Method and apparatus for managing object, device, and medium

Abstract

Provided are a method and apparatus for managing an object, a device, and a medium. The method comprises: respectively acquiring a plurality of responses of a plurality of objects for a media item, wherein the media item comprises a question, and the plurality of responses are respectively encrypted responses of the plurality of objects for the question; determining at least one response type of the plurality of responses; on the basis of the question and the at least one response type of the plurality of responses, dividing the plurality of objects into at least one group, wherein a group among the at least one group corresponds to a response type among the at least one response type; and on the basis of related data of the objects in the at least one group, determining a sample data set used for updating an object grouping model. In this way, users can be divided into groups without knowing specific information in a user response. Furthermore, related data of users in each group can be used for determining a machine learning model, thereby achieving a desired purpose.

Description

Method, apparatus, device and medium for managing objects TECHNICAL FIELD

[0001] Exemplary implementations of the present disclosure generally relate to object management, and in particular, to a method, apparatus, device and computer-readable storage medium for managing objects in a recommendation system. BACKGROUND

[0002] Machine learning techniques have been widely used in various application environments. In a recommendation scenario, a machine learning model can be utilized to build a recommendation model, and then various objects can be recommended with media items that can be of interest. The machine learning model can be trained by utilizing historical data. However, detailed historical data is not always available for performing the training process due to data protection and other reasons. This results in unsatisfactory performance of the machine learning model.

[0003] SUMMARY

[0004] In a first aspect of the present disclosure, a method for managing objects is provided. In the method, a plurality of responses of a plurality of objects to a media item are respectively obtained, the media item comprising a question, and the plurality of responses are respectively encrypted responses of the plurality of objects to the question. At least one response type of the plurality of responses is determined. Based on the question and the at least one response type of the plurality of responses, the plurality of objects are divided into at least one group, a group in the at least one group corresponding to a response type in the at least one response type. Based on related data of the objects in the at least one group, a sample data set for updating an object grouping model is determined.

[0005] In a second aspect of the present disclosure, an apparatus for managing objects is provided. The apparatus comprises: an obtaining module configured to respectively obtain a plurality of responses of a plurality of objects to a media item, the media item comprising a question, and the plurality of responses being respectively encrypted responses of the plurality of objects to the question; a type determining module configured to determine at least one response type of the plurality of responses; a dividing module configured to divide, based on the question and the at least one response type of the plurality of responses, the plurality of objects into at least one group, a group in the at least one group corresponding to a response type in the at least one response type; and a data set determining module configured to determine, based on related data of the objects in the at least one group, a sample data set for updating an object grouping model.

[0006] In a third aspect of the present disclosure, an electronic device is provided. The electronic device comprises: at least one processing unit; and at least one memory coupled to the at least one processing unit and storing instructions for execution by the at least one processing unit, the instructions, when executed by the at least one processing unit, causing the electronic device to perform the method according to the first aspect of the present disclosure.

[0007] In a fourth aspect of the disclosure, a computer readable storage medium is provided, having stored thereon a computer program which, when executed by a processor, causes the processor to implement the method according to the first aspect of the disclosure.

[0008] In a fifth aspect of the disclosure, a computer program product is provided, comprising a computer program which, when executed by a processor, implements the method according to the first aspect of the disclosure.

[0009] It is to be understood that the contents described in this section are not intended to limit the key features or important features of the implementations of the disclosure, nor to limit the scope of the disclosure. Other features of the disclosure will become apparent from the following description. BRIEF DESCRIPTION OF DRAWINGS

[0010] The above and other features, advantages, and aspects of the implementations of the disclosure will become more apparent by describing in detail some implementations thereof with reference to the attached drawings in which:

[0011] FIG. 1 illustrates a block diagram of an application environment according to one example implementation of the disclosure;

[0012] FIG. 2 illustrates a block diagram for managing objects according to some implementations of the disclosure;

[0013] FIG. 3 illustrates a block diagram for determining a type of response according to some implementations of the disclosure;

[0014] FIG. 4 illustrates a block diagram for determining similar objects based on object grouping according to some implementations of the disclosure;

[0015] FIG. 5 illustrates a block diagram of a process for recommending media items to objects according to some implementations of the disclosure;

[0016] FIG. 6 illustrates a flowchart of a method for managing objects according to some implementations of the disclosure;

[0017] FIG. 7 illustrates a block diagram of an apparatus for managing objects according to some implementations of the disclosure; and

[0018] FIG. 8 illustrates a block diagram of a device capable of implementing the implementations of the disclosure. DETAILED DESCRIPTION

[0019] Implementations of the present disclosure will be described more fully hereinafter with reference to the accompanying drawings. While several implementations of the present disclosure are described, it should be understood that the present disclosure can be embodied in many other forms and should not be construed as limited to the implementations set forth herein; rather, these implementations are provided so that this disclosure will be thorough and complete, and fully convey the scope of the present disclosure to those skilled in the art. It should be understood that the drawings and implementations described are for illustrative purposes only and are not intended to limit the scope of the present disclosure.

