games, gourmets, etc. At this time, the amplitude coefficient corresponding to the behavior type of the target interaction behavior may be negative values such as-1, -2, and the like, and the amplitude coefficient corresponding to the tag type of the interest tag corresponding to the target interaction behavior may be any value between 0 and 1. Assume that the target interaction behavior is: the user cancels the attention, and the amplitude coefficient corresponding to the corresponding behavior type can be-5; the target interaction behavior is: the bubble ring is withdrawn, and the amplitude coefficient corresponding to the corresponding behavior type can be minus 5 minutes; the interaction behavior of the target interaction behavior is: clicking a dislike button, wherein the corresponding amplitude coefficient of the corresponding behavior type can be-100; the interaction behavior of the target interaction behavior is: for video scoring (10 points, assuming x points), the amplitude coefficient corresponding to the corresponding behavior type may be x-6; the interaction behavior of the target interaction behavior is: desense, the amplitude coefficient corresponding to the corresponding behavior type may be-3, and so on.
The embodiment of the invention provides a scheme, and the target interaction behavior corresponding to the target account is determined; determining a target interest tag for the target interaction behavior from the interest tags associated with the target account; each interest tag is provided with a weight representing the interest matching degree of the interest tag and the target account; and updating the weight of the target interest tag based on the interaction attribute of the target interaction behavior. Therefore, in the scheme, after the interactive behavior is generated in the target account, the weight of each interest tag associated with the target account can be updated according to the behavior interactive attribute of the interactive behavior, so that when content recommendation is performed on the target account based on the interest tags, the situation that the recommended content is not matched with the current interest point of the user can be avoided, and the user experience is improved.
Optionally, as shown in fig. 2, in another embodiment of the present invention, the method may further include the following steps S201 to S202:
s201, when a preset weight adjusting condition is met, acquiring a target time difference of a target account;
the target time difference is the difference between the current time and the last time of the interactive action generated by the target account;
it can be understood that it is reasonable to update the weight of each interest tag associated with the target account, where the update is performed when the target account generates an interactive behavior, or when a preset weight condition is met. The preset weight adjustment condition may include: when content recommendation needs to be performed on the target account, or when the difference between the time generated by the last interactive action and the current time reaches a preset threshold value, and the like.
When any interactive behavior is generated by the target account, the multimedia client can record the generation time of the interactive behavior. For example, in one implementation, when the execution subject is a multimedia client, obtaining the target time difference of the target account may include: and acquiring the last time of generating the interactive behavior of the target account from the local record information about the interactive behavior, and taking the difference value between the current time and the acquired last time of generating the interactive behavior as the target time difference of the target account.
S202, aiming at each interest tag associated with the target account, determining an adjustment parameter used when the weight of the interest tag is adjusted according to the target time difference, and performing predetermined operation on the determined adjustment parameter and the weight of the interest tag to obtain the updated weight of the interest tag.
It is contemplated that over time, the points of interest of the user of the target account may change, for example: over time, the associated interest tags of the target account may fade away until the retirement is completed. Then, when the weight of the interest tag of the target account is updated, the influence of time on the weight is considered, so that the weight of each interest tag can be attenuated over time, and the higher matching degree of the weight of each interest tag and the interest of the target account can be ensured.
For example, in an implementation manner, determining an adjustment parameter used when performing weight adjustment on the interest tag according to the target time difference, and performing a predetermined operation on the determined adjustment parameter and the weight of the interest tag to obtain an updated weight of the interest tag may include:
calculating an adjustment parameter utilized when the weight of the interest tag is adjusted by adopting a preset parameter calculation formula according to the target time difference;
and multiplying the determined adjusting parameter by the weight of the interest tag to obtain the updated weight of the interest tag.
