CN113486989B

CN113486989B - Object identification method, device, readable medium and equipment based on knowledge graph

Info

Publication number: CN113486989B
Application number: CN202110892035.0A
Authority: CN
Inventors: 刘江冬; 钟黎; 易坤
Original assignee: Beijing ByteDance Network Technology Co Ltd
Current assignee: Beijing ByteDance Network Technology Co Ltd
Priority date: 2021-08-04
Filing date: 2021-08-04
Publication date: 2024-04-09
Anticipated expiration: 2041-08-04
Also published as: CN113486989A

Abstract

The disclosure relates to a knowledge-graph-based object recognition method, a knowledge-graph-based object recognition device, a knowledge-graph-based object recognition readable medium and a knowledge-graph-based object recognition device, and relates to the technical field of electronic information processing, wherein the knowledge-graph-based object recognition method comprises the following steps: according to a pre-established game knowledge graph, determining a target game vector used for representing a target game and an object vector to be identified used for representing an object to be identified, determining the correlation degree between the object to be identified and the target game according to the object vector to be identified, the target game vector and a pre-trained identification model, wherein the identification model is obtained by training a seed object vector used for representing a seed object of the target game and the target game vector, the seed object vector is determined according to the game knowledge graph, and if the correlation degree between the object to be identified and the target game meets a preset condition, the object to be identified is determined to be the target object of the target game. According to the method and the device, the target object can be effectively identified without interaction behavior between the object to be identified and the target game, and the efficiency and the accuracy of object identification are improved.

Description

Object identification method, device, readable medium and equipment based on knowledge graph

Technical Field

The present disclosure relates to the field of electronic information processing technologies, and in particular, to a method, an apparatus, a readable medium, and a device for object identification based on a knowledge graph.

Background

With the continuous development of electronic information technology, various game applications appear in the application market. In the game operation process, the exposure of the game can be increased by putting multimedia content, so that the activity of the game is improved. In order to improve the accuracy of the delivery, a target delivery platform suitable for the game is generally selected from a plurality of delivery platforms, and the delivery is oriented on the target delivery platform. In the prior art, a method for identifying a target object and accurately putting the target object based on a knowledge graph exists, however, the method requires interaction between the object and a game, and for a newly-built game, the interaction between the object and the game does not exist, or little interaction exists, and the problem of cold start of the knowledge graph can occur, so that the target object cannot be identified.

Disclosure of Invention

This summary is provided to introduce a selection of concepts in a simplified form that are further described below in the detailed description. This summary is not intended to identify key features or essential features of the claimed subject matter, nor is it intended to be used to limit the scope of the claimed subject matter.

In a first aspect, the present disclosure provides a knowledge-graph-based object recognition method, the method comprising:

determining a target game vector for representing a target game and an object vector to be identified for representing an object to be identified according to a pre-established game knowledge graph;

determining the correlation degree of the object to be identified and the target game according to the object vector to be identified, the target game vector and a pre-trained identification model, wherein the identification model is obtained by training according to a seed object vector for representing the target game seed object and the target game vector, and the seed object vector is determined according to the game knowledge graph;

and if the correlation degree between the object to be identified and the target game meets a preset condition, determining that the object to be identified is the target object of the target game.

In a second aspect, the present disclosure provides an object recognition apparatus based on a knowledge-graph, the apparatus comprising:

the vector determining module is used for determining a target game vector for representing a target game and an object vector to be identified for representing an object to be identified according to a pre-established game knowledge graph;

The relevance determining module is used for determining the relevance of the object to be identified and the target game according to the object vector to be identified, the target game vector and a pre-trained identification model, wherein the identification model is obtained by training according to a seed object vector of a seed object for representing the target game and the target game vector, and the seed object vector is determined according to the game knowledge graph;

and the identification module is used for determining the object to be identified as the target object of the target game if the correlation degree between the object to be identified and the target game meets a preset condition.

In a third aspect, the present disclosure provides a computer readable medium having stored thereon a computer program which when executed by a processing device performs the steps of the method of the first aspect of the present disclosure.

In a fourth aspect, the present disclosure provides an electronic device comprising:

a storage device having a computer program stored thereon;

processing means for executing said computer program in said storage means to carry out the steps of the method of the first aspect of the disclosure.

Through the technical scheme, the method and the device for identifying the object vector and the target game vector determine the object vector to be identified according to the pre-established game knowledge graph. And then determining the correlation degree of the object to be identified and the target game according to the object vector to be identified, the target game vector and a pre-trained identification model, wherein the identification model is obtained by training according to a seed object vector of a seed object for representing the target game and the target game vector. And finally, under the condition that the correlation degree between the object to be identified and the target game meets the preset condition, determining that the object to be identified is the target object of the target game. The method and the device obtain the vector used for representing the object to be identified and the target game through the game knowledge graph, and then identify the vector by utilizing the identification model, so as to determine whether the object to be identified is the target object. The target object can be effectively identified without interaction behavior between the object to be identified and the target game, the problem of cold start of object identification is solved, and the efficiency and accuracy of object identification are improved.

Additional features and advantages of the present disclosure will be set forth in the detailed description which follows.

Drawings

The above and other features, advantages, and aspects of embodiments of the present disclosure will become more apparent by reference to the following detailed description when taken in conjunction with the accompanying drawings. The same or similar reference numbers will be used throughout the drawings to refer to the same or like elements. It should be understood that the figures are schematic and that elements and components are not necessarily drawn to scale. In the drawings:

FIG. 1 is a flowchart illustrating a knowledge-graph based object recognition method, in accordance with an exemplary embodiment;

FIG. 2 is a schematic diagram of a game knowledge-graph, according to an example embodiment;

FIG. 3 is a flowchart illustrating a training recognition model, according to an exemplary embodiment;

FIG. 4 is a flowchart illustrating another training recognition model, according to an example embodiment;

FIG. 5 is a schematic diagram of an identification model, according to an example embodiment;

FIG. 6 is a flowchart illustrating one method of establishing a game knowledge-graph, in accordance with an exemplary embodiment;

FIG. 7 is a flowchart illustrating another method of establishing a game knowledge-graph, in accordance with an illustrative embodiment;

FIG. 8 is a flowchart illustrating another knowledge-graph based object recognition method, in accordance with an illustrative embodiment;

FIG. 9 is a schematic diagram of an object game atlas, according to an example embodiment;

FIG. 10 is a block diagram of an object recognition device based on a knowledge-graph, shown in accordance with an exemplary embodiment;

FIG. 11 is a block diagram of another knowledge-graph based object recognition apparatus, shown in accordance with an exemplary embodiment;

FIG. 12 is a block diagram of another knowledge-graph based object recognition apparatus, shown in accordance with an exemplary embodiment;

fig. 13 is a block diagram of an electronic device, according to an example embodiment.

Detailed Description

Embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. While certain embodiments of the present disclosure have been shown in the accompanying drawings, it is to be understood that the present disclosure may be embodied in various forms and should not be construed as limited to the embodiments set forth herein, but are provided to provide a more thorough and complete understanding of the present disclosure. It should be understood that the drawings and embodiments of the present disclosure are for illustration purposes only and are not intended to limit the scope of the present disclosure.

It should be understood that the various steps recited in the method embodiments of the present disclosure may be performed in a different order and/or performed in parallel. Furthermore, method embodiments may include additional steps and/or omit performing the illustrated steps. The scope of the present disclosure is not limited in this respect.

The term "including" and variations thereof as used herein are intended to be open-ended, i.e., including, but not limited to. The term "based on" is based at least in part on. The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments. Related definitions of other terms will be given in the description below.

It should be noted that the terms "first," "second," and the like in this disclosure are merely used to distinguish between different devices, modules, or units and are not used to define an order or interdependence of functions performed by the devices, modules, or units.

It should be noted that references to "one", "a plurality" and "a plurality" in this disclosure are intended to be illustrative rather than limiting, and those of ordinary skill in the art will appreciate that "one or more" is intended to be understood as "one or more" unless the context clearly indicates otherwise.

The names of messages or information interacted between the various devices in the embodiments of the present disclosure are for illustrative purposes only and are not intended to limit the scope of such messages or information.

Fig. 1 is a flowchart illustrating a knowledge-graph-based object recognition method, as shown in fig. 1, according to an exemplary embodiment, the method including the steps of:

step 101, determining a target game vector for representing a target game and an object vector to be identified for representing an object to be identified according to a pre-established game knowledge graph.