[0020] In the description of implementations of the present disclosure, the term "includes" and its derivatives mean "including but not limited to". The term "based on" means "based at least in part on". The term "one implementation" or "the implementation" means "at least one implementation". The term "some implementations" means "at least some implementations". Other explicit or implicit definitions can also be included below. As used herein, the term "model" can represent the relationship between various data. For example, the above-mentioned relationship can be obtained based on various technical solutions known at present and / or to be developed in the future.

[0021] It can be understood that the data involved in the technical solutions of the present disclosure (including but not limited to the data itself, the acquisition or use of the data) should comply with the requirements of relevant laws and regulations and relevant provisions.

[0022] It can be understood that before using the technical solutions disclosed by the embodiments of the present disclosure, the type, use range, use scenario, etc. of the personal information involved in the present disclosure should be informed to the user and the authorization of the user should be obtained through appropriate means according to relevant laws and regulations.

[0023] For example, in response to receiving the active request of the user, the user is sent prompt information to explicitly prompt the user that the operation requested to be executed will require the acquisition and use of the personal information of the user. Thus, the user can voluntarily choose whether to provide personal information to the software or hardware such as electronic device, application program, server or storage medium, etc. that executes the operation of the technical solutions of the present disclosure according to the prompt information.

[0024] As an optional but non-limiting implementation, in response to receiving the active request of the user, the way of sending prompt information to the user, for example, can be the way of pop-up window, and the prompt information can be presented in the form of text in the pop-up window. In addition, the pop-up window can also carry selection controls for the user to select "agree" or "disagree" to provide personal information to the electronic device.

[0025] It can be understood that the above-mentioned notification and acquisition of user authorization process is only illustrative, and does not limit the implementations of the present disclosure, and other ways that meet the relevant laws and regulations can also be applied to the implementations of the present disclosure.

[0026] The term "in response to" as used herein refers to a state in which a corresponding event occurs or a condition is satisfied. It will be appreciated that the timing of the execution of a subsequent action performed in response to the event or condition is not necessarily strongly correlated with the time at which the event occurs or the condition is satisfied. For example, in some cases, the subsequent action can be performed immediately upon the occurrence of the event or the satisfaction of the condition; in other cases, the subsequent action can be performed after a period of time has elapsed since the occurrence of the event or the satisfaction of the condition.

[0027] Example Environment

[0028] In a recommendation scenario, a machine learning model can be utilized to construct a recommendation model, and then media items that can be of interest to each object can be recommended to the object. For ease of description, details of object management are described below only with a media sharing application as a specific application environment. Alternatively and / or additionally, the object management process can be performed in other application environments (e.g., a social application, a shopping application, etc.).

[0029] FIG. 1 illustrates a block diagram 100 of an application environment according to one example implementation of the present disclosure. As shown in FIG. 1, one or more media items 120 can be provided to an object 110 in a media sharing application. Here, the object 110 can be a user in the media sharing application, and the media item 120 can include, but is not limited to, a short video, a video, text, an image, and / or other rich media data, etc. In a recommendation scenario, the media item 120 can include one or more questions 130. For example, the media item 120 can be presented in a rich media format, and the question 130 can be a dialog box in the rich media, and the object 110 can respond 140 by selecting an option in the question 130.

[0030] Suppose the media item 120 is a media item promoting a new car, the question 130 can include, for example, "Do you have a driver's license?". At this time, option 1 can correspond to "Yes", and option 2 can correspond to "No". It should be understood that the question and the options here are only exemplary, and in other recommendation scenarios, the question and the options can have different content.

[0031] In the scenario of recommending a car, in general, it is more inclined to recommend the corresponding media item to a user who has a driver's license, so as to make the recommended media item more in line with the user's needs. However, for data security purposes, the response 140 provided by the user is usually stored in an encrypted format, thereby ensuring that the user's personally relevant information is not leaked. At this time, it is not known whether the user has a driver's license, so the relevant information of "whether to have a driver's license" cannot be used in the recommendation process. This results in lower accuracy of the recommendation, and the recommended media item is difficult to meet the user's needs.

[0032] At this time, how to manage the objects in a more effective way, for example, it is expected that the objects can be divided into multiple groups according to some similar attributes. Further, different media items can be recommended to the users in different groups, thereby improving the performance and accuracy of the recommendation.

[0033] Summary of object management

[0034] In order to at least partially solve the deficiencies in the prior art, according to one exemplary implementation of the present disclosure, a method for managing objects is proposed. A summary according to one exemplary implementation of the present disclosure is described with reference to FIG. 2, which shows a block diagram 200 for managing objects according to some implementations of the present disclosure. As shown in FIG. 2, a plurality of objects (e.g., object 210) can be respectively acquired for a plurality of responses (e.g., response 220) to a media item.

