Similar object determination method and apparatus, electronic device, and storage medium

Abstract

Embodiments of the present application provide a similar object determination method and device, electronic equipment and storage medium, belonging to the field of artificial intelligence. The method comprises: obtaining candidate characteristic values of multiple object attributes for each candidate object respectively; determining data distribution information of the object attributes according to the candidate characteristic values; sorting the multiple object attributes according to the data distribution information to obtain attribute sorting information; obtaining target characteristic values of the multiple object attributes of a target object, expanding the target characteristic values of the object attributes of the target object in turn according to the attribute sorting information, and obtaining at least one target characteristic range of the object attributes; and matching the target characteristic values of the at least one target characteristic range and the target characteristic values not expanded with the candidate characteristic values of the corresponding object attributes, and determining a similar object of the target object from the multiple candidate objects according to a matching result. The embodiments of the present application can determine key information affecting similar object searching, and improve the explainability and accuracy of the determination method.

Description

Technical Field

[0001] This application relates to the field of artificial intelligence technology, and in particular to a method and apparatus for determining similar objects, an electronic device, and a storage medium. Background Technology

[0002] With the widespread adoption of the internet, information resources have expanded exponentially, leading to information overload. Users often become lost in a sea of ​​information, unable to find the information they need. Therefore, various information recommendation technologies have emerged, which identify similar objects based on the information of a given object and then recommend that object. One such technology involves vectorizing all information about an object to form a composite vector. Vector similarity is then calculated on the composite vectors of each object to identify objects with high similarity. However, because the processed information makes it difficult to determine the key information influencing object similarity, the interpretability is weak, resulting in low accuracy in similar object searches. Therefore, improving the interpretability and accuracy of similar object determination methods has become a pressing technical problem. Summary of the Invention

[0003] The following is an overview of the subject matter described in detail herein. This overview is not intended to limit the scope of the claims.

[0004] The main objective of this application is to propose a method, apparatus, electronic device, and storage medium for determining similar objects, aiming to identify key information affecting the search for similar objects and improve the interpretability and accuracy of the determination method.

[0005] To achieve the above objectives, a first aspect of this application proposes a method for determining similar objects, the method comprising:

[0006] For each candidate object, obtain candidate feature values ​​for multiple object attributes;

[0007] The data distribution information of the object attribute is determined based on the candidate feature values, and the data distribution information is used to indicate the degree of fluctuation of the candidate feature values ​​of multiple candidate objects under the object attribute;

[0008] Based on the data distribution information, the attributes of multiple objects are sorted to obtain attribute sorting information;

[0009] Obtain target feature values ​​of multiple object attributes of a target object, and expand the target feature values ​​of the object attributes of the target object sequentially according to the attribute sorting information to obtain at least one target feature range of the object attributes;

[0010] Based on at least one of the target feature ranges and the unexpanded target feature values, a matching is performed with the candidate feature values ​​of the corresponding object attributes, and similar objects of the target object are determined from a plurality of candidate objects based on the matching results.

[0011] The method for determining similar objects according to the embodiments provided in this application has at least the following beneficial effects:

[0012] After obtaining candidate feature values ​​for multiple object attributes for each candidate object, data distribution information indicating the fluctuation level of the candidate feature values ​​is determined based on these values. Then, the multiple object attributes are sorted according to the data distribution information to obtain attribute ranking information. Next, the target feature values ​​of the multiple object attributes of the target object are sequentially expanded based on the attribute ranking information to obtain at least one target feature range for the object attributes. Finally, the at least one target feature range and the unexpanded target feature values ​​are matched with the corresponding candidate feature values ​​of the object attributes. Based on the matching results, similar objects of the target object are determined from multiple candidate objects. Therefore, by combining at least one expanded target feature range and the unexpanded target feature values ​​with the corresponding candidate feature values ​​of the object attributes for matching, similar objects of the target object can be determined from multiple candidate objects. This process identifies key information affecting the search for similar objects, improving the interpretability and accuracy of the determination method.

[0013] In some embodiments, before determining the data distribution information of the object attribute based on the candidate feature values, the method further includes:

[0014] Obtain target feature values ​​of multiple object attributes of the target object, match the target feature values ​​with the corresponding candidate feature values ​​of the object attributes, and determine the candidate object containing the candidate feature values ​​that match all the target feature values ​​as the similar object of the target object.

[0015] In some embodiments, determining the data distribution information of the object attributes based on the candidate feature values ​​includes:

[0016] When the object attribute is a continuous feature attribute, the standard deviation of the candidate feature values ​​of multiple candidate objects under the object attribute is calculated to obtain the data distribution information of the object attribute.

[0017] In some embodiments, determining the data distribution information of the object attributes based on the candidate feature values ​​includes:

[0018] When the object attribute is a discrete feature attribute, information entropy calculation is performed on the candidate feature values ​​of multiple candidate objects under the object attribute to obtain the data distribution information of the object attribute.

