CN117851653A

CN117851653A - Object matching method, device, electronic equipment, storage medium and program product

Info

Publication number: CN117851653A
Application number: CN202410126633.0A
Authority: CN
Inventors: 刘心元
Original assignee: Baidu China Co Ltd
Current assignee: Baidu China Co Ltd
Priority date: 2024-01-29
Filing date: 2024-01-29
Publication date: 2024-04-09

Abstract

The disclosure provides an object matching method, an object matching device, electronic equipment, a storage medium and a program product, and relates to the technical fields of clustering algorithm, feature matching and information pushing. The method comprises the following steps: acquiring user characteristics of a user object logged in a target website; determining a first correlation degree between the user characteristics and approximate vermicelli clusters related to the category to which the user objects belong under each candidate anchor object, and carrying out clustering treatment on all vermicelli objects under each anchor object in advance to obtain vermicelli clusters of different categories under each anchor object; determining a second correlation degree between the user characteristics and the clustering centers respectively represented by the candidate anchor objects; determining to obtain a comprehensive correlation degree based on the first correlation degree and the second correlation degree; and determining a target candidate anchor object matched with the user object according to the comprehensive relevance. The method can enable the determined target candidate anchor object to be matched with the actual requirement of the user object.

Description

Object matching method, device, electronic equipment, storage medium and program product

Technical Field

The disclosure relates to the technical field of data processing, in particular to the technical fields of clustering algorithm, feature matching and information pushing, and particularly relates to an object matching method, an object matching device, electronic equipment, a computer readable storage medium and a computer program product.

Background

With the continuous improvement of the economic level and the continuous enrichment of substances, people continuously pursue mental entertainment and relaxation after working.

Taking various live broadcast and video platforms as examples, video creators, anchor and live broadcast personnel attract or gather users interested in the video content by issuing various created video contents.

It is a matter of urgent need for those skilled in the art how to help users accessing these platforms to find their interested video creator, anchor or live person among many candidates quickly and accurately.

Disclosure of Invention

The embodiment of the disclosure provides an object matching method, an object matching device, electronic equipment, a computer readable storage medium and a computer program product.

In a first aspect, an embodiment of the present disclosure provides an object matching method, including: acquiring user characteristics of a user object logged in a target website; determining a first correlation degree between the user characteristics and approximate fan class clusters which are related to the class to which the user objects belong under each candidate anchor object; clustering all vermicelli objects under each anchor object in advance to obtain vermicelli clusters of different categories under each anchor object, wherein the approximate vermicelli clusters are vermicelli clusters with a category which is more than a preset association degree with the category to which the user object belongs; determining a second correlation degree between the user characteristics and the clustering centers respectively represented by the candidate anchor objects; determining to obtain a comprehensive correlation degree based on the first correlation degree and the second correlation degree; and determining a target candidate anchor object matched with the user object according to the comprehensive relevance.

In a second aspect, an embodiment of the present disclosure proposes an object matching apparatus, including: a user feature acquisition unit configured to acquire a user feature of a user object logged in to a target website; a first relevance determining unit configured to determine a first relevance between the user feature and an approximate fan cluster related to a category to which the user object belongs under each candidate anchor object; clustering all vermicelli objects under each anchor object in advance to obtain vermicelli clusters of different categories under each anchor object, wherein the approximate vermicelli clusters are vermicelli clusters with a category which is more than a preset association degree with the category to which the user object belongs; a second correlation determination unit configured to determine a second correlation between the user feature and a cluster center represented by each candidate anchor object, respectively; a comprehensive relevance determining unit configured to determine a resultant comprehensive relevance based on the first relevance and the second relevance; and a matching anchor object determining unit configured to determine a target candidate anchor object matching the user object according to the integrated correlation degree.

In a third aspect, an embodiment of the present disclosure provides an electronic device, including: at least one processor; and a memory communicatively coupled to the at least one processor; wherein the memory stores instructions executable by the at least one processor to enable the at least one processor to implement the object matching method as described in the first aspect when executed.

In a fourth aspect, embodiments of the present disclosure provide a non-transitory computer-readable storage medium storing computer instructions for enabling a computer to implement the object matching method as described in the first aspect when executed.

In a fifth aspect, embodiments of the present disclosure provide a computer program product comprising a computer program which, when executed by a processor, is capable of implementing the steps of the object matching method as described in the first aspect.

The object matching scheme provided by the disclosure is based on the obtained user characteristics of the user object of the login target website, on one hand, whether the user object is matched with a fan cluster under a certain anchor object in the category by determining the correlation between the user characteristics and fan clusters similar to the category to which the user object belongs under each candidate anchor object, on the other hand, the comprehensive matching degree of the user object and the candidate anchor object is more intuitively checked by determining the correlation between the user characteristics and the clustering centers respectively represented by each candidate anchor object, and further, the target candidate anchor object matched with the user object is determined by integrating the correlation between the two matching degrees of the two different dimension characterizations, so that the determined target candidate anchor object and the user object have higher matching degree and the matching result more meet the actual requirements of the user.