[0035] Here, the media item includes a question, and the plurality of responses are respectively encrypted responses of the plurality of objects to the question. Specifically, the question can be embedded inside the media item, or attached at the end of the media item, etc. For example, for a media item in a rich media format, the question and the corresponding options can be embedded in the rich media data. For example, for a media item in a video format, a dialog box format question and options can be presented at the end of the video for the user to select. It should be understood that in some cases, the original response (in unencrypted format) of the user to the question cannot be acquired. Instead, the original response is encrypted using an encryption algorithm, and the encrypted response is, for example, in a garbled format and invisible.

[0036] Further, at least one response type (e.g., response type 230) of the plurality of responses can be determined. For the question in FIG. 1, although the specific content of the response (i.e., "yes" or "no") cannot be determined, the response type of the response can still be determined. That is, for the two encrypted responses, it can be determined whether the types of the two responses are the same. Further, based on the question and the at least one response type of the plurality of responses, the plurality of objects can be divided into at least one group (e.g., group 240). Specifically, a group in the at least one group corresponds to a response type in the at least one response type, that is, there can be a one-to-one correspondence between the response type and the group. Further, based on the related data (e.g., 250) of the objects in the at least one group, a sample data set for updating an object group model 260 can be determined.

[0037] With the example implementation of the present disclosure, the user can be divided into groups without knowing the specific information in the user's answer. Further, the related data of the users in each group can be utilized to determine a machine learning model, and thus achieve the desired purpose. Specifically, in the above example, the users can be divided into two groups: a group of users with a driver's license, and a group of users without a driver's license, without knowing the specific content of the encrypted answer. Although it is not known at this time whether the users in a certain group have a driver's license, the user groups can still be obtained, so as to recommend different media items to different user groups.

[0038] Detailed process of object management

[0039] Having described an overview of some implementations according to the present disclosure, in the following, more details about object management will be described. According to some implementations of the present disclosure, a question is a question raised for an object, associated with a recommendation target, which can include a plurality of options associated with an answer, and the answer can specify an option among the plurality of options. For example, in the scenario of recommending a car, the question may, for example, include "Do you have a driver's license?" In the scenario of recommending a travel product, the question may, for example, include "Do you have a driver's license?", and so on.

[0040] It should be understood that although the above only shows the case where the question includes two options of "yes" and "no", alternatively and / or additionally, the question can include more options. For example, in the scenario of recommending a new car, the question may, for example, include "When do you plan to buy a new car?" At this time, the options may, for example, include: Option 1: within half a year; Option 2: within one year; and Option 3: uncertain, and so on. In this way, the user can be allowed to provide an answer to the question by selecting the desired option without the need for complex input operations. It should be understood that in order to ensure the security of the information in the answer, the answer is stored and transmitted in an encrypted format, and thus the specific content of the answer input by the user cannot be obtained from the answer.

[0041] According to some implementations of the present disclosure, in the process of determining at least one answer type, a mapping function can be established, and the mapping function can be utilized to determine the answer type of an answer among a plurality of answers. The mapping function can be established by utilizing a variety of encoding technical solutions, which maps a first answer and a second answer among the plurality of answers to the same answer type, and maps the first answer and a third answer among the plurality of answers to different answer types, a first unencrypted answer corresponding to the first answer is the same as a second unencrypted answer corresponding to the second answer, and the first unencrypted answer is different from a third unencrypted answer corresponding to the third answer.

[0042] In other words, the first answer and the second answer have the same content (e.g., both are "yes") and can be mapped to the same answer type. The first answer and the third answer have different content (e.g., "yes" and "no") and can be mapped to different answer types. In this way, although the content of each user's specific answer cannot be determined, the answer types of the respective answers can still be utilized to group the users.

[0043] FIG. 3 illustrates a block diagram 300 for determining answer types, according to some implementations of the present disclosure, which illustrates answers of multiple objects to a certain question. The first column can represent the objects 310 (e.g., 4 objects are represented as Ol, 02, 03, and 04, respectively). The second column can represent the selected options of the respective objects (e.g., the objects Ol to 04 select the options Al, A2, A2, and A3, respectively, i.e., the objects 02 and 03 both select the option A2). The third column can represent the encrypted answers 330 generated by utilizing an encryption algorithm (e.g., a conventional encryption algorithm), in which the options Al, A2, A2, and A3 are encrypted as El, E2, E3, and E4, respectively. Although the objects 02 and 03 both select the option A2 at this time, the encrypted answers E2 and E3 are not the same.

