CN112215684A

CN112215684A - Clustering method and device for target controllable objects

Info

Publication number: CN112215684A
Application number: CN202011187274.8A
Authority: CN
Inventors: 马文博; 赵宇; 蒋冰; 庄伟超; 吴国祖; 金朝林; 危毅; 黄耀鹏; 叶雯文; 李家昌; 邓辉; 葛文辉; 冯艺超
Original assignee: Tencent Technology Shenzhen Co Ltd
Current assignee: Tencent Technology Shenzhen Co Ltd
Priority date: 2020-10-30
Filing date: 2020-10-30
Publication date: 2021-01-12
Anticipated expiration: 2040-10-30
Also published as: CN112215684B

Abstract

The embodiment of the application provides a clustering method and a device for target controllable objects, which relate to the technical field of computers, and the method comprises the following steps: aiming at a target controllable object set to be clustered, respectively obtaining an operation state information set of each target controllable object; respectively aiming at each target controllable object, determining an operation state attribute corresponding to each operation state information sequence in an operation state information set of the target controllable object according to at least one operation state information in the corresponding operation state information sequence, and obtaining an operation state attribute set of each target controllable object; and determining the similarity of different target controllable objects based on the positive running state attribute subset and the negative running state attribute subset in the running state attribute set of each target controllable object, and clustering each target controllable object based on the similarity. The accuracy and the universality of clustering are improved.

Description

Clustering method and device for target controllable objects

Technical Field

The embodiment of the application relates to the technical field of computers, in particular to a clustering method and device for target controllable objects.

Background

With the development of computer technology and network technology, services provided by service providers to users through networks are more and more sophisticated. The application program facilitator can provide a service for downloading the application for the user, and recommend other applications after the user downloads the application, the shopping facilitator provides services such as commodity browsing and shopping for the user, and recommends other commodities after the user purchases any commodity, the video facilitator provides services such as video watching, downloading and uploading for the user, and can recommend other videos after the user watches any video, so in order to accurately recommend the user, the similarity between the controllable object used by the user and other controllable objects to be recommended needs to be determined, clustering is carried out based on the similarity, and the recommendation of other controllable objects in the clustering set is facilitated.

In the related art, the similarity is generally determined based on the attribute of the controllable object, for example, if the controllable object used by the user 1 is a video application, other controllable objects recommended to the user 1 are also video applications. However, the similarity between different controllable objects is determined only by the attributes of the controllable objects, which may cause other controllable objects recommended to the user to have the same attribute or similar attributes, resulting in that the recommended other controllable objects are not of interest to the user, and further, the recommendation effectiveness of the other controllable objects is low.

Disclosure of Invention

The embodiment of the application provides a clustering method and a clustering device for target controllable objects, which are used for accurately determining the similarity between different target controllable objects based on positive running state attributes and negative running state attributes corresponding to different target controllable objects, and further carrying out effective recommendation.

In one aspect, an embodiment of the present application provides a method for clustering target controllable objects, where the method includes:

aiming at a target controllable object set to be clustered, respectively obtaining an operation state information set of each target controllable object, wherein the operation state information set of each target controllable object comprises at least one operation state information sequence, each operation state information sequence comprises operation state information of the target controllable object in at least two continuous time segments when the target controllable object is controlled by the same control object, each operation state information is determined based on the control action of the control object on the target controllable object, and the operation state information represents a used state or an unused state;

respectively aiming at each target controllable object, determining an operation state attribute corresponding to each operation state information sequence in an operation state information set of the target controllable object according to at least one operation state information in the corresponding operation state information sequence, and obtaining an operation state attribute set of each target controllable object, wherein the operation state attribute comprises a positive operation state attribute or a negative operation state attribute;

and determining the similarity of different target controllable objects based on the positive running state attribute subset and the negative running state attribute subset in the running state attribute set of each target controllable object, and clustering each target controllable object based on the similarity.

In one aspect, an embodiment of the present application provides a clustering apparatus for target controllable objects, including:

an operation state information set obtaining unit, configured to obtain, for a target controllable object set to be clustered, an operation state information set of each target controllable object, respectively, where the operation state information set of each target controllable object includes at least one operation state information sequence, each operation state information sequence includes operation state information of the target controllable object in at least two consecutive time segments when the target controllable object is controlled by the same control object, each operation state information is determined based on a control behavior of the control object on the target controllable object, and the operation state information represents a used state or an unused state;

the running state attribute set determining unit is used for determining a running state attribute corresponding to each running state information sequence in the running state information set of the target controllable object according to at least one piece of running state information in the corresponding running state information sequence aiming at each target controllable object, and obtaining a running state attribute set of each target controllable object, wherein the running state attribute comprises a positive running state attribute or a negative running state attribute;

and the clustering unit is used for determining the similarity of different target controllable objects based on the positive running state attribute subset and the negative running state attribute subset in the running state attribute set of each target controllable object and clustering each target controllable object based on the similarity.

Optionally, the clustering unit is specifically configured to:

taking a first set value corresponding to the running state attribute set of each target controllable object as a first coordinate value of a first coordinate axis in a clustering coordinate system, and taking a second set value corresponding to the running state attribute set of each target controllable object as a second coordinate value of a second coordinate axis in the clustering coordinate system;

determining the coordinate position of each target controllable object in the clustering coordinate system based on the first coordinate value and the second coordinate value corresponding to each target controllable object;

similarity of different target controllable objects is determined based on the coordinate position of each target controllable object in the cluster coordinate system.

In one aspect, embodiments of the present application provide a computer device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, where the processor implements the steps of the clustering method for the target controllable objects when executing the computer program.

In one aspect, embodiments of the present application provide a computer-readable storage medium storing a computer program executable by a computer device, when the program is run on the computer device, causing the computer device to perform a clustering method for a target controllable object.

In one aspect, embodiments of the present application provide a computer program product, which includes a computer program or instructions, and when the computer program or instructions are executed, the computer program or instructions implement a clustering method that at least one processor is capable of executing a target controllable object.

In the embodiment of the application, for each target controllable object to be clustered, an operation state information set of each target controllable object is respectively obtained, the set represents operation state information in at least two continuous time segments when each target controllable object is controlled by a plurality of control objects, an operation state sequence corresponding to each control object is determined based on each time segment and the corresponding operation state information, and the operation state sequence of each control object forms an operation information set of one target control object.

Further, since the running state information set of each target controllable object includes the running state information corresponding to each time segment, and the running state information has different representation meanings, based on different running state information and corresponding representation meanings, it is possible to determine the running state attribute of the running state information sequence corresponding to the corresponding target controllable object when each control object controls, specifically, the running state attribute is classified as a positive running state attribute or a negative running state attribute, that is, each control object controlling any one target controllable object corresponds to one running state attribute.

After the running state attributes of all the running state information sequences corresponding to each target operable object are determined, a running state attribute set of each target operable object can be obtained, wherein the running state attribute set comprises a positive running state subset and a negative running state subset.

In the embodiment of the application, the similarity of different target controllable objects is determined based on the positive running state attribute subset and the negative running state attribute subset in the running state attribute set of each target controllable object, and each target controllable object is clustered based on the similarity.

