CN111475741A - Method and device for determining user interest tag - Google Patents
Method and device for determining user interest tag Download PDFInfo
- Publication number
- CN111475741A CN111475741A CN201910066383.5A CN201910066383A CN111475741A CN 111475741 A CN111475741 A CN 111475741A CN 201910066383 A CN201910066383 A CN 201910066383A CN 111475741 A CN111475741 A CN 111475741A
- Authority
- CN
- China
- Prior art keywords
- category
- sequence
- determining
- identification
- target
- 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.)
- Pending
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/50—Network services
- H04L67/55—Push-based network services
Abstract
The embodiment of the application discloses a method and a device for determining a user interest tag. One embodiment of the above method comprises: acquiring target historical behavior data of a target user, wherein the target historical behavior data comprises at least one object identification sequence; determining a target category identification according to at least one object identification sequence, a first preset corresponding relation between the object identification and the category identification and the similarity between the predetermined category identifications; and determining the interest tag of the target user according to the target category identification. According to the implementation method, the interest tag of the user can be determined according to the historical behavior data of the user.
Description
Technical Field
The embodiment of the application relates to the technical field of computers, in particular to a method and a device for determining a user interest tag.
Background
Most advertising information is pushed according to audiences when being pushed, so audience targeting is particularly important for information pushing. Behavioral targeting is an important way of targeting in information push. The core of behavior orientation is to mine long-term and short-term interests of a user according to behavior information of the user in a period of time. And directionally pushing information according to the mined interests.
Disclosure of Invention
The embodiment of the application provides a method and a device for determining a user interest tag.
In a first aspect, an embodiment of the present application provides a method for determining a user interest tag, including: acquiring target historical behavior data of a target user, wherein the target historical behavior data comprises at least one object identification sequence; determining a target category identification according to the at least one object identification sequence, the first preset corresponding relation between the object identification and the category identification and the similarity between the predetermined category identifications; and determining the interest tag of the target user according to the target category identification.
In some embodiments, the determining the target category identifier according to the at least one object identifier sequence, the preset correspondence between the object identifier and the category identifier, and the similarity between the predetermined category identifiers includes: determining a category identification sequence corresponding to the at least one object identification sequence according to the first preset corresponding relation; counting the occurrence times of the category identifications in the category identification sequence; and determining the target category identification according to the counted occurrence times and the similarity between the category identifications.
In some embodiments, the at least one object identification sequence comprises at least one of: the object identifiers in the first object identifier sequence are executed with a first operation, the object identifiers in the second object identifier sequence are executed with a second operation, the object identifiers in the third object identifier sequence are executed with a third operation, and different object identifier sequences correspond to different weights; and the step of determining the target category identifier according to the counted occurrence frequency and the similarity between the category identifiers comprises the following steps: determining the weight corresponding to the object identification sequence as the weight corresponding to the category identification sequence corresponding to the object identification sequence; and determining the target category identification according to the weight corresponding to the category identification sequence, the counted occurrence times and the similarity between the category identifications.
In some embodiments, the object identifier is an item identifier, the first operation is a browse operation, the second operation is a join shopping cart operation, and the third operation is a purchase operation.
In some embodiments, the similarity between the category identifications is determined by: obtaining sample historical behavior data of at least one user, wherein the sample historical behavior data comprises the occurrence time of historical behaviors and the target identification; determining a sample object identification sequence corresponding to at least one user according to the occurrence time; determining a sample category identification sequence corresponding to the sample object identification sequence according to the sample object identification sequence and the first preset corresponding relation; determining a vector sequence according to the sample category identification sequence and a pre-established vector output model, wherein the vector output model is used for representing the corresponding relation between the category identification subsequence and the vector sequence; and determining the similarity between the category identifications according to the vector sequence.
In some embodiments, the determining the sample object identification sequence corresponding to the at least one user according to the occurrence time includes: for sample historical behavior data of a single user, sequencing object identifications from early to late according to occurrence time; taking the object identifier with the earliest occurrence time as a target object identifier; adding the target object identification into a sample object identification sequence; based on the target object identification, the following dividing steps are executed: determining the time length between the occurrence time corresponding to the target object identifier and the next object identifier in the sequence; in response to determining that the duration is less than the preset duration, adding the next object into the sample object identification sequence where the target object identification is located; in response to determining that the duration is greater than or equal to a preset duration, adding a next object to the new sample object identification sequence; taking the next object identification as a target object identification; in response to determining that the target object identifier is at the last of the above ordering, ending the dividing step; and in response to determining that the target object identifier is not at the last position of the ordering, continuing to perform the partitioning step.
In some embodiments, the historical behavior comprises a preset operation; and the method further comprises at least one of: deleting a sample object identification sequence which does not comprise the object identification of the executed preset operation; and deleting the sample object identification sequences with the number of the object identifications smaller than a preset threshold value.
In some embodiments, the historical behavior comprises a preset operation; and the determining the vector sequence according to the sample category identification sequence and the pre-established vector output model comprises: for the sample category identification sequence, determining an object identification of a preset operation executed in the corresponding sample object identification sequence; taking the category identification corresponding to the determined object identification as a label of the sample category identification sequence; and inputting the sample category identification sequence and the label into the vector output model to determine the vector sequence, wherein the vector output model is also used for representing the corresponding relation among the sample category identification sequence, the label and the vector sequence.
In some embodiments, the historical behavior comprises a preset operation, and the object identifier is an article identifier; and the determining the vector sequence according to the sample category identification sequence and the pre-established vector output model comprises: determining a sample updating category identification sequence corresponding to the sample article identification sequence according to a second preset corresponding relation between the article identification and the brand identification and the first preset corresponding relation; and inputting the sample updating category identification sequence into the vector output model to determine a vector sequence, wherein the vector output model is also used for representing the corresponding relation between the sample updating category identification sequence and the vector sequence.
In a second aspect, an embodiment of the present application provides an apparatus for determining a tag of interest of a user, including: the device comprises a first acquisition unit, a second acquisition unit and a third acquisition unit, wherein the first acquisition unit is configured to acquire target historical behavior data of a target user, and the target historical behavior data comprises at least one object identification sequence; a first sequence category identifier determining unit configured to determine a target category identifier according to the at least one object identifier sequence, a first preset corresponding relationship between the object identifier and the category identifier, and a similarity between predetermined category identifiers; and the interest label determining unit is configured to determine the interest label of the target user according to the target category identification.