[0019] In some embodiments, when the multiple object attributes include continuous feature attributes and discrete feature attributes, the step of sorting the multiple object attributes according to the data distribution information to obtain attribute sorting information includes:

[0020] The object attributes belonging to continuous feature attributes are sorted according to the data distribution information to obtain continuous sorting information;

[0021] The object attributes belonging to discrete feature attributes are sorted according to the data distribution information to obtain discrete sorting information;

[0022] The discrete sorting information and the continuous sorting information are merged to obtain the attribute sorting information.

[0023] In some embodiments, the step of sequentially expanding the target feature values ​​of the object attributes of the target object according to the attribute sorting information to obtain at least one target feature range of the object attributes includes:

[0024] Based on the attribute sorting information, reference attributes are sequentially determined from the object attributes of the target object;

[0025] When the reference attribute is a continuous feature attribute, the attribute difference is obtained based on the maximum and minimum values ​​of the candidate feature values ​​of all candidate objects under the reference attribute;

[0026] The parameter deviation value is determined based on the attribute difference and the preset reference coefficient;

[0027] Based on the parameter deviation value, the target feature value of the target object under the reference attribute is expanded to obtain the target feature range of the reference attribute.

[0028] In some embodiments, the step of sequentially expanding the target feature values ​​of the object attributes of the target object according to the attribute sorting information to obtain at least one target feature range of the object attributes includes:

[0029] Based on the attribute sorting information, reference attributes are sequentially determined from the object attributes of the target object;

[0030] When the reference attribute belongs to a discrete feature attribute, the candidate feature values ​​of all candidate objects under the reference attribute are classified to obtain each parameter group;

[0031] Based on the number of candidate feature values ​​within each parameter group, sort each parameter group to obtain parameter sorting information;

[0032] Based on the parameter sorting information, the reference feature is determined from the candidate feature values ​​of all the candidate objects under the reference attribute;

[0033] Based on the reference features, the target feature values ​​of the target object under the reference attributes are expanded to obtain the target feature range of the reference attributes.

[0034] To achieve the above objectives, a second aspect of this application provides a similar object determination apparatus, the apparatus comprising:

[0035] The parameter acquisition module is used to obtain candidate feature values ​​of multiple object attributes for each candidate object.

[0036] A feature distribution module is used to determine the data distribution information of the object attribute based on the candidate feature values, wherein the data distribution information is used to indicate the degree of fluctuation of the candidate feature values ​​of multiple candidate objects under the object attribute;

[0037] The attribute sorting module is used to sort multiple object attributes according to the data distribution information to obtain attribute sorting information;

[0038] The feature expansion module is used to obtain target feature values ​​of multiple object attributes of the target object, and expand the target feature values ​​of the object attributes of the target object in sequence according to the attribute sorting information to obtain at least one target feature range of the object attributes;

[0039] The feature matching module is used to match the candidate feature values ​​of the corresponding object attributes with at least one target feature range and the target feature value without expansion, and to determine similar objects of the target object from multiple candidate objects based on the matching results.

[0040] The similar object determination apparatus provided in the embodiments of this application has at least the following beneficial effects: After the similar object determination apparatus uses a parameter acquisition module to acquire candidate feature values ​​of multiple object attributes for each candidate object, it uses a feature distribution module to determine data distribution information of the object attributes that indicates the fluctuation degree of the candidate feature values ​​based on the candidate feature values; then, it uses an attribute sorting module to sort the multiple object attributes according to the data distribution information to obtain attribute sorting information; next, it uses a feature expansion module to sequentially expand the target feature values ​​of the multiple object attributes of the target object according to the attribute sorting information to obtain at least one target feature range of the object attributes; finally, it uses a feature matching module to match the at least one target feature range and the unexpanded target feature values ​​with the corresponding candidate feature values ​​of the object attributes, and determines similar objects of the target object from multiple candidate objects based on the matching results. The similar object determination apparatus provided in the embodiments of this application can determine similar objects of the target object from multiple candidate objects by combining the expanded target feature range and the unexpanded target feature values ​​with the corresponding candidate feature values ​​of the object attributes, thereby realizing the determination of key information affecting the search for similar objects and improving the interpretability and accuracy of the determination method.

[0041] To achieve the above objectives, a third aspect of the present application provides an electronic device, the electronic device including a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the method described in the first aspect.

[0042] To achieve the above objectives, a fourth aspect of the present application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the method described in the first aspect. Attached Figure Description

[0043] Figure 1 This is an optional flowchart of the similar object determination method provided in the embodiments of this application;

[0044] Figure 2 This is another optional flowchart of the similar object determination method provided in the embodiments of this application;

[0045] Figure 3 yes Figure 1 A flowchart of a specific method for step S102;

[0046] Figure 4 yes Figure 1 Another specific method flowchart for step S102;

[0047] Figure 5 yes Figure 1A detailed flowchart of step S103;

[0048] Figure 6 yes Figure 1 A detailed flowchart of step S104;

[0049] Figure 7 yes Figure 1 Another specific flowchart for step S104;

[0050] Figure 8 This is a schematic diagram of the structure of the similar object determination device provided in the embodiments of this application;

[0051] Figure 9 This is a schematic diagram of the hardware structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0052] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0053] It should be noted that although functional modules are divided in the device schematic diagram and a logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in a different order than the module division in the device or the order in the flowchart. The terms "first," "second," etc., in the specification, claims, and the aforementioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence.