It should be understood that the description in this section is not intended to identify key or critical features of the embodiments of the disclosure, nor is it intended to be used to limit the scope of the disclosure. Other features of the present disclosure will become apparent from the following specification.

Drawings

Other features, objects and advantages of the present disclosure will become more apparent upon reading of the detailed description of non-limiting embodiments, made with reference to the following drawings:

FIG. 1 is an exemplary system architecture in which the present disclosure may be applied;

FIG. 2 is a flowchart of an object matching method according to an embodiment of the present disclosure;

FIG. 3 is a flowchart of a method for clustering vermicelli objects to obtain vermicelli clusters according to an embodiment of the present disclosure;

FIG. 4 is a flowchart of another object matching method provided by an embodiment of the present disclosure;

fig. 5 is a schematic branch diagram of an information pushing method according to an embodiment of the present disclosure;

fig. 6 is a flow chart of an object matching and matching object pushing method combined with a specific application scenario according to an embodiment of the present disclosure;

FIG. 7 is a diagram of a network architecture combined with a ranking model according to an embodiment of the present disclosure;

fig. 8 is a block diagram of an object matching device according to an embodiment of the present disclosure;

Fig. 9 is a schematic structural diagram of an electronic device adapted to perform an object matching method according to an embodiment of the present disclosure.

Detailed Description

Exemplary embodiments of the present disclosure are described below in conjunction with the accompanying drawings, which include various details of the embodiments of the present disclosure to facilitate understanding, and should be considered as merely exemplary. Accordingly, one of ordinary skill in the art will recognize that various changes and modifications of the embodiments described herein can be made without departing from the scope and spirit of the present disclosure. Also, descriptions of well-known functions and constructions are omitted in the following description for clarity and conciseness. It should be noted that, without conflict, the embodiments of the present disclosure and features of the embodiments may be combined with each other.

In the technical scheme of the disclosure, the related processes of collecting, storing, using, processing, transmitting, providing, disclosing and the like of the personal information of the user accord with the regulations of related laws and regulations, and the public order colloquial is not violated.

FIG. 1 illustrates an exemplary system architecture 100 to which embodiments of the subject matching methods, apparatus, electronic devices, and computer-readable storage media of the present disclosure may be applied.

As shown in fig. 1, a system architecture 100 may include terminal devices 101, 102, 103, a network 104, and a server 105. The network 104 is used as a medium to provide communication links between the terminal devices 101, 102, 103 and the server 105. The network 104 may include various connection types, such as wired, wireless communication links, or fiber optic cables, among others.

The user may interact with the server 105 via the network 104 using the terminal devices 101, 102, 103 to receive or send messages or the like. Various applications for implementing information communication between the terminal devices 101, 102, 103 and the server 105, such as a video on demand application, a live broadcast application, an instant messaging application, and the like, may be installed on the terminal devices.

The terminal devices 101, 102, 103 and the server 105 may be hardware or software. When the terminal devices 101, 102, 103 are hardware, they may be various electronic devices with display screens, including but not limited to smartphones, tablets, laptop and desktop computers, etc.; when the terminal devices 101, 102, 103 are software, they may be installed in the above-listed electronic devices, which may be implemented as a plurality of software or software modules, or may be implemented as a single software or software module, which is not particularly limited herein. When the server 105 is hardware, it may be implemented as a distributed server cluster formed by a plurality of servers, or may be implemented as a single server; when the server is software, the server may be implemented as a plurality of software or software modules, or may be implemented as a single software or software module, which is not particularly limited herein.

The server 105 can provide various services through various built-in applications, for example, a live broadcast application that can provide a live video service, and the server 105 can achieve the following effects when running the live broadcast application: firstly, receiving a user feature of a user object requesting to log in to a target website through terminal devices 101, 102, 103 through a network 104; then, determining a first correlation degree between the user characteristic and an approximate vermicelli cluster which is related to the category to which the user object belongs under each candidate anchor object, and carrying out clustering treatment on all vermicelli objects under each anchor object in advance to obtain vermicelli clusters of different categories under each anchor object, wherein the approximate vermicelli cluster is a vermicelli cluster which has a correlation degree exceeding a preset correlation degree with the category to which the user object belongs; then, determining a second correlation degree between the user characteristic and the clustering center represented by each candidate anchor object; next, determining to obtain a comprehensive correlation degree based on the first correlation degree and the second correlation degree; and finally, determining a target candidate anchor object matched with the user object according to the comprehensive relevance.

Further, the server 105 may push the determined access identifier of the target candidate anchor object to the terminal device 101, 102, 103 as part of the subsequent response information through the network 104, and the terminal device 101, 102, 103 presents the access identifier to the user object through a display screen.