In a third aspect, an embodiment of the present application provides a method for pushing information, including: acquiring information to be pushed and an interest tag corresponding to the information to be pushed; determining a target user for receiving information to be pushed according to the interest tag; and pushing the information to be pushed to a target user.
In a fourth aspect, an embodiment of the present application provides a server, including: one or more processors; a storage device, on which one or more programs are stored, which, when executed by the one or more processors, cause the one or more processors to implement the method as described in any of the embodiments of the first aspect.
In a fifth aspect, the present application provides a computer-readable medium, on which a computer program is stored, which when executed by a processor implements the method as described in any one of the embodiments of the first aspect.
According to the method and the device for determining the user interest tag, the target historical behavior data of the target user are firstly obtained. And then, determining the target category identification according to at least one object identification sequence included in the target historical behavior data, a first preset corresponding relation between the object identification and the category identification and the similarity between the predetermined category identifications. And finally, determining the interest tag of the target user according to the target category identification. According to the method, the interest tag of the user can be determined according to the historical behavior data of the user.
Drawings
Other features, objects and advantages of the present application will become more apparent upon reading of the following detailed description of non-limiting embodiments thereof, made with reference to the accompanying drawings in which:
FIG. 1 is an exemplary system architecture diagram in which one embodiment of the present application may be applied;
FIG. 2 is a flow diagram of one embodiment of a method for determining user interest tags, according to the present application;
FIG. 3 is a flow diagram of another embodiment of a method for determining user interest tags according to the present application;
FIG. 4 is a flow diagram of yet another embodiment of a method for determining user interest tags according to the present application;
FIG. 5 is a flow diagram for one embodiment of a method for pushing information, according to the present application;
FIG. 6 is a schematic block diagram illustrating one embodiment of an apparatus for determining tags of interest to a user according to the present application;
FIG. 7 is a block diagram of a computer system suitable for use in implementing a server according to embodiments of the present application.
Detailed Description
The present application will be described in further detail with reference to the following drawings and examples. It is to be understood that the specific embodiments described herein are merely illustrative of the relevant invention and not restrictive of the invention. It should be noted that, for convenience of description, only the portions related to the related invention are shown in the drawings.
It should be noted that the embodiments and features of the embodiments in the present application may be combined with each other without conflict. The present application will be described in detail below with reference to the embodiments with reference to the attached drawings.
Fig. 1 shows an exemplary system architecture 100 to which embodiments of the present method for determining user interest tags or an apparatus for determining user interest tags may be applied.
As shown in fig. 1, the system architecture 100 may include terminal devices 101, 102, 103, a network 104, and a server 105. The network 104 serves as a medium for providing communication links between the terminal devices 101, 102, 103 and the server 105. Network 104 may include various connection types, such as wired, wireless communication links, or fiber optic cables, to name a few.
The user may use the terminal devices 101, 102, 103 to interact with the server 105 via the network 104 to receive or send messages or the like. The terminal devices 101, 102, 103 may have various communication client applications installed thereon, such as a web browser application, a shopping application, a search application, an instant messaging tool, a mailbox client, social platform software, and the like.
When the terminal devices 101, 102, 103 are hardware, they may be various electronic devices having a display screen and supporting online shopping, including but not limited to smart phones, tablet computers, e-book readers, MP3 players (Moving Picture Experts Group Audio L layer III, motion Picture Experts compression standard Audio layer 3), MP4(Moving Picture Experts Group Audio L layer IV, motion Picture Experts compression standard Audio layer 4) players, laptop portable computers, desktop computers, and the like.
The server 105 may be a server that provides various services, such as a background server that provides support for shopping websites displayed on the terminal devices 101, 102, 103. The backend server may analyze and otherwise process data such as historical behavior data of the user, and feed back a processing result (e.g., a vector or an interest) to the terminal device 101, 102, 103.
The server may be hardware or software. When the server 105 is hardware, it may be implemented as a distributed server cluster composed of a plurality of servers, or may be implemented as a single server. When the server 105 is software, it may be implemented as multiple pieces of software or software modules (e.g., to provide distributed services), or as a single piece of software or software module. And is not particularly limited herein.
It should be noted that the method for determining the user interest tag provided in the embodiment of the present application is generally performed by the server 105, and accordingly, the apparatus for determining the user interest tag is generally disposed in the server 105.
It should be understood that the number of terminal devices, networks, and servers in fig. 1 is merely illustrative. There may be any number of terminal devices, networks, and servers, as desired for implementation.
With continued reference to FIG. 2, a flow 200 of one embodiment of a method for determining user interest tags in accordance with the present application is shown. The method for determining the user interest tag comprises the following steps:
In this embodiment, an executing subject (for example, the server 105 shown in fig. 1) of the method for determining the user interest tag may acquire the target historical behavior data of the target user through a wired connection manner or a wireless connection manner. The target user may be a user to whom a message needs to be pushed. The target historical behavior data may be the target user's historical behavior data on a certain website or a certain application, and may include at least one object identification sequence. The object may be an item, music, news, etc. The object identification may be an item identification, an identification of music, an identification of news, and the like. For example, when the target historical behavior data includes the historical behavior data of the user at the shopping website, it may include the item identifiers browsed and the corresponding browsing time, the item identifiers purchased and the corresponding purchasing time, the item identifiers concerned and the corresponding concerning time, the item identifiers joined to the shopping cart and the corresponding joining time, the item identifiers collected and the corresponding collecting time, and so on. When the target historical behavior data includes the historical behavior data of the user at the news website, the target historical behavior data may include a browsed news identifier, a corresponding browsing time, and the like.
In this embodiment, the execution subject may locally store a category system, which may include a plurality of object identifiers included under a primary category, a secondary category, a tertiary category, and a tertiary category, where the category system may include a plurality of object identifiers included under the primary category, the secondary category under the primary category may include "home appliance", "mother and baby appliance", and the like, and the secondary category under the "home appliance" may include "large home appliance", "living appliance", "small kitchen" and the like, and the secondary category under the "mother and baby appliance" may include "feeding article", "mom special area", "children's bedroom", and the like, and the tertiary category under the "large home appliance" may include "refrigerator", "washing machine", "air conditioner", "flat tv", and the like, and the tertiary category under the "feeding article" may include "nipple", "milk warming disinfection", "milk sucker", "teether pacifier", and "and the like, and the third category under each of the three-level home and home tv may include a plurality of object identifiers, for example, the object identifier including" XXX L E7E 75 inch large smart bottle nipple 4K 55K (IPS lcd screen) smart home tv 55K 55).