The game knowledge graph comprises a plurality of nodes and at least one edge, wherein the nodes comprise: object nodes, game nodes, and content nodes, each edge being used to characterize an association between two nodes at both ends of the edge.

For example, to identify a target object of a target game from a large number of objects, a game knowledge graph may be first established. It is understood that a game ecology includes a plurality of objects, games and contents, and a game knowledge graph can characterize the game ecology and reflect the association among the plurality of objects, games and contents. The game knowledge graph can comprise a plurality of nodes, and the nodes are divided into three types: an object node for representing an object, a game node for representing a game, and a content node for representing content. Wherein an object may be understood as a delivery platform for delivering multimedia content, through which a user may download a game, enter a game, etc. The delivery platform may be, for example, an application program, a group of application programs corresponding to the same server, a page in the application programs, or the like. An object may also be understood as a drop zone where a user may download a game, enter a game, etc. The delivery area may be, for example, an area covered by a local area network, an area covered by a base station, an area served by an operator, or the like. An object may also be understood as a terminal device through which a user may download a game, enter a game, etc. An object may also be understood as a user, and the specific meaning of the object is not specifically limited by the present disclosure. Content is understood to be any content presented by an object, and may be, for example, any of text (e.g., news, novels, blogs, etc.), audio (e.g., music, broadcast, audio books, etc.), video (e.g., movies, television shows, short videos, etc.). The game knowledge graph also comprises at least one edge, if any two nodes have an association, an edge exists between the two nodes, namely each edge can represent the association between the two nodes at the two ends of the edge, and further, the width or the numerical value of each edge can also represent the attribute of the association between the two nodes at the two ends of the edge. For example, there is an edge between the content node representing the a content and the game node representing the a game, which edge is used to characterize the a game mentioned in the a content. Further, if the content of a is a blog, in which 70% of the text is describing game a, the value of the edge may be 0.7. In one implementation, each node may further include an attribute of the node (e.g., the object node includes a representation of the corresponding object), and in another implementation, the attribute of each node may be used as an attribute node, and an edge may be established between the attribute node and the corresponding node. Taking the game knowledge graph shown in fig. 2 as an example, the game knowledge graph includes: the node A representing the object A, the node B representing the object B, the node A representing the game A, the node B representing the game B, the node a representing the content a, the node B representing the content B, the node A1 representing the image of the object A, the node A1 representing the label of the game A, and the node A1 representing the label of the content a are included. The one-way arrow is between the node A1 and the node A, which indicates that the node A1 is an attribute node of the node A, the two-way arrow is between the node B and the node A, which indicates that the node A and the node B have an association (for example, in a scene that the object is an application program, the object A and the object B can use the same server). A one-way arrow is between node a and node A1, indicating that node A1 is an attribute node of node a, and a two-way arrow is between node B and node a, indicating that node a and node B have similar associations (e.g., game a and game B belong to the same game developer). A one-way arrow is between node a and node a1, indicating that node a1 is an attribute node of node a, and a two-way arrow is between node b and node a, indicating that node a and node b have similar associations (e.g., content a and content b describe the same game). It should be noted that, in the scenario where the object is a user, the information required for establishing the game knowledge graph is obtained under the condition that the user is authorized, or is actively submitted after the user reads the related description, or the terminal device is necessarily sent to the server when the user uses the terminal device.

To identify the object to be identified, an object vector to be identified for characterizing the object to be identified and a target game vector for characterizing the target game may be determined by the game knowledge graph. The object to be identified can be an object represented by any object node in the game knowledge graph. In one implementation, a vector corresponding to each node in the game knowledge graph may be determined by using a graph neural network (english: graph Neural Networks, abbreviated: GNN), a graph roll neural network (english: graph Convolutional Network, abbreviated: GCN), or graph sage (english: graph SAmple and aggreGatE), etc., so as to obtain an object vector to be identified corresponding to an object node representing an object to be identified, and a target game vector corresponding to a game node representing a target game. In another implementation manner, a plurality of sub-graphs capable of reflecting the association among the game nodes in a plurality of dimensions of objects, contents, games and the like can be established according to the game knowledge graph, and a vector corresponding to each game node is determined by using a preset graph representation algorithm and the plurality of sub-graphs. And establishing a plurality of sub-graphs capable of reflecting the association of a plurality of dimensions such as games, contents and the like between each object node and the game node according to the game knowledge graph, and finally determining a vector corresponding to each object node by combining a vector corresponding to each game node and the plurality of sub-graphs, thereby obtaining an object vector to be identified corresponding to the object node representing the object to be identified and a target game vector corresponding to the game node representing the target game.

Step 102, determining the correlation degree between the object to be identified and the target game according to the object vector to be identified, the target game vector and a pre-trained identification model, wherein the identification model is obtained by training according to a seed object vector for representing the target game seed object and the target game vector, and the seed object vector is determined according to a game knowledge graph.

Step 103, if the correlation degree between the object to be identified and the target game meets the preset condition, determining that the object to be identified is the target object of the target game.

For example, a seed object vector for characterizing a seed object of a target game may be determined according to a game knowledge graph, and then an identification model may be trained in advance using the seed object vector and the target game vector, and the identification model may be understood as a model capable of identifying the degree of correlation of any one object with the target game. The recognition model may be a classification model in a Lookalike manner, for example, may be: random Forest (English: random Forest, abbreviated: RF), adaboost model, xgboost model, etc., and the recognition model may also be convolutional neural network (English: convolutional Neural Networks, abbreviated: CNN), which is not specifically limited in this disclosure. The seed objects are a plurality of, and can be understood as a delivery platform for delivering multimedia content of the target game, a delivery platform for providing an access interface of the target game, a terminal device provided with the target game, or a user downloading and using the target game, and the like. For example, before the target game is put on shelf, some users can try to play the target game by using internal measurement and other modes, and the users are seed objects. For another example, the multimedia content of the target game may be launched on a portion of the launching platforms, which may be used as seed objects, before the target game is launched. The recognition model may be trained with the seed object vector as a positive sample and the object vector of the non-seed object as a negative sample. After training to obtain the recognition model, the object vector to be recognized and the target game vector may be input into the recognition model to determine the correlation degree of the object to be recognized and the target game according to the output of the recognition model. It is understood that the recognition model can output a correlation score of the object to be recognized and the target game as a correlation of the object to be recognized and the target game.

Finally, if the correlation degree between the object to be identified and the target game meets the preset condition, the object to be identified can be determined to be the target object of the target game. The preset condition may be that the correlation degree between the object to be identified and the target game is greater than or equal to a preset threshold value, and the object to be identified may be determined to be the target object. The preset condition may be a maximum specified number (for example, 1000) of correlations, the correlations between the plurality of objects to be identified and the target game may be determined first, and then the specified number of objects to be identified with the maximum correlations may be used as the target object. After the target object is determined, the multimedia content of the target game may be targeted for the target object.

Because the game knowledge graph characterizes the whole game ecology, the association among the objects, games and contents can be reflected, the interaction behavior between the object to be identified and the target game is not needed, and meanwhile, the vector capable of representing the object to be identified and the target game is utilized, so that the target object is effectively identified, the problem of cold start of object identification is solved, and the efficiency of object identification is improved. Meanwhile, as the game knowledge graph can reflect the association among the object, the game and the content, the object vector to be identified and the target game vector represent the object to be identified and the target game from three dimensions of the object, the game and the content, and for the identification model, more dimensional information is obtained for identification, so that the correlation degree between the object to be identified and the target game can be identified more accurately, and the accuracy of object identification is improved.

In summary, the present disclosure first determines an object vector to be identified and a target game vector according to a game knowledge graph established in advance. And then determining the correlation degree of the object to be identified and the target game according to the object vector to be identified, the target game vector and a pre-trained identification model, wherein the identification model is obtained by training according to a seed object vector of a seed object for representing the target game and the target game vector. And finally, under the condition that the correlation degree between the object to be identified and the target game meets the preset condition, determining that the object to be identified is the target object of the target game. The method and the device obtain the vector used for representing the object to be identified and the target game through the game knowledge graph, and then identify the vector by utilizing the identification model, so as to determine whether the object to be identified is the target object. The target object can be effectively identified without interaction behavior between the object to be identified and the target game, the problem of cold start of object identification is solved, and the efficiency and accuracy of object identification are improved.