It should be noted that, in addition to being acquired from the terminal devices 101, 102, 103 through the network 104, the user characteristics may be stored in advance in the server 105 in various ways, or may be obtained by analyzing and processing the historical behavior information of the user object collected by the server 105 according to the history. Thus, when the server 105 detects that such data has been stored locally, it may choose to obtain the data directly from the local, in which case the exemplary system architecture 100 may not include the terminal devices 101, 102, 103 and the network 104.

Since computing, processing and feature matching operations on various information contained in numerous candidate anchor agents require more computing resources and stronger computing power, the object matching method provided in the subsequent embodiments of the present disclosure is generally performed by the server 105 having stronger computing power and more computing resources, and accordingly, the object matching device is also generally disposed in the server 105. However, it should be noted that, when the terminal devices 101, 102, 103 also have the required computing capability and computing resources, the terminal devices 101, 102, 103 may also complete each operation performed by the server 105 through the live broadcast application installed thereon, and further output the same result as the server 105. Especially in the case where there are a plurality of terminal devices having different computing capabilities at the same time, but when the live broadcast application determines that the terminal device has a stronger computing capability and more computing resources remain, the terminal device may be allowed to perform the above-mentioned computation, so that the computing pressure of the server 105 is appropriately reduced, and accordingly, the object matching device may also be provided in the terminal devices 101, 102, 103. In this case, the exemplary system architecture 100 may also not include the server 105 and the network 104.

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.

Referring to fig. 2, fig. 2 is a flowchart of an object matching method according to an embodiment of the disclosure, wherein a flowchart 200 includes the following steps:

step 201: acquiring user characteristics of a user object logged in a target website;

this step aims at legally acquiring, by an execution subject of the object matching method (e.g., the server 105 shown in fig. 1), a user feature that can be read by the execution subject via a user object that is legally logged onto a target website (which may also be referred to as a target site, a target address, a target web page, a target page, or a target platform).

It should be understood that the user characteristics may be characteristics (usually, may be represented in a text or image manner that is recognized by a person, or may be represented in a character string, a number, a vector, or a matrix that is recognized by a computer) that are carried in login information of the user object in logging in the target website, or may be some personalized information that the user object actively informs when the website requests after logging in the target website, or may be obtained by analyzing and processing, by the execution subject, historical behavior information of the user object in the target website that is legally collected after the user object is pre-authorized, which is not limited specifically herein, and may be flexibly selected according to practical situations.

Step 202: determining a first correlation degree between the user characteristics and approximate fan class clusters which are related to the class to which the user objects belong under each candidate anchor object;

on the basis of step 201, this step aims at determining, by the execution subject, a first correlation between the user feature and the approximate fan class cluster under each candidate anchor object, which is related to the class to which the user object belongs. The approximate fan cluster is a fan cluster having a degree of association exceeding a preset degree of association with a category to which the user object belongs (for example, when the upper limit of the degree of association is 1, the preset degree of association is set to 80%), and the category to which the user object belongs is determined based on the user characteristics and is used for representing which type of host the user object is interested in, for example, a singing class, a raping class, a smiling class, a live broadcast with goods class, a history comment class, a game class and the like.

In order to achieve the above purpose, clustering processing is needed to be performed on all the vermicelli objects under each anchor object in advance, so as to obtain vermicelli clusters of different categories under each anchor object. In the clustering process, various common clustering algorithms including K-means can be adopted, and proper clustering center numbers can be set by self under the condition that various categories contained in live broadcast contents of the main broadcasting object are known, so that clustering operation can be performed more accurately, and a more accurate clustering result can be obtained.

An implementation, including but not limited to, may be seen in a flow 300 shown in fig. 3, comprising the steps of:

step 301: acquiring all vermicelli objects under each anchor object;

step 302: determining the target clustering center number according to the reason category of the vermicelli which becomes the main broadcasting object and the type number of the video resource content of the main broadcasting object;

step 303: and clustering the vermicelli objects according to the target cluster center number to obtain vermicelli clusters with the number of the target cluster centers.

In this embodiment, according to the reason category of the vermicelli as the anchor object and the type number of the video resource content characterizing the anchor object, a reasonable clustering center number is obtained by self-calculation, that is, the target clustering center number, so that the target clustering center is used to guide the subsequent clustering operation, and the situation that the vermicelli cluster clustered by using the randomly set clustering center number is unreasonable is avoided.

Of course, in addition to the clustering method provided in the embodiment shown in fig. 3, when the target cluster center number cannot be accurately determined, the clustering operation may be performed by adopting a random setting or a test cluster center number selected only from daily experience.

It should be understood that the first relevance obtained in this step is used to characterize whether the user object matches a fan cluster under a certain anchor object in the category to which the user object belongs, if the category of interest of the user object has a higher matching degree with a fan cluster under a certain anchor, it can be obtained that the user object has a higher relevance between the user object and the fan cluster, that is, the corresponding anchor object will be more likely to be interested by the user object because of the fan cluster.

