阅读说明：本技术 一种对象标签确定方法及装置 (Object label determination method and device ) 是由 毋安安 于 2021-08-17 设计创作，主要内容包括：本发明实施例提供了一种对象标签确定方法及装置,涉及数据处理技术领域,其中,上述方法包括：确定用于描述目标多媒体对象的对象信息。根据对象信息与预先构建的针对应用场景的知识图谱中实体的实体信息,在知识图谱中确定与目标多媒体对象存在关联关系的目标实体,目标实体包括：与目标多媒体对象直接存在关联关系的实体和/或与目标多媒体对象间接存在关联关系的实体,应用场景中的实体包括：应用场景中的多媒体对象和/或多媒体对象的关联信息。选择目标实体的全部或部分实体信息,作为目标多媒体对象的对象标签。应用本发明实施例提供的方案确定对象标签,可以提高多媒体对象的对象标签的丰富程度。(The embodiment of the invention provides a method and a device for determining an object label, which relate to the technical field of data processing, wherein the method comprises the following steps: object information describing the target multimedia object is determined. Determining a target entity having an association relation with a target multimedia object in a knowledge graph according to the object information and entity information of an entity in the knowledge graph which is constructed in advance and aims at an application scene, wherein the target entity comprises: the entity directly having an association relation with the target multimedia object and/or the entity indirectly having an association relation with the target multimedia object, the entity in the application scene includes: multimedia objects in the application scene and/or associated information of the multimedia objects. And selecting all or part of entity information of the target entity as an object tag of the target multimedia object. The scheme provided by the embodiment of the invention is applied to determine the object tags, so that the richness of the object tags of the multimedia objects can be improved.)

1. A method for object tag determination, the method comprising:

determining object information for describing a target multimedia object;

determining a target entity in the knowledge graph, which has an association relation with the target multimedia object, according to the object information and entity information of an entity in a knowledge graph which is constructed in advance and aims at an application scene, wherein the target entity comprises: an entity directly associated with the target multimedia object and/or an entity indirectly associated with the target multimedia object, wherein the entity in the application scene comprises: multimedia objects and/or associated information of multimedia objects in the application scene;

and selecting all or part of entity information of the target entity as an object tag of the target multimedia object.

2. The method according to claim 1, wherein the determining, in the knowledge graph, a target entity associated with the target multimedia object according to the object information and entity information of an entity in a pre-constructed knowledge graph for the application scene comprises:

determining a first entity directly associated with the target multimedia object in the knowledge graph according to the object information and entity information of the entities in the knowledge graph;

determining a second entity directly and/or indirectly having an association relation with the first entity according to the association relation among the entities in the knowledge graph;

and obtaining a target entity comprising the first entity and the second entity.

3. The method of claim 2, wherein determining the first entity in the knowledge-graph that is directly associated with the target multimedia object according to the object information and entity information of the entity in the knowledge-graph comprises:

in the knowledge graph, determining an entity of which all or part of entity information is contained in the object information as a first entity directly associated with the target multimedia object;

and/or

And in the knowledge graph, determining all or part of entities with the same entity information and object information semanteme as a first entity directly associated with the target multimedia object.

4. The method according to claim 2, wherein the determining a second entity directly and/or indirectly associated with a first entity according to the association relationship between the entities in the knowledge-graph comprises:

selecting an incidence relation directly pointing to the first entity and/or indirectly pointing to the first entity from incidence relations belonging to a preset relation type among the entities in the knowledge graph;

and determining a second entity which has direct and/or indirect incidence relation with the first entity according to the selected incidence relation.

5. The method according to claim 2, wherein the determining a second entity directly and/or indirectly associated with a first entity according to the association relationship between the entities in the knowledge-graph comprises:

selecting incidence relations with relation weights larger than preset weights and directly pointing to the first entity and/or indirectly pointing to the first entity from incidence relations among entities in the knowledge graph;

and determining a second entity which has direct and/or indirect incidence relation with the first entity according to the selected incidence relation.

6. An object tag determination apparatus, characterized in that the apparatus comprises:

an information determination module for determining object information for describing a target multimedia object;

an entity determining module, configured to determine, in the knowledge graph, a target entity having an association relationship with the target multimedia object according to the object information and entity information of an entity in a pre-constructed knowledge graph for an application scenario, where the target entity includes: an entity directly associated with the target multimedia object and/or an entity indirectly associated with the target multimedia object, wherein the entity in the application scene comprises: multimedia objects and/or associated information of multimedia objects in the application scene;

and the label selection module is used for selecting all or part of the entity information of the target entity as the object label of the target multimedia object.

7. The apparatus of claim 6, wherein the entity determination module comprises:

a first entity determining submodule, configured to determine, in the knowledge graph, a first entity that directly has an association relationship with the target multimedia object according to the object information and entity information of an entity in the knowledge graph;

the second entity determining submodule is used for determining a second entity which has direct and/or indirect incidence relation with the first entity according to the incidence relation among the entities in the knowledge graph;

and the entity obtaining submodule is used for obtaining a target entity containing the first entity and the second entity.

8. The apparatus of claim 7, wherein the first entity determination submodule is specifically configured to:

in the knowledge graph, determining an entity of which all or part of entity information is contained in the object information as a first entity directly associated with the target multimedia object;

and/or

And in the knowledge graph, determining all or part of entities with the same entity information and object information semanteme as a first entity directly associated with the target multimedia object.

9. The apparatus of claim 7, wherein the second entity determination submodule is specifically configured to:

selecting an incidence relation directly pointing to the first entity and/or indirectly pointing to the first entity from incidence relations belonging to a preset relation type among the entities in the knowledge graph;

and determining a second entity which has direct and/or indirect incidence relation with the first entity according to the selected incidence relation.

10. The apparatus of claim 7, wherein the second entity determination submodule is specifically configured to:

selecting incidence relations with relation weights larger than preset weights and directly pointing to the first entity and/or indirectly pointing to the first entity from incidence relations among entities in the knowledge graph;

and determining a second entity which has direct and/or indirect incidence relation with the first entity according to the selected incidence relation.

11. An electronic device is characterized by comprising a processor, a communication interface, a memory and a communication bus, wherein the processor and the communication interface are used for realizing mutual communication by the memory through the communication bus;

a memory for storing a computer program;

a processor for implementing the method steps of any one of claims 1 to 5 when executing a program stored in the memory.

12. A computer-readable storage medium, characterized in that a computer program is stored in the computer-readable storage medium, which computer program, when being executed by a processor, carries out the method steps of any one of the claims 1-5.

Technical Field

The present invention relates to the field of data processing technologies, and in particular, to a method and an apparatus for determining an object tag.

Background

Different multimedia objects, such as audio, video, etc., may have different object tags, with the object tag of each multimedia object being used to describe the multimedia object. For example, in the case that the multimedia object is a video, the object tag may include a category of the video, such as a tv show, a movie, and the like, and may also include a name of a lead actor of the video, such as a dragon, a week, and the like. According to the object tags, the multimedia objects can be classified, searched, and other multimedia objects with similar object tags can be recommended to the user.

In the prior art, the object labels of the multimedia objects are often manually added. Due to the fact that the efficiency of manually adding the object tags is low, under the condition that the number of multimedia objects is large, the number of the object tags which are manually added is limited, and the richness degree of the object tags is low.

Disclosure of Invention

The embodiment of the invention aims to provide a method and a device for determining object tags, so as to improve the richness of the object tags of a multimedia object. The specific technical scheme is as follows:

in a first aspect, an embodiment of the present invention provides an object tag determination method, where the method includes:

determining object information for describing a target multimedia object;

determining a target entity in the knowledge graph, which has an association relation with the target multimedia object, according to the object information and entity information of an entity in a knowledge graph which is constructed in advance and aims at an application scene, wherein the target entity comprises: an entity directly associated with the target multimedia object and/or an entity indirectly associated with the target multimedia object, wherein the entity in the application scene comprises: multimedia objects and/or associated information of multimedia objects in the application scene;

and selecting all or part of entity information of the target entity as an object tag of the target multimedia object.

In an embodiment of the present invention, the determining, in the knowledge graph, a target entity associated with the target multimedia object according to the object information and entity information of an entity in a pre-constructed knowledge graph for an application scenario includes:

determining a first entity directly associated with the target multimedia object in the knowledge graph according to the object information and entity information of the entities in the knowledge graph;

determining a second entity directly and/or indirectly having an association relation with the first entity according to the association relation among the entities in the knowledge graph;

and obtaining a target entity comprising the first entity and the second entity.

In an embodiment of the present invention, the determining, in the knowledge-graph, a first entity directly associated with the target multimedia object according to the object information and entity information of the entity in the knowledge-graph includes:

in the knowledge graph, determining an entity of which all or part of entity information is contained in the object information as a first entity directly associated with the target multimedia object;

and/or

And in the knowledge graph, determining all or part of entities with the same entity information and object information semanteme as a first entity directly associated with the target multimedia object.

In an embodiment of the present invention, the determining, according to the association relationship between the entities in the knowledge-graph, a second entity that directly and/or indirectly has an association relationship with the first entity includes:

selecting an incidence relation directly pointing to the first entity and/or indirectly pointing to the first entity from incidence relations belonging to a preset relation type among the entities in the knowledge graph;

and determining a second entity which has direct and/or indirect incidence relation with the first entity according to the selected incidence relation.

In an embodiment of the present invention, the determining, according to the association relationship between the entities in the knowledge-graph, a second entity that directly and/or indirectly has an association relationship with the first entity includes:

selecting incidence relations with relation weights larger than preset weights and directly pointing to the first entity and/or indirectly pointing to the first entity from incidence relations among entities in the knowledge graph;

and determining a second entity which has direct and/or indirect incidence relation with the first entity according to the selected incidence relation.

In a second aspect, an embodiment of the present invention provides an object tag determination apparatus, where the apparatus includes:

an information determination module for determining object information for describing a target multimedia object;

an entity determining module, configured to determine, in the knowledge graph, a target entity having an association relationship with the target multimedia object according to the object information and entity information of an entity in a pre-constructed knowledge graph for an application scenario, where the target entity includes: an entity directly associated with the target multimedia object and/or an entity indirectly associated with the target multimedia object, wherein the entity in the application scene comprises: multimedia objects and/or associated information of multimedia objects in the application scene;

and the label selection module is used for selecting all or part of the entity information of the target entity as the object label of the target multimedia object.

In an embodiment of the present invention, the entity determining module includes:

a first entity determining submodule, configured to determine, in the knowledge graph, a first entity that directly has an association relationship with the target multimedia object according to the object information and entity information of an entity in the knowledge graph;

the second entity determining submodule is used for determining a second entity which has direct and/or indirect incidence relation with the first entity according to the incidence relation among the entities in the knowledge graph;

and the entity obtaining submodule is used for obtaining a target entity containing the first entity and the second entity.

In an embodiment of the present invention, the first entity determining sub-module is specifically configured to:

in the knowledge graph, determining an entity of which all or part of entity information is contained in the object information as a first entity directly associated with the target multimedia object;

and/or

And in the knowledge graph, determining all or part of entities with the same entity information and object information semanteme as a first entity directly associated with the target multimedia object.

In an embodiment of the present invention, the second entity determining sub-module is specifically configured to:

selecting an incidence relation directly pointing to the first entity and/or indirectly pointing to the first entity from incidence relations belonging to a preset relation type among the entities in the knowledge graph;

and determining a second entity which has direct and/or indirect incidence relation with the first entity according to the selected incidence relation.

In an embodiment of the present invention, the second entity determining sub-module is specifically configured to:

selecting incidence relations with relation weights larger than preset weights and directly pointing to the first entity and/or indirectly pointing to the first entity from incidence relations among entities in the knowledge graph;

and determining a second entity which has direct and/or indirect incidence relation with the first entity according to the selected incidence relation.

In a third aspect, an embodiment of the present invention provides an electronic device, including a processor, a communication interface, a memory, and a communication bus, where the processor and the communication interface complete communication between the memory and the processor through the communication bus;

a memory for storing a computer program;

a processor for implementing the method steps of any of the first aspect when executing a program stored in the memory.

In a fourth aspect, the present invention provides a computer-readable storage medium, in which a computer program is stored, and the computer program, when executed by a processor, implements the method steps of any one of the first aspect.

In a fifth aspect, embodiments of the present invention also provide a computer program product comprising instructions, which when run on a computer, cause the computer to perform the method steps of any of the first aspects described above.

The object tag determining method provided by the embodiment of the invention determines object information for describing a target multimedia object, determines a target entity directly and/or indirectly associated with the target multimedia object in a knowledge graph according to the object information and entity information of an entity in the knowledge graph which is constructed in advance and aims at an application scene, and selects all or part of entity information of the target entity as an object tag of the target multimedia object.

As can be seen from the above description, since the target entity is an entity that has an association relationship with the target multimedia object directly or indirectly, the entity information of the target entity may be considered to have an association relationship with the target multimedia object and may be used to describe the target multimedia object, and therefore, the entity information of the target entity may be used as an object tag of the target multimedia object. And determining the entity information of the target entity as the object tag of the target multimedia object, so that the number of the object tags can be increased, and the richness of the object tags can be improved.

Drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below.

Fig. 1 is a schematic flowchart of a first object tag determination method provided in an embodiment of the present invention;

FIG. 2 is a schematic diagram of an object tag provided in an embodiment of the present invention;

fig. 3 is a schematic flowchart of a second object tag determination method provided in an embodiment of the present invention;

fig. 4 is a schematic flowchart of a third object tag determination method provided in an embodiment of the present invention;

fig. 5 is a schematic structural diagram of a first object tag determination apparatus provided in an embodiment of the present invention;

fig. 6 is a schematic structural diagram of a second object tag determination apparatus provided in an embodiment of the present invention;

fig. 7 is a schematic structural diagram of an electronic device provided in an embodiment of the present invention.

Detailed Description

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived from the embodiments given herein by one of ordinary skill in the art, are within the scope of the invention.

Due to the low richness of object tags of multimedia objects in the prior art. To solve the problem, embodiments of the present invention provide an object tag determination method and apparatus.

In an embodiment of the present invention, a method for determining an object tag is provided, where the method includes:

object information describing the target multimedia object is determined.

And determining a target entity in the knowledge graph, which has an association relation with the target multimedia object, according to the object information and entity information of an entity in the knowledge graph which is constructed in advance and aims at the application scene. Wherein, the target entity comprises: an entity directly having an association relationship with the target multimedia object and/or an entity indirectly having an association relationship with the target multimedia object, where the entity in the application scenario includes: and the multimedia objects in the application scene and/or the associated information of the multimedia objects.

And selecting all or part of entity information of the target entity as the object tag of the target multimedia object.

As can be seen from the above description, since the target entity is an entity that has an association relationship with the target multimedia object directly or indirectly, the entity information of the target entity may be considered to have an association relationship with the target multimedia object and may be used to describe the target multimedia object, and therefore, the entity information of the target entity may be used as an object tag of the target multimedia object. And determining the entity information of the target entity as the object tag of the target multimedia object, so that the number of the object tags can be increased, and the richness of the object tags can be improved.

The following describes a method and an apparatus for determining an object tag according to an embodiment of the present invention with reference to specific embodiments.

Referring to fig. 1, a flowchart of a first object tag determination method according to an embodiment of the present invention is shown, where the method includes the following steps S101 to S103.

S101: object information describing the target multimedia object is determined.

Specifically, the target multimedia object may be video, audio, image, or the like.

The object information may be: the object name, the current object tag, the attribute information, etc. of the target multimedia object, where the attribute information may include: the type of target multimedia object, the source of the target multimedia object, etc.

For example, in the case that the target multimedia object is a video, the types of the target multimedia object may be: movies, television shows, short videos, etc., the sources of the above target multimedia objects may be: various video websites, etc. In the case that the target multimedia object is an audio, the type of the target multimedia object may be: pop songs, classical music, etc., the sources of the above target multimedia objects may be: various music websites, etc.

S102: and determining a target entity in the knowledge graph, which has an association relation with the target multimedia object, according to the object information and entity information of an entity in the knowledge graph which is constructed in advance and aims at the application scene.

Wherein, the target entity comprises: and the entity directly has the association relation with the target multimedia object and/or the entity indirectly has the association relation with the target multimedia object. The entities in the application scenario include: and the multimedia objects in the application scene and/or the associated information of the multimedia objects.

Specifically, the knowledge graph includes nodes corresponding to the entities and edges between the nodes, and the edges between the nodes indicate that the entities corresponding to the nodes on both sides of the edges directly have an association relationship. The knowledge graph may further include weights of association relationships between entities, where a higher weight indicates a stronger association relationship between entities, and the weights of association relationships between different entities may be the same or different.

In addition, the relationship type of the relationship represented by the association relationship between different entities is different, for example, the relationship type of the relationship may be a synonymous relationship type in which the meanings represented by the entity information characterizing two entities having an association relationship are the same, and an inclusion relationship type in which one entity includes the other entity, or the like. The association relationship between the entities may be a directed relationship or an undirected relationship. For example, if the relationship existing between the entity M and the entity N is a synonymous relationship type, the meaning of the entity information of the entity M is the same as that of the entity information of the entity N for the entity M, and the meaning of the entity information of the entity N is the same as that of the entity information of the entity M for the entity N, the relationship may be a undirected relationship. If the association relationship between the entity M and the entity N is an inclusion relationship type, it indicates that the entity M includes the entity N, but the entity N does not include the entity M, so the association relationship may be a directed relationship.

The entity information of the entity may include: entity name, entity type, entity attributes, etc. The entity attributes may include the type of entity, the source of the entity, and the like.

In addition, in order to ensure that an entity having an association relation with the target multimedia object exists in the knowledge graph, the knowledge graph is a knowledge graph aiming at an application scene. For example, in the case that the target multimedia object is a video, the knowledge graph may be a video knowledge graph. Then, the multimedia object included in the entity included in the video knowledge graph is a video in an application scene, and the associated information of the multimedia object is: a director, screenplay, etc. of the multimedia object.

In addition, the target entity may be an entity representing preset content in an entity in the knowledge graph, which has a direct and/or indirect association relationship with the target multimedia object. For example, in the case where the target multimedia object is audio, the preset content may be a distribution company, a singer, a word writer, a music writer, or the like.

The target entity may also be an entity with a popularity degree higher than a preset popularity degree in an entity directly and/or indirectly associated with the target multimedia object in the knowledge graph. The popularity may be expressed in terms of the number of searches the entity has historically been searched by the user, the number of clicks the entity has historically clicked on by the user, and so on. The entity with high popularity is used as the target entity, so that the object tag determined according to the entity information of the target entity can be more easily identified by the user.

The target entity can also be an entity selected by a user through an input entity selection instruction in the entities which directly and/or indirectly have the association relationship with the target multimedia object in the knowledge graph.

In an embodiment of the present invention, an entity having an entity name identical to the object name of the target multimedia object may be selected from the knowledge graph as the target entity.

For example, when the object name of the target multimedia object is the month mi transmission, the target entity is an entity whose entity name included in the knowledge graph is the month mi transmission.

In another embodiment of the present invention, the target entity may also be determined through steps S102A-S102C, which is not detailed in the embodiment of the present invention.

S103: and selecting all or part of entity information of the target entity as the object tag of the target multimedia object.

In an embodiment of the present invention, all entity information of the target entity may be used as the object tag of the target multimedia object.

Only information belonging to a preset type in the entity information may be selected as the object tag of the target multimedia object, for example, the preset type may be an entity name.

And receiving an instruction of selecting entity information by a user, and taking the entity information selected by the user as the object tag of the target multimedia object.

In addition, the resulting object tag may be stored in a tag result data table.

Referring to fig. 2, a schematic diagram of an object tag according to an embodiment of the present invention is provided.

The monthly MI passage table target multimedia objects are contained in the figure, with the other rectangular boxes representing the individual object labels. In the case where the target multimedia object is a month mi transmission, the target entity is an entity named "month mi transmission" included in the knowledge map. The entity information of the month mi entity may be used as an object tag of the target multimedia object, and includes attribute information of the entity, such as a source of the entity, i.e., an M video website in the figure, and an entity type of the entity, such as an ancient drama and a continental drama in the figure. In addition, the names of main actors in the moon MI biography are also contained in the object tags, including SULI and Liutao.

As can be seen from the above description, since the target entity is an entity that has an association relationship with the target multimedia object directly or indirectly, the entity information of the target entity may be considered to have an association relationship with the target multimedia object and may be used to describe the target multimedia object, and therefore, the entity information of the target entity may be used as an object tag of the target multimedia object. And determining the entity information of the target entity as the object tag of the target multimedia object, so that the number of the object tags can be increased, and the richness of the object tags can be improved.

In an embodiment of the present invention, after determining to obtain the object tag of the target multimedia object, the object tag may be output for other users or organizations to use, or according to the object tag of each multimedia object, the multimedia object including the same object tag may be recommended to the user, so as to attract the user to view other multimedia objects.

In addition, the popularity of the target multimedia object can be calculated according to the popularity of the target entity and the entity information, and the popularity of the target entity having an association relation with the target multimedia object is higher, and the popularity of the target multimedia object is higher.

Specifically, an average value, a weighted average value, and the like of the popularity of each target entity may be calculated as the popularity of the target multimedia object. When the weighted average is calculated as the popularity of the target multimedia object, the weight of the relationship between the target entity and the target multimedia object may be used as the weight of the popularity corresponding to the target entity when the weighted average is calculated.

Referring to fig. 3, a flowchart of a second object tag determination method provided in the embodiment of the present invention is shown, and compared with the foregoing embodiment shown in fig. 1, the foregoing step S102 may be implemented by the following steps S102A-S102C.

S102A: and according to the object information and the entity information of the entities in the knowledge graph, determining a first entity which is directly associated with the target multimedia object in the knowledge graph.

Specifically, the first entity directly associated with the target multimedia object may be an entity for representing the target multimedia object. For example, in the case that the target multimedia object is a television drama soldier assault, the first entity may be an entity for representing the soldier assault on a television drama.

In one embodiment of the invention, the first entity may be determined from the entities contained in the knowledge-graph by traversing the knowledge-graph. The number of the first entities may be 1 or more.

In another embodiment of the present invention, the first entity having direct association relationship with the target multimedia object may be determined through the following steps a and/or B.

Step A: in the knowledge graph, an entity in which all or part of entity information is included in the object information is determined as a first entity having a direct association with the target multimedia object.

In an embodiment of the present invention, each entity in the knowledge graph may be traversed, and an entity whose part or all of the entity information is the same as the object information may be determined as the first entity, or an entity whose part or all of the entity information is a part of the object information may be determined as the first entity.

Specifically, the object information may be subjected to word segmentation, and the word segmentation result is compared with part or all of the entity information of the entity, so as to determine a first entity in the entity.

The word segmentation processing can be implemented by a word segmentation mode in the prior art, which is not limited in the embodiment of the present invention.

The process of determining the first entity through step a above may be referred to as a structured matching process.

And B: and in the knowledge graph, determining all or part of entities with the same entity information and the object information semanteme as the first entities directly associated with the target multimedia object.

In an embodiment of the present invention, semantic analysis may be performed on the object information, semantic analysis may be performed on all or part of the entity information of each entity, semantics of part or all of the entity information of each entity are respectively compared with semantics of the object information, and an entity having the same semantics of all or part of the entity information and the object information is determined as the first entity.

Specifically, the semantic analysis may be performed in a manner in the prior art, which is not limited in the embodiment of the present invention.

The process of determining the first entity by means of step B above may be referred to as an unstructured matching process.

S102B: and determining a second entity which directly and/or indirectly has an association relation with the first entity according to the association relation among the entities in the knowledge graph.

In an embodiment of the present invention, an entity that has an association relationship with the first entity directly in the knowledge-graph, that is, an edge exists directly between the first entity and the knowledge-graph may be used as the second entity.

The second entity may also be an entity that is indirectly associated with the first entity in the knowledge-graph, that is, an entity that has an edge with the first entity through another entity.

Specifically, other entities having an association relationship directly with the first entity may be determined as the second entity.

Further, on the basis of the determined second entity, an entity, other than the determined second entity and the first entity, having an association relationship directly with the determined second entity is used as a new second entity. And determining a second entity indirectly having an association relation with the target multimedia object by analogy.

In addition, the distances between different second entities and the target multimedia object are different, that is, the minimum number of entities through which the association relationship indirectly exists between different second entities and the target multimedia object is different, for example, if the association relationship directly exists between the target multimedia object and the entity X and does not directly exist between the target multimedia object and the entity Y, but the association relationship directly exists between the entity X and the entity Y, and the association relationship indirectly exists between the entity Y and the target multimedia object through 1 entity X, the distance between the entity Y and the target multimedia object is 1.

The larger the distance between the second entity and the target multimedia object is, the smaller the degree of association between the second entity and the target multimedia object is, so that in the embodiment of the present invention, an entity whose distance between the second entity and the target multimedia object is smaller than the preset distance can be selected as the second entity from the second entities indirectly associated with the target multimedia object, and thus, the degree of association between the second entity and the target multimedia object obtained by determination can be improved.

S102C: and obtaining a target entity comprising the first entity and the second entity.

As can be seen from the above description, since the first entity has an association relationship directly with the target multimedia object and the second entity has an association relationship directly and/or indirectly with the first entity, the second entity has an association relationship indirectly with the target multimedia object through the first entity, and thus both the first entity and the second entity can be target entities having an association relationship directly and/or indirectly with the target multimedia object.

In another embodiment of the present invention, the above step S102B can be realized by the following steps C to D.

And C: and selecting the incidence relation directly pointing to the first entity and/or indirectly pointing to the first entity from the incidence relations belonging to the preset relation type among the entities in the knowledge graph.

Wherein, the preset relationship type may be: the entity information representing two entities with the incidence relation represents synonymy relation types with the same meaning, one entity representing the incidence relation is an upper relation type of a higher concept of the other entity, one entity representing the incidence relation belongs to the belonging relation type of the other entity, and the like.

Specifically, only the association relationship belonging to the preset relationship type in the knowledge graph may be retained, and the association relationship directly pointing to the first entity and/or indirectly pointing to the first entity may be selected from the retained association relationships.

Wherein the relationship directly pointing to the first entity is: one side of the incidence relation is the incidence relation of the first entity. The relationship indirectly pointing to the first entity is: and directing the incidence relation of the first entity through other entities.

Step D: and determining a second entity which has direct and/or indirect incidence relation with the first entity according to the selected incidence relation.

Specifically, the association relationship between the second entity and the first entity is the selected association relationship, so the determined association relationship between the second entity and the first entity is the association relationship of the preset type, and the association relationship between the second entity and the target multimedia object is the association relationship of the preset type.

As can be seen from the above, since there are many types of association relationships between entities, entity information of an entity that does not have any association relationship with a target multimedia object is applicable to an object tag as the target multimedia object. The association between the second entity and the target multimedia object determined by the scheme provided by the embodiment of the invention is the association of the preset relationship type, so that the matching degree between the obtained object tag and the target multimedia object is improved.

In still another embodiment of the present invention, the above step S102B may be realized by the following steps E to F.

Step E: and selecting the incidence relation which has the relation weight larger than the preset weight and directly points to the first entity and/or indirectly points to the first entity from the incidence relations among the entities in the knowledge graph.

Specifically, only the association relationship with the relationship weight greater than the preset weight in the knowledge graph may be retained, and the association relationship directly pointing to the first entity and/or indirectly pointing to the first entity may be selected from the retained association relationship.

Step F: and determining a second entity which has direct and/or indirect incidence relation with the first entity according to the selected incidence relation.

Specifically, the association between the second entity and the first entity is the selected association, so that the determined relationship weight of the association between the second entity and the first entity is greater than the preset weight, and therefore, the relationship weight of the association between the second entity and the target multimedia object is greater than the preset weight, and the association degree between the second entity and the target multimedia object is higher.

As can be seen from the above, the degree of association between the second entity and the target multimedia object selected by the embodiment of the present invention is higher, so that the second entity with the higher degree of association is used as the target entity, and the obtained entity information of the target entity is used as the object tag, so that the degree of association between the obtained object tag and the target multimedia object is higher, and the obtained object tag is more accurate.

Referring to fig. 4, a flowchart of a third method for determining an object tag according to an embodiment of the present invention is shown.

The target multimedia object data and the entity data in the knowledge graph are used as source data. The object tag of the target multimedia object data may be determined through an offline task when the target multimedia object data is history data, and may be determined through a real-time task when the target multimedia object data is newly generated data.

Specifically, the offline task and the real-time task determine the object tag in the same manner. The steps C-D and/or the steps E-F are realized through an analysis module, so that the association relation in the knowledge graph is selected. And selecting entities in the knowledge graph through the analysis module. And obtaining the target entity from the selected knowledge graph through the graph background. Thereby obtaining an object label and obtaining a label result data table.

The label result data table obtained through the offline task can be output in the form of offline data, the label result data table obtained through the real-time task can be output in the form of real-time information, and query service can be provided for a user through the label result data table.

Corresponding to the foregoing object tag determining method, referring to fig. 5, a schematic structural diagram of a first object tag determining apparatus provided in an embodiment of the present invention is shown, where the apparatus includes:

an information determining module 501, configured to determine object information for describing a target multimedia object;

an entity determining module 502, configured to determine, in the knowledge graph, a target entity having an association relationship with the target multimedia object according to the object information and entity information of an entity in a pre-constructed knowledge graph for an application scenario, where the target entity includes: an entity directly associated with the target multimedia object and/or an entity indirectly associated with the target multimedia object, wherein the entity in the application scene comprises: multimedia objects and/or associated information of multimedia objects in the application scene;

a tag selecting module 503, configured to select all or part of the entity information of the target entity as an object tag of the target multimedia object.

As can be seen from the above description, since the target entity is an entity that has an association relationship with the target multimedia object directly or indirectly, the entity information of the target entity may be considered to have an association relationship with the target multimedia object and may be used to describe the target multimedia object, and therefore, the entity information of the target entity may be used as an object tag of the target multimedia object. And determining the entity information of the target entity as the object tag of the target multimedia object, so that the number of the object tags can be increased, and the richness of the object tags can be improved.

Referring to fig. 6, which is a schematic structural diagram of a second object tag determination apparatus according to an embodiment of the present invention, the entity determination module 502 includes:

a first entity determining sub-module 502A, configured to determine, according to the object information and entity information of an entity in a knowledge graph, a first entity in the knowledge graph, where the first entity directly has an association relationship with the target multimedia object;

the second entity determining submodule 502B is configured to determine, according to an association relationship between entities in the knowledge graph, a second entity having an association relationship with the first entity directly and/or indirectly;

the entity obtaining sub-module 502C is configured to obtain a target entity including the first entity and the second entity.

As can be seen from the above description, since the first entity has an association relationship directly with the target multimedia object and the second entity has an association relationship directly and/or indirectly with the first entity, the second entity has an association relationship indirectly with the target multimedia object through the first entity, and thus both the first entity and the second entity can be target entities having an association relationship directly and/or indirectly with the target multimedia object.

In an embodiment of the present invention, the first entity determining sub-module 502A is specifically configured to:

in the knowledge graph, determining an entity of which all or part of entity information is contained in the object information as a first entity directly associated with the target multimedia object;

and/or

And in the knowledge graph, determining all or part of entities with the same entity information and object information semanteme as a first entity directly associated with the target multimedia object.

In an embodiment of the present invention, the second entity determining sub-module 502B is specifically configured to:

selecting an incidence relation directly pointing to the first entity and/or indirectly pointing to the first entity from incidence relations belonging to a preset relation type among the entities in the knowledge graph;

and determining a second entity which has direct and/or indirect incidence relation with the first entity according to the selected incidence relation.

As can be seen from the above, since there are many types of association relationships between entities, entity information of an entity that does not have any association relationship with a target multimedia object is applicable to an object tag as the target multimedia object. The association between the second entity and the target multimedia object determined by the scheme provided by the embodiment of the invention is the association of the preset relationship type, so that the matching degree between the obtained object tag and the target multimedia object is improved.

In an embodiment of the present invention, the second entity determining sub-module 502B is specifically configured to:

selecting incidence relations with relation weights larger than preset weights and directly pointing to the first entity and/or indirectly pointing to the first entity from incidence relations among entities in the knowledge graph;

and determining a second entity which has direct and/or indirect incidence relation with the first entity according to the selected incidence relation.

As can be seen from the above, the degree of association between the second entity and the target multimedia object selected by the embodiment of the present invention is higher, so that the second entity with the higher degree of association is used as the target entity, and the obtained entity information of the target entity is used as the object tag, so that the degree of association between the obtained object tag and the target multimedia object is higher, and the obtained object tag is more accurate.

An embodiment of the present invention further provides an electronic device, as shown in fig. 7, including a processor 701, a communication interface 702, a memory 703 and a communication bus 704, where the processor 701, the communication interface 702, and the memory 703 complete mutual communication through the communication bus 704,

a memory 703 for storing a computer program;

the processor 701 is configured to implement the method steps of any of the above object tag determination methods when executing the program stored in the memory 703.

When the electronic device provided by the embodiment of the present invention is applied to determine an object tag, since the target entity is an entity that directly or indirectly has an association relationship with the target multimedia object, it can be considered that the entity information of the target entity has an association relationship with the target multimedia object and can be used to describe the target multimedia object, and therefore, the entity information of the target entity can be used as the object tag of the target multimedia object. And determining the entity information of the target entity as the object tag of the target multimedia object, so that the number of the object tags can be increased, and the richness of the object tags can be improved.

The communication bus mentioned in the above terminal may be a Peripheral Component Interconnect (PCI) bus, an Extended Industry Standard Architecture (EISA) bus, or the like. The communication bus may be divided into an address bus, a data bus, a control bus, etc. For ease of illustration, only one thick line is shown, but this does not mean that there is only one bus or one type of bus.

The communication interface is used for communication between the terminal and other equipment.

The Memory may include a Random Access Memory (RAM) or a non-volatile Memory (non-volatile Memory), such as at least one disk Memory. Optionally, the memory may also be at least one memory device located remotely from the processor.

The Processor may be a general-purpose Processor, and includes a Central Processing Unit (CPU), a Network Processor (NP), and the like; the device can also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other Programmable logic device, a discrete Gate or transistor logic device, or a discrete hardware component.

In yet another embodiment of the present invention, a computer-readable storage medium is further provided, in which a computer program is stored, and the computer program, when executed by a processor, implements the object tag determination method described in any of the above embodiments.

When determining an object tag by applying the computer program stored in the computer-readable storage medium provided in this embodiment, since the target entity is an entity that has an association relationship with the target multimedia object directly or indirectly, it can be considered that the entity information of the target entity has an association relationship with the target multimedia object and can be used to describe the target multimedia object, and therefore, the entity information of the target entity can be used as the object tag of the target multimedia object. And determining the entity information of the target entity as the object tag of the target multimedia object, so that the number of the object tags can be increased, and the richness of the object tags can be improved.

In a further embodiment of the present invention, there is also provided a computer program product containing instructions which, when run on a computer, cause the computer to perform the object label determination method as described in any of the above embodiments.

When the computer program product provided by this embodiment is executed to determine the object tag, since the target entity is an entity that has an association relationship directly or indirectly with the target multimedia object, it can be considered that the entity information of the target entity has an association relationship with the target multimedia object and can be used to describe the target multimedia object, and therefore the entity information of the target entity can be used as the object tag of the target multimedia object. And determining the entity information of the target entity as the object tag of the target multimedia object, so that the number of the object tags can be increased, and the richness of the object tags can be improved.

In the above embodiments, the implementation may be wholly or partially realized by software, hardware, firmware, or any combination thereof. When implemented in software, may be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When loaded and executed on a computer, cause the processes or functions described in accordance with the embodiments of the invention to occur, in whole or in part. The computer may be a general purpose computer, a special purpose computer, a network of computers, or other programmable device. The computer instructions may be stored in a computer readable storage medium or transmitted from one computer readable storage medium to another, for example, from one website site, computer, server, or data center to another website site, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, Digital Subscriber Line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device, such as a server, a data center, etc., that incorporates one or more of the available media. The usable medium may be a magnetic medium (e.g., floppy Disk, hard Disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., Solid State Disk (SSD)), among others.

It is noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.

All the embodiments in the present specification are described in a related manner, and the same and similar parts among the embodiments may be referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, as for the apparatus, the electronic device, the computer-readable storage medium and the computer program product, since they are substantially similar to the method embodiments, the description is relatively simple, and in relation to them, reference may be made to the partial description of the method embodiments.

The above description is only for the preferred embodiment of the present invention, and is not intended to limit the scope of the present invention. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention shall fall within the protection scope of the present invention.