In the category system, category identifiers are preset in each of the primary category, the secondary category and the tertiary category. The execution subject may determine the category identifier corresponding to the object identifier in each object identifier sequence according to the category system. In this embodiment, the category identifier may be a category identifier of a third-level category. The execution subject may record a correspondence between the object identifier and the category identifier as a first preset correspondence. The first preset correspondence may exist in a form of a list or an algorithm.
In this embodiment, the execution subject may further obtain a similarity between predetermined category identifiers. And then, determining the target category identification by combining at least one object identification sequence and the first preset corresponding relation. Taking the historical behavior data of the user at the shopping website as an example, the execution subject may first determine the item identifier with the largest number of browsing times in the at least one object identifier sequence. Then, the category identification corresponding to the article identification is determined. And finally, taking the category identification and the category identification with the highest similarity to the category identification as target category identifications. Or, taking the historical behavior data of the user in the news website as an example, the execution subject may first determine the item identifier with the largest number of browsing times in the at least one object identifier sequence. Then, the category identification corresponding to the article identification is determined. And finally, taking the category identification and the N category identifications with the highest similarity to the category identifications as target category identifications.
In this embodiment, the execution subject may determine the interest tag of the target user according to the target category identifier. For example, the execution subject may identify the target category as a corresponding category name as an interest tag of the target user. Or, the execution subject may select at least one category name from the category names corresponding to the target category identifier as the interest tag of the target user. For example, the interest tags of the target users performing the determination may include "flat tv", "gutta percha placard", and the like.
The method for determining the user interest tag provided by the above embodiment of the application first obtains target historical behavior data of a target user. And then, determining the target category identification according to at least one object identification sequence included in the target historical behavior data, a first preset corresponding relation between the object identification and the category identification and the similarity between the predetermined category identifications. And finally, determining the interest tag of the target user according to the target category identification. According to the method, the interest tag of the user can be determined according to the historical behavior data of the user.
With continued reference to FIG. 3, a flow 300 of another embodiment of a method for determining user interest tags in accordance with the present application is shown. As shown in fig. 3, the method of the present embodiment includes the following steps:
Wherein the target historical behavior data comprises at least one object identification sequence. The principle of this step is similar to that of step 201, and this embodiment is not described again.
The execution subject may first determine, according to a first preset correspondence, at least one category identification sequence corresponding to at least one object identification sequence included in the target historical behavior data. Specifically, the execution main body may determine, according to the first preset correspondence, a category identifier corresponding to each object identifier in each object identifier sequence. Then, according to the arrangement relation among the object identifications in each object identification sequence, arranging all kinds of object identifications according to the same arrangement relation to obtain at least one category identification sequence.
After obtaining the various types of target identification sequences, the execution subject may count the occurrence times of the various types of target identifications.
And step 304, determining the target category identification according to the counted occurrence times and the similarity between the category identifications.
The execution main body can determine the target category identification according to the counted occurrence frequency and the similarity between the category identifications. For example, the execution subject may use, as the target category identifier, the category identifier that appears most frequently and the category identifier that has the highest similarity to the above-described category identifiers. Or, the execution subject may determine the top N category identifiers with higher occurrence times according to the occurrence times. Then, the category identifier with the highest similarity of the N category identifiers is used as the target category identifier.
In some optional implementations of this embodiment, the at least one object identification sequence included in the target historical behavior data may include at least one of: a first object identification sequence, a second object identification sequence, and a third object identification sequence. And the object identifiers in the first object identifier sequence are used for executing a first operation by the target user. The object identification in the second object identification sequence is executed by the target user to perform the second operation, and the object identification in the third object identification sequence is executed by the target user to perform the third operation. And different object identification sequences correspond to different weights. For the historical behavior data of the target user on the music website, the first operation may be a listening operation, the second operation may be a collecting operation, and the third operation may be a downloading operation.
The step 304 may be implemented by the following steps not shown in fig. 3: determining the weight corresponding to the object identification sequence as the weight corresponding to the category identification sequence corresponding to the object identification sequence; and determining the target category identification according to the weight corresponding to the category identification sequence, the counted occurrence times and the similarity between the category identifications.
In this implementation, the execution subject may first determine, according to the weight corresponding to the object identifier sequence, the weight corresponding to the category identifier sequence corresponding to the object identifier sequence. And then, determining the target category identification according to the weight corresponding to the category identification sequence, the counted occurrence frequency and the similarity between the category identifications. For example, for each category identification sequence, the execution subject may first multiply the occurrence number of each category identification in the category identification sequence by the weight corresponding to the category identification sequence to obtain a product. Then, the different products of the same category identification are added to obtain a sum value corresponding to each category identification. And finally, determining the target category identification according to the sum of the category identifications and the similarity of the category identifications.
In some optional implementation manners of this embodiment, the object identifier is an item identifier, that is, the object identifier sequence is an item identifier sequence. The execution principal may determine the target category identification according to the following steps not shown in fig. 3: and determining the brand mark corresponding to the article mark according to the second preset corresponding relation between the article mark and the brand mark. And combining the brand mark with the category mark to obtain an updated category mark. Thereby obtaining an updated category identification sequence corresponding to the item identification sequence. Then, the occurrence number of each update category identification in the update category identification sequence is determined. And for each updating category identification, determining the scores of other updating category identifications according to the occurrence frequency of the updating category identification and the similarity between the updating category identification and other updating category identifications. It will be appreciated that the similarity between the update category identifications is the same as the similarity between the category identifications. And then, adding the scores of the same updating category identification to obtain the sum of the scores of all the updating category identifications. Then, the updating category identifications are arranged according to the sequence from big to small of the score sum, and a score sum sequence of the updating category identifications is obtained. The execution subject may use the previous preset number of update category identifications in the score sum sequence as the target category identification.
For example, the target historical behavior data includes a first item identification sequence (A)1、A2、A3……An) Second object identification sequence (B)1、B2、B3……Bn) And a third object identification sequence (C)1、C2、C3……Cn). Identifying a sequence (A) for a first item1、A2、A3……An) The execution subject may count the number of occurrences of each item identifier in the sequence of item identifiers. For example, performing subject statistics results in A12 times, A2… … A1 timen3 times. Then, an update category identification sequence (D) corresponding to the item identification sequence is determined1、D2、D3……Dn) And determining the occurrence frequency of each article identifier as the occurrence frequency of the corresponding update category identifierAnd (4) counting. Namely to obtain D12 times, D21 time … … Dn3 times. The executive agent may then be in accordance with D1Similarity with other update category identifiers and D1The score of other update category identifications is calculated. For example, the executive agent may calculate the score for other update category identifications according to the following formulam＝sim(D1,Dm) Number of occurrences. Wherein D ismScore for the mth update category identificationmIs DmScore of (2), sim (D)1,Dm) Is D1And DmSimilarity between them, the number of occurrences is D1The number of occurrences of (c). It is understood that m is [1, n ]]A natural number in between. Identification sequence (D) for update categories1、D2、D3……Dn) After the scores of other update category identifiers are calculated, the scores of the update category identifiers can be counted, and the sum of the scores of the update category identifiers is calculated. And finally, arranging the updated category identifications according to the sequence of the scores from large to small to obtain the updated category identification score sum sequence.