Step 203: determining a second correlation degree between the user characteristics and the clustering centers respectively represented by the candidate anchor objects;

on the basis of step 202, this step aims at determining, by the above-mentioned executing body, a second degree of correlation between the user characteristics and the cluster centers represented by the candidate anchor objects, respectively. It should be understood that the second correlation degree obtained in this step is used to more intuitively determine the comprehensive matching degree of the user object and the candidate anchor object, that is, the cluster center represented by each candidate anchor object should be determined comprehensively according to each fan cluster included in the candidate anchor object, that is, the cluster center of the candidate anchor object should also be the cluster center between each fan cluster, so that the candidate anchor object can be characterized on the overall level.

Specifically, the user feature may include a user feature graph vector expressed in the form of a graph vector, and the first correlation degree and the second correlation degree respectively represent a vector similarity degree between the graph vectors.

Step 204: determining to obtain a comprehensive correlation degree based on the first correlation degree and the second correlation degree;

based on the step 202 and the step 203, the step aims to determine the comprehensive relevance by the execution body based on the determined first relevance and the determined second relevance to integrate the parameters for representing the relevance from different layers and different dimensions respectively, so as to better represent the relevance between the user object and each candidate anchor object through the comprehensive relevance.

Considering that the contribution degree of the first correlation degree and the second phase Guan Du to the finally obtained comprehensive correlation degree are different, the corresponding weighting weights can be set by combining different contribution degrees under different application scenes, so that the calculated comprehensive correlation degree is more in line with the actual situation through a weighting calculation method by means of reasonable weighting weights.

Step 205: and determining a target candidate anchor object matched with the user object according to the comprehensive relevance.

Based on step 204, this step aims at determining, by the above-mentioned executing body, a target candidate anchor object matching with the user object according to the comprehensive relevance, that is, according to the magnitude of the comprehensive relevance of each candidate anchor object to the user object, the candidate anchor object having a larger comprehensive relevance tends to be determined as the target candidate anchor object.

According to the object matching method provided by the embodiment of the disclosure, based on the obtained user characteristics of the user object of the login target website, on one hand, whether the user object is matched with a fan cluster under a certain anchor object in the category by determining the correlation between the user characteristics and fan clusters similar to the category to which the user object belongs under each candidate anchor object, on the other hand, the comprehensive matching degree of the user object and the candidate anchor object is more intuitively checked by determining the correlation between the user characteristics and the clustering centers respectively represented by each candidate anchor object, and further, the target candidate anchor object matched with the user object is determined by integrating the correlation between the two different dimension characterizations, so that the determined target candidate anchor object and the user object have higher matching degree, and the matching result better meets the actual requirements of the user.

To enhance the understanding of the solution provided in the present application, the present disclosure further provides a more specific object matching method through fig. 4, where the process 400 includes the following steps:

step 401: acquiring user characteristics of a user object logged in a target website;

the present step corresponds to step 201 in the flowchart 200, and the corresponding content is referred to the development description of step 201, and the detailed description is not repeated here.

Step 402: calculating cosine similarity between user characteristics and approximate fan clusters under each candidate anchor object, and determining a first correlation according to the calculated cosine similarity result;

step 403: calculating cosine similarity between the user characteristics and clustering centers respectively represented by the candidate anchor objects, and determining a second correlation according to the calculated cosine similarity result;

in comparison with steps 202-203, steps 402-403 correspond to a further specific implementation manner of calculating the cosine similarity between the two and determining the corresponding correlation according to the calculated cosine similarity. Of course, in some other possible implementations, other ways of calculating the correlation besides the cosine similarity may be used, which are not listed here.

Step 404: acquiring a first weight corresponding to the first correlation degree and a second weight corresponding to the second phase Guan Du;

step 405: obtaining a comprehensive correlation degree according to the first correlation degree weighted by the first weight and the second correlation degree weighted by the second weight;

based on step 204, step 404 and step 405 specifically provide an implementation manner of weighting by using a weighting calculation method, and obtaining the integrated correlation by adding the two weighted correlations.

Step 406: outputting the matching degree sequence of each candidate anchor object and the user object by using a sequence model fused with the comprehensive relevance;

step 407: and determining the candidate anchor objects meeting the preset requirements as target candidate anchor objects according to the matching degree sequencing.

Based on step 205, step 406 and step 407 further combine the ranking model of the actual application scenario, and provide an implementation manner of additionally fusing the calculated comprehensive relevance into the ranking model, and ranking the matching degree of each candidate anchor object and the user object based on the ranking influencing factors including the comprehensive relevance by means of the ranking model, so as to finally obtain the matching degree ranking sequence.

It should be understood that, in this embodiment, there is no causal or dependency relationship between a preferred relevance calculating scheme provided by steps 402 and 403, a preferred weighted thought-based comprehensive relevance calculating scheme provided by steps 404-405, and a preferred combined ranking model determining scheme provided by steps 406-407, and different embodiments may be formed in a manner of independently replacing only their corresponding upper-level schemes, and this embodiment is merely presented as one preferred embodiment that includes them at the same time.