As can be seen from the above, in the clustering method for different target controllable objects in the embodiment of the present application, the similarity is determined based on the positive operating state and the negative operating state corresponding to the different target controllable objects, and the positive operating state and the negative operating state are determined based on the control behavior of the control object, that is, the control behavior similarity between the different target controllable objects is determined from the control behavior of each target controllable object, and clustering is performed through the similarity, that is, a plurality of target controllable objects having similar control behavior characteristics are clustered into one cluster, so that recommendation, analysis, and other uses are facilitated.

In the embodiment of the application, different target controllable objects with similar control behaviors are clustered into a cluster, so that the barrier of clustering according to attributes in the related technology is broken through, the accuracy and the universality of clustering are improved, the effectiveness of recommending other target controllable objects is further improved, and the feeling of the control objects is improved.

Drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed to be used in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without inventive exercise.

Fig. 1(a) is a schematic view of an application scenario provided in an embodiment of the present application;

fig. 1(b) is a schematic view of an application scenario provided in an embodiment of the present application;

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

FIG. 3 is an illustration of a set of operating state information for each target controllable object according to an embodiment of the present disclosure;

fig. 4 is a schematic diagram of a clustering coordinate system provided in an embodiment of the present application;

fig. 5 is a schematic view of an application scenario provided in an embodiment of the present application;

fig. 6 is a schematic diagram of an application scenario provided in an embodiment of the present application;

fig. 7 is a schematic structural diagram of a target controllable object apparatus according to an embodiment of the present application;

fig. 8 is a schematic structural diagram of a computer device according to an embodiment of the present invention.

Detailed Description

In order to make the objects, technical solutions and advantages of the present application more apparent, the technical solutions in the embodiments of the present application will be described clearly and completely with reference to the accompanying drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application. In the present application, the embodiments and features of the embodiments may be arbitrarily combined with each other without conflict. Also, while a logical order is shown in the flow diagrams, in some cases, the steps shown or described may be performed in an order different than here.

For convenience of understanding, terms referred to in the embodiments of the present application are explained below.

In addition, it should be understood that the terms "system" and "network" in the embodiments of the present application may be used interchangeably. "at least one" means one or more, "a plurality" means two or more. "and/or" describes the association relationship of the associated objects, meaning that there may be three relationships, e.g., a and/or B, which may mean: a exists alone, A and B exist simultaneously, and B exists alone, wherein A and B can be singular or plural. The character "/" generally indicates that the former and latter associated objects are in an "or" relationship. "at least one of the following" or similar expressions refer to any combination of these items, including any combination of the singular or plural items. For example, at least one (one) of a, b, or c, may represent: a, b, c, a-b, a-c, b-c, or a-b-c, wherein a, b, c may be single or multiple. And, unless stated to the contrary, the embodiments of the present application refer to the ordinal numbers "first", "second", etc., for distinguishing a plurality of objects, and do not limit the sequence, timing, priority, or importance of the plurality of objects. For example, the first set of values and the second set of values are merely to distinguish between the different sets of numerical representations, and are not to indicate a difference in priority, degree of importance, or the like between the two sets of numerical values.

Target controllable object: refers to an object that can be controlled and can realize different use states based on different control behaviors, and the object can be an application program App, namely when the App is controlled, the App can be a user in a use state or can be in an unused state; the object may be a video, music, text, or the like, or may be an online product, and the like, and is not limited herein.

In the embodiment of the present application, the usage state information of the target controllable object is used to represent different usage states, and the different usage states correspond to different usage state information. For example, if the target controllable object is an App, the App use status information may be open status information, which indicates that the App is in a used status.

The control object is: the target controllable object can be controlled by the target controllable object, the target controllable object can be a user of the target controllable object, the user operates the target controllable object, the operations form control behaviors, and different control behaviors cause the target controllable object to have different running state information, so that representations of different running state information are obtained.

In the embodiment of the present application, the control behavior generally includes two kinds, one is a behavior for changing the target controllable object from the unused state to the used state, and the other is a behavior for changing the target controllable object from the used state to the unused state.

For example, when the target controllable object is an App, the control behaviors may be a start App behavior, an install App behavior, a close App behavior, and an unload App behavior, where the start App behavior, the install App behavior, are behaviors that transform the App from an unused state to a used state, the close App behavior, and the unload App behavior are behaviors that transform the target controllable object from a used state to an unused state.

Running state information sequence: the information sequence is formed by collecting the running state information of the target controllable object in at least two consecutive time segments, and the information sequence is arranged according to the time sequence, wherein the time sequence can be the sequence from far to near according to the collected running state information, or the sequence from near to far, or other time sequences.

The running state attribute is as follows: the attribute corresponding to each running state information sequence can include a positive running state attribute and a negative running state attribute, each running state information sequence corresponds to one running state attribute, and different running state arrangements in the running state information sequences can cause different corresponding running state attributes.

Having introduced the above terms, the concepts of the present application will now be described based on the problems presented in the related art.

With the increasingly powerful functions of terminal devices such as smart phones and tablet computers, controllable objects displayed on terminals such as smart phones and tablet computers are increasingly diversified. The diversification of the controllable object types brings rich and diverse user experiences to the terminal users, and meanwhile, the difficulty in selection is increased to the terminal users. In order to rapidly and conveniently select controllable objects meeting the requirements or preferences of terminal users from various controllable objects, recommendation services for various controllable objects are developed.

The principle of recommendation service for various types of controllable objects is to recommend based on the similarity between various types of controllable objects, so the similarity between various types of controllable objects needs to be determined first.

In the related art, the method for determining the similarity between various types of controllable objects is determined based on the attributes of the controllable objects, for example, the controllable object 1 is installed in the terminal device of the user 1, the controllable object 1 is a video application App1, and the other controllable objects similar to the controllable object 1 are determined to be video applications App 2; the user 2 has the controllable object 2 installed in the terminal device, the controllable object 2 is a shopping application App1, and the other controllable object determined to be similar to the controllable object 2 is a shopping application App 2.

As can be seen from the above, in the related art, although the similarity between different controllable objects can be determined based on the attributes of the controllable objects, clustering is performed based on the similarity, and recommendation of other controllable objects is performed through the clustering result, the attributes of the controllable objects recommended in the related art and the controllable objects operated by the controllable objects are similar, the occurrence probability of redundant controllable objects is high, and further, the effectiveness of recommendation of the controllable objects is low, and the applicability is poor.

Based on the above problem, the inventor of the present application first proposes a method for clustering target controllable objects, in the embodiment of the present application, for each target controllable object to be clustered, an operation state information set of each target controllable object is respectively obtained, where the set represents operation state information in at least two consecutive time segments when each target controllable object is controlled by a plurality of control objects, an operation state sequence corresponding to each control object is determined based on each time segment and the corresponding operation state information, and the operation state sequence of each control object forms an operation information set of one target controllable object.

After introducing the inventive concept of the present application, first a system architecture diagram applicable to the present application is introduced, as shown in fig. 1(a), the system architecture at least includes at least one terminal device 101 and at least one server 102, a plurality of controllable objects may be operated in the terminal device 101, the server 102 may obtain an operation state information set of each controllable object, the server 102 determines a positive operation state attribute subset and a negative operation state attribute subset in the operation state attribute set of each controllable object based on obtaining the operation state information set of each controllable object, determines similarities of different target controllable objects, and clusters each target controllable object based on the similarities.