[0044] The fourth column can represent the answer types 340 determined by utilizing the mapping function described above, as shown in FIG. 4, in which the objects 02 and 03 both select the option A2, and the answer types of the selected options of the objects 02 and 03 are both T2. By utilizing the example implementations of the present disclosure, for a certain question, the objects 02 and 03 select the same option A2, and the option A2 has a unique answer type. In this way, the objects 02 and 03 can be divided into the same group. At this time, based on the answer types Tl, T2, and T3, the object Ol can be divided into a first group, the objects 02 and 03 can be divided into a second group, and the object 03 can be divided into a third group.

[0045] According to some implementations of the present disclosure, in the process of dividing the groups, different questions can be distinguished. Specifically, a plurality of binary tuples respectively corresponding to the plurality of objects can be established by utilizing the question and at least one answer type. Then, the plurality of binary tuples can be clustered to divide the plurality of objects into at least one group. In this way, in the process of determining the clusters of the objects, the groups of the objects can be divided at a more fine-grained granularity, thereby preparing more abundant training data for the subsequent training process. Specifically, the binary tuple can be represented by <question, type>, in which question can represent the identification of the question, and type can represent the identification of the answer type.

[0046] According to some implementations of the present disclosure, in the process of determining the sample dataset, a matching degree between the object and the recommendation target of the media item can be determined, and based on the matching degree and the related data of the object, the sample dataset is generated. Here, the matching degree can represent whether the response type of the object is helpful to achieve the recommendation target of the media item. For example, in the scenario of recommending a vehicle, the recommendation target is to improve the promotion, sales, etc. of the vehicle. At this time, for the question "Do you have a driver's license", the object selecting the option "Yes" has a higher matching degree with the recommendation target, and the object selecting the option "No" has a lower matching degree with the recommendation target. In this way, the corresponding sample data (e.g., determining positive samples and negative samples) can be further determined according to whether it is helpful to improve the recommendation target.

[0047] According to some implementations of the present disclosure, the matching degree is determined based on events associated with the object and the recommendation target. Here, the events can be associated with the historical interactions of the object, for example, in the scenario of recommending a car, it is assumed that it is found that a plurality of users in a certain group have purchased a new car in the later stage, or various accessories, services, etc. matched with the new car model, it can be inferred that the users in the group have a driver's license.

[0048] Here, the events are obtained by suitable means with the permission of the user. For example, the events can come from inside the media sharing application (e.g., the user's subscription, subscription events, etc. inside the application); alternatively and / or additionally, the events can come from outside the media sharing application (e.g., from a car seller, etc.). In this way, the specific content of the response submitted by the user can be inferred in a more accurate manner, thereby improving the performance of the subsequent processing process.

[0049] According to some implementations of the present disclosure, the positive sample data and the negative sample data can be determined based on the matching degree between the object and the recommendation target. Here, it is not necessary to collect the specific content of the events associated with each object, but only based on the related events of part of the objects, it can be determined that the objects in the group belong to positive samples or negative samples. With the example implementations of the present disclosure, the matching degree between the object and the recommendation target can be determined by the response type, and thus the positive sample data and / or the negative sample data are generated accordingly.

[0050] Specifically, in the process of generating the sample dataset, in response to determining that the matching degree satisfies the predetermined condition, positive sample data in the sample dataset is generated for the related data of each object in the group including the object among the at least one group. At this time, the positive sample data can include multiple dimensions, such as the identification of the object, the identification of the question, the identification of the response type, and a label "1" representing a positive sample, and the like. Alternatively and / or additionally, in response to determining that the matching degree does not satisfy the predetermined condition, negative sample data in the sample dataset is generated for the related data of each object in the group including the object among the at least one group. At this time, the negative sample data can include multiple dimensions, such as the identification of the object, the identification of the question, the identification of the response type, and a label "0" representing a negative sample, and the like.

[0051] According to some implementations of the present disclosure, the object grouping model can be updated with the sample dataset, so that the object grouping model describes the association between the grouping of the object and the response type. It should be understood that although the specific content in the response cannot be directly obtained from the user, the classification of the response can be inferred through the response type. In this way, the trained object grouping model can extract the knowledge in the positive and negative sample data, and further determine the grouping of the object based on the knowledge.

[0052] It should be understood that although the above only illustrates the process of processing one question, alternatively and / or additionally, the technical solutions described above can be used to process multiple questions respectively, and thus obtain the grouping of the object associated with each response type of each question. More details are described with reference to FIG. 4, which illustrates a block diagram 400 for determining similar objects based on object grouping according to some implementations of the present disclosure. As shown in FIG. 4, there can be questions 410 (e.g., denoted as Q1),..., and question 412 (e.g., denoted as QN).

[0053] In the following, only the question 410 is described as an example. It is assumed that the question 410 includes 3 options, and the responses from multiple objects are mapped to response types 420 and 422 respectively. At this time, the response types 420 and 422 correspond to object groups 430 and 432 respectively. Continuing the above example, it is assumed that the question 410 is "Do you have a driver's license?" The response types 420 and 422 correspond to "Yes" and "No" respectively. At this time, with the object grouping model, users with a driver's license can be determined from the object group 430, and users without a driver's license can be determined from the object group 432.