On the basis of any of the above embodiments, after determining the target candidate anchor object according to the comprehensive relevance, different recommended branching schemes may be further selected and executed later, which may be referred to as a branching diagram shown in fig. 5:

step 501 shows a mode one, namely recommending a target candidate anchor object to a user object:

specifically, the target candidate anchor object may be recommended to the user object in a recommendation popup manner by presenting the target candidate anchor object in a personalized recommendation popup, and/or by increasing the probability that video resource content of the target candidate anchor object appears in a resource recommendation list of the user object.

Step 502 and step 503 illustrate a second way to determine a potential fan-object of each candidate anchor object according to the comprehensive correlation between each user object and each candidate anchor object, and then recommend the potential fan-object to the corresponding candidate anchor object.

The method is different from the recommendation method provided by the first method for pushing the target candidate anchor object to the user object, and the second method is to reversely utilize the comprehensive relevance to provide a batch of potential silk noodles with higher relevance to the candidate anchor object, which may be called silk noodles thereof, for the candidate anchor object to plan or adjust subsequent live broadcast content or content category according to the potential silk noodles.

For deepening understanding, the present disclosure further provides a complete set of implementation manner by specifically combining with a live broadcast service scenario:

considering that the conversion efficiency of the anchor fan is obviously higher than that of the non-fan, the conversion of the live broadcast service mainly comprises the actions of watching live broadcast time of a user, interaction of the user and the anchor, appreciation of the anchor by the user and the like. Therefore, how to efficiently mine out the live potential vermicelli is important to the promotion of the live vermicelli scale and the improvement of the subsequent live conversion efficiency. The embodiment provides a live broadcast potential vermicelli mining method, which can effectively improve the attention rate of live broadcast under a live broadcast recommended scene, so that the scale of live broadcast vermicelli is continuously improved, and the conversion efficiency of live broadcast is driven to be improved.

Cluster-like vectors representing different preferences of the anchor fans are generated by clustering the anchor fans, as shown in fig. 6: and respectively screening main class clusters representing the main broadcasting silk noodles and class clusters related to the current user from the clustered class clusters. And finally, respectively calculating the similarity between the current user and the two screened class clusters, and fusing the similarity into the existing sequencing model to estimate the probability of the user focusing on the anchor.

The specific technical details are as follows:

1) And (5) main sowing vermicelli clustering: based on graph vectors of vermicelli users, using a k-means clustering method to cluster the vermicelli of the anchor into a plurality of class clusters, wherein each class cluster represents different preferences of the anchor vermicelli, and as shown in fig. 7, the clustered class clusters are represented by 'seed gcf' for subsequent steps;

2) Cluster-like screening: as shown in fig. 7, the user gcf is used to represent the graph vector of the current user, then the class clusters related to the current user are screened out through an attention mechanism, and the main class clusters representing the main broadcasting vermicelli are screened out through a self-attention mechanism;

3) Similarity calculation: calculating cosine similarity of the current user and the fan cluster related to the user, and measuring the distance between the current user and the related group; and calculating the cosine correlation degree between the current user and the group cluster representing the main broadcasting fan group, and measuring the distance between the current user and the whole main broadcasting fan group. The two similarities respectively represent the possibility that the user becomes a potential fan of the anchor from different dimensions;

4) Fusion with existing ranking model: as shown in fig. 7, 2) and 3) are packaged into a potential vermicelli mining component, the output result of the component and the output result of the existing model are fused by addition, and the probability of the user focusing on the anchor is obtained through sigmoid activation function (S-type function).

With further reference to fig. 8, as an implementation of the method shown in the foregoing figures, the present disclosure provides an embodiment of an object matching apparatus, where an embodiment of the apparatus corresponds to the embodiment of the method shown in fig. 2, and the apparatus may be specifically applied to various electronic devices.

As shown in fig. 8, the object matching apparatus 800 of the present embodiment may include: a user feature acquisition unit 801, a first relevance determination unit 802, a second relevance determination unit 803, an integrated relevance determination unit 804, a matching anchor object determination unit 805. Wherein, the user feature acquisition unit 801 is configured to acquire a user feature of a user object logged in a target website; a first relevance determining unit 802 configured to determine a first relevance between the user feature and an approximate fan cluster related to the category to which the user object belongs under each candidate anchor object; clustering all vermicelli objects under each anchor object in advance to obtain vermicelli clusters of different categories under each anchor object, wherein the approximate vermicelli clusters are vermicelli clusters with a category which is more than a preset association degree with the category to which the user object belongs; a second correlation determination unit 803 configured to determine a second correlation between the user feature and the cluster center represented by each candidate anchor object, respectively; an integrated correlation determination unit 804 configured to determine an integrated correlation based on the first correlation and the second correlation; the matching anchor object determination unit 805 is configured to determine a target candidate anchor object matching the user object according to the integrated relevance.