[0054] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0055] First, let's analyze some of the terms used in this application:

[0056] Artificial Intelligence (AI) is a new branch of computer science that studies, develops, and applies theories, methods, technologies, and systems to simulate, extend, and expand human intelligence. It aims to understand the essence of intelligence and produce intelligent machines that can react in a way similar to human intelligence. Research in this field includes robotics, speech recognition, image recognition, natural language processing, and expert systems. AI can simulate the information processes of human consciousness and thought. Furthermore, AI utilizes digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceiving the environment, acquiring knowledge, and using that knowledge to achieve optimal results.

[0057] Dataset features can be categorized into continuous features and discrete features based on their values. Continuous features, also known as quantitative features, are typically measured using interval and scales. They have numerous, even infinite, numerical values, and their values ​​are taken from a continuous interval, representing a measurable property of an object, such as a person's height and age, a product's price, air temperature, or an object's length. Continuous features can be sorted and arithmetic operations can be performed on their values.

[0058] In addition, discrete features, also known as qualitative features, are generally defined using nominal or ordered scales. They refer to the characteristics of objects expressed in textual data, such as a person's gender or education level, or the purpose of a product (food, clothing). The value range of such features is limited to a small number of values. The values ​​of discrete features can sometimes be ordered, but they cannot be used for arithmetic operations.

[0059] With the widespread adoption of the internet, information resources have expanded exponentially, leading to information overload. Users often become lost in a sea of ​​information, unable to easily find the information they need. Therefore, various information recommendation technologies have emerged. These technologies identify similar objects based on the target user's information and then recommend relevant information, enabling users to quickly access information that is identical or similar to the target user's information, thus improving the user experience.

[0060] Similarity object search methods in related technologies vectorize all information about objects to form a composite vector, and then calculate vector similarity for each composite vector. For example, they might measure vector similarity using cosine distance or Euclidean distance to identify objects with high similarity. Alternatively, they might train a model to transform the original input data into an n-dimensional dense vector before calculating similarity. However, these existing methods all involve comprehensive information processing. Because this comprehensive processing makes it difficult to determine the key information affecting object similarity, the interpretability is weak, leading to low accuracy in similarity object search. Therefore, improving the interpretability and accuracy of similarity object determination methods has become an urgent technical problem to be solved.

[0061] Based on this, embodiments of this application provide a method and apparatus, electronic device and storage medium for determining similar objects, aiming to determine key information affecting the search for similar objects and improve the interpretability and accuracy of the determination method.

[0062] The similar object determination method, apparatus, electronic device, and storage medium provided in this application are specifically described through the following embodiments. First, the similar object determination method in this application is described.

[0063] The similar object determination method provided in this application relates to the field of artificial intelligence technology. The method includes: for each candidate object, obtaining candidate feature values ​​of multiple object attributes; determining data distribution information of the object attributes based on the candidate feature values, wherein the data distribution information indicates the degree of fluctuation of the candidate feature values ​​of multiple candidate objects under the object attributes; sorting the multiple object attributes according to the data distribution information to obtain attribute sorting information; obtaining target feature values ​​of multiple object attributes of the target object, and sequentially expanding the target feature values ​​of the target object's object attributes according to the attribute sorting information to obtain at least one target feature range of the object attributes; matching the at least one target feature range and the unexpanded target feature values ​​with the corresponding candidate feature values ​​of the object attributes, and determining similar objects of the target object from the multiple candidate objects based on the matching results. According to the embodiments provided in this application, after obtaining candidate feature values ​​of multiple object attributes for each candidate object, data distribution information indicating the fluctuation level of the candidate feature values ​​is determined based on the candidate feature values. Then, the multiple object attributes are sorted according to the data distribution information to obtain attribute sorting information. Next, the target feature values ​​of the multiple object attributes of the target object are sequentially expanded according to the attribute sorting information to obtain at least one target feature range of the object attributes. Finally, the at least one target feature range and the unexpanded target feature values ​​are matched with the corresponding candidate feature values ​​of the object attributes, and similar objects of the target object are determined from multiple candidate objects based on the matching results. By combining the expanded target feature range and the unexpanded target feature values ​​with the corresponding candidate feature values ​​of the object attributes for matching, similar objects of the target object can be determined from multiple candidate objects. This can identify key information affecting the search for similar objects, improving the interpretability and accuracy of the determination method.

[0064] The embodiments of this application can acquire and process relevant data based on artificial intelligence technology. AI refers to the theories, methods, technologies, and application systems that utilize digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results.

[0065] Foundational technologies for artificial intelligence generally include sensors, dedicated AI chips, cloud computing, distributed storage, big data processing, operating / interactive systems, and mechatronics. AI software technologies mainly encompass computer vision, robotics, biometrics, speech processing, natural language processing, and machine learning / deep learning.

[0066] The similar object determination method provided in this application can be applied to a terminal, a server, or software running on either a terminal or a server. In some embodiments, the terminal can be a smartphone, tablet, laptop, desktop computer, etc.; the server can be configured as an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDN), and big data and artificial intelligence platforms; the software can be an application implementing the similar object determination method, but is not limited to the above forms.