In one implementation, the implementation of step 102 may be:

and inputting the object vector to be identified, the target game vector and the object characteristics of the object to be identified into an identification model to obtain the correlation degree between the object to be identified and the target game output by the identification model, wherein the object characteristics are determined according to the object information of the object to be identified.

For example, to identify an object to be identified, object characteristics may also be determined according to object information of the object to be identified. In the scenario that the object is a user, the object information is obtained under the condition that the user is authorized, or is actively submitted after the user reads the related description, or the terminal equipment is necessarily sent to the server when the user uses the terminal equipment. Further, the data related to the personal attribute in the object information is subjected to desensitization processing, for example, a certain type of data can be partially hidden, or a certain type of data can be segmented and divided. Object features may be understood as statistical indicators or trends that reflect the activity of the object to be identified, and may include one or more features. Object features may include a variety of features of an object at different stages. For example, in a scenario where the object is a user, a time period in which the object to be identified obtained by statistics on the basis of the object information browses the content related to the target game may be used as the object feature, a frequency in which the object to be identified obtained by statistics on the basis of the object information browses the content related to the target game may be used as the object feature, and a manner (for example, an open screen manner, an information flow) and a channel (for example, different APP) in which the object to be identified browses the content related to the target game may be used as the object feature. In a scenario where the object is a delivery platform, a duration in which the object to be identified displays content related to the target game, which is obtained by statistics of object information, may be used as an object feature, and a frequency in which the object to be identified displays content related to the target game, which is obtained by statistics of object information, may be used as an object feature. The present disclosure is not particularly limited as to the kind of specific object features included, and the manner in which the object features are obtained. After the object features are obtained, the object features, the object vector to be identified and the target game vector can be spliced and input into the identification model together, so that the correlation between the object to be identified and the target game output by the identification model is obtained. For the recognition model, besides the object vector to be recognized and the target game vector, the object characteristics are combined, so that the correlation degree between the object to be recognized and the target game can be recognized more accurately, and the accuracy of object recognition is further improved.

FIG. 3 is a flowchart illustrating a training recognition model according to an exemplary embodiment, as shown in FIG. 3, including object nodes corresponding to a plurality of sample objects, the sample objects including positive sample objects and negative sample objects, the positive sample objects including seed objects. The method further comprises the steps of:

step 104, training the recognition model for the target game.

Specifically, the implementation manner of step 104 may be:

step 1041, obtaining a sample input set, where the sample input set includes: sample inputs corresponding to each sample object, the sample inputs comprising: and the object vector and the target game vector are determined according to the game knowledge graph and used for representing the sample object.

In step 1042, a sample output set is obtained, the sample output set comprising sample outputs corresponding to each sample input, each sample output comprising a true recognition result of the corresponding sample object.

Step 1043, taking the sample input set as the input of the recognition model, and taking the sample output set as the output of the recognition model, so as to train the recognition model.

For example, in training the recognition model in the above embodiment, it is necessary to acquire a sample input set and a sample output set first. The sample input set comprises sample inputs corresponding to each sample object in the plurality of sample objects, and the sample inputs corresponding to the sample objects are an object vector and a target game vector of the sample object. It should be noted that, the plurality of sample objects includes a plurality of positive sample objects and a plurality of negative sample objects, and the positive sample objects may include seed objects, and further, a ratio of the number of positive sample objects to the number of negative sample objects may be controlled (for example, may be 1:1). The sample output set includes a sample output corresponding to each sample input, each sample output including a true recognition result of the corresponding sample object. The true recognition result of the positive sample object is correct (can be represented as 1), that is, the positive sample object is the target object of the target game, and the true recognition result of the negative sample object is incorrect (can be represented as 0), that is, the negative sample object is not the target object of the target game. The sample input set may then be used as input to the recognition model, and the sample output set may be trained as output to the recognition model, such that when the sample input set is input, the output of the recognition model can be matched to the sample output set. For example, the difference (or mean square error) between the output of the recognition model and the sample output set may be used as a loss function of the recognition model, and the parameters of the neurons in the recognition model, such as weights (english: weight) and offsets (english: bias), may be corrected by using a back propagation algorithm with the aim of reducing the loss function. Repeating the steps until the loss function meets the preset condition, for example, the loss function is smaller than the preset loss threshold value.

In one application scenario, the positive sample object may also include an extension object, which is determined by:

and 1) determining the correlation degree of other games and target games according to other game vectors and target game vectors used for representing the other games, wherein the other games are games except the target games, and the other game vectors are determined according to the game knowledge graph.

And 2) taking other games with the correlation degree larger than or equal to a preset correlation degree threshold value as the correlation games corresponding to the target games.

Step 3) active objects of the related game are used as expansion objects.

For example, in general, a seed object needs to be acquired by means of internal measurement, and when the number of seed objects is too small or the seed objects cannot be acquired, a cold start problem of the identification model may occur. Therefore, the relevance between other games and the target game can be calculated by means of each game vector determined in the game knowledge graph, and then the other games with the relevance greater than or equal to the preset relevance threshold value are taken as relevant games corresponding to the target game, and one or more relevant games can be adopted. The correlation degree between the other games and the target game may be, for example, a cosine similarity between a game vector corresponding to the other games and a game vector of the target game, or a Jaccard similarity between a game vector corresponding to the other games and a game vector of the target game, which is not particularly limited in this disclosure. Finally, the active object of the related game can be used as an extension object, and the extension object is used as a positive sample object for training the identification model, so that the problem of cold start of the identification model caused by too few seed objects can be solved. Specifically, the active object of the related game may be determined directly according to the history data acquired by the related game, or an object meeting a specified condition (for example, continuously going online for N days or using time exceeds M hours) may be determined as the active object, or the active object may be screened out from the game knowledge graph, where an edge is provided between an object node corresponding to the active object and a game node corresponding to the related game, and a numerical value of the edge meets the specified condition.

FIG. 4 is a flowchart illustrating another training recognition model, as shown in FIG. 4, in accordance with an exemplary embodiment, an implementation of step 1043 may include:

and step A, clustering the sample input sets according to a preset clustering algorithm to obtain a plurality of groups of sample input subsets, wherein each group of sample input subsets corresponds to one group of sample output subsets.

And B, inputting the set of sample input subsets into the recognition sub-models corresponding to the set of sample input subsets, which are included in the recognition model, aiming at each set of sample input subsets, and taking the sample output subsets corresponding to the set of sample input subsets as the output of the recognition sub-models corresponding to the set of sample input subsets so as to train the recognition sub-models corresponding to the set of sample input subsets.

The output of the recognition model is determined according to the output of the recognition sub-model corresponding to each group of sample input subsets.

For example, in the case that the sample object covers multiple object groups, the sample input sets may be clustered according to a preset clustering algorithm to obtain multiple groups of sample input subsets, where each group of sample input subsets includes a portion of sample inputs in the sample input set, which may be understood as the portion of sample inputs belong to the same object group. Accordingly, each set of sample input subsets corresponds to a set of sample output subsets, and the sample output subsets comprise sample outputs corresponding to the sample inputs. The clustering algorithm may be, for example, DBSCAN (english: density-Based Spatial Clustering of Applications with Noise, chinese: density-based clustering method with noise), which is not specifically limited in this disclosure.

Further, the recognition model may include recognition sub-models corresponding to each group of sample input subsets, that is, the recognition sub-models correspond to the sample input subsets one by one, and the structure of the recognition model may be as shown in fig. 5. For each set of sample input subsets, the set of sample input subsets may be input to a corresponding recognition sub-model, and the sample output subset corresponding to the set of sample input subsets may be output as a corresponding recognition sub-model to train the corresponding recognition sub-model such that when the set of sample input subsets is input, the output of the corresponding recognition sub-model may be matched with the sample output subset corresponding to the set of sample input subsets. For example, the difference (or mean square error) between the output of the corresponding recognition sub-model and the corresponding subset of sample outputs may be used as a loss function of the corresponding recognition sub-model, and the parameter of the neuron in the corresponding recognition sub-model may be corrected by using a back propagation algorithm with the goal of reducing the loss function. Repeating the steps until the loss function meets the preset condition, for example, the loss function is smaller than the preset loss threshold value.