In the present embodiment, in the object matching apparatus 800: the specific processes and technical effects of the user feature obtaining unit 801, the first relevance determining unit 802, the second relevance determining unit 803, the comprehensive relevance determining unit 804, and the matching anchor object determining unit 805 may refer to the relevant descriptions of steps 201 to 205 in the corresponding embodiment of fig. 2, and are not described herein.

In some optional implementations of the present embodiment, the object matching apparatus 800 further includes: a clustering operation unit configured to obtain vermicelli clusters of different categories under each anchor object, the clustering operation unit being further configured to:

acquiring all vermicelli objects under each anchor object;

determining the number of target clustering centers according to the number of reasons for becoming the anchor objects and the number of types of video resource contents of the anchor objects;

and clustering the vermicelli objects according to the target cluster center number to obtain vermicelli clusters with the number of the target cluster centers.

In some optional implementations of the present embodiment, the first relevance determining unit 802 may be further configured to:

calculating cosine similarity between user characteristics and approximate fan clusters under each candidate anchor object, and determining a first correlation according to the calculated cosine similarity result;

Correspondingly, the second relatedness-determining unit 803 may be further configured to:

and calculating cosine similarity between the user characteristics and the clustering centers respectively represented by the candidate anchor objects, and determining a second correlation according to the calculated cosine similarity result.

In some optional implementations of this embodiment, the user features include user feature graph vectors represented in graph vector form, and the corresponding first and second correlations each represent a degree of vector similarity between the graph vectors.

In some optional implementations of the present embodiment, the integrated relevance determining unit 804 may be further configured to:

acquiring a first weight corresponding to the first correlation degree and a second weight corresponding to the second phase Guan Du; the first weight and the second weight are respectively obtained by determining the contribution degree of the corresponding relevance to the obtained comprehensive relevance under the actual application scene;

and obtaining the comprehensive correlation degree according to the first correlation degree weighted by the first weight and the second correlation degree weighted by the second weight.

In some optional implementations of the present embodiment, the matching anchor object determination unit 805 may be further configured to:

Outputting the matching degree sequence of each candidate anchor object and the user object by using the sequence model fused with the comprehensive relevance;

according to the matching degree sequencing, determining candidate anchor objects meeting preset requirements as target candidate anchor objects; wherein, the preset requirements include: at least one of the first N bits in the matching degree sequence, the matching degree exceeding a preset value and the first M% in the matching degree sequence, wherein N and M are positive numbers.

In some optional implementations of the present embodiment, the object matching apparatus 800 may further include:

and the first recommending unit is configured to recommend the target candidate anchor object to the user object.

In some optional implementations of the present embodiment, the first recommendation unit may be further configured to:

presenting the target candidate anchor object in a personalized recommendation popup to recommend the target candidate anchor object to a user object in a recommendation popup mode;

and/or

The probability of video resource content of the target candidate anchor object appearing in the resource recommendation list of the user object is increased.

the potential fan object determining unit is configured to determine potential fan objects of the candidate anchor objects according to comprehensive relativity between the user objects and the candidate anchor objects;

And the second recommending unit is configured to recommend the potential vermicelli object to the corresponding candidate anchor object.

The object matching device provided in this embodiment is based on the obtained user characteristics of the user object logging in the target website, and on one hand, by determining the correlation between the user characteristics and the fan clusters of each candidate anchor object, which are similar to the category to which the user object belongs, it is possible to see whether the user object is matched with a fan cluster of a certain anchor object in the category, on the other hand, by determining the correlation between the user characteristics and the cluster centers represented by each candidate anchor object, it is possible to more intuitively see the comprehensive matching degree of the user object and the candidate anchor object, and further, by integrating the correlation between two different dimension characterizations, it is possible to determine the target candidate anchor object matched with the user object, so that the determined target candidate anchor object and the user object have a higher matching degree, and the matching result more accords with the actual requirement of the user.

According to an embodiment of the present disclosure, the present disclosure further provides an electronic device including: at least one processor; and a memory communicatively coupled to the at least one processor; the memory stores instructions executable by the at least one processor to enable the at least one processor to implement the object matching method described in any of the embodiments above when executed.

According to an embodiment of the present disclosure, there is also provided a readable storage medium storing computer instructions for enabling a computer to implement the object matching method described in any of the above embodiments when executed.

According to an embodiment of the present disclosure, the present disclosure further provides a computer program product, which, when executed by a processor, is capable of implementing the object matching method described in any of the above embodiments.

Fig. 9 shows a schematic block diagram of an example electronic device 900 that may be used to implement embodiments of the present disclosure. Electronic devices are intended to represent various forms of digital computers, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. The electronic device may also represent various forms of mobile devices, such as personal digital processing, cellular telephones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions, are meant to be exemplary only, and are not meant to limit implementations of the disclosure described and/or claimed herein.

As shown in fig. 9, the apparatus 900 includes a computing unit 901 that can perform various appropriate actions and processes according to a computer program stored in a Read Only Memory (ROM) 902 or a computer program loaded from a storage unit 908 into a Random Access Memory (RAM) 903. In the RAM 903, various programs and data required for the operation of the device 900 can also be stored. The computing unit 901, the ROM 902, and the RAM 903 are connected to each other by a bus 904. An input/output (I/O) interface 905 is also connected to the bus 904.