[0067] This application can be used in a wide variety of general-purpose or special-purpose computer system environments or configurations. Examples include: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, and distributed computing environments including any of the above systems or devices. This application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform specific tasks or implement specific abstract data types. This application can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0068] The embodiments of this application will now be described in conjunction with the accompanying drawings.

[0069] Firstly, Figure 1 This is an optional flowchart of the similar object determination method provided in the embodiments of this application. Figure 1 The method may include, but is not limited to, steps S101 to S105.

[0070] Step S101: For each candidate object, obtain the candidate feature values ​​of multiple object attributes respectively;

[0071] Step S102: Determine the data distribution information of the object attribute based on the candidate feature values. The data distribution information is used to indicate the degree of fluctuation of the candidate feature values ​​of multiple candidate objects under the object attribute.

[0072] Step S103: Sort multiple object attributes according to data distribution information to obtain attribute sorting information;

[0073] Step S104: Obtain the target feature values ​​of multiple object attributes of the target object; according to the attribute sorting information, sequentially expand the target feature values ​​of the object attributes of the target object to obtain at least one target feature range of the object attributes.

[0074] Step S105: Match the target feature range and the unexpanded target feature value with the candidate feature values ​​of the corresponding object attributes, and determine similar objects of the target object from multiple candidate objects based on the matching results.

[0075] Steps S101 to S105 of this application embodiment involve obtaining candidate feature values ​​of multiple object attributes for each candidate object, performing calculations to obtain data distribution information indicating the fluctuation level of the candidate feature values; then, sorting the multiple object attributes based on the data distribution information; furthermore, obtaining target feature values ​​of multiple object attributes of the target object, and sequentially expanding the target feature values ​​of at least one object attribute according to the sorting of the object attributes to obtain at least one target feature range. By combining at least one expanded target feature range with the unexpanded target feature value, and performing matching processing with the corresponding candidate feature values ​​of the object attributes, similar objects to the target object can be determined from multiple candidate objects. Compared to the information integration processing method used in related technologies, which vectorizes all information of an object to form a comprehensive vector, this application embodiment focuses on key information affecting object similarity. Through this application embodiment, similar objects to the target object can be determined, as well as key information affecting the search for similar objects can be identified. Through this key information, it is possible to know which attributes are similar between the target object and the recommended similar objects, improving the interpretability, accuracy, and reliability of the determination method.

[0076] It should be noted that the embodiments of this application can be used to find historical users similar to new users. Specifically, when the candidate object is a historical user and the target object is a new user, candidate feature values ​​of multiple user attributes for each historical user are obtained, where user attributes can be user height, age, gender, blood type, etc.; then, the candidate feature values ​​of multiple user attributes are calculated and processed to obtain data distribution information indicating the degree of fluctuation of the candidate feature values; next, the multiple user attributes are sorted with reference to the data distribution information; then, target feature values ​​of multiple user attributes of the target user are obtained, and the target feature values ​​of at least one user attribute are expanded sequentially according to the sorting of user attributes to obtain at least one target feature range; finally, by combining at least one expanded target feature range and the unexpanded target feature value, matching processing is performed with the corresponding candidate feature values ​​of user attributes to determine similar users to the new user from multiple historical users.

[0077] It is understood that the application scenarios described in the embodiments of the present invention are for the purpose of more clearly illustrating the technical solutions of the embodiments of the present invention, and do not constitute a limitation on the technical solutions provided by the embodiments of the present invention. Those skilled in the art will recognize that with the emergence of new application scenarios, the technical solutions provided by the embodiments of the present invention are equally applicable to similar technical problems. Furthermore, with the emergence of new application scenarios, different candidate objects can be set according to different application scenarios, and similarly, different object attributes can be obtained according to different candidate objects. Therefore, this application does not impose specific limitations on the application scenarios, candidate objects, and object attributes in the method for determining similar objects.

[0078] It should be noted that in all specific embodiments of this application, when processing data related to user identity or characteristics, such as user information, user behavior data, user historical data, and user location information, user permission or consent is obtained first. Furthermore, the collection, use, and processing of this data comply with relevant laws, regulations, and standards of the relevant countries and regions. In addition, when embodiments of this application require access to sensitive personal information of users, separate permission or consent from the user is obtained through pop-ups or redirects to confirmation pages. Only after obtaining the user's separate permission or consent is the necessary user-related data for the proper functioning of the embodiments of this application obtained.

[0079] Please see Figure 2 , Figure 2 This is another optional flowchart of the similar object determination method provided in the embodiments of this application. In some embodiments, the similar object determination method may include, but is not limited to, steps S101 and S201:

[0080] Step S101: For each candidate object, obtain the candidate feature values ​​of multiple object attributes respectively;

[0081] Step S201: Obtain target feature values ​​of multiple object attributes of the target object; match the target feature values ​​with the corresponding candidate feature values ​​of the object attributes; and determine the candidate objects containing candidate feature values ​​that match all target feature values ​​as similar objects of the target object.