In the embodiment of the present application, a client is installed in the terminal device 101, and the client may correspond to the controllable object. The client in the terminal device 101 may be a browser client, a video application client, an application client such as a software store, etc. The client in the terminal apparatus 101 is a client of each application, that is, each controllable independent can be run through the terminal apparatus 101, and the state information data of each controllable object in the terminal apparatus 101 is reported to the server 102.

The terminal device 101 may be, but is not limited to, a smart phone, a tablet computer, a notebook computer, a desktop computer, a smart speaker, a smart television, a smart watch, and the like.

Further, the terminal device 101 may install each application client in an active manner (for example, by downloading through a software store application) or a passive manner (pre-installed), and after each application client is started, monitor the state information change of each target controllable object on the interface of the terminal device 101.

In this embodiment, the server 102 is an electronic device providing computing power, and the server 102 performs data analysis and statistics according to the state information data of each target controllable object reported by the terminal device 101 to obtain a clustering result. The server 102 may be an independent physical server, a server cluster or a distributed system formed by a plurality of physical servers, or a cloud server providing basic cloud computing services such as a cloud service, a cloud database, cloud computing, a cloud function, cloud storage, a Network service, cloud communication, a middleware service, a domain name service, a security service, a Content Delivery Network (CDN), a big data and artificial intelligence platform, and the like. The terminal device 101 and the server 102 may be directly or indirectly connected through wired or wireless communication, and the application is not limited herein.

Further, as shown in fig. 1(b), explaining the structure of the terminal device 101 and the server 102 exemplarily, the terminal device 101 may include one or more processors 1011, a memory 1012, an I/O interface 1013 interacting with the server 102, and a display panel 1014, etc. The terminal device 101 may be, but is not limited to, a smart phone, a tablet computer, a notebook computer, a desktop computer, a smart speaker, a smart television, a smart watch, and the like.

In the embodiment of the present application, the server 102 may include one or more processors 1021, a memory 1022, and an I/O interface 1023 for interaction with the terminal apparatus 101. In addition, server 102 may also configure database 1024.

In this embodiment, after the server 102 performs data analysis and statistics, the result may be stored in the memory 1022, or may be stored in another storage device, which is not limited herein.

Illustratively, a software store client is installed in the terminal device 101, and when the terminal device 101 downloads and installs the software store application, a software development kit SDK is installed at the same time, and is used to monitor the status information of each target controllable object in the terminal device 101 and report the status information of each target controllable object to the server 102.

Specifically, the process of the server 102 for processing the running state information of each target controllable object reported by the terminal device 101 is as follows: aiming at a target controllable object set to be clustered, respectively obtaining an operation state information set of each target controllable object, wherein the operation state information set of each target controllable object comprises at least one operation state information sequence, each operation state information sequence comprises operation state information of the target controllable object in at least two continuous time segments when the target controllable object is controlled by the same control object, each operation state information is determined based on the control action of the control object on the target controllable object, and the operation state information represents a used state or an unused state; respectively aiming at each target controllable object, determining an operation state attribute corresponding to each operation state information sequence in an operation state information set of the target controllable object according to at least one operation state information in the corresponding operation state information sequence, and obtaining an operation state attribute set of each target controllable object, wherein the operation state attribute comprises a positive operation state attribute or a negative operation state attribute; and determining the similarity of different target controllable objects based on the positive running state attribute subset and the negative running state attribute subset in the running state attribute set of each target controllable object, and clustering each target controllable object based on the similarity.

In this embodiment of the application, after the clustering result is determined, information of other target controllable objects of the same clustering cluster may be obtained based on the existing target controllable objects in the terminal device 101, and recommendation is performed.

Of course, the method provided in the embodiment of the present application is not limited to be used in the application scenarios shown in fig. 1(a) and fig. 1(b), and may also be used in other possible application scenarios, and the embodiment of the present application is not limited thereto. Functions that can be implemented by each device of the application scenarios shown in fig. 1(a) and fig. 1(b) will be described together in the following method embodiments, and will not be described in detail herein.

Based on the above design concept and the above application scenario, the method of the embodiment of the present application is specifically described below.

As shown in fig. 2, the present application provides a clustering method for target controllable objects, which is specifically executed by a computer device, and optionally, the computer device may be an electronic device with computing capability, such as a terminal device or a server, where a specific execution subject of the clustering method is not limited herein.

The method shown in fig. 2 specifically includes:

step S201, for a set of target controllable objects to be clustered, obtaining an operation state information set of each target controllable object, respectively, where the operation state information set of each target controllable object includes at least one operation state information sequence, each operation state information sequence includes operation state information of the target controllable object in at least two consecutive time segments when the target controllable object is controlled by the same control object, each operation state information is determined based on a control behavior of the control object on the target controllable object, and the operation state information represents a used state or an unused state.

Specifically, in this embodiment of the present application, the target controllable object set to be clustered may be a set of all target controllable objects, or a set formed based on target controllable objects matched with the setting conditions, for example, a target controllable object set to be clustered formed by target controllable objects operating in terminal devices of the same model, or a target controllable object set to be clustered formed by target controllable objects operating in terminal devices located in the same geographic area, and of course, other methods for defining a target controllable object set are also provided, which are not described herein again.

In this embodiment of the present application, each target controllable object corresponds to one running state information set, the state information obtained by the set includes running state information when each target controllable object is controlled, each target controllable object has a corresponding relationship with a plurality of control objects, that is, different control objects can control the same target controllable object, and perform a control behavior operation on the target controllable object, and the running state of the target controllable object is changed or not changed through the control behavior operation.

Furthermore, the timeliness of each state information in the set, that is, the state information generated within a set time, may be limited to be the set of operating state information corresponding to the target controllable object in the present application. For example, the state information of the target controllable object 1 is the state information at the first time, the state information at the second time, and the state information at the third time; the state information of the target controllable object 2 is the state information at the fourth time, and the time interval between the fourth time and the first time is two years, and the correlation between the state information within two years is poor, so that the timeliness of each state information needs to be limited.

Therefore, in the embodiment of the present application, each piece of status information within the set time interval range may constitute a status information set, and for example, 3 months before the current time may be used as the set time interval, and status information within 3 months before the current time may constitute a status information set. In the embodiment of the present application, the time intervals may be time intervals of the same time length, or may be different, and are not limited herein.

That is, in the embodiment of the present application, the state information sets of each target controllable object are generated in the same time period, so that the relevance between the state information of different target controllable objects can be better determined.

The state information of each target controllable object in a plurality of time segments can form an operation state information sequence, so that each target controllable object at least comprises one operation state information sequence, and the operation state information in the sequence is sorted according to a certain sequence, wherein the sequence can be a random sequence or a set sequence.

Optionally, in this embodiment of the application, the latest state information may better represent the state of the current target controllable object, so in a specific embodiment, the state information may be sorted according to a sequence from near to far in the segment time, and an operation state information sequence is obtained.

Illustratively, as shown in Table 1, t time segments, t-1 time segments, and t-2 time segments are included in Table 1, each time corresponding to the state information of the target controllable object 1.

TABLE 1

Time segmentation	t-2	t-1	t
				Status information	Status information 3	Status information 2	Status information 1

In table 1, the state information of the target controllable object 1 in three consecutive time segments is shown, and based on each state information in table 1, the consecutive time segments corresponding to the state information are sorted from near to far in order to obtain an operating state information sequence, which can be characterized as state information 1, state information 2, and state information 3.