Various components in device 900 are connected to I/O interface 905, including: an input unit 906 such as a keyboard, a mouse, or the like; an output unit 907 such as various types of displays, speakers, and the like; a storage unit 908 such as a magnetic disk, an optical disk, or the like; and a communication unit 909 such as a network card, modem, wireless communication transceiver, or the like. The communication unit 909 allows the device 900 to exchange information/data with other devices through a computer network such as the internet and/or various telecommunications networks.

The computing unit 901 may be a variety of general and/or special purpose processing components having processing and computing capabilities. Some examples of computing unit 901 include, but are not limited to, a Central Processing Unit (CPU), a Graphics Processing Unit (GPU), various specialized Artificial Intelligence (AI) computing chips, various computing units running machine learning model algorithms, a Digital Signal Processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 901 performs the respective methods and processes described above, such as an object matching method. For example, in some embodiments, the object matching method may be implemented as a computer software program tangibly embodied on a machine-readable medium, such as the storage unit 908. In some embodiments, part or all of the computer program may be loaded and/or installed onto the device 900 via the ROM 902 and/or the communication unit 909. When the computer program is loaded into the RAM 903 and executed by the computing unit 901, one or more steps of the object matching method described above may be performed. Alternatively, in other embodiments, the computing unit 901 may be configured to perform the object matching method by any other suitable means (e.g. by means of firmware).

Various implementations of the systems and techniques described here above may be implemented in digital electronic circuitry, integrated circuit systems, field Programmable Gate Arrays (FPGAs), application Specific Integrated Circuits (ASICs), application Specific Standard Products (ASSPs), systems On Chip (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and/or combinations thereof. These various embodiments may include: implemented in one or more computer programs, the one or more computer programs may be executed and/or interpreted on a programmable system including at least one programmable processor, which may be a special purpose or general-purpose programmable processor, that may receive data and instructions from, and transmit data and instructions to, a storage system, at least one input device, and at least one output device.

Program code for carrying out methods of the present disclosure may be written in any combination of one or more programming languages. These program code may be provided to a processor or controller of a general purpose computer, special purpose computer, or other programmable data processing apparatus such that the program code, when executed by the processor or controller, causes the functions/operations specified in the flowchart and/or block diagram to be implemented. The program code may execute entirely on the machine, partly on the machine, as a stand-alone software package, partly on the machine and partly on a remote machine or entirely on the remote machine or server.

In the context of this disclosure, a machine-readable medium may be a tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. The machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

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

The systems and techniques described here can be implemented in a computing system that includes a background component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front-end component (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such background, middleware, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: local Area Networks (LANs), wide Area Networks (WANs), and the internet.

The computer system may include a client and a server. The client and server are typically remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. The server can be a cloud server, also called a cloud computing server or a cloud host, and is a host product in a cloud computing service system, so as to solve the defects of large management difficulty and weak service expansibility in the traditional physical host and virtual private server (VPS, virtual Private Server) service.

According to the technical scheme of the embodiment of the disclosure, based on the obtained user characteristics of the user object of the login target website, on one hand, whether the user object is matched with a fan cluster under a certain anchor object in the category by determining the correlation between the user characteristics and fan clusters similar to the category to which the user object belongs under each candidate anchor object, on the other hand, the comprehensive matching degree of the user object and the candidate anchor object is more intuitively checked by determining the correlation between the user characteristics and the clustering centers respectively represented by each candidate anchor object, and further, the target candidate anchor object matched with the user object is determined by integrating the correlation between the two matching degrees of the two different dimension characterizations, so that the determined target candidate anchor object and the user object have higher matching degree and the matching result more meet the actual requirements of the user.

It should be appreciated that various forms of the flows shown above may be used to reorder, add, or delete steps. For example, the steps recited in the present disclosure may be performed in parallel or sequentially or in a different order, provided that the desired results of the technical solutions of the present disclosure are achieved, and are not limited herein.

The above detailed description should not be taken as limiting the scope of the present disclosure. It will be apparent to those skilled in the art that various modifications, combinations, sub-combinations and alternatives are possible, depending on design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present disclosure are intended to be included within the scope of the present disclosure.

Claims

1. An object matching method, comprising:

acquiring user characteristics of a user object logged in a target website;

determining a first correlation degree between the user characteristics and approximate fan class clusters related to the class to which the user objects belong under each candidate anchor object; clustering all vermicelli objects under each anchor object in advance to obtain vermicelli clusters of different categories under each anchor object, wherein the approximate vermicelli clusters are vermicelli clusters with a category which is more than a preset association degree with the category to which the user object belongs;

determining a second correlation degree between the user characteristics and the clustering centers respectively represented by the candidate anchor objects;

determining to obtain a comprehensive correlation degree based on the first correlation degree and the second correlation degree;

and determining a target candidate anchor object matched with the user object according to the comprehensive relevance.