[0054] According to some implementations of the present disclosure, a respective candidate media item can be recommended to different types of objects. For a certain candidate media item, a candidate answer type associated with a candidate recommendation target of the candidate media item can be determined, the candidate media item including a question. Further, a candidate group of objects associated with the candidate answer type can be determined based on the object grouping model, and the candidate media item can be recommended to the objects in the candidate group. In this way, a media item that is more suitable for a certain type of object can be recommended to the objects of this type, thereby improving the accuracy of data recommendation.

[0055] According to some implementations of the present disclosure, in determining the candidate answer type, candidate related data of a candidate object providing a candidate answer for the candidate media item can be obtained. In response to determining that the candidate related data indicates that a matching degree between the candidate object and the candidate recommendation target of the candidate media item satisfies a predetermined condition, the candidate answer type is determined based on an answer type of the candidate answer. In this way, the candidate answer type can be determined in a more accurate manner.

[0056] It should be appreciated that the above-described method can be used during a cold start of a recommendation system. It should be appreciated that during the cold start, only a small amount of historical data can be obtained, which cannot adjust the recommendation model to a desired accuracy. With the example implementations of the present disclosure, by determining a candidate answer type that is more matched to a candidate media item, objects corresponding to the candidate answer type can be determined based on an object grouping model that has been verified to be reliable. Thereby, the recommendation accuracy during the cold start of the recommendation system is improved.

[0057] For example, the candidate media item 1 can be a media item recommending a self-driving tour product (the media item including a question “Do you have a driver’s license?” or a question expressed in a similar manner). During the cold start, for a small number of users who have purchased the self-driving tour product, the answer type of the user to the question can be obtained. At this time, it can be determined that users with a driver’s license are more matched to the recommendation target of the self-driving tour product, and thus the candidate media item 1 can be recommended to the users with a driver’s license in the object group 430.

[0058] For example, the candidate media item 2 can be a media item recommending a group tour product (the media item including a question “Do you have a driver’s license?” or a question expressed in a similar manner). During the cold start, for a small number of users who have purchased the group tour product, the answer type of the user to the question can be obtained. At this time, it can be determined that users without a driver’s license are more matched to the recommendation target of the group tour product, and thus the candidate media item 2 can be recommended to the users with a driver’s license in the object group 432.

[0059] It should be appreciated that although the process of performing object management is described above by way of example with respect to whether an object has a driver's license, alternatively and / or additionally, other issues can be handled in a similar manner. For example, issue 412 can include N options, and a plurality of objects that answer issue 412 can be divided into N different object groups. Different issues and respective object answers to the issues can be obtained, and the object group model can be updated accordingly. For example, another issue can relate to "your education level," and the options can include: high school, college, graduate school, etc. In this way, given the issue and the answer types, the object group model can be utilized to determine the respective object groups. Further, the education level of respective users can be inferred, and respective media items can be recommended to respective users accordingly.

[0060] FIG. 5 illustrates a block diagram 500 of a process for recommending media items to objects, according to some implementations of the present disclosure. As shown, the recommendation model 510 can include the object classification model 260, and the recommendation can be performed using the process as shown in FIG. 5. The media item 520 can include an issue 522, which can be input to the recommendation model 510 along with respective answer types, in order to determine respective object groups. Further, an object group that is more matched to the recommendation goal of the media item 520 can be selected from the plurality of object groups, and respective objects 530 in the group can be output. At this point, the media item 520 can be recommended to the objects 530, thereby improving the accuracy of the recommendation process.

[0061] Alternatively and / or additionally, for a particular object group, media items that are matched to the object group can be selected from a plurality of media items. Specifically, for the object group 430 that has a driver's license, media items related to new car recommendations, media items related to self-driving tours, and media items related to vehicle maintenance and care, etc. can be recommended preferentially. For example, for the object group 432 that does not have a driver's license, media items for group tours, media items for driving school recommendations, etc. can be recommended preferentially.

[0062] With the exemplary implementations of the present disclosure, the groups of users can be divided without knowing the specific relevant information in the user answers. Further, the machine learning model can be determined using the relevant data of the users in respective groups, and the desired purposes can be achieved accordingly.

[0063] Example Process

[0064] FIG. 6 illustrates a flowchart of a method 600 for managing objects, according to some implementations of the present disclosure. At block 610, a plurality of responses of a plurality of objects to a media item are obtained, respectively, the media item comprising a question, and the plurality of responses are encrypted responses of the plurality of objects to the question, respectively. At block 620, at least one response type of the plurality of responses is determined; at block 630, the plurality of objects are partitioned into at least one group based on the question and the at least one response type of the plurality of responses, a group in the at least one group corresponding to a response type in the at least one response type; and at block 640, a sample dataset for updating an object grouping model is determined based on related data of the objects in the at least one group.