[0082] Understandably, for both candidate and target objects, multiple candidate and target feature values ​​for object attributes are obtained. These candidate and target feature values ​​are then matched. If one or more candidate object feature values ​​match all target feature values, then one or more candidate objects are identified as similar objects to the target object. If similar objects to the target object can be found through steps S101 and S201, the similarity object determination method ends, and further processing such as determining data distribution information of object attributes based on candidate feature values ​​is discontinued. This helps reduce the time required to determine similar objects and improves the efficiency of the determination method. Conversely, if no similar objects to the target object can be found through steps S101 and S201, steps S102 to S105 as described above continue, ensuring the orderly execution of the similarity object determination method.

[0083] Please see Figure 3 , Figure 3 yes Figure 1 A specific method flowchart for step S102 is provided. In some embodiments, step S102 may include, but is not limited to, step S301.

[0084] Step S301: When the object attribute is a continuous feature attribute, the standard deviation of the candidate feature values ​​of multiple candidate objects under the object attribute is calculated to obtain the data distribution information of the object attribute.

[0085] It should be noted that the features of the dataset can be divided into continuous features and discrete features. Based on this, object attributes are also divided into continuous feature attributes and discrete feature attributes. When an object attribute is a continuous feature attribute, the standard deviation of the candidate feature values ​​of multiple candidate objects under that object attribute is calculated. The standard deviation reflects the degree of fluctuation of the candidate feature values. That is to say, when an object attribute is a continuous feature attribute, the data distribution information is the standard deviation of the candidate feature values ​​under that object attribute. The larger the standard deviation, the greater the range of feature fluctuation of that object attribute; the smaller the standard deviation, the smaller the range of feature fluctuation of that object attribute. This step can obtain the data distribution information of continuous feature attributes, providing reliable reference information for subsequent attribute ranking processing.

[0086] For example, when it is necessary to obtain the data distribution information of the age attribute of 10 candidate objects, the age data of the 10 candidate objects is first obtained, and then the standard deviation is calculated based on the standard deviation of the 10 age data to obtain the age standard deviation. The age standard deviation is the data distribution information of the age attribute, which is used to indicate the degree of fluctuation of the age of the 10 candidate objects under the age attribute. This example is only used to further explain this step and does not constitute a limitation on the technical solution provided in the embodiments of this application.

[0087] Please see Figure 4, Figure 4 yes Figure 1 Another specific method flowchart for step S102. In some embodiments, step S102 may include, but is not limited to, step S401.

[0088] Step S401: When the object attribute belongs to discrete feature attribute, perform information entropy calculation on the candidate feature values ​​of multiple candidate objects under the object attribute to obtain the data distribution information of the object attribute.

[0089] It should be noted that when an object attribute is a discrete feature attribute, the information entropy of the candidate feature values ​​of multiple candidate objects under that object attribute is calculated. This information entropy reflects the degree of fluctuation of the candidate feature values. In other words, when an object attribute is a discrete feature attribute, the data distribution information is the information entropy of the candidate feature values ​​under that object attribute. The larger the information entropy, the greater the range of feature fluctuation for that object attribute; conversely, the smaller the information entropy, the smaller the range of feature fluctuation for that object attribute. This step allows us to obtain the data distribution information of the discrete feature attribute, providing reliable reference information for subsequent attribute sorting.

[0090] It should be noted that the formula for information entropy is: Where p k This represents the probability of each feature value occurring, and N represents the number of feature values.

[0091] For example, when it is necessary to obtain the data distribution information of the gender attributes of 10 candidate objects, it is known that 5 candidate objects are male and 5 candidate objects are female. The probability of being male is 0.5, and the probability of being female is 0.5. According to the information entropy formula mentioned above, the information entropy of the gender attribute can be calculated to be 1. The information entropy with a value of 1 is the data distribution information of the gender attribute, used to indicate the characteristic fluctuation range of the gender attribute. This example is only used to further explain this step and does not constitute a limitation on the technical solution provided in the embodiments of this application.

[0092] Please see Figure 5 , Figure 5 yes Figure 1 A specific flowchart of step S103 is provided. In some embodiments, step S103 may include, but is not limited to, steps S501 to S503.

[0093] Step S501: Sort the object attributes that belong to continuous feature attributes according to the data distribution information to obtain continuous sorting information;

[0094] Step S502: Sort the object attributes that belong to discrete feature attributes according to the data distribution information to obtain discrete sorting information;

[0095] Step S503: Merge the discrete sorting information and the continuous sorting information to obtain the attribute sorting information.

[0096] Steps S501 to S503 as shown in the embodiments of this application sort continuous feature attributes and discrete feature attributes respectively by calculating the data distribution information to obtain continuous sorting information and discrete sorting information. Then, the continuous sorting information and discrete sorting information are merged to obtain attribute sorting information. The sorting is based on data-driven reference information, which improves the reliability and interpretability of the sorting results.