The preset parameter calculation formula comprises the following steps:
Ti＝e^-aix；
wherein, TiThe adjustment parameter used when the weight adjustment is performed for the ith associated interest tag, e ^ characterizing the exponential function, aiX is the hyper-parameter of the ith associated interest tag and x is the target time difference. Wherein, the hyperparameter aiMay be any value between 0 and 1
Accordingly, based on the above parameter calculation formula, the update formula of the weight of any interest tag can be exemplarily shown as follows:
Score(i)＝score(i)*e^-aix
wherein score (i) is the updated weight of the ith interest tag, e ^ -aix is the adjustment parameter used when the weight adjustment is performed on the ith associated interest tag, the operation symbol of the predetermined operation is represented, the exponential function is represented by e ^ aiX is the hyper-parameter of the ith associated interest tag and x is the target time difference. Wherein, the hyperparameter aiAnd may be any value between 0 and 1.
Therefore, in the scheme, after the interactive behavior is generated in the target account, the weight of each interest tag associated with the target account can be updated according to the behavior interactive attribute of the interactive behavior, so that when content recommendation is performed on the target account based on the interest tags, the situation that the recommended content is not matched with the current interest point of the user can be avoided, and the user experience is improved. In addition, the interest points of the users of the target accounts are considered to change along with time, the weights of the interest labels are updated based on the time information, the weights of the interest labels can be further guaranteed to be matched with the current interests of the users, and therefore the matching degree of the recommended content and the current interest points of the users is further improved.
Based on the above weight updating method, as shown in fig. 3, an embodiment of the present invention further provides a content recommendation method, which may include the following steps:
s301, when detecting that a target account meets a content recommendation condition, determining each interest tag associated with the target account;
in order to facilitate content recommendation for the target account, the content recommendation for the target account may be performed when it is detected that the target account satisfies the content recommendation condition. Also, there may be various ways of satisfying the content recommendation condition, for example: the method comprises the steps of playing any multimedia data through a multimedia client for a preset time when a target account logs in the multimedia client, receiving a recommendation request for content recommendation for the target account and the like after the target account logs in the multimedia client.
Accordingly, the trigger condition for executing the content recommendation process may be an automatic trigger or a manual trigger, for example: the method comprises the steps that after a target account logs in a multimedia client, any multimedia data is played through the multimedia client to reach a preset time, the content recommendation process can be automatically triggered, when the target account logs in the multimedia client, the content recommendation process can be automatically triggered, after the target account logs in the multimedia client, the multimedia data is played through the multimedia client to any moment, the content recommendation process is manually triggered, and the like.
It is understood that, when the triggering condition for executing the content recommendation process is manual triggering, the manually triggered function button may be located on a user setting interface of the electronic device running the multimedia client, or may be located at any position of a playing interface of the multimedia client, such as the top end or the bottom end. The form of the manually activated function button is not particularly limited.
For the implementation of determining each interest tag associated with the target account, reference may be made to related descriptions in step S102 in the foregoing embodiment, which is not described herein again.
S302, based on the weight of each interest tag, screening the interest tags required to be utilized by the content recommendation from each interest tag;
the weight of each interest tag may be updated based on the updating method in the above embodiment.
Considering the weight of each interest tag, the interest matching degree between the interest tag and the target account may be characterized, and then, in order to improve the matching degree of the recommendation, the interest tags that are required to be utilized when content recommendation is performed on the target account may be selected from the interest tags based on the weight of the interest tags associated with the target account. There are various ways of weighting the interest tags, for example, the weight of the interest tag may be a value within 0-100, or 0-10, or 0-1, and the higher the weight is, the higher the matching degree of the interest tag corresponding to the weight with the interest of the target account is.
In addition, the interaction type of the interaction behavior generated by the target account at the multimedia client may be a positive interest type, such as a video is praised, or a video uploading author is concerned, or a negative interest type, such as: cancel praise, or cancel attention, etc. In order to improve the matching degree of the recommended content and the interest of the target account and avoid recommending the content with lower interest points for the target account, the scheme provided by the embodiment of the invention can reduce the weight of the target interest label aiming at the interactive behavior representing the negative interest type when the interactive type of the interactive behavior of the target account is the negative interest type.