Optionally, in this embodiment of the application, different clustering strategies correspond to different time segment information, and a clustering strategy may also be understood as a clustering rule, where each clustering rule corresponds to at least two consecutive target time segment information, and the number of the target time segment information is not limited herein.

And when the clustering rule is acquired, respectively acquiring an operation state information set of each target controllable object based on the clustering rule, wherein each operation state information sequence comprises operation state information in at least two continuous target time segments corresponding to at least two continuous target time segment information when the target controllable objects are controlled by the same control object.

Illustratively, the time segment information corresponding to the clustering rule 1 is a time segment of 1 month and each week, and based on the clustering rule 1, an operation state information set of each target controllable object is respectively obtained, and each operation state information sequence includes operation state information in a time segment of one month and each week corresponding to at least two continuous target time segment information when the target controllable objects are controlled by the same control object.

And the time segmentation information corresponding to the clustering rule 2 is 3 months per month time segmentation, then based on the clustering rule 2, respectively obtaining the running state information set of each target controllable object, wherein each running state information sequence comprises the running state information in the 3 months per month time segmentation corresponding to at least two continuous target time segmentation information when the target controllable object is controlled by the same control object.

In the embodiment of the present application, each piece of operation state information represents a used state or an unused state of the corresponding target controllable object, and the use state of the corresponding target controllable object can be changed by the control behavior of different control objects.

It can be considered that, in two adjacent continuous time segments, the control object does not change the operation behavior of the target controllable object, or after one operation behavior occurs, other operation behaviors are not performed, and it can be considered that the operation state information corresponding to the two continuous time segments does not change. For example, in the t-1 time segment, the control object performs the start operation on the target controllable object 1, and in the t time segment, the control object does not perform the control operation on the target controllable object 1, and the state information of the target controllable object corresponding to the t-1 time segment is the same as the state information of the target controllable object corresponding to the t time segment.

Further, in two adjacent continuous time segments, the control object changes the operation behavior of the target controllable object, or after the operation behavior does not occur, the operation behavior is performed, and it can be considered that the operation state information corresponding to the two continuous time segments changes. For example, in the t-1 time segment, the control object performs an opening operation on the target controllable object 1, and in the t time segment, the control object performs a deletion operation on the target controllable object 1, so that the state information of the target controllable object corresponding to the t-1 time segment is different from the state information of the target controllable object corresponding to the t time segment.

The at least one operation state information sequence corresponding to each target controllable object may be a sequence having the same operation state information or a sequence having different operation state information.

Having introduced the above, the information content comprised by the set of target controllable objects to be clustered is explained below with reference to the schematic diagram of fig. 3.

For example, in the embodiment of the present application, the target controllable objects to be clustered are App1, App2, App3 and App4, and the control objects of App1 are user 1 and user 2; control objects of App2 are user 3, user 4, and user 5; the control object of App3 is user 6.

Specifically, in the embodiment of the application, the user 1 installs the App1 in the t time segment, the running state information of the t time segment is installation, the user 1 uninstalls the App1 in the t-1 time segment, and the running state information of the t time segment is uninstallation, so that for the user 1 corresponding to the App1, the obtained running state information sequence is (installation, uninstallation) in the order of the time segments from near to far.

Similarly, based on the above principle, the operation state information sequence corresponding to the user 2 is determined as (on, on), the operation state information sequence corresponding to the user 3 is determined as (install, on), the operation state information sequence corresponding to the user 4 is determined as (uninstall ), the operation state information sequence corresponding to the user 5 is determined as (install, on), and the operation state information sequence corresponding to the user 6 is determined as (uninstall, install).

It is possible to determine a set of operating state information for each App, specifically, for App1, set of operating state information is (install, uninstall), (open )), for App2, set of operating state information is ((install, open), (uninstall ), (install, open)), and for App3, set of operating state information is (uninstall ).

Step S202, respectively aiming at each target controllable object, determining an operation state attribute corresponding to each operation state information sequence in an operation state information set of the target controllable object according to at least one operation state information in the corresponding operation state information sequence, and obtaining an operation state attribute set of each target controllable object, wherein the operation state attribute comprises a positive operation state attribute or a negative operation state attribute.

Specifically, in the embodiment of the present application, for each target controllable object, an attribute of each operation state information sequence corresponding to the target controllable object can be determined, and according to the attribute of the operation state information sequence, an operation state attribute set of the target controllable object is determined.

In the embodiment of the application, because the running state information in each running state information sequence has different state representations, the running state information sequence attribute can be determined through the state representation corresponding to each running state information in the sequence.

In this embodiment of the present application, different operation state information sequences may have the same attribute or different attributes, and a specific attribute is determined based on at least one operation state information state representation in the operation state information sequence.

For example, in the embodiment of the present application, the operation state information sequence attribute may be determined based on all the operation state information in the operation state information sequence, may be determined based on setting several pieces of operation state information, or may be determined based on the operation state information at the set sequence position.

Optionally, in this embodiment of the application, since the latest operating state information can represent the latest state of the target controllable object, for example, the used state or the unused state, the operating state information sequence attribute may be determined based on the latest operating state information in the operating state information sequence.

In the embodiment of the application, the attribute of the running state information sequence may be determined based on the corresponding relationship between the representation and the attribute of the running state information.

In an optional embodiment, in the embodiment of the present application, in order to determine the running state information sequence attribute based on different use states, the different use states may be first converted into state identifiers, and then the running state information sequence attribute is determined according to a correspondence between the state identifiers and the attributes.

Optionally, in this embodiment of the application, the used state is represented by a first state identifier, and the used state is represented by a second state identifier, but of course, the used state may be represented by a second state identifier, and the used state may be represented by a first state identifier, which is not limited herein.

In this embodiment of the present application, the state identifier has a corresponding relationship with the attribute, and the relationship may be represented by a corresponding relationship table, which is not limited herein.

Therefore, in the real-time example of the application, the latest running state information is determined, then the state identification is calibrated, after the state identification is obtained, the corresponding attribute is determined, and the attribute is used as the running state information sequence attribute.

For example, in the embodiment of the present application, the target controllable object is taken as an App for explanation, and specifically, for the App, the use state may be a start state, an uninstalled state, or an uninstalled state.

Further, a starting state of the App is defined, where the starting state of the App refers to that if the last operation of the user on the App in the t time period is installation, the App is considered to have started the App once, or if the last operation of the user on the App in the t time period is starting, the App is considered to have started the App once.

The uninstalled or uninstalled state of the App refers to that the App is considered to be uninstalled by the user if the App is not started within a fixed time range (for example, 31 days), that is, the state of the App is considered to be an uninstalled or uninstalled state.

In the embodiment of the application, the starting state of the App is represented by a first state identifier, and the uninstalling or uninstalling state of the App is represented by a second state identifier.

For an App, the different running state information sequence attributes may be classified as positive running state attributes or negative running state attributes, the positive running state attributes indicating that the App is used or installed, e.g., the App is installed during the t time period, or indicating that the App is started, e.g., the App is started during the t time period.

The negative running state attribute characterizes whether the App is unloaded or unused, e.g., the App is unloaded in time t segments or the App has not been started up to date 30.