Finally, the output of the recognition model may be determined based on the output of the recognition sub-model corresponding to each set of sample input subsets, i.e. the recognition model may be understood as a voting system comprising a plurality of recognition sub-models. For example, the average value of the outputs of the recognition sub-models corresponding to the sample input subsets of each group may be used as the output of the recognition model, the maximum value of the outputs of the recognition sub-models corresponding to the sample input subsets of each group may be used as the output of the recognition model, and the outputs of the recognition sub-models corresponding to the sample input subsets of each group may be weighted and summed to be used as the output of the recognition model. The present disclosure is not particularly limited thereto. The sample object covers various object groups, and the correspondingly trained identification model has strong generalization capability, so that the correlation degree between the objects in the various object groups and the target game can be accurately identified.

In one application scenario, each set of sample inputs a recognition sub-model corresponding to a subset, which is a tree integration model, comprising a plurality of tree models. The implementation of step B may include:

first, for each tree model, the set of sample input subsets is randomly sampled to obtain a sampled input subset, the sampled input subset comprising a smaller number of sample inputs than the set of sample input subsets.

Then, each sample input included in the sample input subset is randomly sampled to obtain a sample input corresponding to the sample input, and the sample input belongs to the sample input.

Finally, taking the sample input corresponding to each sample input included in the sample input subset as the input of the tree model, and taking the sample output corresponding to each sample input included in the sample input subset as the output of the tree model to train the tree model.

Wherein the output of the tree integration model is determined from the output of each tree model.

For example, each of the above-described recognition sub-models may be a tree integration model (alternatively referred to as an integrated tree model), such as: may be a random forest, adaboost model, xgboost model, etc., in which a plurality of tree models are included, which may be, for example, classification trees. The process of training for each recognition sub-model may be a process of jointly training a plurality of tree models included in the recognition sub-model.

For each tree model, the set of sample input subsets may be randomly sampled to obtain a sample input subset, where the sample input subset includes a smaller number of sample inputs than the set of sample input subsets, i.e., the sample input subset includes a portion of the sample inputs in the set of sample input subsets. For example, the set of sample input subsets includes 100 sample inputs, random sampling is performed, and the resulting sample input subset may include 75 sample inputs. And then randomly sampling each sample input included in the sample input subset to obtain a sample input corresponding to the sample input, wherein the sample input belongs to the sample input, namely the dimension of the sample input is smaller than that of the sample input. For example, a subset of sample inputs includes 75 sample inputs, each sample input being a 256-dimensional vector. Each sample input is randomly sampled, and 75 sample inputs corresponding to the 75 sample inputs are obtained, and each sample input can be a 200-dimensional vector. And taking the sample input corresponding to each sample input included in the sample input subset as the input of the tree model, and taking the sample output corresponding to each sample input included in the sample input subset as the output of the tree model so as to train the tree model, so that when the sample input is input, the output of the corresponding tree model can be matched with the corresponding sample output. For example, the difference (or mean square error) between the output of the corresponding tree model and the output of the corresponding sample may be used as a loss function of the corresponding tree model, and the parameter of the neuron in the corresponding tree model may be corrected by using a back propagation algorithm with the goal of reducing the loss function. Repeating the steps until the loss function meets the preset condition, for example, the loss function is smaller than the preset loss threshold value.

Finally, the output of the tree-integrated model may be determined from the output of each tree model, i.e. the tree-integrated model may be understood as a voting system comprising a plurality of tree models. For example, the average value of the outputs of each tree model may be used as the output of the tree integrated model, the maximum value of the outputs of each tree model may be used as the output of the tree integrated model, and the outputs of each tree model may be weighted and summed to be used as the output of the tree integrated model. The present disclosure is not particularly limited thereto. Because each group of sample input subsets are randomly sampled and each sample input is randomly sampled, the generalization capability of the tree integration model can be improved in two dimensions of sample number and sample characterization, so that the recognition model can accurately recognize the correlation degree of an object and a target game.

FIG. 6 is a flow chart illustrating a method of establishing a game knowledge-graph, as shown in FIG. 6, according to an exemplary embodiment, by:

and step C, acquiring object information of a plurality of objects, game information of a plurality of games and content information of a plurality of contents. The content is any one of text, audio and video, the objects comprise objects to be identified and seed objects, and the games comprise target games.

For example, a game knowledge graph may be established prior to identifying the object to be identified. Specifically, object information of a plurality of objects, game information of a plurality of games, and content information of a plurality of contents may be acquired first. Wherein the number of objects, the number of games, and the number of contents are not related to each other. It is understood that the game ecology includes a plurality of objects, games and contents, where the plurality of objects includes the object to be identified and the seed object, and may also include other objects. The plurality of games includes the target game described above, and may also include other games (such as the related games mentioned above).

Specifically, the object information may include, for example, an object image, an object identifier, an object group type, and the like of the corresponding object. It should be noted that, in the scenario that the object is a user, the object information is obtained under the condition that the corresponding user is authorized, or the corresponding user is actively submitted after reading the related description, or the terminal device is necessarily sent to the server when the user uses the terminal device. Further, the data related to the personal attribute in the object information is subjected to desensitization processing, for example, a certain type of data can be partially hidden, or a certain type of data can be segmented and divided. The game information may include game tags (e.g., poker, changing, playing, basketball, round playing, role playing, in-time strategy, etc.), game categories (e.g., cards, games, combat, sports, etc.), game developers, etc. for the corresponding game. The content information may include content tags, content categories, content providers, etc. of the corresponding content, wherein the content tags and content categories may be shared with the game tags and game categories, i.e., the corresponding content tags and content categories may be determined from the games involved in the content. For example, if game a is referred to in content a, then the content tag and content category of content a may be the same as the game a game tag and game category.

After obtaining the object information, game information, and content information, the information may be subjected to data cleansing and normalization. Specifically, the process of data cleaning may be deleting information with more data loss, or may be completed by interpolation. The present disclosure is not particularly limited thereto.

And D, establishing an object node corresponding to each object, a game node corresponding to each game and a content node corresponding to each content.

And E, establishing edges among the plurality of nodes according to a preset association rule.

For example, corresponding nodes can be respectively built for each object, each game and each content in the game knowledge graph, namely, the game knowledge graph at least comprises three types of nodes: an object node for representing an object, a game node for representing a game, and a content node for representing content. Further, the game knowledge graph may further include an object attribute node for representing object information, a game attribute node for representing game information, and a content attribute node for representing content information. Then, according to a preset association rule, an edge can be established between two nodes with association, so that a game knowledge graph is obtained.

FIG. 7 is a flowchart illustrating another method of establishing a game knowledge-graph, as shown in FIG. 7, in accordance with an exemplary embodiment, the implementation of step E may include:

step E1, determining a game association object meeting association rules with each object according to the object information of each object, wherein the game association object comprises: at least one of an object, a game and a content, and establishing edges between nodes corresponding to the object and nodes corresponding to the game related object at the object nodes corresponding to the object.

And E2, determining a content association object meeting association rules with the content according to the content information of each content, wherein the content association object comprises: and establishing edges between the nodes corresponding to the content and the nodes corresponding to the content association objects.

And E3, determining an associated game meeting the association rule according to the game information of each game, and establishing edges between game nodes corresponding to the game and game nodes corresponding to the associated game.

The specific way of establishing edges in the game knowledge graph may first determine association rules. The association rules may include associations of multiple dimensions: first association between objects: such as an association that the object has with the object, a similarity of the object to the object information of the object exceeding a first threshold (e.g., 75%), etc. Second association between object and game: such as object download, use, consumed games. Third association between object and content: such as object browsing, commenting, praying, sharing content. Fourth association of content with content: for example, the similarity of the content information of the content to the content exceeds a second threshold (e.g., 60%), and the content information of the content to the content includes the same content tag (or content category, content provider). Fifth association between content and game: the similarity of the content information, e.g., content, to the game information of the game exceeds a third threshold (e.g., 50%). Sixth association of game with game: for example, the similarity of the game to the game information of the game exceeds a fourth threshold (e.g., 80%), and the same game tag (or game category, game developer) is included in the game to the game information of the game.

Then, according to the association rules, an object, game, content (i.e., game association object) satisfying the association rules (e.g., first association, second association, third association) with each object can be determined, and a corresponding edge can be established. Games, content (i.e., content association objects) that satisfy association rules (e.g., fourth association, fifth association) with each content are determined and corresponding edges are established. An association game satisfying the association rule (i.e., sixth association) with each game is determined, and a corresponding edge is established. In this way, the edges in the game knowledge graph can describe the association between objects, games, and content from multiple dimensions.

It should be noted that the execution sequence in the above embodiment is only for illustration, and the execution sequence among the steps E1, E2, and E3 may be set according to specific requirements, which is not specifically limited in this disclosure.