In an exemplary implementation manner, the screening of the interest tags required to be utilized by the content recommendation from the interest tags based on the weights of the interest tags may include: and sequencing the weights of the interest tags in a descending manner, and taking the interest tags corresponding to the sequenced weights in the preset ranking as the interest tags required to be utilized by the content recommendation.
For example, in another implementation manner, screening, from the interest tags, the interest tags that need to be utilized by the content recommendation based on the weight of each interest tag may include: and screening the interest tags with the weight larger than a preset threshold value from the interest tags, and taking the interest tags with the weight larger than the preset threshold value as the interest tags required to be utilized by the content recommendation.
And S303, recommending the content for the target account based on the interest tag obtained by screening.
It will be appreciated that the target account may have different points of interest at different stages, so that the target account may generate different interactive behaviors on the multimedia client, for example: the video content can be praised in the previous stage, or praise can be cancelled in the next stage, etc. Then, in order to make the recommended content match the current interest point of the target account, content recommendation may be performed for the target account based on the filtered interest tags when performing content recommendation.
For example, in one implementation, the recommending content for the target account based on the filtered interest tag may include: and according to the screened interest tags, determining multimedia data associated with the screened interest tags, and recommending the associated multimedia data for the target account. There may be a plurality of multimedia data associated with any filtered interest tag. For example: when the interest tag is a video uploading author A, the multimedia data associated with the screened interest tag can be a video uploaded by the video uploading author A in the multimedia client.
In addition, in this embodiment, an implementation manner of determining the multimedia data associated with the filtered interest tag is not specifically limited. For example: in order to improve the matching degree of the recommended content and the interest of the target account, when determining the multimedia data associated with the filtered interest tag, the determination may be made according to the historical play amount of each associated multimedia data, and the like.
It is understood that the execution subject of content recommendation may be the same as or different from the execution subject of updating the weight of the target interest tag. For example: when the execution subject of the content recommendation is the multimedia client, it is reasonable that the execution subject of updating the weight of the target interest tag is executed, and the execution subject may be the multimedia client or the server.
According to the scheme provided by the embodiment of the invention, when the target account is detected to meet the content recommendation condition, each interest tag associated with the target account is determined; based on the weight of each interest tag, screening the interest tags required to be utilized by the content recommendation from each interest tag; and recommending the content for the target account based on the screened interest tags. Therefore, when the content recommendation is performed on the target account, the weight of the screened interest tag is updated based on the interaction behavior of the target account, so that the degree of interest matching between the screened interest tag and the user of the target account is high, and the user experience is improved.
With respect to the above embodiment, as shown in fig. 4, an embodiment of the present invention further provides a device for updating a weight of an interest tag, including:
the data acquisition module 410 is configured to determine a target interaction behavior corresponding to a target account;
a tag determination module 420, configured to determine, from the interest tags associated with the target account, a target interest tag for the target interaction behavior; each interest tag is provided with a weight representing the interest matching degree of the interest tag and the target account;
a weight updating module 430, configured to update the weight of the target interest tag based on the interaction attribute of the target interaction behavior.
Optionally, the weight updating module 430 includes:
the weight determining submodule is used for determining the adjustment amount of the weight of the target interest tag based on the interaction attribute of the target interaction behavior;
and the data updating submodule is used for updating the weight of the target interest tag according to the adjustment quantity of the weight of the target interest tag.
Optionally, the weight determining submodule is specifically configured to determine, from a preset association relationship, an amplitude coefficient corresponding to an interaction attribute of the target interaction behavior;
calculating the adjustment amount of the weight of the target interest tag according to the determined amplitude coefficient;
the preset association relationship is an association relationship between each interaction attribute and an amplitude coefficient, and the amplitude coefficient corresponding to any interaction attribute is as follows: coefficients for adjusting the weights of the interest tags when generating the interactive behavior with the interactive attribute.