Further, in this embodiment of the present application, the operation behavior of the user that may correspond to the forward running state attribute is: the user has unloaded the App but installed it at t fragmentation time; the negative running state attribute may correspond to an operation behavior of the user as follows: the user has installed the App once, but uninstalls the App at t-segment time, and as can be seen from the above, the positive running state attribute and the negative running state attribute need to be determined based on running state information of a plurality of continuous time segments, such as t-segment time, t-1 time segment time, and the like.

In an optional embodiment, in the embodiment of the present application, the forward running state attribute includes that the user continuously uses App; the method comprises the following steps that a user adds an App, namely the user does not install the App at the time of t-1, and the user newly installs the App at the time of t; the user reloads the App, namely the user initially installs the App, then the user unloads the App in t-1 or t-2 time segments or other segments, and then reinstalls the App in t time segments.

In another optional embodiment, in the embodiment of the present application, the negative-going operating state attribute includes that the user uninstalls the App in a time slice t, and it may be that the user installs the App in any time slice before t-1, and performs an uninstalling control action in the time slice t.

Further, in order to better represent each attribute, the operation behavior of the user in the attribute can be marked, for example, a letter C indicates that the user continuously uses the App, a letter N indicates that the user newly adds the App, a letter U indicates that the user installs the App in any time segment before t-1, the user performs the unloading control behavior in the time segment at t, and a letter R indicates that the user reloads the App.

Therefore, it can be determined through the letters that when the operating state of App determined by the latest control action of the user corresponds to the control action of the forward operating state attribute letter C, N, R, the attribute of the operating state information sequence of App is determined to be the forward operating state attribute; and when the running state of the App determined by the latest control action of the user corresponds to the control action of the negative running state attribute letter U, determining that the attribute of the running state information sequence of the App is the negative running state attribute.

In the embodiment of the present application, the starting state of the App is represented by a first state identifier, and the uninstalled or uninstalled state of the App is represented by a second state identifier.

TABLE 2

By the state identification of each time segment, the attribute letters corresponding to different running state information sequences can be determined, and further the running state information sequences can be determined.

In table 2, the letter I indicates that no App is installed or started in consecutive time segments, so that the user is considered to ignore the App, the letter I can be considered to represent an undifferentiated behavior, and the letter I can also be considered to be a negative-running-state attribute.

Further, the first state identifier and the second state identifier can be assigned, so that the state of the App in each time segment can be conveniently represented, the first state identifier is assigned to be 1, the second state identifier is assigned to be 0, and then the table 2 can be updated to be the table 3.

TABLE 3

Further, in the embodiment of the present application, each operation state information sequence attribute can be determined by table 2 or table 3.

Of course, in the embodiment of the present application, the running state information sequence attribute is determined only by the running state information of the time segments from t-2 to t, and the running state information sequence attribute may also be determined by the running state information of more time segments.

In the embodiment of the present application, it can be determined through tables 2 and 3 that the attribute letter C is determined based on the assignments corresponding to the operating state information of two consecutive time segments, so that the assignment corresponding to the operating state information of any two time segments matches with the attribute letter C, and thus can correspond to one attribute letter C, and similarly, the attribute letter U is determined based on the assignment corresponding to the operating state information of two consecutive time segments, so that the assignment corresponding to the operating state information of any two time segments matches with the attribute letter U, and thus can correspond to one attribute letter U, and thus can obtain a plurality of corresponding attribute letters based on the operating state information of the time segments greater than t-2 to t, in the embodiment of the present application, 1 attribute letter can be represented as a one-layer control behavior state of the user for the App, and 2 attribute letters can be represented as a two-layer control behavior state of the user for the App, and so on.

Illustratively, in the embodiment of the present application, as shown in table 4, the running state information sequence attribute is determined by the running state information of the time segment from t-3 to t.

TABLE 4

In Table 4, two-level behavioral states are represented by two attribute letters, wherein the preceding letter of the two attribute letters is determined based on at least two consecutive time segments t-2 to t-1, and the following letters are determined based on two consecutive time segments t-1 to t.

Specifically, the preceding letter C in the CC is determined based on 1, 1 corresponding to two consecutive time segments from t-2 to t-1, and the following letter C in the CC is determined based on 1, 1 corresponding to two consecutive time segments from t-1 to t.

The prefix letter C in the CU is determined based on 1 and 1 corresponding to two continuous time segments from t-2 to t-1, and the subsequent letter U in the CU is determined based on 1 and 0 corresponding to two continuous time segments from t-1 to t.

The prefix letter N in NC is determined based on 0, 1 corresponding to t-3 to t-1 continuous time segments, and the subsequent letter C in NC is determined based on 1, 1 corresponding to t-1 to t continuous time segments.

The determination method of the attribute letter sequence corresponding to the behavior states of the other two layers is similar to the determination method in the above contents, and is not described herein again.

After the operating state attribute of each operating state information sequence corresponding to each target controllable object is determined, a different operating state attribute set corresponding to each target controllable object can be determined.

Specifically, the different running state attribute sets may include a positive running state attribute subset and a negative running state attribute subset, that is, the running state information sequences with positive running state attributes form the positive running state attribute subset, and the running state information sequences with negative running state attributes form the positive running state attribute subset.

Step S203, determining the similarity of different target controllable objects based on the positive running state attribute subset and the negative running state attribute subset in the running state attribute set of each target controllable object, and clustering each target controllable object based on the similarity.

Specifically, in the embodiment of the present application, for each target controllable object, the positive running state attribute subset and the negative running state attribute subset in the running state attribute set can represent control behavior characteristics of different control objects corresponding to the target controllable object, and the similarity of the different target controllable objects can be determined and clustered by the control behavior characteristics of the control objects of the different target controllable objects.

In the embodiment of the application, the clustering characteristic values of the target controllable objects can be determined based on the positive running state attribute subset and the negative running state attribute subset in the running state attribute set of each target controllable object, and the similarity is determined and clustered based on the clustering characteristic values of different target controllable objects.

Further, in the embodiment of the present application, the running state attribute set of each target controllable object includes two subsets, so that the clustering feature values of the two subsets can be determined respectively.

Each subset is composed of a plurality of operating state information sequences, so in order to be able to determine the cluster characteristic value of each subset, it is also necessary to determine the value of each operating state information sequence.

There are various methods for determining the value of each operation state information sequence, and one optional method is to determine the value of each operation state information in each operation state information sequence, and then add the values of each operation state information to obtain the value of each operation state information sequence; another optional method is to take each operation state information sequence as a whole to determine the value of the operation state information sequence, specifically, the value of the operation state information sequence may be determined by a sequence-to-sequence value conversion method, that is, the value of the operation state information sequence is determined by a preset sequence-to-sequence value conversion relationship; there is also an optional method, since the operation state information can represent two states, the two states are respectively assigned to form a sequence with two state assignments, the sequence is a binary sequence, and the values of the operation state information sequence are obtained through binary sequence conversion.

That is to say, in the embodiment of the present application, for each target controllable object, a state value of each operation state information sequence is determined according to each operation state information in the corresponding operation state information sequence and a corresponding relationship between the obtained operation state information and the state value; and determining a first set value of a positive running state attribute subset and a second set value of a negative running state attribute subset in the corresponding running state attribute set based on the state value of each running state information sequence.