FIG. 8 is a flowchart illustrating another knowledge-graph-based object recognition method, according to an exemplary embodiment, as shown in FIG. 8, the implementation of step 101 may include:

in step 1011, a game vector for characterizing each game is determined based on the game knowledge-graph.

For example, a game vector for characterizing each game may be determined from a game knowledge-graph. Specifically, the method can be realized by the following steps:

And 4) determining an object game map according to the edges between the object nodes in the game knowledge map and the game nodes, determining an object content map according to the edges between the object nodes and the content nodes in the game knowledge map, and determining a game map according to the edges between the game nodes and the game nodes in the game knowledge map.

And 5) determining each game vector according to a preset graph representation algorithm, an object game graph, an object content graph and a game graph.

The object game atlas is used for representing the association of any two game nodes in the object dimension, the object content atlas is used for representing the association of any two game nodes in the content dimension, and the game atlas is used for representing the association of any two game nodes in the game dimension.

For example, an object game map for representing association of any two game nodes in an object dimension, an object content map for representing association of any two game nodes in a content dimension, and a game map for representing association of any two game nodes in a game dimension may be extracted from game knowledge maps. It may be understood that, the object game map, the object content map and the game map each include a game node corresponding to each of the plurality of games, and if any two games have a relationship in the object dimension, an edge may be established between two game nodes corresponding to the two games in the object game map. Likewise, if there is an association between the two games in the content (or game) dimension, then an edge may be established between the two game nodes corresponding to the two games in the object content graph (or game graph).

Specifically, the game A and the game B are related in object dimension, and it is understood that the contact ratio of objects related to the game A and the game B is greater than a preset fifth threshold (for example: 30%), namely, the number of object nodes with edges between game nodes corresponding to the game A and between game nodes corresponding to the game B in the game knowledge graph is more than 30% of the number of objects. An edge can be established between the game node corresponding to the game a and the game node corresponding to the game B in the object game map, and further, the overlap ratio of the object can be used as the width of the edge or the value of the edge.

The association between the game a and the game B in the content dimension can be understood that the coincidence degree of the objects associated with the content a and the content B is greater than a preset sixth threshold (for example, 20%), wherein the game a is related to the game a, and the game B is related to the game B in the content B, that is, the number of object nodes with edges between the content nodes corresponding to the content a in the game knowledge graph at the same time, and the number of object nodes with edges between the content nodes corresponding to the content B is more than 20% of the number of objects. An edge can be established between the content node corresponding to the content a and the content node corresponding to the content b in the object content map, and further, the coincidence degree of the object can be used as the width of the edge or the numerical value of the edge.

The game A and the game B are related in the game dimension, and the similarity of the game A and the game B is understood to exceed a seventh threshold (for example, 50%), namely, the side between the game node corresponding to the game A and the game node corresponding to the game B in the game knowledge graph represents that the similarity of the game A and the game B exceeds 50%. Then an edge may be established between the game node corresponding to game a and the game node corresponding to game B in the game map, and further, the similarity of the objects may also be used as the width of the edge, or the value of the edge.

After the object game map, the object content map, and the game map are obtained, each game vector may be determined according to a preset graph representation algorithm. The graph representation algorithm can represent the graph in a manner of an adjacency list or an adjacency linked list, so that a vector for representing each node in the graph is obtained. Specifically, the object game map may be input to a map representation algorithm to obtain a first game vector corresponding to each game, the object content map may be input to a map representation algorithm to obtain a second game vector corresponding to each game, and the game map may be input to a map representation algorithm to obtain a third game vector corresponding to each game. Finally, the first game vector, the second game vector, and the third game vector corresponding to each game may be averaged as the game vector corresponding to the game. The first game vector, the second game vector, and the third game vector corresponding to each game may be spliced to be the game vector corresponding to the game. The maximum value of the first game vector, the second game vector, and the third game vector corresponding to each game may also be used as the game vector corresponding to the game. The manner in which the game vector is specifically determined is not particularly limited by the present disclosure. In this way, the game vector corresponding to each game can characterize the game from multiple dimensions, including a greater amount of information than if the game was described with only game information.

Step 1012, determining an object vector for characterizing each object based on the game knowledge-graph and each game vector.

For example, after each game vector is derived, the game knowledge-graph may be further combined to determine an object vector that characterizes each object. Specifically, the method can be realized by the following steps:

in one application scenario, step 1012 may be implemented by:

and 6) determining an object game map according to the edges between the object nodes in the game knowledge map and the game nodes, and determining an object content map according to the edges between the object nodes and the content nodes in the game knowledge map.

Step 7) determining each object vector according to each game vector, the object game map and the object content map.

The object game atlas is used for representing the association of each object node and each game node in the game dimension, and the object content atlas is used for representing the association of each object node and each game node in the content dimension.

For example, an object game graph for characterizing the association of each object node with each game node in the game dimension and an object content graph for characterizing the association of each object node with each game node in the content dimension may be extracted from the game knowledge graph. It may be understood that, in the object game map and the object content map, each object node corresponding to each object and each game node corresponding to each game are included, if a certain object is associated with a certain game in a game dimension, an edge may be established between the object node corresponding to the object and the game node corresponding to the game in the object game map. Similarly, if the game is associated with the game in the content dimension, an edge may be established between the object node corresponding to the object and the game node corresponding to the game in the object content map. The object game pattern in step 1011 represents a different pattern from the object game pattern in step 1012, and similarly, the object content pattern in step 1011 represents a different pattern from the object content pattern in step 1012.

Specifically, the association between the object a and the game a in the game dimension may be understood as that the object a downloads, uses, or consumes the game a, that is, there is an edge between the object node corresponding to the object a and the game node corresponding to the game a in the game knowledge graph. An edge may be established between the object node corresponding to the object a and the game node corresponding to the game a in the object game map, and further, any measure (e.g., use duration, use frequency, etc.) of the object a may be downloaded, used, or consumed by the game a as the width of the edge, or the value of the edge.

The association between the object A and the game A in the content dimension can be understood as that the object A browses, reviews, praise or shares the content a, wherein the content a relates to the game A, namely an edge exists between an object node corresponding to the object A and a content node corresponding to the content a in the game knowledge graph. An edge may be established between the object node corresponding to the object a and the game node corresponding to the game a in the object content map, and further, any measure (such as forwarding times, number of praise, etc.) of browsing, commenting, praise, or sharing the content a by the object a may be used as the width of the edge, or the value of the edge.

After the object game atlas and the object content atlas are obtained, each object vector may be determined from the object game atlas, the object content atlas, and each game vector previously determined. Specifically, a first object vector corresponding to each object may be determined according to the object game map, while a second object vector corresponding to each object may be determined according to the object content map. Taking the object game map shown in fig. 9 as an example, if there is an edge between the node a corresponding to the object a and the node B corresponding to the game B in the object game map, and there is an edge between the node B corresponding to the object B and only the node B, then the game vector corresponding to the game a and the game vector corresponding to the game B may be summed to form the first object vector corresponding to the object a, and the game vector corresponding to the game B may be the first object vector corresponding to the object B. Further, if the value of the edge existing between the node a and the node a is 0.8 and the value of the edge existing between the node a and the node B is 0.2, the game vector corresponding to the game a and the game vector corresponding to the game B may be weighted and summed according to weights of 0.8 and 0.2 to be the first object vector corresponding to the object a.

The manner of determining the second object vector from the object content map and each game vector is the same as the manner of determining the first object vector from the object game map and each game vector, and will not be described again here. Finally, the first object vector and the second object vector corresponding to each object may be averaged to be the object vector corresponding to the object. The first object vector and the second object vector corresponding to each object may be spliced to form the object vector corresponding to the object. The maximum value of the first object vector and the second object vector corresponding to each object can also be used as the object vector corresponding to the object. The manner in which the object vector is specifically determined is not specifically limited by the present disclosure. In this way, the object vector corresponding to each object can be characterized from multiple dimensions by using the game vector, reflecting the association of each object with each game in different dimensions, and containing more information than the way of describing the object by only object information.

In step 1013, an object vector for representing the object to be identified is taken as an object vector to be identified, and a game vector for representing the target game is taken as a target game vector.

For example, after each game vector and each object vector are obtained according to steps 1011 to 1012, an object vector for representing an object to be identified may be taken as the object vector to be identified, and a game vector for representing a target game may be taken as the target game vector. Further, an object vector for characterizing the seed object may also be used as the seed object vector.