Optionally, the interaction attribute of the target interaction behavior comprises a plurality of attributes; calculating an adjustment amount of the weight of the target interest tag according to the determined amplitude coefficient, including:
and fusing the amplitude coefficients corresponding to the determined interactive attributes according to a preset fusion mode to obtain the adjustment quantity of the weight of the target interest tag.
Optionally, the apparatus further comprises:
the time obtaining module is used for obtaining a target time difference of the target account when a preset weight adjusting condition is met, wherein the target time difference is a difference value between the current time and the last time of the interactive action generated by the target account;
and the parameter calculation module is used for determining an adjustment parameter used when the weight of the interest tag is adjusted according to the target time difference for each interest tag associated with the target account, and performing predetermined operation on the determined adjustment parameter and the weight of the interest tag to obtain the updated weight of the interest tag.
With respect to the above embodiment, as shown in fig. 5, an embodiment of the present invention further provides a content recommendation apparatus, including:
a condition detection module 510, configured to determine, when it is detected that a target account meets a content recommendation condition, each interest tag associated with the target account;
a tag screening module 520, configured to screen, from each interest tag, an interest tag that needs to be utilized for the content recommendation based on the weight of each interest tag; updating the weight of each interest label based on any one of the interest label updating methods;
and the content recommending module 530 is configured to recommend content to the target account based on the filtered interest tag.
An embodiment of the present invention further provides an electronic device, as shown in fig. 6, including a processor 601, a communication interface 602, a memory 603, and a communication bus 604, where the processor 601, the communication interface 602, and the memory 603 complete mutual communication through the communication bus 604,
a memory 603 for storing a computer program;
the processor 601 is configured to implement the steps of any one of the weight updating method and the content recommendation method for the interest tag when executing the program stored in the memory 603.
The communication bus mentioned in the above terminal may be a Peripheral Component Interconnect (PCI) bus, an Extended Industry Standard Architecture (EISA) bus, or the like. The communication bus may be divided into an address bus, a data bus, a control bus, etc. For ease of illustration, only one thick line is shown, but this does not mean that there is only one bus or one type of bus.
The communication interface is used for communication between the terminal and other equipment.
The Memory may include a Random Access Memory (RAM) or a non-volatile Memory (non-volatile Memory), such as at least one disk Memory. Optionally, the memory may also be at least one memory device located remotely from the processor.
The Processor may be a general-purpose Processor, and includes a Central Processing Unit (CPU), a Network Processor (NP), and the like; the Integrated Circuit may also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other Programmable logic device, a discrete Gate or transistor logic device, or a discrete hardware component.
In yet another embodiment of the present invention, a computer-readable storage medium is further provided, in which a computer program is stored, and the computer program, when executed by a processor, implements the method for updating the weight of the interest tag or the method for recommending content according to any of the above embodiments.
In yet another embodiment of the present invention, there is also provided a computer program product containing instructions which, when run on a computer, cause the computer to perform the method for updating the weight of an interest tag or the method for recommending content as described in any of the above embodiments.
In the above embodiments, the implementation may be wholly or partially realized by software, hardware, firmware, or any combination thereof. When implemented in software, may be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When loaded and executed on a computer, cause the processes or functions described in accordance with the embodiments of the invention to occur, in whole or in part. The computer may be a general purpose computer, a special purpose computer, a network of computers, or other programmable device. The computer instructions may be stored in a computer readable storage medium or transmitted from one computer readable storage medium to another, for example, from one website site, computer, server, or data center to another website site, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, Digital Subscriber Line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device, such as a server, a data center, etc., that incorporates one or more of the available media. The usable medium may be a magnetic medium (e.g., floppy Disk, hard Disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., Solid State Disk (SSD)), among others.
It is noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
All the embodiments in the present specification are described in a related manner, and the same and similar parts among the embodiments may be referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, for embodiments such as the apparatus, the electronic device, and the storage medium, since they are substantially similar to the method embodiments, the description is relatively simple, and for the relevant points, reference may be made to the partial description of the method embodiments.
The above description is only for the preferred embodiment of the present invention, and is not intended to limit the scope of the present invention. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention shall fall within the protection scope of the present invention.