[0097] Specifically, object attributes are divided into continuous and discrete feature attributes. The standard deviation of the continuous feature attributes and the information entropy of the discrete feature attributes are calculated separately. Continuous feature attributes are sorted according to the rule that a larger standard deviation results in a higher ranking and a smaller standard deviation results in a lower ranking, thus obtaining continuous ranking information. Discrete feature attributes are sorted according to the rule that a larger information entropy results in a higher ranking and a smaller information entropy results in a lower ranking, thus obtaining discrete ranking information. Finally, the continuous and discrete ranking information are merged according to a ranking merging rule to obtain attribute ranking information. It should be noted that in the process of matching discrete features, it is easy to obtain completely identical matching results (e.g., male / female), while in the process of matching continuous features, it is less likely to obtain completely identical matching results (e.g., height). Therefore, the ranking merging rule generally adopted is that the ranking of continuous feature attributes takes precedence over the ranking of discrete feature attributes. However, it is understood that those skilled in the art can also adopt a ranking merging rule that prioritizes discrete feature attributes over continuous feature attributes based on the actual application situation, and this application does not impose specific restrictions on this.

[0098] Please see Figure 6 , Figure 6 yes Figure 1 A specific flowchart of step S104 is provided. In some embodiments, step S104 may include, but is not limited to, steps S601 to S604.

[0099] Step S601: Based on the attribute sorting information, determine the reference attributes sequentially from the object attributes of the target object;

[0100] Step S602: When the reference attribute is a continuous feature attribute, the attribute difference is obtained based on the maximum and minimum values ​​of the candidate feature values ​​of all candidate objects under the reference attribute.

[0101] Step S603: Determine the parameter deviation value based on the attribute difference and the preset reference coefficient;

[0102] Step S604: Based on the parameter deviation value, expand the target feature value of the target object under the reference attribute to obtain the target feature range of the reference attribute.

[0103] Steps S601 to S604, as illustrated in this embodiment, involve sequentially determining reference attributes from the object attributes of the target object using reference attribute sorting information. When the reference attribute is a continuous feature attribute, the maximum and minimum values ​​of the candidate feature values ​​of all candidate objects under the reference attribute are obtained, and the difference is calculated to obtain the attribute difference. A preset reference coefficient is multiplied by the attribute difference to obtain a parameter deviation value. The target feature values ​​of the target object under the reference attribute are then expanded using the parameter deviation value to obtain the target feature range of the reference attribute. By systematically selecting reference attributes using reliable reference information and then rationally expanding the target feature values ​​using the parameter deviation value, similar objects are systematically searched, ensuring the reliability and interpretability of the similar object determination method.

[0104] It is understood that those skilled in the art can determine the preset reference coefficients according to actual application needs. Therefore, this application does not impose specific restrictions on the values ​​of the preset reference coefficients.

[0105] For example, considering 5 candidate objects, if the attribute sorting information is in the order of age, height, gender, and blood type, then based on this sorting information, the reference attribute is first determined from the target object's attributes as age. If the target feature value of the target object to be searched is 14 years old, and the ages of the 5 candidate objects are 12, 13, 15, 16, and 17 years old respectively, directly matching the target feature value 14 among the candidate objects will not find similar objects. Therefore, the target feature value 14 needs to be expanded. In the first round of similar object determination, firstly, the maximum age of all candidate objects under the age attribute (17) and the minimum age (12) are obtained; then, the maximum and minimum ages are subtracted to obtain the age attribute difference of 5; subsequently, when the preset reference coefficient is 0.1, the age attribute difference of 5 is multiplied by the preset reference coefficient 0.1 to obtain the parameter deviation value of 0.5; based on the parameter deviation value of 0.5, the target feature value 14 of the target object under the age attribute is expanded to the target feature range of 13.5 to 14.5. Then, based on the target feature range obtained from the first expansion, matching is performed on the 5 candidate objects. At this point, no similar object can still be found from the 5 candidate objects. Therefore, based on the attribute ranking information, the target feature value of the second-ranked height attribute is selected for expansion. The expansion of the target feature value under the height attribute will not be elaborated further here. If a similar object can be matched after expanding the target feature value of the height attribute, the similar object determination process ends. If a similar object still cannot be matched, the next target feature value of the gender attribute is selected for expansion. If a similar object still cannot be matched after the first expansion of the target feature values ​​under each object attribute, a second round of similar object determination is performed. The target feature values ​​under each object attribute are expanded and matched a second time according to the ranking of the object attributes, and so on.

[0106] Understandably, in the second round of similar object identification, based on the parameter deviation value of 0.5, the target feature range obtained from the first expansion (13.5 to 14.5) needs to be expanded a second time, resulting in a target feature range of 13 to 15. Matching is then performed on the five candidate objects based on this second expanded target feature range. If two similar objects are matched, the process of finding similar objects in the age attribute ends. This process continues in the same manner; other cases will not be elaborated upon.

[0107] Please see Figure 7 , Figure 7 yes Figure 1 Another specific flowchart of step S104. In some embodiments, step S104 may include, but is not limited to, steps S701 to S705.

[0108] Step S701: Based on the attribute sorting information, determine the reference attributes sequentially from the object attributes of the target object;

[0109] Step S702: When the reference attribute belongs to a discrete feature attribute, classify the candidate feature values ​​of all candidate objects under the reference attribute to obtain the parameter groupings.