Fig. 10 is a block diagram of an object recognition apparatus based on a knowledge-graph, according to an exemplary embodiment, and as shown in fig. 10, the apparatus 200 includes:

the vector determining module 201 is configured to determine a target game vector for characterizing a target game and an object vector to be identified for characterizing an object to be identified according to a pre-established game knowledge graph.

The relevance determining module 202 is configured to determine a relevance between the object to be identified and the target game according to the object vector to be identified, the target game vector, and a pre-trained identification model, where the identification model is obtained by training a seed object vector representing a seed object of the target game and the target game vector, and the seed object vector is determined according to a game knowledge graph.

The identifying module 203 is configured to determine that the object to be identified is a target object of the target game if the correlation between the object to be identified and the target game meets a preset condition.

In one application scenario, the relevance determining module 202 may be configured to:

In another application scenario, a game knowledge graph includes a plurality of nodes and at least one edge, the plurality of nodes including: object nodes, game nodes, and content nodes. Each edge is used to characterize an association between two nodes at both ends of the edge.

Fig. 11 is a block diagram of another knowledge-graph-based object recognition apparatus according to an exemplary embodiment, in which object nodes include object nodes corresponding to a plurality of sample objects, the sample objects include positive sample objects and negative sample objects, and the positive sample objects include seed objects, as shown in fig. 11. The apparatus 200 may further include:

training module 204 is configured to train the recognition model for the target game.

The training module 204 may be configured to perform the following steps:

step a, obtaining a sample input set, wherein the sample input set comprises: sample inputs corresponding to each sample object, the sample inputs comprising: and the object vector and the target game vector are determined according to the game knowledge graph and used for representing the sample object.

And b, acquiring a sample output set, wherein the sample output set comprises sample outputs corresponding to each sample input, and each sample output comprises a real identification result of a corresponding sample object.

Step c, taking the sample input set as the input of the recognition model, and taking the sample output set as the output of the recognition model so as to train the recognition model.

In another application scenario, the positive sample object may also include an extension object, which is determined in the following manner.

Step 3) active objects of the related game are used as expansion objects.

In yet another application scenario, step c may include:

step c1, clustering the sample input sets according to a preset clustering algorithm to obtain a plurality of groups of sample input subsets, wherein each group of sample input subsets corresponds to one group of sample output subsets.

And c2, inputting the set of sample input subsets into the recognition sub-models corresponding to the set of sample input subsets and included in the recognition model, and taking the sample output subset corresponding to the set of sample input subsets as the output of the recognition sub-model corresponding to the set of sample input subsets so as to train the recognition sub-model corresponding to the set of sample input subsets.

In yet another application scenario, each set of sample inputs a recognition sub-model corresponding to a subset, which is a tree integration model, comprising a plurality of tree models. The implementation of step c2 may include:

In one implementation, the game knowledge graph is established by:

And d, acquiring object information of a plurality of objects, game information of a plurality of games and content information of a plurality of contents. The content is any one of text, audio and video, the objects comprise objects to be identified and seed objects, and the games comprise target games.

And e, establishing an object node corresponding to each object, a game node corresponding to each game and a content node corresponding to each content.

And f, establishing edges among the plurality of nodes according to a preset association rule.

In another implementation, the implementation of step f may include:

step f1, determining a game association object meeting association rules with each object according to the object information of each object, wherein the game association object comprises: at least one of an object, a game and a content, and establishing edges between nodes corresponding to the object and nodes corresponding to the game related object at the object nodes corresponding to the object.

Step f2, according to the content information of each content, determining a content association object meeting the association rule with the content, wherein the content association object comprises: and establishing edges between the nodes corresponding to the content and the nodes corresponding to the content association objects.

And f3, determining an associated game meeting the association rule according to the game information of each game, and establishing edges between game nodes corresponding to the game and game nodes corresponding to the associated game.

Fig. 12 is a block diagram of another knowledge-graph-based object recognition apparatus, as shown in fig. 12, according to an exemplary embodiment, the vector determination module 201 may include:

a first determining submodule 2011 is used for determining a game vector used for representing each game according to the game knowledge graph.

A second determination submodule 2012 determines an object vector for characterizing each object according to the game knowledge graph and each game vector.

A third determining submodule 2013 is configured to take an object vector for representing an object to be identified as an object vector to be identified, and take a game vector for representing a target game as a target game vector.

In one application scenario, the first determination submodule 2011 may be configured to perform the following steps:

In one application scenario, the second determination submodule 2012 may be used to perform the following steps:

The specific manner in which the various modules perform the operations in the apparatus of the above embodiments have been described in detail in connection with the embodiments of the method, and will not be described in detail herein.

Referring now to fig. 13, there is shown a schematic diagram of an electronic device (e.g., an execution body, which may be a terminal device or a server in the above-described embodiments) 300 suitable for implementing embodiments of the present disclosure. The terminal devices in the embodiments of the present disclosure may include, but are not limited to, mobile terminals such as mobile phones, notebook computers, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), in-vehicle terminals (e.g., in-vehicle navigation terminals), and the like, and stationary terminals such as digital TVs, desktop computers, and the like. The electronic device shown in fig. 13 is merely an example and should not impose any limitations on the functionality and scope of use of embodiments of the present disclosure.

As shown in fig. 13, the electronic device 300 may include a processing means (e.g., a central processing unit, a graphics processor, etc.) 301 that may perform various appropriate actions and processes according to a program stored in a Read Only Memory (ROM) 302 or a program loaded from a storage means 308 into a Random Access Memory (RAM) 303. In the RAM 303, various programs and data required for the operation of the electronic apparatus 300 are also stored. The processing device 301, the ROM 302, and the RAM 303 are connected to each other via a bus 304. An input/output (I/O) interface 305 is also connected to bus 304.

In general, the following devices may be connected to the I/O interface 305: input devices 306 including, for example, a touch screen, touchpad, keyboard, mouse, camera, microphone, accelerometer, gyroscope, etc.; an output device 307 including, for example, a Liquid Crystal Display (LCD), a speaker, a vibrator, and the like; storage 308 including, for example, magnetic tape, hard disk, etc.; and communication means 309. The communication means 309 may allow the electronic device 300 to communicate with other devices wirelessly or by wire to exchange data. While fig. 13 shows an electronic device 300 having various means, it is to be understood that not all of the illustrated means are required to be implemented or provided. More or fewer devices may be implemented or provided instead.

In particular, according to embodiments of the present disclosure, the processes described above with reference to 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 non-transitory computer readable medium, the computer program comprising program code for performing the method shown in the flow chart. In such an embodiment, the computer program may be downloaded and installed from a network via a communication device 309, or installed from a storage device 308, or installed from a ROM 302. The above-described functions defined in the methods of the embodiments of the present disclosure are performed when the computer program is executed by the processing means 301.

It should be noted that the computer readable medium described in the present disclosure may be a computer readable signal medium or a computer readable storage medium, or any combination of the two. The computer readable storage medium can be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or a combination of any 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 context of this disclosure, 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 the present disclosure, however, the computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, with the computer-readable program code embodied therein. Such a propagated data signal may take any of a variety of forms, including, but not limited to, electro-magnetic, optical, or any suitable combination of the foregoing. 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: electrical wires, fiber optic cables, RF (radio frequency), and the like, or any suitable combination of the foregoing.

In some embodiments, the terminal devices, servers, may communicate using any currently known or future developed network protocol, such as HTTP (HyperText Transfer Protocol ), and may be interconnected with any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network ("LAN"), a wide area network ("WAN"), the internet (e.g., the internet), and peer-to-peer networks (e.g., ad hoc peer-to-peer networks), as well as any currently known or future developed networks.

The computer readable medium may be contained in the electronic device; or may exist alone without being incorporated into the electronic device.

The computer readable medium carries one or more programs which, when executed by the electronic device, cause the electronic device to: determining a target game vector for representing a target game and an object vector to be identified for representing an object to be identified according to a pre-established game knowledge graph; determining the correlation degree of the object to be identified and the target game according to the object vector to be identified, the target game vector and a pre-trained identification model, wherein the identification model is obtained by training according to a seed object vector for representing the target game seed object and the target game vector, and the seed object vector is determined according to the game knowledge graph; and if the correlation degree between the object to be identified and the target game meets a preset condition, determining that the object to be identified is the target object of the target game.

Computer program code for carrying out operations of the present disclosure may be written in one or more programming languages, including, but not limited to, 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 program code may execute entirely on the subject computer, partly on the subject computer, as a stand-alone software package, partly on the subject computer and partly on a remote computer or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the object computer through any kind of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or may be connected to an external computer (for example, through the internet using an internet service provider).

The flowcharts 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 disclosure. 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 modules described in the embodiments of the present disclosure may be implemented in software or hardware. The name of the module is not limited to the module itself in some cases, and for example, the vector determination module may be also described as "a module that determines an object vector to be identified and a target game vector".

The functions described above herein may be performed, at least in part, by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that may be used include: a Field Programmable Gate Array (FPGA), an Application Specific Integrated Circuit (ASIC), an Application Specific Standard Product (ASSP), a system on a chip (SOC), a Complex Programmable Logic Device (CPLD), and the like.

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.

According to one or more embodiments of the present disclosure, example 1 provides an object recognition method based on a knowledge-graph, including: determining a target game vector for representing a target game and an object vector to be identified for representing an object to be identified according to a pre-established game knowledge graph; determining the correlation degree of the object to be identified and the target game according to the object vector to be identified, the target game vector and a pre-trained identification model, wherein the identification model is obtained by training according to a seed object vector for representing the target game seed object and the target game vector, and the seed object vector is determined according to the game knowledge graph; and if the correlation degree between the object to be identified and the target game meets a preset condition, determining that the object to be identified is the target object of the target game.

According to one or more embodiments of the present disclosure, example 2 provides the method of example 1, the determining a relevance of the object to be identified to the target game according to the object vector to be identified, the target game vector, and a pre-trained identification model, including: and inputting the object vector to be identified, the target game vector and the object characteristics of the object to be identified into the identification model to obtain the correlation degree between the object to be identified and the target game output by the identification model, wherein the object characteristics are determined according to the object information of the object to be identified.

According to one or more embodiments of the present disclosure, example 3 provides the method of example 1 or example 2, the game knowledge-graph comprising a plurality of nodes and at least one edge, the plurality of nodes comprising: object nodes, game nodes and content nodes; each edge is used to characterize an association between two nodes at both ends of the edge.

According to one or more embodiments of the present disclosure, example 4 provides the method of example 3, the object nodes comprising object nodes corresponding to a plurality of sample objects, the sample objects comprising a positive sample object and a negative sample object, the positive sample object comprising the seed object; the method further comprises the steps of: training the recognition model for the target game; the training the recognition model for the target game includes: obtaining a sample input set, the sample input set comprising: sample inputs corresponding to each of the sample objects, the sample inputs comprising: the object vector used for representing the sample object and the target game vector are determined according to the game knowledge graph; obtaining a sample output set, wherein the sample output set comprises sample outputs corresponding to each sample input, and each sample output comprises a real identification result of the corresponding sample object; the sample input set is used as the input of the recognition model, and the sample output set is used as the output of the recognition model to train the recognition model.

In accordance with one or more embodiments of the present disclosure, example 5 provides the method of example 4, the positive sample object further comprising an extension object, the extension object being determined by: determining the relevance of other games and the target game according to other game vectors used for representing the other games and the target game, wherein the other games are games except the target game, and the other game vectors are determined according to the game knowledge graph; taking the other games with the correlation degree larger than or equal to a preset correlation degree threshold value as the correlation games corresponding to the target games; and taking the active object of the related game as the extension object.

In accordance with one or more embodiments of the present disclosure, example 6 provides the method of example 4, the taking the sample input set as an input to the recognition model and the sample output set as an output of the recognition model to train the recognition model, comprising: clustering the sample input sets according to a preset clustering algorithm to obtain a plurality of groups of sample input subsets, wherein each group of sample input subsets corresponds to one group of sample output subsets; for each group of sample input subsets, inputting the group of sample input subsets into the recognition sub-model corresponding to the group of sample input subsets, which is included in the recognition model, and taking the sample output subset corresponding to the group of sample input subsets as the output of the recognition sub-model corresponding to the group of sample input subsets so as to train the recognition sub-model corresponding to the group of sample input subsets; and the output of the recognition model is determined according to the output of the recognition submodel corresponding to each group of the sample input subsets.

Example 7 provides the method of example 6, according to one or more embodiments of the present disclosure, wherein each set of the recognition sub-models corresponding to the sample input subset is a tree integration model comprising a plurality of tree models; the inputting the set of sample input subsets into the recognition sub-model corresponding to the set of sample input subsets and the outputting the sample output subset corresponding to the set of sample input subsets as the output of the recognition sub-model corresponding to the set of sample input subsets to train the recognition sub-model corresponding to the set of sample input subsets includes: randomly sampling the set of sample input subsets for each of the tree models to obtain a sampled input subset, the sampled input subset comprising a smaller number of the sample inputs than the set of sample input subsets; randomly sampling each sample input included in the sample input subset to obtain a sample input corresponding to the sample input, wherein the sample input belongs to the sample input; taking a sample input corresponding to each sample input included in the sample input subset as an input of the tree model, and taking a sample output corresponding to each sample input included in the sample input subset as an output of the tree model to train the tree model; the output of the tree integration model is determined from the output of each of the tree models.

Example 8 provides the method of example 3, according to one or more embodiments of the present disclosure, the game knowledge-graph being established by: acquiring object information of a plurality of objects, game information of a plurality of games and content information of a plurality of contents; establishing object nodes corresponding to each object, game nodes corresponding to each game and content nodes corresponding to each content; and establishing edges among the plurality of nodes according to a preset association rule.

According to one or more embodiments of the present disclosure, example 9 provides the method of example 8, the establishing edges between the plurality of nodes according to a preset association rule, comprising: determining a game association object meeting the association rule with the object according to the object information of each object, wherein the game association object comprises: at least one of an object, a game and a content, and establishing an edge between the nodes corresponding to the object and the nodes corresponding to the game related object; according to the content information of each content, determining a content association object meeting the association rule with the content, wherein the content association object comprises: game and/or content, and establishing edges between content nodes corresponding to the content and nodes corresponding to the content association objects; and determining an associated game meeting the association rule according to the game information of each game, and establishing edges between game nodes corresponding to the associated game and game nodes corresponding to the associated game.

According to one or more embodiments of the present disclosure, example 10 provides the method of example 9, the determining a target game vector for characterizing a target game and an object vector to be identified for characterizing an object to be identified according to a pre-established game knowledge-graph, comprising: determining a game vector for representing each game according to the game knowledge graph; determining an object vector for representing each object according to the game knowledge graph and each game vector; and taking the object vector used for representing the object to be identified as the object vector to be identified, and taking the game vector used for representing the target game as the target game vector.

Example 11 provides the method of example 10, according to one or more embodiments of the present disclosure, the determining a game vector for characterizing each of the games according to the game knowledge-graph, comprising: determining an object game map according to the edges between the object nodes and the game nodes in the game knowledge map, determining an object content map according to the edges between the object nodes and the content nodes in the game knowledge map, and determining a game map according to the edges between the game nodes and the game nodes in the game knowledge map; determining each game vector according to a preset graph representation algorithm, the object game graph, the object content graph and the game graph; the object game atlas is used for representing the association of any two game nodes in the object dimension, the object content atlas is used for representing the association of any two game nodes in the content dimension, and the game atlas is used for representing the association of any two game nodes in the game dimension.

Example 12 provides the method of example 10, according to one or more embodiments of the present disclosure, the determining an object vector for characterizing each of the objects from the game knowledge-graph and each of the game vectors, comprising: determining an object game map according to the edges between the object nodes in the game knowledge map and the game nodes, and determining an object content map according to the edges between the object nodes and the content nodes in the game knowledge map; determining each of the object vectors according to each of the game vectors, the object game atlas and the object content atlas; the object game atlas is used for representing the association of each object node and each game node in a game dimension, and the object content atlas is used for representing the association of each object node and each game node in a content dimension.

Example 13 provides an object recognition apparatus based on a knowledge-graph, according to one or more embodiments of the present disclosure, comprising: the vector determining module is used for determining a target game vector for representing a target game and an object vector to be identified for representing an object to be identified according to a pre-established game knowledge graph; the relevance determining module is used for determining the relevance of the object to be identified and the target game according to the object vector to be identified, the target game vector and a pre-trained identification model, wherein the identification model is obtained by training according to a seed object vector of a seed object for representing the target game and the target game vector, and the seed object vector is determined according to the game knowledge graph; and the identification module is used for determining the object to be identified as the target object of the target game if the correlation degree between the object to be identified and the target game meets a preset condition.

According to one or more embodiments of the present disclosure, example 14 provides a computer-readable medium having stored thereon a computer program which, when executed by a processing device, implements the steps of the methods described in examples 1 to 12.

Example 15 provides an electronic device according to one or more embodiments of the present disclosure, comprising: a storage device having a computer program stored thereon; processing means for executing the computer program in the storage means to implement the steps of the method described in examples 1 to 12.

The foregoing description is only of the preferred embodiments of the present disclosure and description of the principles of the technology being employed. It will be appreciated by persons skilled in the art that the scope of the disclosure referred to in this disclosure is not limited to the specific combinations of features described above, but also covers other embodiments which may be formed by any combination of features described above or equivalents thereof without departing from the spirit of the disclosure. Such as those described above, are mutually substituted with the technical features having similar functions disclosed in the present disclosure (but not limited thereto).

Moreover, although operations are depicted in a particular order, this should not be understood as requiring that such operations be performed in the particular order shown or in sequential order. In certain circumstances, multitasking and parallel processing may be advantageous. Likewise, while several specific implementation details are included in the above discussion, these should not be construed as limiting the scope of the present disclosure. Certain features that are described in the context of separate embodiments can also be implemented in combination in a single embodiment. Conversely, various features that are described in the context of a single embodiment can also be implemented in multiple embodiments separately or in any suitable subcombination.

Although the subject matter has been described in language specific to structural features and/or methodological acts, it is to be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or acts described above. Rather, the specific features and acts described above are example forms of implementing the claims. The specific manner in which the various modules perform the operations in the apparatus of the above embodiments have been described in detail in connection with the embodiments of the method, and will not be described in detail herein.

Claims

1. An object recognition method based on a knowledge graph, which is characterized by comprising the following steps:

If the correlation degree between the object to be identified and the target game meets a preset condition, determining that the object to be identified is a target object of the target game;

the game knowledge graph includes a plurality of nodes and at least one edge, the plurality of nodes including: object nodes, game nodes and content nodes; each edge is used for representing that the two nodes at two ends of the edge have an association; the object to be identified comprises an object represented by any object node in the game knowledge graph; the games corresponding to the game nodes comprise the target games.

2. The method of claim 1, wherein the determining the relevance of the object to be identified to the target game based on the object vector to be identified, the target game vector, and a pre-trained recognition model comprises:

and inputting the object vector to be identified, the target game vector and the object characteristics of the object to be identified into the identification model to obtain the correlation degree between the object to be identified and the target game output by the identification model, wherein the object characteristics are determined according to the object information of the object to be identified.

3. The method of claim 1, wherein the object nodes comprise object nodes corresponding to a plurality of sample objects, the sample objects comprising positive sample objects and negative sample objects, the positive sample objects comprising the seed objects;

The method further comprises the steps of: training the recognition model for the target game;

the training the recognition model for the target game includes:

obtaining a sample input set, the sample input set comprising: sample inputs corresponding to each of the sample objects, the sample inputs comprising: the object vector used for representing the sample object and the target game vector are determined according to the game knowledge graph;

obtaining a sample output set, wherein the sample output set comprises sample outputs corresponding to each sample input, and each sample output comprises a real identification result of the corresponding sample object;

the sample input set is used as the input of the recognition model, and the sample output set is used as the output of the recognition model to train the recognition model.

4. A method according to claim 3, wherein the positive sample object further comprises an extension object, the extension object being determined by:

determining the relevance of other games and the target game according to other game vectors used for representing the other games and the target game, wherein the other games are games except the target game, and the other game vectors are determined according to the game knowledge graph;

Taking the other games with the correlation degree larger than or equal to a preset correlation degree threshold value as the correlation games corresponding to the target games;

and taking the active object of the related game as the extension object.

5. A method according to claim 3, wherein said taking the sample input set as input to the recognition model and the sample output set as output from the recognition model to train the recognition model comprises:

clustering the sample input sets according to a preset clustering algorithm to obtain a plurality of groups of sample input subsets, wherein each group of sample input subsets corresponds to one group of sample output subsets;

for each group of sample input subsets, inputting the group of sample input subsets into the recognition sub-model corresponding to the group of sample input subsets, which is included in the recognition model, and taking the sample output subset corresponding to the group of sample input subsets as the output of the recognition sub-model corresponding to the group of sample input subsets so as to train the recognition sub-model corresponding to the group of sample input subsets;

and the output of the recognition model is determined according to the output of the recognition submodel corresponding to each group of the sample input subsets.

6. The method of claim 5, wherein the recognition sub-model corresponding to each set of the sample input subsets is a tree integration model comprising a plurality of tree models; the inputting the set of sample input subsets into the recognition sub-model corresponding to the set of sample input subsets and the outputting the sample output subset corresponding to the set of sample input subsets as the output of the recognition sub-model corresponding to the set of sample input subsets to train the recognition sub-model corresponding to the set of sample input subsets includes:

randomly sampling the set of sample input subsets for each of the tree models to obtain a sampled input subset, the sampled input subset comprising a smaller number of the sample inputs than the set of sample input subsets;

randomly sampling each sample input included in the sample input subset to obtain a sample input corresponding to the sample input, wherein the sample input belongs to the sample input;

taking a sample input corresponding to each sample input included in the sample input subset as an input of the tree model, and taking a sample output corresponding to each sample input included in the sample input subset as an output of the tree model to train the tree model;

The output of the tree integration model is determined from the output of each of the tree models.

7. The method of claim 1, wherein the game knowledge-graph is established by:

acquiring object information of a plurality of objects, game information of a plurality of games and content information of a plurality of contents;

establishing object nodes corresponding to each object, game nodes corresponding to each game and content nodes corresponding to each content;

and establishing edges among the plurality of nodes according to a preset association rule.

8. The method of claim 7, wherein establishing edges between the plurality of nodes according to a preset association rule comprises:

determining a game association object meeting the association rule with the object according to the object information of each object, wherein the game association object comprises: at least one of an object, a game and a content, and establishing an edge between the nodes corresponding to the object and the nodes corresponding to the game related object;

according to the content information of each content, determining a content association object meeting the association rule with the content, wherein the content association object comprises: game and/or content, and establishing edges between content nodes corresponding to the content and nodes corresponding to the content association objects;

And determining an associated game meeting the association rule according to the game information of each game, and establishing edges between game nodes corresponding to the associated game and game nodes corresponding to the associated game.

9. The method of claim 8, wherein determining a target game vector for characterizing a target game and an object vector to be identified for characterizing an object to be identified based on a pre-established game knowledge-graph, comprises:

determining a game vector for representing each game according to the game knowledge graph;

determining an object vector for representing each object according to the game knowledge graph and each game vector;

and taking the object vector used for representing the object to be identified as the object vector to be identified, and taking the game vector used for representing the target game as the target game vector.

10. The method of claim 9, wherein said determining a game vector for characterizing each of said games based on said game knowledge-graph comprises:

determining an object game map according to the edges between the object nodes and the game nodes in the game knowledge map, determining an object content map according to the edges between the object nodes and the content nodes in the game knowledge map, and determining a game map according to the edges between the game nodes and the game nodes in the game knowledge map;

Determining each game vector according to a preset graph representation algorithm, the object game graph, the object content graph and the game graph;

11. The method of claim 9, wherein said determining an object vector for characterizing each of said objects from said game knowledge-graph and each of said game vectors comprises:

determining an object game map according to the edges between the object nodes in the game knowledge map and the game nodes, and determining an object content map according to the edges between the object nodes and the content nodes in the game knowledge map;

determining each of the object vectors according to each of the game vectors, the object game atlas and the object content atlas;

the object game atlas is used for representing the association of each object node and each game node in a game dimension, and the object content atlas is used for representing the association of each object node and each game node in a content dimension.

12. An object recognition device based on a knowledge-graph, the device comprising:

the identification module is used for determining the object to be identified as the target object of the target game if the correlation degree between the object to be identified and the target game meets a preset condition;

13. A computer readable medium on which a computer program is stored, characterized in that the program, when being executed by a processing device, carries out the steps of the method according to any one of claims 1-11.

14. An electronic device, comprising:

a storage device having a computer program stored thereon;

processing means for executing said computer program in said storage means to carry out the steps of the method according to any one of claims 1-11.