Claims (10)
1. A method for updating the weight of an interest tag is characterized in that: the method comprises the following steps:
determining a target interaction behavior corresponding to a target account;
determining a target interest tag for the target interaction behavior from the interest tags associated with the target account; each interest tag is provided with a weight representing the interest matching degree of the interest tag and the target account;
and updating the weight of the target interest tag based on the interaction attribute of the target interaction behavior.
2. The method of claim 1, wherein updating the weight of the target interest tag based on the interaction attribute of the target interaction behavior comprises:
determining an adjustment amount of the weight of the target interest tag based on the interaction attribute of the target interaction behavior;
and updating the weight of the target interest tag according to the adjustment amount of the weight of the target interest tag.
3. The method of claim 2, wherein determining an adjustment amount for the weight of the target interest tag based on the interaction attribute of the target interaction behavior comprises:
determining an amplitude coefficient corresponding to the interaction attribute of the target interaction behavior from a preset incidence relation;
calculating the adjustment amount of the weight of the target interest tag according to the determined amplitude coefficient;
the preset association relationship is an association relationship between each interaction attribute and an amplitude coefficient, and the amplitude coefficient corresponding to any interaction attribute is as follows: coefficients for adjusting the weights of the interest tags when generating the interactive behavior with the interactive attribute.
4. The method of claim 3, wherein the interaction attribute of the target interaction behavior comprises a plurality; calculating an adjustment amount of the weight of the target interest tag according to the determined amplitude coefficient, including:
and fusing the amplitude coefficients corresponding to the determined interactive attributes according to a preset fusion mode to obtain the adjustment quantity of the weight of the target interest tag.
5. The method of claim 1, further comprising:
when a preset weight adjustment condition is met, acquiring a target time difference of the target account, wherein the target time difference is a difference value between the current time and the last time of the target account for generating the interactive behavior;
and determining an adjustment parameter used when the weight of the interest tag is adjusted according to the target time difference for each interest tag associated with the target account, and performing predetermined operation on the determined adjustment parameter and the weight of the interest tag to obtain an updated weight of the interest tag.
6. A content recommendation method, comprising:
when detecting that a target account meets a content recommendation condition, determining each interest tag associated with the target account;
based on the weight of each interest tag, screening the interest tags required to be utilized by the content recommendation from each interest tag; wherein the weight of each interest tag is updated based on the updating method of any one of claims 1-5;
and recommending the content for the target account based on the screened interest tags.
7. An apparatus for updating weights of interest tags, comprising:
the data acquisition module is used for determining a target interaction behavior corresponding to the target account;
a tag determination module, configured to determine, from the interest tags associated with the target account, a target interest tag for the target interaction behavior; each interest tag is provided with a weight representing the interest matching degree of the interest tag and the target account;
and the weight updating module is used for updating the weight of the target interest tag based on the interaction attribute of the target interaction behavior.
8. A content recommendation apparatus, comprising:
the condition detection module is used for determining each interest tag associated with the target account when the target account is detected to meet the content recommendation condition;
the tag screening module is used for screening the interest tags required to be utilized by the content recommendation from the interest tags based on the weight of each interest tag; wherein the weight of each interest tag is updated based on the updating method of any one of claims 1-5;
and the content recommendation module is used for recommending the content for the target account based on the interest tag obtained by screening.
9. An electronic device is characterized by comprising a processor, a communication interface, a memory and a communication bus, wherein the processor and the communication interface are used for realizing mutual communication by the memory through the communication bus;
a memory for storing a computer program;
a processor for performing the method steps of any one of claims 1 to 5 or the method steps of claim 6 when executing a program stored in the memory.