In some optional implementations of this embodiment, the first operation is a browse operation, the second operation is a shopping cart adding operation, and the third operation is a purchase operation.
And 305, determining the interest tag of the target user according to the target category identification.
The principle of this step is similar to that of step 203, and this embodiment is not described again.
According to the method for determining the interest tag of the user, provided by the embodiment of the application, the interest tag of the user can be determined according to different operations of the user on the object identifier, so that the accuracy of determining the interest tag of the user is improved.
With continued reference to FIG. 4, a flow 400 of yet another embodiment of a method for determining user interest tags in accordance with the present application is shown. As shown in fig. 4, the method of this embodiment may determine the similarity between the category identifiers through the following steps:
In this embodiment, the execution subject for determining the similarity between the category identifiers may be the same as or different from the execution subject in the embodiment shown in fig. 2. If not, the execution subject determining the similarity between the category identifiers may send the determined similarity to the execution subject in the embodiment shown in fig. 2.
For a shopping website, the historical behavior data may include an item identification for which the historical behavior is intended and a time of occurrence of the historical behavior. Such historical behavior may include browsing, joining a shopping cart, purchasing, and the like. For example, the historical behavior data may include item identifications browsed by the user over the past month, item identifications joined to the shopping cart, and item identifications purchased. The item identifier here may be a SKU (Stock Keeping Unit) of the item, or may be a title of the item displayed on a page of a shopping site. For example, the identifier of a certain brand of mobile phone may be "XXX full screen game smart phone 6GB +64GB black full internet through 4G dual card dual standby", where "XXX" is the brand name. The occurrence time of the historical behavior may refer to a time when the user performed the historical behavior.
In this embodiment, after obtaining the historical behavior data, the execution subject may arrange the object identifiers corresponding to the historical behaviors of the same user according to the sequence of the occurrence time of the historical behaviors from early to late, so as to obtain a sample object identifier sequence corresponding to the user. Therefore, for the historical behavior data of at least one user, the sample object identification sequence corresponding to the at least one user can be obtained correspondingly.
Specifically, the execution subject may also preset the number of object identifiers in each sample object identifier sequence to obtain a specified number of object identifiers. For example, the execution subject may set the object identifications included in the sample object identification sequence to 50.
In some optional implementations of this embodiment, the step 402 may be specifically implemented by the following steps not shown in fig. 4:
step 4021, for sample historical behavior data of a single user, sequencing object identifications from early to late according to occurrence time.
For each user's sample historical behavior data, the execution principal may order the object identifications from early to late as the historical behavior occurrence time. It will be appreciated that the execution body may order the occurrence times of the same historical behaviors in order from early to late. For example, the execution body may sort the occurrence times of the browsing operations in order from early to late. The occurrence time of the operation of adding the shopping cart is sorted according to the sequence from morning to evening.
Step 4022, the object identifier with the earliest occurrence time is used as the target object identifier.
For each ordering, the execution principal may identify the object that occurs the earliest as the target object identification.
Step 4023, add the target object identification to the sample object identification sequence.
The execution body may then add the target object identification to the sample object identification sequence.
Step 4024, based on the target object identification, executing the following dividing steps: determining the time length between the occurrence time corresponding to the target object identifier and the next object identifier in the sequence; in response to determining that the duration is less than the preset duration, adding the next object into the sample object identification sequence where the target object identification is located; in response to determining that the duration is greater than or equal to a preset duration, adding a next object to the new sample object identification sequence; taking the next object identification as a target object identification; and ending the dividing step in response to determining that the target object identifier is at the last position of the sorting.
The execution body may divide the above ordering into a plurality of sample object identification sequences according to the dividing step. Specifically, the execution principal may first determine a time duration between occurrence times of the target object identifier and the next object identifier in the ranking. And if the duration is less than the preset duration, adding the next object into the sample object identification sequence where the target object identification is located. And if the time length is greater than or equal to the preset time length, adding the next object into the new sample object identification sequence. And the next object identification as the new target object identification. And simultaneously judging whether the new target object identification is positioned at the tail bit of the sorting. And if the new target object identification is determined to be positioned at the last position of the sorting, the sorting is completed, and the dividing step is finished. If it is determined that the new target object id is not located at the last of the rankings, step 4025 is performed.
Step 4025, in response to determining that the target object identifier is not located at the last position of the above sequence, continuing to perform the dividing step.
If the new target object identification is not located at the last position of the sorting, the dividing step is continuously executed.
In some optional implementations of the embodiment, the historical behavior includes a preset operation. The above method may further comprise at least one of the following not shown in fig. 4: deleting a sample object identification sequence which does not comprise the object identification of the executed preset operation; and deleting the sample object identification sequences with the number of the object identifications smaller than a preset threshold value.
In this implementation, the execution subject may determine whether each sample object identifier sequence includes an object identifier for which a preset operation is executed. The preset operation may be a purchase operation. If not, the execution principal may delete the sample object identification sequence. The execution subject may further determine the number of each object id in the sample object id sequence, and if the number is smaller than a preset threshold, the execution subject may delete the sample object id sequence.
For each obtained sample object identification sequence, the execution subject may determine, according to the first preset corresponding relationship, a category identification corresponding to each object identification in the sample object identification sequence. And then the sample category identification sequence corresponding to the sample object identification sequence can be obtained.
In this embodiment, the vector output model may be used to represent a correspondence between the category identifier subsequence and the vector sequence. Specifically, the vector output model may include a correspondence table, and the correspondence table may be pre-established by a skilled person based on correspondence between a large number of category identification sequences and vector sequences. The vector output model can also be a neural network or a model formed by combining the neural networks, such as a word2vec algorithm and a fastText algorithm. The word2vec algorithm can quickly and effectively express a word into a vector form through an optimized training model according to a given corpus. fastText is a text classification and vectorization tool introduced in 2016 by fair (facebook AI research). The fastText algorithm includes two parts: classifiers and skip-gram models. Wherein the classifier can be used for classifying the text content; the skip-gram model can be used to convert text, words, sentences into vectors.