[0065] According to some implementations of the present disclosure, determining the at least one response type comprises: establishing a mapping function that maps a first response and a second response in the plurality of responses to a same response type, and maps a first response and a third response in the plurality of responses to different response types, a first unencrypted response corresponding to the first response being the same as a second unencrypted response corresponding to the second response, and the first unencrypted response being different from a third unencrypted response corresponding to the third response; and determining the response type of a response in the plurality of responses using the mapping function.

[0066] According to some implementations of the present disclosure, partitioning the plurality of objects into the at least one group comprises: establishing a plurality of pairs respectively corresponding to the plurality of objects using the question and the at least one response type; and clustering the plurality of pairs to partition the plurality of objects into the at least one group.

[0067] According to some implementations of the present disclosure, determining the sample dataset comprises: determining a matching degree between the object and a recommended target of the media item; and generating the sample dataset based on the matching degree and the related data of the object.

[0068] According to some implementations of the present disclosure, generating the sample dataset comprises at least one of: in response to determining that the matching degree satisfies a predetermined condition, generating positive sample data in the sample dataset for the related data of each object in a group in the at least one group that includes the object; and in response to determining that the matching degree does not satisfy the predetermined condition, generating negative sample data in the sample dataset for the related data of each object in a group in the at least one group that includes the object.

[0069] According to some implementations of the present disclosure, the matching degree is determined based on an event associated with the object and the recommended target.

[0070] According to some implementations of the present disclosure, the method further comprises: updating the object grouping model using the sample dataset, such that the object grouping model describes an association between the groups of objects and the response types.

[0071] According to some implementations of the present disclosure, the method further includes: determining a candidate answer type associated with a candidate recommendation target of a candidate media item, the candidate media item including a question; determining, based on the object grouping model, a candidate group of objects associated with the candidate answer type; and recommending, to the objects in the candidate group, the candidate media item.

[0072] According to some implementations of the present disclosure, determining the candidate answer type includes: obtaining candidate correlation data of a candidate object providing a candidate answer for the candidate media item; and in response to determining that the candidate correlation data indicates that a degree of matching between the candidate object and the candidate recommendation target of the candidate media item satisfies a predetermined condition, determining the candidate answer type based on an answer type of the candidate answer.

[0073] According to some implementations of the present disclosure, the question is a question posed to the objects in association with the recommendation target, the question includes a plurality of options associated with the answer, and the answer specifies an option of the plurality of options.

[0074] Example apparatuses and devices

[0075] FIG. 7 illustrates a block diagram of an apparatus 700 for managing objects according to some implementations of the present disclosure. The apparatus 700 includes: an obtaining module 710 configured to obtain a plurality of answers of a plurality of objects respectively for a media item, the media item including a question, and the plurality of answers being encrypted answers of the plurality of objects respectively for the question; a type determining module 720 configured to determine at least one answer type of the plurality of answers; a dividing module 730 configured to divide the plurality of objects into at least one group based on the question and the at least one answer type of the plurality of answers, a group of the at least one group corresponding to an answer type of the at least one answer type; and a dataset determining module 740 configured to determine a sample dataset for updating an object grouping model based on correlation data of the objects in the at least one group.

[0076] According to some implementations of the present disclosure, the type determining module includes: an establishing module configured to establish a mapping function that maps a first answer and a second answer of the plurality of answers to a same answer type and maps a first answer and a third answer of the plurality of answers to different answer types, a first unencrypted answer corresponding to the first answer being the same as a second unencrypted answer corresponding to the second answer and the first unencrypted answer being different from a third unencrypted answer corresponding to the third answer; and an answer type determining module configured to determine the answer type of the answers of the plurality of answers using the mapping function.

[0077] According to some implementations of the present disclosure, the dividing module includes: a creating module configured to create a plurality of pairs respectively corresponding to the plurality of objects using the question and the at least one answer type; and a clustering module configured to cluster the plurality of pairs to divide the plurality of objects into the at least one group.

[0078] According to some implementations of the present disclosure, the dataset determining module includes: a matching degree determining module configured to determine a matching degree between the object and the recommended target of the media item; and a generating module configured to generate the sample dataset based on the matching degree and the related data of the object.

[0079] According to some implementations of the present disclosure, the generating module includes: a first generating module configured to generate, in response to determining that the matching degree satisfies a predetermined condition, positive sample data in the sample dataset for the related data of each object in a group including the object among the at least one group; and a second generating module configured to generate, in response to determining that the matching degree does not satisfy the predetermined condition, negative sample data in the sample dataset for the related data of each object in the group including the object among the at least one group.

[0080] According to some implementations of the present disclosure, the matching degree is determined based on events associated with the object and the recommended target.