10. A computer-readable storage medium, characterized in that a computer program is stored in the computer-readable storage medium, which computer program, when being executed by a processor, carries out the method steps of any one of the claims 1-5 or the method steps of claim 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011619811.1A CN112765400B (en) | 2020-12-31 | 2020-12-31 | Weight updating method, content recommending method, device and equipment for interest labels |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011619811.1A CN112765400B (en) | 2020-12-31 | 2020-12-31 | Weight updating method, content recommending method, device and equipment for interest labels |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112765400A true CN112765400A (en) | 2021-05-07 |
| CN112765400B CN112765400B (en) | 2024-04-23 |
Family
ID=75698250
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011619811.1A Active CN112765400B (en) | 2020-12-31 | 2020-12-31 | Weight updating method, content recommending method, device and equipment for interest labels |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112765400B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113365095A (en) * | 2021-06-15 | 2021-09-07 | 北京百度网讯科技有限公司 | Live broadcast resource recommendation method and device, electronic equipment and storage medium |
| CN114329187A (en) * | 2021-12-08 | 2022-04-12 | 北京五八信息技术有限公司 | Recommendation method and device of content object, electronic equipment and readable medium |
| CN115146156A (en) * | 2022-06-28 | 2022-10-04 | 北京百度网讯科技有限公司 | Linkage adjusting method, device, equipment and medium for interest point situation |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2007280057A (en) * | 2006-04-06 | 2007-10-25 | Sony Corp | Data processor, data processing method, and program |
| CN103399883A (en) * | 2013-07-19 | 2013-11-20 | 百度在线网络技术（北京）有限公司 | Method and system for performing personalized recommendation according to user interest points/concerns |
| KR20160108701A (en) * | 2015-03-05 | 2016-09-20 | 주식회사 벤플 | Method for producing space network considering diversity and accuracy on recommendation |
| CN106339502A (en) * | 2016-09-18 | 2017-01-18 | 电子科技大学 | Modeling recommendation method based on user behavior data fragmentation cluster |
| CN108920503A (en) * | 2018-05-28 | 2018-11-30 | 哈尔滨工程大学 | A kind of micro- video personalized recommendation algorithm based on social networks degree of belief |
| CN109544396A (en) * | 2019-01-10 | 2019-03-29 | 腾讯科技（深圳）有限公司 | Account recommended method, device, server, terminal and storage medium |
| CN110929052A (en) * | 2019-12-03 | 2020-03-27 | 北京奇艺世纪科技有限公司 | Multimedia resource recommendation method and device, electronic equipment and storage medium |
| CN111191124A (en) * | 2019-12-23 | 2020-05-22 | 东软集团股份有限公司 | Method and device for determining interest label weight, storage medium and electronic equipment |
| CN111680219A (en) * | 2020-06-09 | 2020-09-18 | 腾讯科技（深圳）有限公司 | Content recommendation method, device, equipment and readable storage medium |
| CN111767429A (en) * | 2020-06-29 | 2020-10-13 | 北京奇艺世纪科技有限公司 | Video recommendation method and device and electronic equipment |
| CN112069415A (en) * | 2020-08-13 | 2020-12-11 | 中国海洋大学 | Interest point recommendation method based on heterogeneous attribute network characterization learning |
-
2020
- 2020-12-31 CN CN202011619811.1A patent/CN112765400B/en active Active
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2007280057A (en) * | 2006-04-06 | 2007-10-25 | Sony Corp | Data processor, data processing method, and program |
| CN103399883A (en) * | 2013-07-19 | 2013-11-20 | 百度在线网络技术（北京）有限公司 | Method and system for performing personalized recommendation according to user interest points/concerns |
| KR20160108701A (en) * | 2015-03-05 | 2016-09-20 | 주식회사 벤플 | Method for producing space network considering diversity and accuracy on recommendation |
| CN106339502A (en) * | 2016-09-18 | 2017-01-18 | 电子科技大学 | Modeling recommendation method based on user behavior data fragmentation cluster |
| CN108920503A (en) * | 2018-05-28 | 2018-11-30 | 哈尔滨工程大学 | A kind of micro- video personalized recommendation algorithm based on social networks degree of belief |