[0110] Step S703: Sort each parameter group according to the number of candidate feature values ​​within the parameter group to obtain parameter sorting information;

[0111] Step S704: Based on the parameter sorting information, determine the reference feature from the candidate feature values ​​of all candidate objects under the reference attribute;

[0112] Step S705: Based on the reference features, expand the target feature values ​​of the target object under the reference attributes to obtain the target feature range of the reference attributes.

[0113] Steps S701 to S705, as illustrated in this embodiment, determine the reference attribute using reference attribute sorting information. When the reference attribute is a discrete feature attribute, the candidate feature values ​​of all candidate objects under the reference attribute are categorized into parameter groups. These parameter groups are then sorted from highest to lowest order based on the number of candidate feature values ​​within each group. Based on this sorting, reference feature values ​​are determined from the candidate feature values ​​of all candidate objects under the reference attribute. The target feature values ​​are then expanded to obtain the target feature range of the reference attribute. By systematically selecting reference attributes using reliable reference information and then selecting reference feature values ​​that appear frequently to reasonably expand the target feature values, similar objects can be found systematically, ensuring the reliability and interpretability of the similar object determination method.

[0114] Secondly, please refer to Figure 8 This application embodiment also provides a similar object determination device 800, which can implement the above-described similar object determination method. The device includes:

[0115] The parameter acquisition module 810 is used to acquire candidate feature values ​​of multiple object attributes for each candidate object.

[0116] The feature distribution module 820 is used to determine the data distribution information of object attributes based on candidate feature values. The data distribution information is used to indicate the degree of fluctuation of candidate feature values ​​of multiple candidate objects under the object attributes.

[0117] The attribute sorting module 830 is used to sort the attributes of multiple objects according to the data distribution information to obtain attribute sorting information;

[0118] The feature extension module 840 is used to obtain the target feature values ​​of multiple object attributes of the target object, and expand the target feature values ​​of the object attributes of the target object in turn according to the attribute sorting information to obtain at least one target feature range of the object attributes.

[0119] The feature matching module 850 is used to match the target feature range and the unextended target feature value with the candidate feature value of the corresponding object attribute, and to determine the similar object of the target object from multiple candidate objects based on the matching result.

[0120] Therefore, after the similar object determination device 800 uses the parameter acquisition module 810 to acquire candidate feature values ​​of multiple object attributes for each candidate object, the feature distribution module 820 determines the data distribution information of the object attributes, which indicates the fluctuation degree of the candidate feature values, based on the candidate feature values. Then, the attribute sorting module 830 sorts the multiple object attributes according to the data distribution information to obtain attribute sorting information. Next, the feature expansion module 840 sequentially expands the target feature values ​​of the multiple object attributes of the target object according to the attribute sorting information, obtaining at least one target feature range of the object attributes. Finally, the feature matching module 850 matches the at least one target feature range and the unexpanded target feature values ​​with the corresponding candidate feature values ​​of the object attributes, and determines similar objects of the target object from multiple candidate objects based on the matching results. The similar object determination device 800 provided in this application embodiment can determine similar objects of the target object from multiple candidate objects by combining at least one expanded target feature range and the unexpanded target feature values ​​with the corresponding candidate feature values ​​of the object attributes, thereby determining key information affecting the search for similar objects and improving the interpretability and accuracy of the determination method.

[0121] The specific implementation of the similar object determination device 800 is basically the same as the specific embodiment of the similar object determination method described above, and will not be repeated here.

[0122] In addition, another embodiment of this application provides a similar object determination device 800, which can implement the above-described similar object determination method, wherein:

[0123] The feature matching module 850 is also used to obtain target feature values ​​of multiple object attributes of the target object before determining the data distribution information of object attributes based on candidate feature values, and to match the target feature values ​​with the corresponding candidate feature values ​​of the object attributes, and to determine the candidate objects containing candidate feature values ​​that match all target feature values ​​as similar objects of the target object.

[0124] In addition, the feature distribution module 820 is also used to calculate the standard deviation of the candidate feature values ​​of multiple candidate objects under the object attribute when the object attribute belongs to continuous feature attributes, so as to obtain the data distribution information of the object attribute.

[0125] In addition, the feature distribution module 820 is also used to perform information entropy calculation on the candidate feature values ​​of multiple candidate objects under the object attribute when the object attribute belongs to discrete feature attribute, so as to obtain the data distribution information of the object attribute.

[0126] In addition, the attribute sorting module 830 is also used to sort the object attributes belonging to continuous feature attributes according to the data distribution information when multiple object attributes include continuous feature attributes and discrete feature attributes, to obtain continuous sorting information; to sort the object attributes belonging to discrete feature attributes according to the data distribution information, to obtain discrete sorting information; and to merge the discrete sorting information and the continuous sorting information to obtain attribute sorting information.

[0127] In addition, the feature expansion module 840 is also used to determine reference attributes from the object attributes of the target object in sequence according to the attribute sorting information. When the reference attribute is a continuous feature attribute, the attribute difference is obtained according to the maximum and minimum values ​​of the candidate feature values ​​of all candidate objects under the reference attribute. The parameter deviation value is determined according to the attribute difference and the preset reference coefficient. The target feature value of the target object under the reference attribute is expanded according to the parameter deviation value to obtain the target feature range of the reference attribute.