[0081] According to some implementations of the present disclosure, the apparatus further includes: an updating module configured to update the object grouping model using the sample dataset, so that the object grouping model describes an association relationship between the grouping of the object and the answer type.

[0082] According to some implementations of the present disclosure, the apparatus further includes: a candidate type determining module configured to determine a candidate answer type associated with a candidate recommended target of a candidate media item, the candidate media item including the question; a candidate group determining module configured to determine, based on the object grouping model, a candidate group of the object associated with the candidate answer type; and a recommending module configured to recommend the candidate media item to the object in the candidate group.

[0083] According to some implementations of the present disclosure, the candidate type determining module includes: a data obtaining module configured to obtain candidate related data of a candidate object providing a candidate answer for the candidate media item; and a matching degree based determining module configured to determine, in response to determining that the candidate related data indicates that a matching degree between the candidate object and the candidate recommended target of the candidate media item satisfies a predetermined condition, the candidate answer type based on an answer type of the candidate answer.

[0084] According to some implementations of the present disclosure, the question is a question posed to the object in association with the recommendation target, the question including a plurality of options associated with an answer, and the answer specifying an option of the plurality of options.

[0085] FIG. 8 illustrates a block diagram of a device 800 that is capable of implementing the various implementations of the present disclosure. It should be understood that the computing device 800 illustrated in FIG. 8 is merely an example and should not be construed as any limitation of the functionality and scope of the implementations described herein. The computing device 800 illustrated in FIG. 8 can be used to implement the methods described above.

[0086] As illustrated in FIG. 8, the computing device 800 is in the form of a general- purpose computing device. Components of the computing device 800 can include, but are not limited to, one or more processors or processing units 810, a memory 820, a storage device 830, one or more communication units 840, one or more input devices 850, and one or more output devices 860. The processing unit 810 can be a real or virtual processor and is capable of executing various processing in accordance with programs stored in the memory 820. In a multi-processing system, multiple processing units execute computer-executable instructions in parallel to improve the processing power of the computing device 800.

[0087] The computing device 800 typically includes a plurality of computer storage media. Such media can be removable, non-removable, or a combination thereof. The memory 820 can be volatile (such as, for example, registers, cache, RAM), non-volatile (such as, for example, ROM, EEPROM, flash memory), or some combination thereof. The storage device 830 can be a removable or non-removable media, and can include machine-readable media, such as flash drives, disks, or any other media capable of storing information and / or data (e.g., training data for training) and accessible by the computing device 800.

[0088] The computing device 800 can further include additional removable / non-removable, volatile / non-volatile storage media. Although not shown in FIG. 8, a disk drive or other computer-readable media drive can be provided for reading from or writing to a removable, non-removable, volatile, or non-volatile media. In these instances, each drive can be connected to the bus (not shown) by one or more data media interfaces. The memory 820 can include a computer program product 825 having one or more program modules configured to carry out the various methods or actions of the implementations of the present disclosure.

[0089] The communication unit 840 enables communications with other computing devices over a communication medium. Additionally, the functionality of the components of the computing device 800 can be implemented in a single computing cluster or a plurality of computer machines that are capable of communicating with each other through a communication connection. Thus, the computing device 800 can operate in a networked environment using logical connections to one or more other servers, network personal computers (PCs), or another network node.

[0090] The input device 850 can be one or more input devices, such as a mouse, a keyboard, a trackball, etc. The output device 860 can be one or more output devices, such as a display, a speaker, a printer, etc. The computing device 800 can also communicate with one or more external devices (not shown), such as a storage device, a display device, etc., through the communication unit 840, as needed, communicate with one or more devices that enable a user to interact with the computing device 800, or communicate with any devices (e.g., a network card, a modem, etc.) that enable the computing device 800 to communicate with one or more other computing devices. Such communication can be carried out via an input / output (I / O) interface (not shown).

[0091] According to an example implementation of the present disclosure, a computer readable storage medium is provided, having stored thereon computer executable instructions, wherein the computer executable instructions are executed by a processor to implement the method described above. According to an example implementation of the present disclosure, a computer program product is also provided, which is tangibly stored on a non-transitory computer readable medium and includes computer executable instructions, wherein the computer executable instructions are executed by a processor to implement the method described above. According to an example implementation of the present disclosure, a computer program product is provided, having stored thereon a computer program, which when executed by a processor implements the method described above.

[0092] Various aspects of the disclosure are now described with reference to the drawings. In general, the drawings described herein relate to a method, an apparatus, a device, and a computer program product implemented according to the present disclosure. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer readable program instructions.

[0093] These computer-readable program instructions can be provided to a processing unit of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus to produce a machine such that, when executed by the processing unit of the computer or other programmable data processing apparatus, they create means for implementing the functions / actions specified in one or more blocks of the flowchart and / or block diagram. These computer-readable program instructions can also be stored in a computer-readable storage medium that causes a computer, programmable data processing apparatus, and / or other device to operate in a particular manner. Thus, the computer-readable medium storing the instructions comprises an article of manufacture that includes instructions for implementing aspects of the functions / actions specified in one or more blocks of the flowchart and / or block diagram.

[0094] Computer-readable program instructions can be loaded onto a computer, other programmable data processing apparatus, or other device to cause a series of operational steps to be performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, thereby causing the instructions that execute on the computer, other programmable data processing apparatus, or other device to perform the functions / actions specified in one or more boxes of a flowchart and / or block diagram.

[0095] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of an instruction, which contains one or more executable instructions for implementing the specified logical function. In some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, may be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.

[0096] Various implementations of this disclosure have been described above. These descriptions are exemplary and not exhaustive, nor are they limited to the disclosed implementations. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described implementations. The terminology used herein is chosen to best explain the principles, practical applications, or improvements to technology in the market, or to enable others skilled in the art to understand the various implementations disclosed herein.

Claims

1. A method for managing objects, comprising: Multiple responses from multiple objects to a media item are obtained, the media item including a question, and the multiple responses are encrypted responses from the multiple objects to the question; Determine at least one response type from the plurality of responses; Based on the question and the at least one response type of the plurality of responses, the plurality of objects are divided into at least one group, wherein the group in the at least one group corresponds to the response type in the at least one response type; as well as Based on the relevant data of the objects in the at least one group, a sample dataset is determined for updating the object grouping model.

2. The method of claim 1, wherein determining the at least one response type comprises: A mapping function is established, which maps the first and second responses among the plurality of responses to the same response type, and maps the first and third responses among the plurality of responses to different response types. The first unencrypted response corresponding to the first response is the same as the second unencrypted response corresponding to the second response, and the first unencrypted response is different from the third unencrypted response corresponding to the third response. as well as The mapping function is used to determine the response type of the responses among the plurality of responses.

3. The method of claim 1, wherein dividing the plurality of objects into the at least one group comprises: Using the question and the at least one response type, establish multiple tuples corresponding to the multiple objects respectively; as well as The plurality of binary tuples are clustered to divide the plurality of objects into at least one group.

4. The method of claim 1, wherein determining the sample dataset comprises: Determine the degree of matching between the object and the recommendation target of the media item; as well as The number of samples is generated based on the matching degree and the relevant data of the object. According to the collection.

5. The method of claim 4, wherein generating the sample dataset comprises at least one of the following: In response to determining that the matching degree meets a predetermined condition, positive sample data in the sample dataset is generated for the relevant data of each object in the group that includes the object in the at least one group; and In response to determining that the matching degree does not meet a predetermined condition, negative sample data in the sample dataset is generated for the relevant data of each object in the group that includes the object in the at least one group.

6. The method of claim 4, wherein the degree of matching is determined based on events associated with the object and the recommendation target.

7. The method of claim 4, further comprising: The object grouping model is updated using the sample dataset so that it describes the association between object grouping and response type.

8. The method of claim 7, further comprising: Determine the candidate response type associated with the candidate recommendation target of the candidate media item, wherein the candidate media item includes the question; Based on the object grouping model, candidate groups of objects associated with the candidate response type are determined; as well as Recommend the candidate media items to the objects in the candidate group.

9. The method of claim 8, wherein determining the candidate response type comprises: Obtain candidate-related data for candidate objects that provide candidate responses to the candidate media items; as well as In response to determining that the candidate-related data indicates that the degree of matching between the candidate object and the candidate recommendation target of the candidate media item meets a predetermined condition, the candidate response type is determined based on the response type of the candidate response.

10. The method of claim 4, wherein the question is a question posed to the object and associated with the recommendation objective, the question including a plurality of options associated with the response, and the response specifying an option among the plurality of options.

11. An apparatus for managing objects, comprising: The acquisition module is configured to acquire multiple responses from multiple objects to a media item, the media item including a question, and the multiple responses being encrypted responses from the multiple objects to the question. A type determination module is configured to determine at least one response type of the plurality of responses; A partitioning module is configured to divide the plurality of objects into at least one group based on the question and at least one response type of the plurality of responses, wherein the group in the at least one group corresponds to the response type in the at least one response type; as well as A dataset determination module is configured to determine a sample dataset for updating the object grouping model based on relevant data of objects in the at least one group.

12. An electronic device, comprising: At least one processing unit; as well as At least one memory, coupled to the at least one processing unit and storing instructions for execution by the at least one processing unit, the instructions causing the electronic device to perform the method according to any one of claims 1 to 10 when executed by the at least one processing unit.

13. A computer-readable storage medium having a computer program stored thereon, the computer program causing the processor to implement the method according to any one of claims 1 to 10 when executed by a processor.

14. A computer program product comprising a computer program, wherein the computer program, when executed by a processor, implements the method according to any one of claims 1 to 10.

Images