| CN109544396A (en) * | 2019-01-10 | 2019-03-29 | 腾讯科技（深圳）有限公司 | Account recommended method, device, server, terminal and storage medium |
| CN110929052A (en) * | 2019-12-03 | 2020-03-27 | 北京奇艺世纪科技有限公司 | Multimedia resource recommendation method and device, electronic equipment and storage medium |
| CN111191124A (en) * | 2019-12-23 | 2020-05-22 | 东软集团股份有限公司 | Method and device for determining interest label weight, storage medium and electronic equipment |
| CN111680219A (en) * | 2020-06-09 | 2020-09-18 | 腾讯科技（深圳）有限公司 | Content recommendation method, device, equipment and readable storage medium |
| CN111767429A (en) * | 2020-06-29 | 2020-10-13 | 北京奇艺世纪科技有限公司 | Video recommendation method and device and electronic equipment |
| CN112069415A (en) * | 2020-08-13 | 2020-12-11 | 中国海洋大学 | Interest point recommendation method based on heterogeneous attribute network characterization learning |
Non-Patent Citations (1)
| Title |
|---|
| 刘勇;吴翔宇;解本巨;: "基于动态用户画像的信息推荐研究", 计算机系统应用, 15 June 2018 (2018-06-15), pages 236 - 239 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113365095A (en) * | 2021-06-15 | 2021-09-07 | 北京百度网讯科技有限公司 | Live broadcast resource recommendation method and device, electronic equipment and storage medium |
| CN114329187A (en) * | 2021-12-08 | 2022-04-12 | 北京五八信息技术有限公司 | Recommendation method and device of content object, electronic equipment and readable medium |
| CN115146156A (en) * | 2022-06-28 | 2022-10-04 | 北京百度网讯科技有限公司 | Linkage adjusting method, device, equipment and medium for interest point situation |
| CN115146156B (en) * | 2022-06-28 | 2023-08-15 | 北京百度网讯科技有限公司 | Point-of-interest behavior linkage adjustment method, device, equipment and medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112765400B (en) | 2024-04-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112765400B (en) | Weight updating method, content recommending method, device and equipment for interest labels | |
| US10803111B2 (en) | Live video recommendation by an online system | |
| US9892109B2 (en) | Automatically coding fact check results in a web page | |
| US10528912B2 (en) | User profile improvement system | |
| CN110659388B (en) | To-be-recommended information screening method and device, electronic equipment and storage medium | |
| CN107430630A (en) | For assembling and presenting method, system and the medium of the content related to particular video frequency game | |
| KR102066773B1 (en) | Method, apparatus and system for content recommendation | |
| CN109451333B (en) | Bullet screen display method, device, terminal and system | |
| CN109165691B (en) | Training method and device for model for identifying cheating users and electronic equipment | |
| US8954868B2 (en) | Guided profile editing system | |
| CN112364202A (en) | Video recommendation method and device and electronic equipment | |
| CN112669078A (en) | Behavior prediction model training method, device, equipment and storage medium | |
| CN112989158A (en) | Method, device and storage medium for identifying webpage crawler behavior | |
| CN112883275B (en) | Live broadcast room recommendation method, device, server and medium | |
| CN113743963A (en) | Abnormal recognition model training method, abnormal object recognition device and electronic equipment | |
| CN116821513B (en) | Parameter searching method, device, equipment and medium under recommended scene | |
| US9565224B1 (en) | Methods, systems, and media for presenting a customized user interface based on user actions | |
| CN110996142A (en) | Video recall method and device, electronic equipment and storage medium | |
| WO2021048902A1 (en) | Learning model application system, learning model application method, and program | |
| CN112804566A (en) | Program recommendation method, device and computer readable storage medium | |
| CN113468434B (en) | Resource recommendation method, device, readable medium and equipment | |
| CN110633408A (en) | Recommendation method and system for intelligent business information | |
| CN110839178B (en) | Content prediction method and device and computer readable storage medium | |
| CN111683143B (en) | Message pushing method and device, electronic equipment and computer readable storage medium | |
| CN114154014A (en) | Video cold start recommendation method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |