阅读说明：本技术 数据标注方法、装置、电子设备及计算机可读存储介质 (Data annotation method and device, electronic equipment and computer readable storage medium ) 是由 徐晨 于 2019-02-25 设计创作，主要内容包括：本申请提供了一种数据标注方法、装置、电子设备及计算机可读存储介质,其中,该方法包括：获取至少一个列表信息,每个列表信息中包括至少一个对象,且至少一个列表信息中部分或者全部列表信息携带一个或多个主题信息；基于列表信息确定与各个第一对象相关联的第二对象,第一对象为至少一个列表信息中未标注标签信息的对象；基于第二对象的标签信息确定各个第一对象的标签信息。本申请通过列表信息中已标注标签信息的对象,对列表信息中未标注标签信息的对象进行标注的方式,能够更加准确的确定出列表信息中各个对象的标签信息,进而,通过该标签信息就能够对列表信息中的各个对象进行相应的学习分类。(The application provides a data annotation method, a data annotation device, electronic equipment and a computer-readable storage medium, wherein the method comprises the following steps: acquiring at least one piece of list information, wherein each piece of list information comprises at least one object, and part or all of the at least one piece of list information carries one or more pieces of subject information; determining second objects associated with the first objects based on the list information, wherein the first objects are objects which are not labeled with label information in the at least one list information; tag information of the respective first objects is determined based on tag information of the second objects. According to the method and the device, the label information of each object in the list information can be more accurately determined by labeling the object labeled with the label information in the list information, and then, corresponding learning classification can be carried out on each object in the list information through the label information.)

1. A method for annotating data, comprising:

acquiring at least one piece of list information, wherein each piece of list information comprises at least one object, and part or all of the at least one piece of list information carries one or more pieces of subject information;

determining second objects associated with the first objects based on the list information, wherein the first objects are objects which are not labeled with label information in the at least one list information, the number of the second objects is one or more, and the second objects comprise objects labeled with label information;

determining tag information of each of the first objects based on tag information of the second object.

2. The method of claim 1, wherein determining a second object associated with a first object based on the list information comprises:

determining topic distribution information based on the list information, wherein the topic distribution information represents distribution information of each object in the list information in the one or more implicit topics, and the implicit topics are implicit topics contained in the at least one list information;

determining a second object associated with the first object based on the topic distribution information.

3. The method of claim 2, wherein determining topic distribution information based on the list information comprises:

inputting the list information into a text topic model for processing;

and taking the processing result as the theme distribution information.

4. The method of claim 2, wherein the topic distribution information is a target vector group, the target vector group comprises at least one target vector, and one object in the list information corresponds to one target vector, and each of the target vectors represents a probability value of an object belonging to each implicit topic.

5. The method of claim 4, wherein determining a second object associated with the first object based on the topic distribution information comprises:

determining a first target vector corresponding to the first object in the target vector group;

calculating similarity between the first target vector and a second target vector, wherein the second target vector is other target vectors except the first target vector;

determining a second object associated with the first object according to the similarity.

6. The method of claim 5, wherein the similarity is multiple, and wherein determining the second object associated with the first object based on the similarity comprises:

determining a second target vector corresponding to a target similarity in the similarities, wherein the target similarity is a similarity which is greater than a preset similarity in the similarities;

and determining the object to which the second target vector corresponding to the target similarity belongs as the second object.

7. The method of claim 1, wherein determining the tag information for each of the first objects based on the tag information for the second object comprises:

repeatedly executing the following steps until the label information of each object in the list information is marked:

determining labeled objects in second objects associated with the first objects Ai, wherein the labeled objects are objects labeled with label information, I sequentially takes 1 to I, and I is the number of the first objects in the list information;

and determining the label information of the first object Ai by using the label information of the labeled object.

8. The method of claim 7, wherein determining tag information for the first object Ai using the tag information for the tagged objects comprises:

judging whether the number of the marked objects in the second object exceeds a preset number value or not;

if yes, determining the label information of the first object Ai by using the label information of the labeled object.

9. The method of claim 7 or 8, wherein the tagged objects are a plurality of objects, and wherein determining the tag information of the first object Ai using the tag information of the tagged objects comprises:

based on the similarity, sequencing the marked objects according to a preset sequencing sequence to obtain a sequencing result;

calculating the weight of the labeled object based on the sorting result;

determining the weight as the weight of the label information corresponding to the labeled object;

calculating the sum of the weights of the label information of the same type in the label information corresponding to the plurality of labeled objects to obtain the weight values of various types of label information;

determining label information of the first object Ai based on the weight values of the various types of label information.

10. The method of claim 9, wherein determining the tag information of the first object Ai based on the weight values of the various types of tag information comprises:

and determining label information corresponding to a target weight value in the weight values of the various types of label information as the label information of the first object Ai, wherein the target weight value is a numerical value which is greater than a preset weight value in the weight values of the various types of label information.

11. The method of claim 9, wherein computing the weight of the labeled object based on the ranking result comprises:

and calculating the weight of the labeled object by using a formula W/(logR +1), wherein W is the weight, and R is the ranking value of the labeled object in the ranking result.

12. The method of claim 1, wherein after obtaining at least one list information, the method further comprises:

determining target list information containing objects Bj in the at least one list information, wherein i sequentially takes 1 to J, and J is the number of objects in the at least one list information;

and labeling the object Bj based on the subject information of the target list information, and taking the object labeled with the label information as a labeled object.

13. The method of claim 12, wherein determining the target list information containing the object Bj in the at least one list information comprises:

determining list information carrying subject information in the at least one list information;

and determining target list information containing the object Bj in the list information carrying the subject information.

14. The method of claim 13, wherein labeling the object Bj based on the subject information of the target list information comprises:

counting the occurrence times of each topic information carried by the target list information in the target list information;

and marking the object Bj based on the occurrence times.

15. The method of claim 14, wherein labeling the object Bj based on the number of occurrences comprises:

if the occurrence times comprise target occurrence times, determining the subject information corresponding to the target occurrence times as the label information of the object Bj, and determining the object Bj as a labeled object, wherein the target occurrence times are the occurrence times of which the occurrence times are greater than or equal to a preset threshold value.

16. A data annotation device, comprising:

the system comprises an acquisition unit, a processing unit and a processing unit, wherein the acquisition unit is used for acquiring at least one piece of list information, each piece of list information comprises at least one object, and part or all of the at least one piece of list information carries one or more pieces of subject information;

a first determining unit, configured to determine, based on the list information, second objects associated with respective first objects, where the first objects are objects that are not labeled with tag information in the at least one list information, the number of the second objects is one or more, and the second objects include objects labeled with tag information;

a second determining unit configured to determine tag information of each of the first objects based on tag information of the second object.

17. The apparatus of claim 16, wherein the first determining unit comprises:

a first determining module, configured to determine topic distribution information based on the list information, where the topic distribution information represents distribution information of each object in the list information in the one or more implied topics, and the implied topics are implied topics included in the at least one list information;

a second determination module to determine a second object associated with the first object based on the topic distribution information.

18. The apparatus of claim 17, wherein the first determining module is configured to:

inputting the list information into a text topic model for processing;

and taking the processing result as the theme distribution information.

19. The apparatus of claim 17, wherein the topic distribution information is a target vector group, the target vector group comprises at least one target vector, and one object in the list information corresponds to one target vector, and each of the target vectors represents a probability value of an object belonging to each implied topic.

20. The apparatus of claim 19, wherein the second determining module is configured to:

determining a first target vector corresponding to the first object in the target vector group;

calculating similarity between the first target vector and a second target vector, wherein the second target vector is other target vectors except the first target vector;

determining a second object associated with the first object according to the similarity.

21. The apparatus of claim 20, wherein the similarity is multiple, and wherein the second determining module is further configured to:

determining a second target vector corresponding to a target similarity in the similarities, wherein the target similarity is a similarity which is greater than a preset similarity in the similarities;

and determining the object to which the second target vector corresponding to the target similarity belongs as the second object.

22. The apparatus of claim 16, wherein the second determining unit comprises:

repeatedly executing the following steps by using a third determining module and a fourth determining module until label information of each object in the list information is marked:

the third determining module is configured to determine a labeled object in a second object associated with the first object Ai, where the labeled object is an object labeled with label information, I sequentially takes 1 to I, and I is the number of the first objects in the list information;

the fourth determining module is configured to determine the tag information of the first object Ai by using the tag information of the labeled object.

23. The apparatus of claim 22, wherein the fourth determining module is further configured to:

judging whether the number of the marked objects in the second object exceeds a preset number value or not;

if yes, determining the label information of the first object Ai by using the label information of the labeled object.

24. The apparatus according to claim 22 or 23, wherein the labeled objects are plural, and the fourth determining module is further configured to:

based on the similarity, sequencing the marked objects according to a preset sequencing sequence to obtain a sequencing result;

calculating the weight of the labeled object based on the sorting result;

determining the weight as the weight of the label information corresponding to the labeled object;

calculating the sum of the weights of the label information of the same type in the label information corresponding to the plurality of labeled objects to obtain the weight values of various types of label information;

determining label information of the first object Ai based on the weight values of the various types of label information.

25. The apparatus of claim 24, wherein the fourth determining module is further configured to:

and determining label information corresponding to a target weight value in the weight values of the various types of label information as the label information of the first object Ai, wherein the target weight value is a numerical value which is greater than a preset weight value in the weight values of the various types of label information.

26. The apparatus of claim 24, wherein the fourth determining module is further configured to:

and calculating the weight of the labeled object by using a formula W/(logR +1), wherein W is the weight, and R is the ranking value of the labeled object in the ranking result.

27. The apparatus of claim 16, wherein after obtaining at least one list information, the apparatus further comprises:

a third determining unit, configured to determine, in the at least one piece of list information, target list information including an object Bj, where i sequentially takes 1 to J, and J is the number of objects in the at least one piece of list information;

and the labeling unit is used for labeling the object Bj based on the subject information of the target list information and taking the object labeled with the label information as a labeled object.

28. The apparatus of claim 27, wherein the third determining unit is configured to:

determining list information carrying subject information in the at least one list information;

and determining target list information containing the object Bj in the list information carrying the subject information.

29. The apparatus of claim 28, wherein the labeling unit is configured to:

counting the occurrence times of each topic information carried by the target list information in the target list information;

and marking the object Bj based on the occurrence times.

30. The apparatus of claim 29, wherein the labeling unit is further configured to:

if the occurrence times comprise target occurrence times, determining the subject information corresponding to the target occurrence times as the label information of the object Bj, and determining the object Bj as a labeled object, wherein the target occurrence times are the occurrence times of which the occurrence times are greater than or equal to a preset threshold value.

31. An electronic device, comprising: a processor, a storage medium and a bus, the storage medium storing machine-readable instructions executable by the processor, the processor and the storage medium communicating via the bus when the electronic device is running, the processor executing the machine-readable instructions to perform the steps of the data annotation method according to any one of claims 1 to 15.

32. A computer-readable storage medium, having stored thereon a computer program for performing, when executed by a processor, the steps of the data annotation method according to any one of claims 1 to 15.

Technical Field

The present application relates to the technical field of data processing, and in particular, to a data annotation method, an apparatus, an electronic device, and a computer-readable storage medium.

Background

Currently, for list data, in an existing data labeling method, labeling of each list item in the list data may be implemented by labeling the list data, and generally, the list items in the list data do not have complete attribute information or label information, such as a song list. When the user creates the song list, the style and type information of the song list can be marked. But lacks style label information for the songs in the menu. If a song (i.e., the list item described above) lacks corresponding style label information, then the song (i.e., the list item) will not be able to be correspondingly learned for classification.

Disclosure of Invention

In view of the above, embodiments of the present application provide a data labeling method, an apparatus, an electronic device, and a computer-readable storage medium, in which label information of each object in list information can be more accurately determined by labeling an object labeled with label information in the list information with an object not labeled with label information in the list information, and thus each object in the list information can be correspondingly learned and classified by the label information.

According to one aspect of the present application, an electronic device is provided that may include one or more storage media and one or more processors in communication with the storage media. One or more storage media store machine-readable instructions executable by a processor. When the electronic device is operated, the processor communicates with the storage medium through the bus, and the processor executes the machine readable instructions to perform one or more of the following operations:

acquiring at least one piece of list information, wherein each piece of list information comprises at least one object, and part or all of the at least one piece of list information carries one or more pieces of subject information; determining second objects associated with the first objects based on the list information, wherein the first objects are objects which are not labeled with label information in the at least one list information, the number of the second objects is one or more, and the second objects comprise objects labeled with label information; determining tag information of each of the first objects based on tag information of the second object.

In a preferred embodiment of the present application, determining the second object associated with the first object based on the list information comprises: determining topic distribution information based on the list information, wherein the topic distribution information represents distribution information of each object in the list information in the one or more implicit topics, and the implicit topics are implicit topics contained in the at least one list information; determining a second object associated with the first object based on the topic distribution information.

In a preferred embodiment of the present application, determining topic distribution information based on the list information comprises: inputting the list information into a text topic model for processing; and taking the processing result as the theme distribution information.

In a preferred embodiment of the present application, the topic distribution information is a target vector group, the target vector group includes at least one target vector, and one object in the list information corresponds to one target vector, and each of the target vectors represents a probability value that the object belongs to each implicit topic.

In a preferred embodiment of the present application, determining the second object associated with the first object based on the topic distribution information comprises: determining a first target vector corresponding to the first object in the target vector group; calculating similarity between the first target vector and a second target vector, wherein the second target vector is other target vectors except the first target vector; determining a second object associated with the first object according to the similarity.

In a preferred embodiment of the present application, the similarity is multiple, and determining the second object associated with the first object according to the similarity includes: determining a second target vector corresponding to a target similarity in the similarities, wherein the target similarity is a similarity which is greater than a preset similarity in the similarities; and determining the object to which the second target vector corresponding to the target similarity belongs as the second object.

In a preferred embodiment of the present application, determining the tag information of each of the first objects based on the tag information of the second object includes: repeatedly executing the following steps until label information of each object in the list information is marked: determining labeled objects in second objects associated with the first objects Ai, wherein the labeled objects are objects labeled with label information, I sequentially takes 1 to I, and I is the number of the first objects in the list information; and determining the label information of the first object Ai by using the label information of the labeled object.

In a preferred embodiment of the present application, determining the tag information of the first object Ai using the tag information of the labeled object includes: judging whether the number of the marked objects in the second object exceeds a preset number value or not; if yes, determining the label information of the first object Ai by using the label information of the labeled object.

In a preferred embodiment of the present application, the labeled objects are multiple objects, and determining the label information of the first object Ai by using the label information of the labeled objects includes: based on the similarity, sequencing the marked objects according to a preset sequencing sequence to obtain a sequencing result; calculating the weight of the labeled object based on the sorting result; determining the weight as the weight of the label information corresponding to the labeled object; calculating the sum of the weights of the label information of the same type in the label information corresponding to the plurality of labeled objects to obtain the weight values of various types of label information; determining label information of the first object Ai based on the weight values of the various types of label information.

In a preferred embodiment of the present application, determining the tag information of the first object Ai based on the weight values of the various types of tag information includes: and determining label information corresponding to a target weight value in the weight values of the various types of label information as the label information of the first object Ai, wherein the target weight value is a numerical value which is greater than a preset weight value in the weight values of the various types of label information.

In a preferred embodiment of the present application, calculating the weight of the labeled object based on the ranking result comprises: and calculating the weight of the labeled object by using a formula W/(logR +1), wherein W is the weight, and R is the ranking value of the labeled object in the ranking result.

In a preferred embodiment of the present application, after obtaining at least one list information, the method further includes: determining target list information containing objects Bj in the at least one list information, wherein i sequentially takes 1 to J, and J is the number of objects in the at least one list information; and labeling the object Bj based on the subject information of the target list information, and taking the object labeled with the label information as a labeled object.

In a preferred embodiment of the present application, determining the target list information including the object Bj in the at least one list information includes: determining list information carrying subject information in the at least one list information; and determining target list information containing the object Bj in the list information carrying the subject information.

In a preferred embodiment of the present application, labeling the object Bj based on the subject information of the target list information includes: counting the occurrence times of each topic information carried by the target list information in the target list information; and marking the object Bj based on the occurrence times.

In a preferred embodiment of the present application, labeling the object Bj based on the occurrence number includes: if the occurrence times comprise target occurrence times, determining the subject information corresponding to the target occurrence times as the label information of the object Bj, and determining the object Bj as a labeled object, wherein the target occurrence times are the occurrence times of which the occurrence times are greater than or equal to a preset threshold value.

According to another aspect of the present application, there is also provided a data annotation device, including: the system comprises an acquisition unit, a processing unit and a processing unit, wherein the acquisition unit is used for acquiring at least one piece of list information, each piece of list information comprises at least one object, and part or all of the at least one piece of list information carries one or more pieces of subject information; a first determining unit, configured to determine, based on the list information, second objects associated with respective first objects, where the first objects are objects that are not labeled with tag information in the at least one list information, the number of the second objects is one or more, and the second objects include objects labeled with tag information; a second determining unit configured to determine tag information of each of the first objects based on tag information of the second object.

In a preferred embodiment of the present application, the first determining unit includes: a first determining module, configured to determine topic distribution information based on the list information, where the topic distribution information represents distribution information of each object in the list information in the one or more implied topics, and the implied topics are implied topics included in the at least one list information; a second determination module to determine a second object associated with the first object based on the topic distribution information.

In a preferred embodiment of the present application, the first determining module is configured to: inputting the list information into a text topic model for processing; and taking the processing result as the theme distribution information.

In a preferred embodiment of the present application, the topic distribution information is a target vector group, the target vector group includes at least one target vector, and one object in the list information corresponds to one target vector, and each of the target vectors represents a probability value that the object belongs to each implicit topic.

In a preferred embodiment of the present application, the second determining module is configured to: determining a first target vector corresponding to the first object in the target vector group; calculating similarity between the first target vector and a second target vector, wherein the second target vector is other target vectors except the first target vector; determining a second object associated with the first object according to the similarity.

In a preferred embodiment of the present application, the similarity is multiple, and the second determining module is further configured to: determining a second target vector corresponding to a target similarity in the similarities, wherein the target similarity is a similarity which is greater than a preset similarity in the similarities; and determining the object to which the second target vector corresponding to the target similarity belongs as the second object.

In a preferred embodiment of the present application, the second determining unit includes: repeatedly executing the following steps by using a third determining module and a fourth determining module until label information of each object in the list information is marked: the third determining module is configured to determine a labeled object in a second object associated with the first object Ai, where the labeled object is an object labeled with label information, I sequentially takes 1 to I, and I is the number of the first objects in the list information; the fourth determining module is configured to determine the tag information of the first object Ai by using the tag information of the labeled object.

In a preferred embodiment of the present application, the fourth determining module is further configured to: judging whether the number of the marked objects in the second object exceeds a preset number value or not; if yes, determining the label information of the first object Ai by using the label information of the labeled object.

In a preferred embodiment of the present application, the number of labeled objects is multiple, and the fourth determining module is further configured to: based on the similarity, sequencing the marked objects according to a preset sequencing sequence to obtain a sequencing result; calculating the weight of the labeled object based on the sorting result; determining the weight as the weight of the label information corresponding to the labeled object; calculating the sum of the weights of the label information of the same type in the label information corresponding to the plurality of labeled objects to obtain the weight values of various types of label information; determining label information of the first object Ai based on the weight values of the various types of label information.

In a preferred embodiment of the present application, the fourth determining module is further configured to: and determining label information corresponding to a target weight value in the weight values of the various types of label information as the label information of the first object Ai, wherein the target weight value is a numerical value which is greater than a preset weight value in the weight values of the various types of label information.

In a preferred embodiment of the present application, the fourth determining module is further configured to: and calculating the weight of the labeled object by using a formula W/(logR +1), wherein W is the weight, and R is the ranking value of the labeled object in the ranking result.

In a preferred embodiment of the present application, after obtaining at least one list information, the apparatus further includes: a third determining unit, configured to determine, in the at least one piece of list information, target list information including an object Bj, where i sequentially takes 1 to J, and J is the number of objects in the at least one piece of list information; and the labeling unit is used for labeling the object Bj based on the subject information of the target list information and taking the object labeled with the label information as a labeled object.

In a preferred embodiment of the present application, the third determining unit is configured to: determining list information carrying subject information in the at least one list information; and determining target list information containing the object Bj in the list information carrying the subject information.

In a preferred embodiment of the present application, the labeling unit is configured to: counting the occurrence times of each topic information carried by the target list information in the target list information; and marking the object Bj based on the occurrence times.

In a preferred embodiment of the present application, the labeling unit is further configured to: if the occurrence times comprise target occurrence times, determining the subject information corresponding to the target occurrence times as the label information of the object Bj, and determining the object Bj as a labeled object, wherein the target occurrence times are the occurrence times of which the occurrence times are greater than or equal to a preset threshold value.

According to another aspect of the present application, there is also provided an electronic device including: the data annotation device comprises a processor, a storage medium and a bus, wherein the storage medium stores machine-readable instructions executable by the processor, when the electronic device runs, the processor and the storage medium communicate through the bus, and the processor executes the machine-readable instructions to execute the steps of any one of the data annotation methods.

According to another aspect of the present application, there is also provided a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, performs the steps of any of the above-described methods of data annotation.

In this embodiment, at least one list information is first obtained, where each list information includes at least one object, and a part or all of the at least one list information carries one or more topic information; thereafter, second objects associated with the respective first objects may be determined based on the list information; finally, tag information of each of the first objects is determined based on tag information of a second object. According to the above description, the label information of each object in the list information can be more accurately determined by labeling the object labeled with the label information in the list information, and then, corresponding learning classification can be performed on each object in the list information through the label information.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are required to be used in the embodiments will be briefly described below, it should be understood that the following drawings only illustrate some embodiments of the present application and therefore should not be considered as limiting the scope, and for those skilled in the art, other related drawings can be obtained from the drawings without inventive effort.

FIG. 1 is a schematic diagram of an electronic device provided by an embodiment of the present application;

FIG. 2 is a flow chart of a data annotation method provided in an embodiment of the present application;

FIG. 3 is a flow chart illustrating a first alternative data annotation method provided by an embodiment of the present application;

FIG. 4 is a flow chart illustrating a second alternative data annotation method provided by the embodiments of the present application;

FIG. 5 is a flow chart of a third alternative data annotation method provided in the embodiments of the present application;

FIG. 6 is a flow chart of a fourth alternative data annotation method provided in the embodiments of the present application;

FIG. 7 is a flow chart illustrating a fifth alternative data annotation method provided in the embodiments of the present application;

fig. 8 shows a schematic diagram of a data annotation device provided in an embodiment of the present application.

Detailed Description

In order to make the purpose, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it should be understood that the drawings in the present application are for illustrative and descriptive purposes only and are not used to limit the scope of protection of the present application. Additionally, it should be understood that the schematic drawings are not necessarily drawn to scale. The flowcharts used in this application illustrate operations implemented according to some embodiments of the present application. It should be understood that the operations of the flow diagrams may be performed out of order, and steps without logical context may be performed in reverse order or simultaneously. One skilled in the art, under the guidance of this application, may add one or more other operations to, or remove one or more operations from, the flowchart.

In addition, the described embodiments are only a part of the embodiments of the present application, and not all of the embodiments. The components of the embodiments of the present application, generally described and illustrated in the figures herein, can be arranged and designed in a wide variety of different configurations. Thus, the following detailed description of the embodiments of the present application, presented in the accompanying drawings, is not intended to limit the scope of the claimed application, but is merely representative of selected embodiments of the application. All other embodiments, which can be derived by a person skilled in the art from the embodiments of the present application without making any creative effort, shall fall within the protection scope of the present application.

It should be noted that in the embodiments of the present application, the term "comprising" is used to indicate the presence of the features stated hereinafter, but does not exclude the addition of further features.

The terms "service" and "order" are used interchangeably herein to refer to a service request initiated by a passenger, a service requester, a driver, a service provider, a supplier, or the like, or any combination thereof. Accepting the "service" or "order" may be a passenger, a service requester, a driver, a service provider, a supplier, or the like, or any combination thereof. The service may be charged or free.

FIG. 1 illustrates a schematic diagram of exemplary hardware and software components of an electronic device 100 that may implement the data annotation methods provided herein, according to some embodiments of the present application.

The electronic device 100 may be a general purpose computer or a special purpose computer, both of which may be used to implement the data annotation methods of the present application. Although only a single computer is shown, for convenience, the functions described herein may be implemented in a distributed fashion across multiple similar platforms to balance processing loads.

For example, the electronic device 100 may include a network port 110 connected to a network, one or more processors 120 for executing program instructions, a communication bus 130, and a storage medium 140 of different form, such as a disk, ROM, or RAM, or any combination thereof. Illustratively, the computer platform may also include program instructions stored in ROM, RAM, or other types of non-transitory storage media, or any combination thereof. The method of the present application may be implemented in accordance with these program instructions. The electronic device 100 also includes an Input/Output (I/O) interface 150 between the computer and other Input/Output devices (e.g., keyboard, display screen).

The storage medium 140 stores machine-readable instructions executable by the processor 120, when the electronic device is operated, the processor 120 communicates with the storage medium 140 via the bus, and the processor executes the machine-readable instructions to perform the following steps of the data annotation method. In addition, the storage medium may also be referred to as a computer-readable storage medium, on which a computer program is stored, which computer program, when being executed by a processor, performs the steps of the data annotation method described below.

For ease of illustration, only one processor is depicted in electronic device 100. However, it should be noted that the electronic device 100 in the present application may also comprise a plurality of processors, and thus the steps performed by one processor described in the present application may also be performed by a plurality of processors in combination or individually. For example, if the processor of the electronic device 100 executes steps a and B, it should be understood that steps a and B may also be executed by two different processors together or separately in one processor. For example, a first processor performs step a and a second processor performs step B, or the first processor and the second processor perform steps a and B together.

See fig. 2 for a flow chart of a method of data annotation.

The data annotation method shown in fig. 2 is described by taking an application as an example at a server, and the method includes the following steps:

step S202, at least one list information is obtained, each list information comprises at least one object, and part or all of the list information in the at least one list information carries one or more subject information;

in this embodiment, the list information may be a song list, may be a menu list, and the like, and the content of the list information is not particularly limited in this embodiment.

If the list information is a song list, the song list comprises at least one song; if the list information is a menu list, the menu list contains at least one dish name.

In this embodiment, when the at least one list information is multiple, a part of or all of the multiple list information carries one or more topic information. Wherein the subject information is used to determine style information of each listing information.

For example, if the list information is a song list, the subject information corresponding to the song list may be "sports" and "rock"; if the list information is a menu list, the theme information corresponding to the menu list can be "chuanwei" and "spicy" and the like.

Step S204, determining second objects associated with the respective first objects based on the list information, where the first objects are objects that are not labeled with tag information in the at least one list information, the number of the second objects is one or more, and the second objects include objects labeled with tag information;

it should be noted that, in the present embodiment, the list information includes an object (i.e., a first object) that is not labeled with tag information, and includes an object (i.e., a second object) associated with the first object.

For example, if the list information is a list of song sheets, and the list of song sheets includes songs "body surface" and "cool" where "cool" is an object (i.e., a first object) to which label information is not labeled and "body surface" is an object associated with the first object "cool". In the present embodiment, the association means that the degree of similarity between the first object and the second object is high, for example, the "cool and cool" and the "body and face" are both the emotional songs of injury.

Step S206, determining the label information of each first object based on the label information of the second object.

It should be noted that, in this embodiment, the second object includes an object labeled with tag information, and the second object may further include an object not labeled with tag information, and at this time, the tag information of each first object may be determined based on the object labeled with tag information in the second object.

In this embodiment, at least one list information is first obtained, where each list information includes at least one object, and a part or all of the at least one list information carries one or more topic information; thereafter, second objects associated with the respective first objects may be determined based on the list information; finally, tag information of each of the first objects is determined based on tag information of a second object. According to the above description, the label information of each object in the list information can be more accurately determined by labeling the object labeled with the label information in the list information, and then, corresponding learning classification can be performed on each object in the list information through the label information.

In an alternative embodiment, as shown in fig. 3, the step S204 of determining a second object associated with the first object based on the list information includes the steps of:

step S301, determining topic distribution information based on the list information, wherein the topic distribution information represents the distribution information of each object in the list information in the one or more implicit topics, and the implicit topics are the implicit topics contained in the at least one list information;

step S302, determining a second object associated with the first object based on the theme distribution information.

As can be seen from the above description, in this embodiment, at least one piece of list information is first obtained, and then, distribution information of each object in the list information in one or more implicit topics is determined based on the list information. Then, a second object associated with the first object may be determined based on the topic distribution information.

It is assumed that the list information is a list of songs, and the list of songs includes at least one song (i.e., object), at least one song includes a song with labeled information, and a song with unlabeled label information. Based on this, in this embodiment, topic distribution information of each song in one or more implicit topics in the song list may be determined based on the song list, in this case, the one or more implicit topics are implicit topics included in the song list. Songs associated with songs that are not tagged with tag information may then be determined based on the topic distribution information. Finally, labeling the tag information of the song which is not labeled with the tag information according to the song which is labeled with the tag information in the associated songs.

In this embodiment, the topic distribution information can represent a probability value of each object belonging to each implicit topic. Determining the manner of the second object associated with the first object based on the topic distribution information enables a more accurate second object. After the more accurate second object is obtained, when the second object is used for labeling the first object, a more accurate labeling result can be obtained.

Optionally, in this embodiment, in step S301, determining the topic distribution information based on the list information includes the following steps:

firstly, inputting the list information into a text topic model for processing;

then, the processing result is taken as the topic distribution information.

In this embodiment, at least one piece of list information is first obtained, and then, topic distribution information of each object in the list information in one or more implicit topics is determined based on the list information. When determining the topic distribution information, the list information may be input into a text topic model lda (content Dirichlet allocation) for processing, and the processing result is used as the topic distribution information.

In this embodiment, optionally, the topic distribution information may be a target vector group, where the target vector group includes at least one target vector, and one object in the list information corresponds to one target vector, and each vector in the target vectors represents a probability value that the object belongs to each implicit topic.

Assuming that the list information is a list of songs, the list of songs may be regarded as an article containing potential topic information, and each song may be regarded as a word, so that a list of songs may be regarded as a process of determining the topic information of an article, then sampling words from the topic information, and then generating an article. Therefore, in this embodiment, using the text topic model LDA, the whole amount of song list data (i.e., at least one song list), including the song lists with and without labels, is input into the text topic model LDA, and the topic distribution information of each song is trained, for example, the set implicit topic is 200. In this embodiment, the implied theme is not limited to be 200, and other numerical values may be selected, which is not specifically limited in this embodiment.

In this embodiment, the at least one song list may be expressed as:

Playlist 1:song1,song2,song3,song4……songN；

Playlist 2:song1,song2,song3,song4……songM。

here, Playlist 1 is indicated as a song list 1, and Playlist 2 is indicated as a song list 2. The song list 1 contains N songs, "song 1, song2, song3, and song4 … … song" respectively; the song list 1 contains M songs, "song 1, song2, song3, and song4 … … song", respectively.

The vector representation (i.e., the target vector group) of 200 implicit themes for each song in the at least one list is as follows, and the target vectors of only two songs are used as an example for description.

Song1:[num1,num2,……,num200]

Song2:[num1,num2,……,num200]

Specifically, in this embodiment, as can be known from the foregoing description, the topic distribution information may be a target vector group, where the target vector group includes at least one target vector, and each object in the list information corresponds to one target vector, and each vector in the target vectors represents a probability value that the object belongs to each implicit topic.

Then Song1: [ num1, num2, … …, num200] is the target vector for Song1 (i.e., Song1), and Song2: [ num1, num2, … …, num200] is the target vector for Song2 (i.e., Song 2). In the target vector Song1 [ num1, num2, … …, num200], num1 represents the probability value that Song1 belongs to the implied topic 1, and so on, num200 represents the probability value that Song1 belongs to the implied topic 200. In the target vector Song2 [ num1, num2, … …, num200], num1 represents the probability value that Song2 belongs to the implied topic 1, and so on, num200 represents the probability value that Song2 belongs to the implied topic 200.

Optionally, in this embodiment, the step S302 of determining, based on the topic distribution information, a second object associated with the first object includes the following steps:

firstly, determining a first target vector corresponding to the first object in the target vector group;

secondly, calculating the similarity between the first target vector and a second target vector, wherein the second target vector is the other target vector except the first target vector;

finally, a second object associated with the first object is determined according to the similarity.

Specifically, in this embodiment, a first target vector corresponding to the first object may be determined in the target vector group, and then, other target vectors in the target vector group except for the first target vector may be used as the second target vector.

Thereafter, a similarity between the first target vector and the second target vector may be calculated. The similarity may be cosine similarity, and the calculation formula of the cosine similarity is as follows:

where A and B represent the first target vector and the second target vector, respectively.

After calculating the similarity between the first target vector and the second target vector, the second object associated with the first object may be determined according to the similarity.

As shown in fig. 4, if there are a plurality of similarities, the step of determining the second object associated with the first object according to the similarities includes the following steps:

step S401, determining a second target vector corresponding to a target similarity in the similarities, wherein the target similarity is a similarity which is greater than a preset similarity in the similarities;

step S402, determining an object to which a second target vector corresponding to the target similarity belongs as the second object.

Specifically, if the second target vector is plural, a plurality of similarity values will be obtained when calculating the similarity between the first target vector and the second target vector. For example, the first target vector is denoted as a1, and the second target vector is denoted as B1, B2, and B3, then the similarity between the first target vector a1 and the second target vector B1 may be calculated to obtain the similarity C1, and the similarity between the first target vector a1 and the second target vector B2 may be calculated to obtain the similarity C2; and calculating the similarity between the first target vector A1 and the second target vector B3 to obtain the similarity C3.

Among the three calculated similarities, the similarity greater than the preset similarity (i.e., the target similarity) is determined, and if the similarity C1 is determined to be greater than the preset similarity, the similarity C1 is the target similarity.

At this time, the object to which the second target vector B1 corresponding to the similarity C1 belongs may be determined as the second object.

As can be seen from the above description, in this embodiment, each vector in the target vectors represents a probability value that an object belongs to each implied topic, so when calculating the similarity between the first target vector and the second target vector by using the target vector group, the second object associated or similar to the first object can be determined based on the distribution situation of the implied topics corresponding to the respective objects, where the association or similarity refers to the association or similarity of the tag information corresponding to the first object and the second object.

In this embodiment, after the second object associated with the first object is determined according to the method described above, the tag information of each first object may be determined based on the tag information of the second object.

In an alternative embodiment, as shown in fig. 5, in step S206, when determining the tag information of each of the first objects based on the tag information of the second object, the following steps are repeatedly performed until the tag information of each object in the list information is labeled:

step S501, a labeled object is determined in a second object associated with a first object Ai, wherein the labeled object is an object labeled with label information, I sequentially takes 1 to I, and I is the number of the first objects in the list information;

step S502, the label information of the first object Ai is determined by using the label information of the labeled object.

In the present embodiment, the above-described steps S501 and S502 are executed in a loop until the tag information of each object in the tag list information is labeled. In each loop, each first object in the list information is traversed, and the above steps S501 and S502 are performed for each first object.

In an alternative embodiment, as shown in fig. 6, the step S502 of determining the tag information of the first object Ai by using the tag information of the labeled object includes the following steps:

step S601, judging whether the number of the marked objects in the second object exceeds a preset number value or not; if yes, go to step S602, otherwise go to step S603;

step S602, determining the label information of the first object Ai by using the label information of the labeled object;

step S603, judging whether the list information contains unmarked objects; if yes, the process returns to step S501, otherwise, the loop process is ended.

That is, the specific operation flow of step S501 and step S502 is a large outer loop, and each loop is completed. Some objects which are not marked with label information are marked with label information; and circulating the steps until a new marked object song does not exist in a certain circulation, and ending the whole process, wherein the process is exemplified below.

Assume that during the G-th major loop, first, a labeled object is determined in the second object associated with the first object A1. Then, whether the number of the marked objects in the second object exceeds a preset number value is judged. If yes, the label information of the first object a1 is determined by using the label information of the labeled object in the second object. If not, judging whether the list information contains the objects which are not marked, if so, returning to the step S501, namely acquiring the next first object A2; if not, the large loop ends in the G th time.

When labeling the first object a2, the procedure is performed as described above. I.e. the annotated object is determined in the second object associated with the first object a 2. Then, whether the number of the marked objects in the second object exceeds a preset number value is judged. If yes, the label information of the first object a2 is determined by using the label information of the labeled object in the second object. If not, judging whether the list information contains the objects which are not marked, if so, returning to the step S501, namely acquiring the next first object A3; if not, the large loop ends in the G th time.

When labeling the first object a3, the implementation is the same as above, and will not be described in detail here. It should be noted that, in the present embodiment, in the process of the G-th major cycle, after the end of the first object a1, when the first object a2 or the first object A3 is labeled, the first object a1 is still used as an unlabeled object.

According to the description, the mode described above is adopted, so that rapid labeling of a large number of objects can be realized, and meanwhile, each object in each list information can be labeled more accurately, so that the problem of object missing is avoided.

Optionally, in this embodiment, if there are a plurality of labeled objects, then the step S602 of determining the tag information of the first object Ai by using the tag information of the labeled objects includes the following steps:

and step S1, based on the similarity, sorting the marked objects according to a preset sorting order to obtain a sorting result.

Step S2, calculating the weight of the labeled object based on the sorting result.

Optionally, in step S2, the calculating the weight of the labeled object based on the sorting result includes: and calculating the weight of the labeled object by using a formula W/(logR +1), wherein W is the weight, and R is the ranking value of the labeled object in the ranking result.

Step S3, determining the weight as the weight of the label information corresponding to the labeled object.

Step S4, calculating a sum of weights of the same type of label information in the label information corresponding to the plurality of labeled objects, to obtain weights of various types of label information.

At step S5, the tag information of the first object Ai is determined based on the weight values of the various types of tag information.

Optionally, in step S5, the determining the label information of the first object Ai based on the weight values of the various types of label information includes: and determining label information corresponding to a target weight value in the weight values of the various types of label information as the label information of the first object Ai, wherein the target weight value is a numerical value which is greater than a preset weight value in the weight values of the various types of label information.

Specifically, as is apparent from the above description, in the present embodiment, after the second object associated with the first object Ai is determined, the object to which the tag information has been tagged is determined in the second object. Then, the labeled tag information is sorted according to the sequence from high to low through the similarity between the first object Ai and the object labeled with the tag information, and each labeled tag information corresponds to a sorting result, for example, a sorting number.

The weight of the labeled object may then be calculated based on the ranking results. Specifically, the weight may be calculated by the formula W ═ 1/(logR +1), thereby obtaining the weight of each labeled object. After the weight of each labeled object is obtained, the sum of the weights of the label information of the same type in the label information corresponding to the plurality of labeled objects can be calculated, and the weight values of various types of label information can be obtained.

For example, the plurality of labeled objects includes: the marked object D1, the marked object D2 and the marked object D3, wherein the label information corresponding to the marked object D1 is E1 and E2, the label information corresponding to the marked object D2 is E2 and E3, the label information corresponding to the marked object D3 is E1 and E3, the weight of the marked object D1 is W1, the weight of the marked object D2 is W2, and the weight of the marked object D3 is W3.

At this time, the sum of weights of the tag information E1 may be calculated, for example, the sum of weight W1 and weight W3 is calculated, which is N1; the sum of weights of the tag information E2 can be calculated, for example, the sum of weight W1 and weight W2 is calculated, i.e., N2; the sum of weights of the tag information E3 can be calculated, for example, the sum of weight W2 and weight W3 is calculated, i.e., N3. At this time, normalization processing is required for N1, N2, and N3. The normalization processing process comprises the following steps: values of N1/(N1+ N2+ N3), N2/(N1+ N2+ N3), and N3/(N1+ N2+ N3) are calculated, respectively, and the result after the normalization processing is taken as the weight values of the above-described various types of label information, that is, the weight values of label information E1, label information E2, and label information E3.

Finally, label information for the first object Ai can be determined based on the weight value. If N1/(N1+ N2+ N3) and N2/(N1+ N2+ N3) are greater than or equal to the preset weight value, the label information (label information E1, label information E2) corresponding to N1/(N1+ N2+ N3) and N2/(N1+ N2+ N3) is used as the label information of the first object Ai.

It should be noted that, in this embodiment, each first object is performed in the manner described above for each first object, and details are not repeated here. According to the above description, the label information of each object in the list information can be more accurately determined by labeling the object labeled with the label information in the list information, and then, corresponding learning classification can be performed on each object in the list information through the label information.

Note that, in this embodiment, as shown in fig. 7, after step S202, the following steps are further included:

step S701, determining target list information containing objects BJ in the at least one list information, wherein i sequentially takes 1 to J, and J is the number of the objects in the at least one list information;

optionally, in step S701, the determining, in the at least one list information, target list information including the object Bj includes: determining list information carrying subject information in the at least one list information; and determining target list information containing the object Bj in the list information carrying the subject information.

Step S702, labeling the object Bj based on the subject information of the target list information, and using the object labeled with the tag information as a labeled object.

In determining the seed object (i.e., the labeled object), this is done based on the subject information of the at least one list information. As can be seen from the above description, part or all of the at least one list information carries one or more topic information. In order to improve the efficiency of data processing, the list information carrying the subject information may be determined from the at least one list information. And then, labeling the object in the list information according to the list information carrying the subject information. After the objects in the list information are labeled, the objects successfully labeled are called labeled objects (or seed objects).

After the list information carrying the subject information is determined, the target list information containing the object Bj can be determined in the list information carrying the subject information, and then the object Bj is labeled according to the subject information of the target list information, and the object labeled with the label information is used as the labeled object.

When the object Bj is labeled according to the subject information of the target list information, the occurrence frequency of each subject information in the target list information can be counted.

If the occurrence times comprise target occurrence times, determining the subject information corresponding to the target occurrence times as the label information of the object Bj, and determining the object Bj as a labeled object, wherein the target occurrence times are the occurrence times of which the occurrence times are greater than or equal to a preset threshold value.

The above determination process is exemplified below. Assume that the list information is a list of songs, which includes J songs. For the song Bj, in the present embodiment, a list of songs carrying the subject information is first determined in the list of songs, and then a target list of songs containing the song Bj is determined in the list of songs carrying the subject information. And then counting the occurrence times of each topic information carried by the target song list in the target song list.

For example, the song Bj appears in 100 labeled song sheets, and the 100 labeled song sheets are the above-mentioned target song sheet list. Counting and accumulating the theme information of each target song list to obtain a song Bj: sport 90, rock 98, drive 60, etc. Wherein, the motion, rock and drive are the above-mentioned subject information, the motion 90 represents the number of occurrences of the motion in 100 sings with labels, the rock 98 represents the number of occurrences of the rock in 100 sings with labels, and the drive 60 represents the number of occurrences of the drive in 100 sings with labels.

In this embodiment, if the number of occurrences includes a target number of occurrences (i.e., a number of occurrences greater than a preset threshold), the topic information corresponding to the target number of occurrences is determined as the tag information of the song Bj. In this embodiment, the preset threshold may be a preset value, and it is assumed that the preset threshold is a preset ratio value and the number of the target song list, where the preset ratio value may be selected to be 0.6, and in addition, other ratio values may be selected, which is not specifically limited in this embodiment. Based on the above, the label information of the song Bj is sport, rock and roll, and drive.

FIG. 8 is a block diagram illustrating a data annotation device according to some embodiments of the present application, which implements functions corresponding to the steps performed by the above-described method. The apparatus may be understood as the server or the processor of the server, or may be understood as a component that is independent of the server or the processor and implements the functions of the present application under the control of the server, as shown in the figure, the data annotation apparatus may include an obtaining unit 810, a first determining unit 820, and a second determining unit 830.

An obtaining unit 810, configured to obtain at least one piece of list information, where each piece of list information includes at least one object, and a part or all of the at least one piece of list information carries one or more pieces of topic information;

a first determining unit 820, configured to determine, based on the list information, second objects associated with respective first objects, where the first objects are objects that are not labeled with tag information in the at least one list information, the number of the second objects is one or more, and the second objects include objects labeled with tag information;

a second determining unit 830, configured to determine tag information of each of the first objects based on the tag information of the second object.

In this embodiment, at least one list information is first obtained, where each list information includes at least one object, and a part or all of the at least one list information carries one or more topic information; thereafter, second objects associated with the respective first objects may be determined based on the list information; finally, tag information of each of the first objects is determined based on tag information of a second object. According to the above description, the label information of each object in the list information can be more accurately determined by labeling the object labeled with the label information in the list information, and then, corresponding learning classification can be performed on each object in the list information through the label information.

Optionally, the first determining unit includes: a first determining module, configured to determine topic distribution information based on the list information, where the topic distribution information represents distribution information of each object in the list information in the one or more implied topics, and the implied topics are implied topics included in the at least one list information; a second determination module to determine a second object associated with the first object based on the topic distribution information.

Optionally, the first determining module is configured to: inputting the list information into a text topic model for processing; and taking the processing result as the theme distribution information.

Optionally, the topic distribution information is a target vector group, the target vector group includes at least one target vector, and one object in the list information corresponds to one target vector, and each vector in the target vectors represents a probability value that the object belongs to each implicit topic.

Optionally, the second determining module is configured to: determining a first target vector corresponding to the first object in the target vector group; calculating similarity between the first target vector and a second target vector, wherein the second target vector is other target vectors except the first target vector; determining a second object associated with the first object according to the similarity.

Optionally, the similarity is multiple, and the second determining module is further configured to: determining a second target vector corresponding to a target similarity in the similarities, wherein the target similarity is a similarity which is greater than a preset similarity in the similarities; and determining the object to which the second target vector corresponding to the target similarity belongs as the second object.

Optionally, the second determining unit includes: repeatedly executing the following steps by using a third determining module and a fourth determining module until label information of each object in the list information is marked: the third determining module is configured to determine a labeled object in a second object associated with the first object Ai, where the labeled object is an object labeled with label information, I sequentially takes 1 to I, and I is the number of the first objects in the list information; the fourth determining module is configured to determine the tag information of the first object Ai by using the tag information of the labeled object.

Optionally, the fourth determining module is further configured to: judging whether the number of the marked objects in the second object exceeds a preset number value or not; if yes, determining the label information of the first object Ai by using the label information of the labeled object.

Optionally, the number of the labeled objects is multiple, and the fourth determining module is further configured to: based on the similarity, sequencing the marked objects according to a preset sequencing sequence to obtain a sequencing result; calculating the weight of the labeled object based on the sorting result; determining the weight as the weight of the label information corresponding to the labeled object; calculating the sum of the weights of the label information of the same type in the label information corresponding to the plurality of labeled objects to obtain the weight values of various types of label information; determining label information of the first object Ai based on the weight values of the various types of label information.

Optionally, the fourth determining module is further configured to: and determining label information corresponding to a target weight value in the weight values of the various types of label information as the label information of the first object Ai, wherein the target weight value is a numerical value which is greater than a preset weight value in the weight values of the various types of label information.

Optionally, the fourth determining module is further configured to: and calculating the weight of the labeled object by using a formula W/(logR +1), wherein W is the weight, and R is the ranking value of the labeled object in the ranking result.

Optionally, after obtaining the at least one list information, the apparatus further includes: a third determining unit, configured to determine, in the at least one piece of list information, target list information including an object Bj, where i sequentially takes 1 to J, and J is the number of objects in the at least one piece of list information; and the labeling unit is used for labeling the object Bj based on the subject information of the target list information and taking the object labeled with the label information as a labeled object.

Optionally, the third determining unit is configured to: determining list information carrying subject information in the at least one list information; and determining target list information containing the object Bj in the list information carrying the subject information.

Optionally, the labeling unit is configured to: counting the occurrence times of each topic information carried by the target list information in the target list information; and marking the object Bj based on the occurrence times.

Optionally, the labeling unit is further configured to: if the occurrence times comprise target occurrence times, determining the subject information corresponding to the target occurrence times as the label information of the object Bj, and determining the object Bj as a labeled object, wherein the target occurrence times are the occurrence times of which the occurrence times are greater than or equal to a preset threshold value.

The modules may be connected or in communication with each other via a wired or wireless connection. The wired connection may include a metal cable, an optical cable, a hybrid cable, etc., or any combination thereof. The wireless connection may comprise a connection over a LAN, WAN, bluetooth, ZigBee, NFC, or the like, or any combination thereof. Two or more modules may be combined into a single module, and any one module may be divided into two or more units.

In the present application, a computer-readable storage medium is also provided, on which a computer program is stored, which, when being executed by a processor, performs the steps of any of the above-mentioned data annotation methods.

It can be clearly understood by those skilled in the art that, for convenience and brevity of description, the specific working processes of the system and the apparatus described above may refer to corresponding processes in the method embodiments, and are not described in detail in this application. In the several embodiments provided in the present application, it should be understood that the disclosed system, apparatus and method may be implemented in other ways. The above-described apparatus embodiments are merely illustrative, and for example, the division of the modules is merely a logical division, and there may be other divisions in actual implementation, and for example, a plurality of modules or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection of devices or modules through some communication interfaces, and may be in an electrical, mechanical or other form.

The modules described as separate parts may or may not be physically separate, and parts displayed as modules may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.

In addition, functional units in the embodiments of the present application may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit.

The functions, if implemented in the form of software functional units and sold or used as a stand-alone product, may be stored in a non-volatile computer-readable storage medium executable by a processor. Based on such understanding, the technical solution of the present application or portions thereof that substantially contribute to the prior art may be embodied in the form of a software product stored in a storage medium and including instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present application. And the aforementioned storage medium includes: various media capable of storing program codes, such as a U disk, a removable hard disk, a ROM, a RAM, a magnetic disk, or an optical disk.

The above description is only for the specific embodiments of the present application, but the scope of the present application is not limited thereto, and any person skilled in the art can easily conceive of the changes or substitutions within the technical scope of the present application, and shall be covered by the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.