In some optional implementations of the embodiment, the historical behavior includes a preset operation. The step 404 may be specifically realized by the following steps not shown in fig. 4: for the sample category identification sequence, determining an object identification of a preset operation executed in the corresponding sample object identification sequence; taking the category identification corresponding to the determined object identification as a label of the sample category identification sequence; and inputting the sample category identification sequence and the label into a vector output model, and determining a vector sequence.
In this implementation manner, for each sample category identification sequence, the execution subject may determine the object identifier on which the preset operation is executed from the sample object identification sequence corresponding to the sample category identification sequence. The preset operation may be a purchase operation. Then, the execution subject may use the category identifier corresponding to the determined object identifier as a tag of the sample category identifier sequence. Finally, the execution subject can input the sample category identification sequence and the label into the vector output model to determine the vector sequence. The vector output model is also used for representing the corresponding relation among the sample category identification sequence, the label and the vector sequence.
It should be noted that, when the sample object identification sequence includes a plurality of item identifications added to the shopping cart, the executing entity may determine the tag of the sample object identification sequence according to the following steps: in response to determining that the sample object identification sequence includes at least two item identifications joined to the shopping cart, category identifications corresponding to the at least two item identifications joined to the shopping cart are determined. In response to determining that at least two category identifications are the same, determining that the category identification is a tag of the sample object identification sequence. In response to determining that the at least two category identifiers are not the same, determining the number of article identifiers corresponding to the category identifier in the at least two category identifiers in the sample object identifier sequence; and identifying the category with the largest number of corresponding article identifications as the label of the sample object identification sequence.
That is, the executing entity may first determine whether the category identifiers corresponding to the item identifiers of the respective participating shopping carts are the same. If so, such object is identified as a tag for the sample object identification sequence. If not, the execution subject can count the quantity of the article identifications corresponding to the various target identifications. And then, the category identifier with the largest number of corresponding article identifiers is used as a label of the sample object identification sequence.
In some optional implementations of this embodiment, the historical behavior includes a preset operation, and the object identifier is an item identifier. The step 404 may be specifically realized by the following steps not shown in fig. 4: determining a sample updating category identification sequence corresponding to the sample article identification sequence according to a second preset corresponding relation and a first preset corresponding relation between the article identification and the brand identification; and inputting the sample updating category identification sequence into the vector output model, and determining the vector sequence.
In this implementation, the execution main body may obtain a second preset corresponding relationship between the item identifier and the brand identifier. And simultaneously determining the updating category identification corresponding to each article identification in the article identification sequence by combining the first preset corresponding relation. Specifically, the execution subject may merge the category identifier and the brand identifier corresponding to the article identifier, and use the merged identifier as the updated category identifier. For example, the article identifier is "AAA full screen game smartphone 6GB +64GB black full internet through 4G dual card dual standby", and the brand to which the article belongs is AAA. The execution subject may determine the brand identifier to be 123 according to a pre-stored correspondence between the brand name and the brand identifier. Then, through the preset corresponding relationship list between the category identifier and the brand identifier, it is determined that the corresponding second category identifier is 134. The first category identification 134_123 corresponding to the item identification is finally obtained.
Therefore, a sample updating category identification sequence corresponding to the sample article identification sequence can be obtained. The execution agent may then input the sample update category identification sequence into the vector output model, determining a vector sequence. The vector output model is also used for representing the corresponding relation between the sample updating category identification sequence and the vector sequence.
After determining the vector sequence, the execution subject may calculate a distance between vectors according to each vector in the vector sequence, and determine a similarity between category identifiers. The distance may be a cosine distance, a manhattan distance, or the like.
In some optional implementation manners of this embodiment, the execution main body may determine a vector sequence corresponding to the category identifier sequence by using a skip-gram model of the fastText algorithm. Specifically, the execution main body may input the category identification sequence into the trained skip-gram model to obtain a vector corresponding to each category identification in the category identification sequence.
The method for determining the user interest tag provided by the above embodiment of the application may extract a vector of the category identifier corresponding to the item identifier targeted by the user behavior from the historical behavior data of the user, so that the similarity between the category identifiers may be calculated.
With continued reference to fig. 5, a flow 500 of one embodiment of a method for pushing information in accordance with the present application is shown. As shown in fig. 5, the method of the present embodiment includes the following steps:
In this embodiment, an execution subject of the method for pushing information (e.g., the server 105 shown in fig. 1) may acquire information to be pushed and an interest tag corresponding to the information to be pushed in various ways. For example, a user may send information to be pushed and an interest tag corresponding to the information to be pushed to an execution subject by using a terminal. The user can be an advertiser, and the information to be pushed can be various kinds of advertisement information, weather information, traffic information and the like. The interest tag corresponding to the information to be pushed may be a keyword for describing the information to be pushed. When the information to be pushed includes article information, the interest tag may be a category name corresponding to the article information on the sales website.
In this embodiment, the execution main body may store a mapping relationship between the user and the interest tag. The mapping relationship may be determined by any of the embodiments shown in fig. 2 to 4. After obtaining the interest tag corresponding to the information to be pushed, the execution subject may query the mapping relationship to determine at least one user corresponding to the interest tag. The execution subject may take the at least one user as a target user. It is understood that each user may correspond to a plurality of interest tags, and the executing subject may also obtain a plurality of interest tags. The execution main body can set a quantity threshold value according to actual requirements so as to meet different pushing precision. For example, the executing subject may consider the user as the target user only when it is determined that the number of interest tags of the user and the same interest tags in the interest tags corresponding to the information to be pushed is greater than 2.
And step 503, pushing the information to be pushed to the target user.
After determining the target user, the execution subject may push information to be pushed to the target user, so that the target user can receive information related to the interest tag of the target user.
The method for pushing the information can be used for pushing the relevant information to the user in a targeted manner, so that the pushing precision of the information is improved.
With further reference to fig. 6, as an implementation of the method shown in the above-mentioned figures, the present application provides an embodiment of an apparatus for determining a tag of interest of a user, where the embodiment of the apparatus corresponds to the embodiment of the method shown in fig. 2, and the apparatus may be applied to various electronic devices.
As shown in fig. 6, the apparatus 600 for determining a tag of interest of a user of the present embodiment includes: a first obtaining unit 601, a category identification determining unit 602, and an interest label determining unit 603.
A first obtaining unit 601, configured to obtain target historical behavior data of a target user, where the target historical behavior data includes at least one object identification sequence.
A first sequence determining unit 602, configured to determine a target category identifier according to at least one object identifier sequence, a first preset correspondence between object identifiers and category identifiers, and a similarity between predetermined category identifiers.
An interest tag determining unit 603 configured to determine an interest tag of the target user according to the target category identification.
In some optional implementation manners of this embodiment, the category identifier determining unit 603 may further include a first category target identifier determining module, an occurrence count counting module, and a second category identifier determining module.
And the first type target identification determining module is configured to determine a category identification sequence corresponding to at least one object identification sequence according to the first preset corresponding relation.
And the occurrence frequency counting module is configured to count the occurrence frequency of the category identification in the category identification sequence.
And the second category identification determining module is configured to determine the target category identification according to the counted occurrence times and the similarity between the category identifications.
In some optional implementations of this embodiment, the at least one object identification sequence includes at least one of: a first object identification sequence, a second object identification sequence, and a third object identification sequence. The object identifiers in the first object identifier sequence are subjected to a first operation, the object identifiers in the second object identifier sequence are subjected to a second operation, and the object identifiers in the third object identifier sequence are subjected to a third operation. Different object identification sequences correspond to different weights. The second category identification determination module is further configured to: determining the weight corresponding to the object identification sequence as the weight corresponding to the category identification sequence corresponding to the object identification sequence; and determining the target category identification according to the weight corresponding to the category identification sequence, the counted occurrence times and the similarity between the category identifications.
In some optional implementations of this embodiment, the object identifier is an item identifier, the first operation is a browsing operation, the second operation is a shopping cart entering operation, and the third operation is a purchasing operation.
In some optional implementations of this embodiment, the similarity between the category identifiers is determined by the following units not shown in the apparatus 600: the device comprises a second acquisition unit, a first sequence determination unit, a second sequence determination unit, a vector sequence determination unit and a similarity determination unit.
A second obtaining unit configured to obtain sample historical behavior data of at least one user. The sample historical behavior data includes the time of occurrence of the historical behavior and the identification of the object for which it is intended.
And the first sequence determining unit is configured to determine the sample object identification sequence corresponding to at least one user according to the occurrence time.
And the second sequence determining unit is configured to determine a sample category identification sequence corresponding to the sample object identification sequence according to the sample object identification sequence and the first preset corresponding relation.
And the vector sequence determining unit is configured to determine the vector sequence according to the sample category identification sequence and a vector output model established in advance. The vector output model is used for representing the corresponding relation between the category identification subsequence and the vector sequence.
And the similarity determining unit is configured to determine the similarity between the category identifications according to the vector sequence.
In some optional implementations of this embodiment, the first sequence determination unit is further configured to: for sample historical behavior data of a single user, sequencing object identifications from early to late according to occurrence time; taking the object identifier with the earliest occurrence time as a target object identifier; adding the target object identification to the sample object identification sequence; based on the target object identification, performing the following partitioning steps: determining the time length between the occurrence time corresponding to the target object identifier and the next object identifier in the sequence; in response to determining that the duration is less than the preset duration, adding the next object into the sample object identification sequence where the target object identification is located; in response to determining that the duration is greater than or equal to a preset duration, adding a next object to the new sample object identification sequence; taking the next object identification as a target object identification; in response to determining that the target object identifier is at the last position of the ordering, ending the dividing step; in response to determining that the target object identification is not at the last position in the ordering, continuing to perform the partitioning step.
In some optional implementations of the embodiment, the historical behavior includes a preset operation. The apparatus 600 may further include a deletion unit, not shown in fig. 6, configured to: deleting a sample object identification sequence which does not comprise the object identification of the executed preset operation; and/or deleting sample object identification sequences with the number of object identifications smaller than a preset threshold.
In some optional implementations of the embodiment, the historical behavior includes a preset operation. The vector sequence determination unit is further configured to: for the sample category identification sequence, determining an object identification of a preset operation executed in the corresponding sample object identification sequence; taking the category identification corresponding to the determined object identification as a label of the sample category identification sequence; and inputting the sample category identification sequence and the label into a vector output model, and determining the vector sequence, wherein the vector output model is also used for representing the corresponding relation among the sample category identification sequence, the label and the vector sequence.
In some optional implementations of this embodiment, the historical behavior includes a preset operation, and the object identifier is an item identifier. The vector sequence determination unit is further configured to: determining a sample updating category identification sequence corresponding to the sample article identification sequence according to a second preset corresponding relation and a first preset corresponding relation between the article identification and the brand identification; and inputting the sample updating category identification sequence into a vector output model, and determining the vector sequence, wherein the vector output model is also used for representing the corresponding relation between the sample updating category identification sequence and the vector sequence.
The apparatus for determining a tag of interest of a user provided in the above embodiment of the present application first obtains target historical behavior data of a target user. And then, determining the target category identification according to at least one object identification sequence included in the target historical behavior data, a first preset corresponding relation between the object identification and the category identification and the similarity between the predetermined category identifications. And finally, determining the interest tag of the target user according to the target category identification. According to the method, the interest tag of the user can be determined according to the historical behavior data of the user.
It should be understood that the units 601 to 603, which are recorded in the apparatus 600 for determining a tag of interest of a user, correspond to the respective steps in the method described with reference to fig. 2, respectively. Thus, the operations and features described above for the method for determining a tag of interest of a user are equally applicable to the apparatus 600 and the units comprised therein and will not be described in detail here.
Referring now to FIG. 7, shown is a block diagram of a computer system 700 suitable for use in implementing a server according to embodiments of the present application. The server shown in fig. 7 is only an example, and should not bring any limitation to the functions and the scope of use of the embodiments of the present application.
As shown in fig. 7, the computer system 700 includes a Central Processing Unit (CPU)701, which can perform various appropriate actions and processes in accordance with a program stored in a Read Only Memory (ROM)702 or a program loaded from a storage section 708 into a Random Access Memory (RAM) 703. In the RAM 703, various programs and data necessary for the operation of the system 700 are also stored. The CPU 701, the ROM 702, and the RAM 703 are connected to each other via a bus 704. An input/output (I/O) interface 705 is also connected to bus 704.
To the I/O interface 705, AN input section 706 including a keyboard, a mouse, and the like, AN output section 707 including a keyboard such as a Cathode Ray Tube (CRT), a liquid crystal display (L CD), and the like, a speaker, and the like, a storage section 708 including a hard disk and the like, and a communication section 709 including a network interface card such as a L AN card, a modem, and the like, the communication section 709 performs communication processing via a network such as the internet, a drive 710 is also connected to the I/O interface 705 as necessary, a removable medium 711 such as a magnetic disk, AN optical disk, a magneto-optical disk, a semiconductor memory, and the like is mounted on the drive 710 as necessary, so that a computer program read out therefrom is mounted into the storage section 708 as necessary.
In particular, according to an embodiment of the present disclosure, the processes described above with reference to the flowcharts may be implemented as computer software programs. For example, embodiments of the present disclosure include a computer program product comprising a computer program embodied on a machine-readable medium, the computer program comprising program code for performing the method illustrated in the flow chart. In such an embodiment, the computer program can be downloaded and installed from a network through the communication section 709, and/or installed from the removable medium 711. The computer program, when executed by a Central Processing Unit (CPU)701, performs the above-described functions defined in the method of the present application.
It should be noted that the computer readable medium described herein can be a computer readable signal medium or a computer readable storage medium or any combination of the two. A computer readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the foregoing. More specific examples of the computer readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
In the present application, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. In this application, however, a computer readable signal medium may include a propagated data signal with computer readable program code embodied therein, for example, in baseband or as part of a carrier wave. Such a propagated data signal may take many forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wire, fiber optic cable, RF, etc., or any suitable combination of the foregoing.
Computer program code for carrying out operations of the present application may be written in any combination of one or more programming languages, including AN object oriented programming language such as Java, Smalltalk, C + +, and conventional procedural programming languages, such as the "C" programming language or similar programming languages.
The flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present application. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
The units described in the embodiments of the present application may be implemented by software or hardware. The described units may also be provided in a processor, and may be described as: a processor includes a first acquisition unit, a category identification determination unit, and an interest tag determination unit. Where the names of these units do not in some cases constitute a limitation on the unit itself, for example, the first obtaining unit may also be described as a "unit that obtains target historical behavior data of a target user".
As another aspect, the present application also provides a computer-readable medium, which may be contained in the apparatus described in the above embodiments; or may be present separately and not assembled into the device. The computer readable medium carries one or more programs which, when executed by the apparatus, cause the apparatus to: acquiring target historical behavior data of a target user, wherein the target historical behavior data comprises at least one object identification sequence; determining a target category identification according to at least one object identification sequence, a first preset corresponding relation between the object identification and the category identification and the similarity between the predetermined category identifications; and determining the interest tag of the target user according to the target category identification.
The above description is only a preferred embodiment of the application and is illustrative of the principles of the technology employed. It will be appreciated by those skilled in the art that the scope of the invention herein disclosed is not limited to the particular combination of features described above, but also encompasses other arrangements formed by any combination of the above features or their equivalents without departing from the spirit of the invention. For example, the above features may be replaced with (but not limited to) features having similar functions disclosed in the present application.
Claims (13)
1. A method for determining user interest tags, comprising:
acquiring target historical behavior data of a target user, wherein the target historical behavior data comprises at least one object identification sequence;
determining a target category identification according to the at least one object identification sequence, a first preset corresponding relation between the object identification and the category identification and the similarity between the predetermined category identifications;
and determining the interest tag of the target user according to the target category identification.
2. The method according to claim 1, wherein the determining a target category identifier according to the at least one object identifier sequence, the preset correspondence between the object identifiers and the category identifiers, and the similarity between the predetermined category identifiers comprises:
determining a category identification sequence corresponding to the at least one object identification sequence according to the first preset corresponding relation;
counting the occurrence times of the category identifications in the category identification sequence;
and determining the target category identification according to the counted occurrence times and the similarity between the category identifications.
3. The method of claim 2, wherein the at least one object identification sequence comprises at least one of: the object identifiers in the first object identifier sequence are executed with a first operation, the object identifiers in the second object identifier sequence are executed with a second operation, the object identifiers in the third object identifier sequence are executed with a third operation, and different object identifier sequences correspond to different weights; and
the determining the target category identifier according to the counted occurrence times and the similarity between the category identifiers includes:
determining the weight corresponding to the object identification sequence as the weight corresponding to the category identification sequence corresponding to the object identification sequence;
and determining the target category identification according to the weight corresponding to the category identification sequence, the counted occurrence times and the similarity between the category identifications.
4. The method of claim 3, wherein the object identifier is an item identifier, the first operation is a browse operation, the second operation is a join shopping cart operation, and the third operation is a purchase operation.
5. A method according to any of claims 1-3, wherein the similarity between the category labels is determined by:
obtaining sample historical behavior data of at least one user, wherein the sample historical behavior data comprises occurrence time of historical behaviors and targeted object identification;
determining a sample object identification sequence corresponding to at least one user according to the occurrence time;
determining a sample category identification sequence corresponding to the sample object identification sequence according to the sample object identification sequence and the first preset corresponding relation;
determining a vector sequence according to the sample category identification sequence and a pre-established vector output model, wherein the vector output model is used for representing the corresponding relation between the category identification subsequence and the vector sequence;
and determining the similarity between the category identifications according to the vector sequence.
6. The method of claim 5, wherein said determining a sample object identification sequence corresponding to said at least one user according to said occurrence time comprises:
for sample historical behavior data of a single user, sequencing object identifications from early to late according to occurrence time;
taking the object identifier with the earliest occurrence time as a target object identifier;
adding the target object identification to a sample object identification sequence;
based on the target object identification, performing the following partitioning steps: determining the time length between the occurrence time corresponding to the target object identifier and the next object identifier in the sequence; in response to determining that the duration is less than the preset duration, adding the next object into the sample object identification sequence where the target object identification is located; in response to determining that the duration is greater than or equal to a preset duration, adding a next object to a new sample object identification sequence; taking the next object identification as a target object identification; in response to determining that the target object identifier is at the last of the ordering, ending the dividing step;
in response to determining that the target object identification is not at the last position of the ordering, continuing to perform the partitioning step.
7. The method of claim 5, wherein the historical behavior comprises a preset operation; and
the method further comprises at least one of:
deleting a sample object identification sequence which does not comprise the object identification of the executed preset operation;
and deleting the sample object identification sequences with the number of the object identifications smaller than a preset threshold value.