Similarity of different target controllable objects is determined based on the first set value and the second set value corresponding to the set of operating state attributes of each target controllable object.

Further, in the embodiment of the present application, in order to obtain the binary sequence, the running state information of the used state is assigned with the state value 1, and the running state information of the unused state is assigned with the state value 0, so that the binary sequence can be obtained.

Further, after the assignments are assigned, the assignments are also ordered according to a time sequence, so that a binary sequence conforming to a time rule can be obtained.

In the embodiment of the application, the assignments of each piece of operating state information are ordered according to the sequence from near to far of the corresponding time segment, and a corresponding binary state value sequence is obtained.

Illustratively, as shown in table 3, if the behavior state of user 1 for App1 is represented by attribute letter CC, the sequence is sorted from near to far according to the time segmentation, the obtained binary state value sequence is 111, and if the behavior state of user 2 for App2 is represented by attribute letter NC, the sequence is sorted from near to far according to the time segmentation, and the obtained binary state value sequence is 1100.

In the embodiment of the present application, since the behavior states of some control objects correspond to four binary digits and the behavior states of some control objects correspond to three binary digits, in order to obtain a binary sequence with the same digits, the binary sequence with a missing digit is complemented.

Specifically, in the embodiment of the present application, a specific method of bit padding is not limited, and bit padding may be performed using a default value, or bit padding may be performed according to a set bit padding rule.

After each binary sequence is obtained, the binary sequence can be converted into a decimal system, and a sequence value of each running state information sequence is obtained.

For example, if the binary state value sequence is 1100 and the converted decimal number is 12, the sequence value of the running state information sequence corresponding to the attribute letter NC may be determined to be 12.

By the above method, a sequence value of each operation state information sequence can be determined. After determining the sequence value of each sequence of operating state information, set values for two subsets of different attribute states in the set of operating state attributes for each target controllable object may be determined.

In an alternative embodiment, sequence values of the running state information sequences belonging to the same attribute set are added to obtain an aggregate value of different subsets, for example, a forward running state attribute subset includes 5 running state information sequences, and decimal sequence values of the 5 running state information sequences are added to obtain an aggregate value of the forward running state attribute subset; similarly, if the negative running state attribute subset includes 3 running state information sequences, adding the decimal sequence values of the 3 running state information sequences to obtain a set value of the negative running state attribute subset.

In another alternative embodiment, the sequence values of the running state information sequences belonging to the same attribute set may be added and an average value may be calculated to obtain set values of different subsets.

Or in an alternative embodiment, the sequence values of the operation state information sequences belonging to the same attribute subset are subjected to weighted addition, that is, the weight of each operation state information sequence is determined, and then the weighted addition is performed, where the weight of each operation state information sequence may be preset or determined based on the attribute information of the control object, and is not described herein again.

In an alternative embodiment, after obtaining the value of each attribute subset, the weight of each attribute may be fused.

Specifically, a first weight value corresponding to the positive running state attribute subset and a second weight value corresponding to the negative running state attribute subset are determined based on the number of running state information sequences of the positive running state attributes and the number of running state information sequences of the negative running state attributes in the running state attribute set.

That is to say, based on the numbers 1 and 2 of the running state information sequences of the forward running state attribute, it can be determined that the total number of the sequences in the running state attribute set is 3, and it may be considered that the first weight value is determined according to the number 1 and the total number 3, and may be directly divided or multiplied, which is not limited herein.

Adding each state value in a positive running state attribute subset in the corresponding running state attribute set to obtain a positive running state total value, and adding each state value in a negative running state attribute subset in the corresponding running state attribute set to obtain a negative running state total value; and obtaining a first set value according to the positive running state total value and the first weight value, and obtaining a second set value according to the negative running state total value and the second weight value.

The above disclosure therefore discloses a first set value and a second set value in each set of operating state attributes, determining a similarity of different target controllable objects by the first set value and the second set value, and clustering each of the target controllable objects based on the similarity.

In the embodiment of the present application, the method of determining similarity may be determined by a euclidean distance-based method, a machine learning-based method, or other methods.

In an alternative embodiment, the clustering may be performed by using a K-means clustering algorithm kmeans, specifically, K objects are randomly selected as initial clustering centers, that is, a first set value and a second set value of K target controllable objects are selected. The distance between each object and the respective seed cluster center is then calculated, and each object is assigned to the cluster center closest to it. The cluster centers and the objects assigned to them represent a cluster. Once all objects are assigned, the cluster center for each cluster is recalculated based on the objects existing in the cluster. This process will be repeated until some termination condition is met. The termination condition may be any one of the following:

1) no (or minimum number) objects are reassigned to different clusters.

2) No (or minimal) cluster centers change again.

3) The sum of squared errors is locally minimal.

After the optional clustering method is introduced, the first set value and the second set value of different target controllable objects can be projected into one clustering coordinate system, the position of each target controllable object in the clustering coordinate system is determined through the two set values, and clustering is performed through the positions of different clustering coordinate systems.

That is, a first set value corresponding to the operation state attribute set of each target controllable object is used as a first coordinate value of a first coordinate axis in the cluster coordinate system, and a second set value corresponding to the operation state attribute set of each target controllable object is used as a second coordinate value of a second coordinate axis in the cluster coordinate system;

As will be exemplarily explained below with reference to fig. 4, in fig. 4, a cluster coordinate system exists, an origin of the cluster coordinate system is (0,0), a first set value of different target controllable objects is used as a first coordinate value of a first coordinate axis in the cluster coordinate system, a second set value corresponding to the set of operating state attributes of each target controllable object is used as a second coordinate value of a second coordinate axis in the cluster coordinate system, and a coordinate position of each target controllable object in the cluster coordinate system is determined based on the first coordinate value and the second coordinate value corresponding to each target controllable object.

And clustering the coordinate positions in the clustering coordinate system by using an Euclidean distance method or any clustering method to obtain clustering results of different target control objects.

Specifically, in fig. 4, it can be determined from the euclidean distances between the coordinate positions of the different target controllable objects, that the target controllable object 1, the target controllable object 2, and the target controllable object 3 belong to the same cluster set, and that the target controllable object 4, the target controllable object 5, and the target controllable object 6 belong to the same cluster set.

In the embodiment of the application, after the clustering results of different target controllable objects are obtained, the target controllable objects belonging to the same clustering set can be recommended.

Illustratively, as shown in fig. 5, what is characterized in fig. 5 is that the user downloads App1 in the App download App, and when the download is completed, recommended apps, App2, App3 and App4, are displayed in an application interface of the App download App.

In the above embodiment, each App is a target controllable object, and when a user uses an App, the App has different use states, which may be represented by 1 or 0, so that cluster analysis can be performed on each App based on a behavior sequence of each App.

As another example, as shown in fig. 6, when the user purchases a product 1 in the shopping App, and when the shopping behavior is completed, recommended products, namely a product 2, a product 3, a product 4, and a product 5, are displayed in an application interface of the shopping App.

In the above embodiment, each commodity is a target controllable object, and when a user purchases a commodity, each commodity has a different use state, and may be characterized by 1 or 0, for example, 1 represents a purchased state, and 0 represents an unpurchased state, so that the commodities can be clustered and analyzed based on the behavior sequence of the user on the different commodities.

Further, in this embodiment of the application, videos may also be used as one target controllable object, and when a user performs a video watching operation, each video may have a different use state, and may be represented by 1 or 0, for example, 1 indicates that a video is watched, and 0 indicates that the video is not watched, so that each video can be subjected to cluster analysis based on a behavior sequence of each user on different videos.

Based on the same concept, an embodiment of the present application further provides a clustering apparatus for target controllable objects, as shown in fig. 7, including:

an operation state information set obtaining unit 701, configured to obtain, for a target controllable object set to be clustered, an operation state information set of each target controllable object, respectively, where the operation state information set of each target controllable object includes at least one operation state information sequence, each operation state information sequence includes operation state information of the target controllable object in at least two consecutive time segments when the target controllable object is controlled by the same control object, each operation state information is determined based on a control behavior of the control object on the target controllable object, and the operation state information represents a used state or an unused state;

an operation state attribute set determining unit 702, configured to determine, for each target controllable object, an operation state attribute corresponding to each operation state information sequence in the operation state information set of the target controllable object according to at least one operation state information in the corresponding operation state information sequence, and obtain an operation state attribute set of each target controllable object, where the operation state attribute includes a positive operation state attribute or a negative operation state attribute;

a clustering unit 703, configured to determine similarities of different target controllable objects based on the positive running state attribute subset and the negative running state attribute subset in the running state attribute set of each target controllable object, and perform clustering on each target controllable object based on the similarities.

Optionally, the running state attribute set determining unit 702 is specifically configured to:

respectively determining a target running state identifier corresponding to running state information of the latest time segment in a corresponding running state information sequence aiming at each target controllable object, wherein the used state corresponds to a first state identifier, and the unused state corresponds to a second state identifier;

and determining the running state attribute corresponding to each running state information sequence in the running state information set of the target controllable object according to the target state identifier and the obtained corresponding relation between the state identifier and the running state attribute.

Optionally, the operation state information set obtaining unit 701 is specifically configured to:

obtaining a clustering rule, and determining at least two continuous target time segmentation information corresponding to the clustering rule;

and aiming at a target controllable object set to be clustered, respectively obtaining an operation state information set of each target controllable object based on a clustering rule, wherein each operation state information sequence comprises operation state information in at least two continuous target time segments corresponding to at least two continuous target time segment information when the target controllable objects are controlled by the same control object.

Optionally, the running state attribute set determining unit 702 is further configured to:

respectively aiming at each target controllable object, determining the state value of each operation state information sequence according to each operation state information in the corresponding operation state information sequence and the corresponding relation between the obtained operation state information and the state value;

determining a first set value of a positive running state attribute subset and a second set value of a negative running state attribute subset in a corresponding running state attribute set based on the state value of each running state information sequence;

the clustering unit 703 is specifically configured to:

Optionally, the corresponding relationship between the operating state information and the state value is that the operating state information representing the used state is assigned as a state value 1, and the operating state information representing the unused state is assigned as a state value 0;

the running state attribute set determining unit 702 is specifically configured to:

respectively aiming at each target controllable object, assigning the running state information representing the used state in the corresponding running state information sequence to be a state value 1, assigning the running state information representing the unused state to be a state value 0, and sequencing the state values of each running state information according to the corresponding time segment from near to far to obtain a corresponding binary state value sequence;

and carrying out decimal conversion on each binary state value sequence to obtain the state value of the running state information sequence.

determining a first weight value corresponding to the positive running state attribute subset and a second weight value corresponding to the negative running state attribute subset based on the number of running state information sequences of the positive running state attributes and the number of running state information sequences of the negative running state attributes in the running state attribute set;

adding each state value in a positive running state attribute subset in the corresponding running state attribute set to obtain a positive running state total value, and adding each state value in a negative running state attribute subset in the corresponding running state attribute set to obtain a negative running state total value;

and obtaining a first set value according to the positive running state total value and the first weight value, and obtaining a second set value according to the negative running state total value and the second weight value.

Optionally, the clustering unit 703 is specifically configured to:

As can be seen from the above, in the embodiment of the present application, the clustering method and the clustering device for different target controllable objects determine the similarity based on the positive operating state and the negative operating state corresponding to the different target controllable objects, where the positive operating state and the negative operating state are determined based on the control behavior of the control object, that is, the control behavior similarity between the different target controllable objects is determined from the control behavior of each target controllable object, and perform clustering through the similarity, that is, cluster a plurality of target controllable objects having similar control behavior characteristics into one cluster, so as to facilitate recommendation, analysis, and other uses.

Based on the same technical concept, the embodiment of the present application provides a computer device, as shown in fig. 8, including at least one processor 801 and a memory 802 connected to the at least one processor, where a specific connection medium between the processor 801 and the memory 802 is not limited in the embodiment of the present application, and the processor 801 and the memory 802 are connected through a bus in fig. 8 as an example. The bus may be divided into an address bus, a data bus, a control bus, etc.

In the embodiment of the present application, the memory 802 stores instructions executable by the at least one processor 801, and the at least one processor 801 may execute the steps included in the method for clustering target controllable objects by executing the instructions stored in the memory 802.

The processor 801 is a control center of the computer device, and may connect various parts of the computer device by using various interfaces and lines, and create a virtual machine by executing or executing instructions stored in the memory 802 and calling up data stored in the memory 802. Optionally, the processor 801 may include one or more processing units, and the processor 801 may integrate an application processor and a modem processor, wherein the application processor mainly handles operating systems, user interfaces, application programs, and the like, and the modem processor mainly handles wireless communications. It will be appreciated that the modem processor described above may not be integrated into the processor 801. In some embodiments, the processor 801 and the memory 802 may be implemented on the same chip, or in some embodiments, they may be implemented separately on separate chips.

The processor 801 may be a general-purpose processor, such as a Central Processing Unit (CPU), a digital signal processor, an Application Specific Integrated Circuit (ASIC), a field programmable gate array or other programmable logic device, discrete gate or transistor logic, discrete hardware components, or any combination thereof, configured to implement or perform the methods, steps, and logic blocks disclosed in the embodiments of the present Application. A general purpose processor may be a microprocessor or any conventional processor or the like. The steps of a method disclosed in connection with the embodiments of the present application may be directly implemented by a hardware processor, or may be implemented by a combination of hardware and software modules in a processor.

Memory 802, which is a non-volatile computer-readable storage medium, may be used to store non-volatile software programs, non-volatile computer-executable programs, and modules. The Memory 802 may include at least one type of storage medium, and may include, for example, a flash Memory, a hard disk, a multimedia card, a card-type Memory, a Random Access Memory (RAM), a Static Random Access Memory (SRAM), a Programmable Read Only Memory (PROM), a Read Only Memory (ROM), a charge Erasable Programmable Read Only Memory (EEPROM), a magnetic Memory, a magnetic disk, an optical disk, and so on. The memory 802 is any other medium that can be used to carry or store desired program code in the form of instructions or data structures and that can be accessed by a computer, but is not limited to such. The memory 802 in the embodiments of the present application may also be circuitry or any other device capable of performing a storage function for storing program instructions and/or data.

Based on the same inventive concept, embodiments of the present application provide a computer-readable storage medium storing a computer program executable by a computer device, which, when the program is run on the computer device, causes the computer device to perform the steps of the clustering method of target controllable objects.

As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.

The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

It will be apparent to those skilled in the art that various changes and modifications may be made in the present application without departing from the spirit and scope of the application. Thus, if such modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is intended to include such modifications and variations as well.

Claims

1. A method of clustering target controllable objects, the method comprising:

aiming at a target controllable object set to be clustered, respectively obtaining an operation state information set of each target controllable object, wherein the operation state information set of each target controllable object comprises at least one operation state information sequence, each operation state information sequence comprises operation state information of the target controllable object in at least two continuous time segments when the target controllable object is controlled by the same control object, each operation state information is determined based on the control behavior of the control object on the target controllable object, and the operation state information represents a used state or an unused state;

respectively aiming at each target controllable object, determining an operation state attribute corresponding to each operation state information sequence in the operation state information set of the target controllable object according to at least one operation state information in the corresponding operation state information sequence, and obtaining the operation state attribute set of each target controllable object, wherein the operation state attribute comprises a positive operation state attribute or a negative operation state attribute;

determining similarity of different target controllable objects based on the positive running state attribute subset and the negative running state attribute subset in the running state attribute set of each target controllable object, and clustering each target controllable object based on the similarity.

2. The method according to claim 1, wherein the determining, for each target controllable object, an operating state attribute corresponding to each operating state information sequence in the operating state information set of the target controllable object according to at least one operating state information in the corresponding operating state information sequence, respectively, comprises:

respectively determining a target running state identifier corresponding to the running state information of the latest time segment in the running state information sequence for each target controllable object, wherein the used state corresponds to a first state identifier, and the unused state corresponds to a second state identifier;

3. The method according to claim 1, wherein the obtaining the set of operating state information of each target controllable object for the set of target controllable objects to be clustered respectively comprises:

and respectively obtaining an operation state information set of each target controllable object based on the clustering rule, wherein each operation state information sequence comprises operation state information in at least two continuous target time segments corresponding to the at least two continuous target time segment information when the target controllable objects are controlled by the same control object.

4. The method according to claim 2, wherein after determining, for each target controllable object, an operating state attribute corresponding to each operating state information sequence in the operating state information set of the target controllable object according to at least one operating state information in the corresponding operating state information sequence, the method further includes:

respectively aiming at each target controllable object, determining the state value of each running state information sequence according to each running state information in the corresponding running state information sequence and the corresponding relation between the running state information and the state value;

determining a first set value of the positive running state attribute subset and a second set value of the negative running state attribute subset in the corresponding running state attribute set based on the state value of each running state information sequence;

said determining similarities for different target controllable objects based on said positive run state attribute subset and said negative run state attribute subset of said run state attribute sets for each of said target controllable objects comprises:

determining similarity of different target controllable objects based on the first set value and the second set value corresponding to the running state attribute set of each target controllable object.

5. The method according to claim 4, wherein the corresponding relationship between the operating state information and the state value is that the operating state information representing the used state is assigned to the state value 1, and the operating state information representing the unused state is assigned to the state value 0;

the determining, for each target controllable object, a state value of each operating state information sequence according to each operating state information in the corresponding operating state information sequence and a corresponding relationship between the operating state information and the state value includes:

respectively aiming at each target controllable object, assigning all running state information representing the used state in the corresponding running state information sequence to be a state value 1, assigning all running state information representing the unused state to be a state value 0 according to the corresponding relation between the running state information and the state value, and sequencing the state value of each running state information according to the corresponding time segment from near to far to obtain a corresponding binary state value sequence;

6. The method of claim 5, wherein said determining, based on the state value of each of the sequences of operating state information, a first set value of the positive subset of operating state attributes and a second set value of the negative subset of operating state attributes of the corresponding set of operating state attributes comprises:

determining a first weight value corresponding to the positive running state attribute subset and a second weight value corresponding to the negative running state attribute subset based on the number of the running state information sequences of the positive running state attributes and the number of the running state information sequences of the negative running state attributes in the running state attribute set;

adding each state value in the positive running state attribute subset in the corresponding running state attribute set to obtain a positive running state total value, and adding each state value in the negative running state attribute subset in the corresponding running state attribute set to obtain a negative running state total value;

and obtaining the first aggregate value according to the positive running state total value and the first weight value, and obtaining the second aggregate value according to the negative running state total value and the second weight value.

7. The method according to any of claims 4 to 6, wherein the determining the similarity between different target controllable objects based on the first set value and the second set value corresponding to the set of operating state attributes of each target controllable object comprises:

taking the first set value corresponding to the running state attribute set of each target controllable object as a first coordinate value of a first coordinate axis in a clustering coordinate system, and taking the second set value corresponding to the running state attribute set of each target controllable object as a second coordinate value of a second coordinate axis in the clustering coordinate system;

determining a coordinate position of each of the target controllable objects in the cluster coordinate system based on the first coordinate value and the second coordinate value corresponding to each of the target controllable objects;

determining a similarity of different ones of the target controllable objects based on a coordinate position of each of the target controllable objects in the cluster coordinate system.

8. A clustering apparatus of target controllable objects, comprising:

an operation state information set obtaining unit, configured to obtain, for a target controllable object set to be clustered, an operation state information set of each target controllable object, where the operation state information set of each target controllable object includes at least one operation state information sequence, each operation state information sequence includes operation state information of the target controllable object in at least two consecutive time segments when the target controllable object is controlled by the same control object, each operation state information is determined based on a control behavior of the control object on the target controllable object, and the operation state information represents a used state or an unused state;

an operating state attribute set determining unit, configured to determine, for each target controllable object, an operating state attribute corresponding to each operating state information sequence in the operating state information set of the target controllable object according to at least one piece of operating state information in the corresponding operating state information sequence, and obtain the operating state attribute set of each target controllable object, where the operating state attribute includes a positive operating state attribute or a negative operating state attribute;

a clustering unit, configured to determine similarities of different target controllable objects based on the positive running state attribute subset and the negative running state attribute subset in the running state attribute set of each target controllable object, and perform clustering on each target controllable object based on the similarities.

9. The apparatus according to claim 8, wherein the operating state attribute set determining unit is specifically configured to:

10. The apparatus according to claim 8, wherein the operation state information set obtaining unit is specifically configured to:

and aiming at a target controllable object set to be clustered, respectively obtaining an operation state information set of each target controllable object based on the clustering rule, wherein each operation state information sequence comprises operation state information in at least two continuous target time segments corresponding to the information of the at least two continuous target time segments when the target controllable objects are controlled by the same control object.

11. The apparatus of claim 9, wherein the run state attribute set determination unit is further configured to:

respectively aiming at each target controllable object, determining the state value of each running state information sequence according to each running state information in the corresponding running state information sequence and the corresponding relation between the obtained running state information and the state value;

the clustering unit is specifically configured to:

12. The apparatus according to claim 11, wherein the corresponding relationship between the operating status information and the status value is that the operating status information representing the used status is assigned to the status value 1, and the operating status information representing the unused status is assigned to the status value 0;

the running state attribute set determining unit is specifically configured to:

13. The apparatus according to claim 12, wherein the operating state attribute set determining unit is specifically configured to:

14. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the steps of the method of any one of claims 1 to 7 are performed by the processor when the program is executed.

15. A computer-readable storage medium, in which a computer program is stored which is executable by a computer device, and which, when run on the computer device, causes the computer device to carry out the steps of the method as claimed in any one of claims 1 to 7.