2. The method of claim 1, wherein the process of obtaining vermicelli class clusters of different classes under each of the anchor objects comprises:

acquiring all vermicelli objects under each anchor object;

determining a target cluster center number according to the reason category of the vermicelli which becomes the anchor object and the type number of the video resource content of the anchor object;

and clustering the vermicelli objects according to the target clustering center number to obtain vermicelli clusters with the number of the target clustering center number.

3. The method of claim 1, wherein the determining a first degree of correlation between the user characteristic and an approximate fan-class cluster under each candidate anchor object that is related to the class to which the user object belongs comprises:

calculating cosine similarity between the user characteristics and the approximate vermicelli clusters under each candidate anchor object, and determining the first relevance according to the calculated cosine similarity result;

correspondingly, the determining the second relatedness between the user characteristic and the clustering center respectively represented by each candidate anchor object includes:

and calculating cosine similarity between the user characteristics and clustering centers respectively represented by the candidate anchor objects, and determining the second relevance according to the calculated cosine similarity result.

4. The method of claim 1, wherein the user features comprise user feature graph vectors represented in graph vector form, and the first and second correlations each represent a degree of vector similarity between graph vectors.

5. The method of claim 1, wherein the determining the integrated correlation based on the first correlation and the second correlation comprises:

acquiring a first weight corresponding to the first correlation degree and a second weight corresponding to the second phase Guan Du; the first weight and the second weight are respectively obtained by determining the contribution degree of the corresponding correlation degree to the comprehensive correlation degree under the actual application scene;

6. The method of any of claims 1-5, wherein the determining a target candidate anchor object matching the user object based on the integrated relevance comprises:

Determining candidate anchor objects meeting preset requirements as the target candidate anchor objects according to the matching degree sequencing; wherein the preset requirements include: at least one of the first N bits of the matching degree sequence, the matching degree exceeding a preset value and the first M% of the matching degree sequence, wherein N and M are positive numbers.

7. The method of claim 6, further comprising:

recommending the target candidate anchor object to the user object.

8. The method of claim 7, wherein the recommending the target candidate anchor object to the user object comprises:

presenting the target candidate anchor object in a personalized recommendation popup to be recommended to the user object in a recommendation popup mode;

and/or

And increasing the probability that the video resource content of the target candidate anchor object appears in the resource recommendation list of the user object.

9. The method of claim 6, further comprising:

determining potential fan objects of the candidate anchor objects according to the comprehensive correlation degree between the user objects and the candidate anchor objects respectively;

recommending the potential vermicelli object to the corresponding candidate anchor object.

10. An object matching apparatus comprising:

a user feature acquisition unit configured to acquire a user feature of a user object logged in to a target website;

a first relevance determining unit configured to determine a first relevance between the user feature and an approximate fan cluster related to a category to which the user object belongs under each candidate anchor object; clustering all vermicelli objects under each anchor object in advance to obtain vermicelli clusters of different categories under each anchor object, wherein the approximate vermicelli clusters are vermicelli clusters with a category which is more than a preset association degree with the category to which the user object belongs;

a second correlation determination unit configured to determine a second correlation between the user feature and a cluster center represented by each of the candidate anchor objects, respectively;

an integrated correlation determination unit configured to determine an integrated correlation based on the first correlation and the second correlation;

and the matching anchor object determining unit is configured to determine a target candidate anchor object matched with the user object according to the comprehensive relevance.

11. The apparatus of claim 10, further comprising: a clustering operation unit configured to obtain vermicelli clusters of different categories under each of the anchor objects, the clustering operation unit being further configured to:

Acquiring all vermicelli objects under each anchor object;

12. The apparatus of claim 10, wherein the determining first correlation determination unit is further configured to:

correspondingly, the second relatedness-determining unit is further configured to:

13. The apparatus of claim 10, wherein the user features comprise user feature graph vectors represented in graph vector form, and the first and second correlations each represent a degree of vector similarity between graph vectors.

14. The apparatus of claim 10, wherein the integrated relevance determining unit is further configured to:

15. The apparatus according to any of claims 10-14, wherein the matching anchor object determination unit is further configured to:

16. The method of claim 15, further comprising:

17. The apparatus of claim 16, wherein the first recommendation unit is further configured to:

and/or

18. The apparatus of claim 16, further comprising:

a potential fan object determining unit configured to determine a potential fan object of each candidate anchor object according to a comprehensive correlation between each user object and each candidate anchor object, respectively;

19. An electronic device, comprising:

at least one processor; and

a memory communicatively coupled to the at least one processor; wherein,

the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the object matching method of any one of claims 1-9.

20. A non-transitory computer readable storage medium storing computer instructions for causing the computer to perform the object matching method of any one of claims 1-9.

21. A computer program product comprising a computer program which, when executed by a processor, implements the steps of the object matching method according to any of claims 1-9.