8. The method of claim 5, wherein the historical behavior comprises a preset operation; and
determining a vector sequence according to the sample category identification sequence and a vector output model established in advance comprises:
for the sample category identification sequence, determining an object identification of a preset operation executed in the corresponding sample object identification sequence;
taking the category identification corresponding to the determined object identification as a label of the sample category identification sequence;
and inputting the sample category identification sequence and the label into the vector output model, and determining the vector sequence, wherein the vector output model is also used for representing the corresponding relation among the sample category identification sequence, the label and the vector sequence.
9. The method of claim 5, wherein the historical behavior comprises a preset operation, and the object identifier is an item identifier; and
determining a vector sequence according to the sample category identification sequence and a vector output model established in advance comprises:
determining a sample updating category identification sequence corresponding to the sample article identification sequence according to a second preset corresponding relation between the article identification and the brand identification and the first preset corresponding relation;
and inputting the sample updating category identification sequence into the vector output model, and determining the vector sequence, wherein the vector output model is also used for representing the corresponding relation between the sample updating category identification sequence and the vector sequence.
10. An apparatus for determining user interest tags, comprising:
the device comprises a first acquisition unit, a second acquisition unit and a third acquisition unit, wherein the first acquisition unit is configured to acquire target historical behavior data of a target user, and the target historical behavior data comprises at least one object identification sequence;
a category identifier determining unit configured to determine a target category identifier according to the at least one object identifier sequence, a first preset correspondence between object identifiers and category identifiers, and a similarity between predetermined category identifiers;
and the interest label determining unit is configured to determine the interest label of the target user according to the target category identification.
11. A method for pushing information, comprising:
acquiring information to be pushed and an interest tag corresponding to the information to be pushed;
determining a target user receiving the information to be pushed according to the interest tag;
and pushing the information to be pushed to the target user.
12. A server, comprising:
one or more processors;
a storage device having one or more programs stored thereon,
when executed by the one or more processors, cause the one or more processors to implement the method of any one of claims 1-9.
13. A computer-readable medium, on which a computer program is stored which, when being executed by a processor, carries out the method according to any one of claims 1-9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910066383.5A CN111475741A (en) | 2019-01-24 | 2019-01-24 | Method and device for determining user interest tag |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910066383.5A CN111475741A (en) | 2019-01-24 | 2019-01-24 | Method and device for determining user interest tag |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111475741A true CN111475741A (en) | 2020-07-31 |
Family
ID=71743505
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910066383.5A Pending CN111475741A (en) | 2019-01-24 | 2019-01-24 | Method and device for determining user interest tag |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111475741A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| TWI823036B (en) * | 2020-08-19 | 2023-11-21 | 英華達股份有限公司 | Recommended target user selecting method, system, equipment and storage medium |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090276377A1 (en) * | 2008-04-30 | 2009-11-05 | Cisco Technology, Inc. | Network data mining to determine user interest |
| CN102819804A (en) * | 2011-06-07 | 2012-12-12 | 阿里巴巴集团控股有限公司 | Goods information pushing method and device |
| CN104090886A (en) * | 2013-12-09 | 2014-10-08 | 深圳市腾讯计算机系统有限公司 | Method and device for constructing real-time portrayal of user |
| CN105653693A (en) * | 2015-12-30 | 2016-06-08 | 东软集团股份有限公司 | Individualization recommendation method and apparatus |
| CN106919619A (en) * | 2015-12-28 | 2017-07-04 | 阿里巴巴集团控股有限公司 | A kind of commercial articles clustering method, device and electronic equipment |
| CN108629608A (en) * | 2017-03-22 | 2018-10-09 | 腾讯科技（深圳）有限公司 | User data processing method and processing device |
-
2019
- 2019-01-24 CN CN201910066383.5A patent/CN111475741A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090276377A1 (en) * | 2008-04-30 | 2009-11-05 | Cisco Technology, Inc. | Network data mining to determine user interest |
| CN102819804A (en) * | 2011-06-07 | 2012-12-12 | 阿里巴巴集团控股有限公司 | Goods information pushing method and device |
| CN104090886A (en) * | 2013-12-09 | 2014-10-08 | 深圳市腾讯计算机系统有限公司 | Method and device for constructing real-time portrayal of user |
| CN106919619A (en) * | 2015-12-28 | 2017-07-04 | 阿里巴巴集团控股有限公司 | A kind of commercial articles clustering method, device and electronic equipment |
| CN105653693A (en) * | 2015-12-30 | 2016-06-08 | 东软集团股份有限公司 | Individualization recommendation method and apparatus |
| CN108629608A (en) * | 2017-03-22 | 2018-10-09 | 腾讯科技（深圳）有限公司 | User data processing method and processing device |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| TWI823036B (en) * | 2020-08-19 | 2023-11-21 | 英華達股份有限公司 | Recommended target user selecting method, system, equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107172151B (en) | Method and device for pushing information | |
| CN109299994B (en) | Recommendation method, device, equipment and readable storage medium | |
| CN107426328B (en) | Information pushing method and device | |
| US8725592B2 (en) | Method, system, and medium for recommending gift products based on textual information of a selected user | |
| CN105718184A (en) | Data processing method and apparatus | |
| CN110009401A (en) | Advertisement placement method, device and storage medium based on user's portrait | |
| CN107862553A (en) | Advertisement real-time recommendation method, device, terminal device and storage medium | |
| CN106294500B (en) | Content item pushing method, device and system | |
| CN110363604B (en) | Page generation method and device | |
| WO2012031239A2 (en) | User interest analysis systems and methods | |
| WO2016192309A1 (en) | Pushed information processing method, apparatus, and device, and non-volatile computer storage medium | |
| TW201411381A (en) | Labeling Product Identifiers and Navigating Products | |
| US10853424B1 (en) | Content delivery using persona segments for multiple users | |
| TW201905736A (en) | Information push method and system | |
| JP6976207B2 (en) | Information processing equipment, information processing methods, and programs | |
| US20150348097A1 (en) | Autocreated campaigns for hashtag keywords | |
| CN110059256B (en) | System, method and device for displaying information | |
| CN104077707B (en) | A kind of optimization method and device for promoting presentation mode | |
| CN104063799A (en) | Promotion message pushing method and device | |
| CN107977678A (en) | Method and apparatus for output information | |
| WO2015110846A1 (en) | Native creative generation using hashtagged user generated content | |
| CN111104590A (en) | Information recommendation method, device, medium and electronic equipment | |
| CN110428277B (en) | Touch method of recommended product, storage medium and program product | |
| CN111899053A (en) | Online and offline advertisement management system for merchants | |
| KR102438679B1 (en) | Operating method of server for providing media marketing service |
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 |