[0128] In addition, the feature expansion module 840 is also used to determine reference attributes from the object attributes of the target object sequentially according to the attribute sorting information; when the reference attribute belongs to discrete feature attributes, classify the candidate feature values ​​of all candidate objects under the reference attribute to obtain each parameter group; sort each parameter group according to the number of candidate feature values ​​in the parameter group to obtain parameter sorting information; determine the reference feature from the candidate feature values ​​of all candidate objects under the reference attribute based on the parameter sorting information; and expand the target feature value of the target object under the reference attribute according to the reference feature to obtain the target feature range of the reference attribute.

[0129] The specific implementation of the similar object determination device 800 is basically the same as the specific embodiment of the similar object determination method described above, and will not be repeated here.

[0130] Thirdly, embodiments of this application also provide an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the aforementioned method for determining similar objects. This electronic device can be any smart terminal, including tablet computers, in-vehicle computers, etc.

[0131] Please see Figure 9, Figure 9 The hardware structure of an electronic device according to another embodiment is illustrated. The electronic device includes:

[0132] The processor 901 can be implemented using a general-purpose CPU (Central Processing Unit), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this application.

[0133] The memory 902 can be implemented as a read-only memory (ROM), static storage device, dynamic storage device, or random access memory (RAM). The memory 902 can store the operating system and other application programs. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 902 and is called by the processor 901 to execute the similar object determination method of the embodiments of this application, for example, executing the method described above. Figure 1 Method steps S101 to S105, Figure 2 Method steps S101 and S201, Figure 3 Method steps S301 Figure 4 Method steps S401 Figure 5 Method steps S501 to S503, Figure 6 Method steps S601 to S604 and Figure 7 Method steps S701 to S705;

[0134] The 903 input / output interface is used to implement information input and output.

[0135] The communication interface 904 is used to enable communication and interaction between this device and other devices. Communication can be achieved through wired means (such as USB, Ethernet cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.).

[0136] Bus 905 transmits information between various components of the device (e.g., processor 901, memory 902, input / output interface 903, and communication interface 904);

[0137] The processor 901, memory 902, input / output interface 903, and communication interface 904 are connected to each other within the device via bus 905.

[0138] Fourthly, embodiments of this application also provide a computer-readable storage medium storing a computer program that, when executed by a processor, implements the aforementioned method for determining similar objects.

[0139] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs. Furthermore, memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, memory may optionally include memory remotely located relative to the processor, and these remote memories can be connected to the processor via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0140] The non-transient software program and instructions required to implement the similar object determination method in the above embodiments are stored in memory. When executed by one or more processors, the similar object determination method in the above embodiments is executed, for example, the method described above is executed. Figure 1 Method steps S101 to S105, Figure 2 Method steps S101 and S201, Figure 3 Method steps S301 Figure 4 Method steps S401 Figure 5 Method steps S501 to S503, Figure 6 Method steps S601 to S604 and Figure 7 The method steps S701 to S705.

[0141] The similar object determination method, apparatus, electronic device, and storage medium provided in this application, after obtaining candidate feature values ​​of multiple object attributes for each candidate object, determine data distribution information of the object attributes to indicate the fluctuation degree of the candidate feature values ​​based on the candidate feature values; then, sort the multiple object attributes according to the data distribution information to obtain attribute ranking information; next, expand the target feature values ​​of the multiple object attributes of the target object sequentially according to the attribute ranking information to obtain at least one target feature range of the object attributes; finally, match the at least one target feature range and the unexpanded target feature values ​​with the corresponding candidate feature values ​​of the object attributes, and determine similar objects of the target object from multiple candidate objects based on the matching results. By combining at least one expanded target feature range and the unexpanded target feature values ​​with the corresponding candidate feature values ​​of the object attributes for matching processing, similar objects of the target object can be determined from multiple candidate objects, which can identify key information affecting the search for similar objects and improve the interpretability and accuracy of the determination method.

[0142] The embodiments described in this application are for the purpose of more clearly illustrating the technical solutions of the embodiments of this application, and do not constitute a limitation on the technical solutions provided by the embodiments of this application. As those skilled in the art will know, with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by the embodiments of this application are also applicable to similar technical problems.

[0143] Those skilled in the art will understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of this application, and may include more or fewer steps than shown, or combine certain steps, or different steps.

[0144] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0145] Those skilled in the art will understand that all or some of the steps in the methods disclosed above, as well as the functional modules / units in the systems and devices, can be implemented as software, firmware, hardware, or suitable combinations thereof.

[0146] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0147] It should be understood that in this application, "multiple" means two or more. The character " / " generally indicates that the preceding and following objects are in an "or" relationship.

[0148] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of the units described above is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0149] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0150] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0151] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes multiple instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing programs, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0152] The preferred embodiments of the present application have been described above with reference to the accompanying drawings, but this does not limit the scope of the claims of the present application. Any modifications, equivalent substitutions, and improvements made by those skilled in the art without departing from the scope and substance of the embodiments of the present application shall be within the scope of the claims of the present application.

Claims

1. A similar object determination method characterized by, The method comprises: For each candidate object, the candidate characteristic values of multiple object attributes are respectively obtained; Data distribution information of the object attributes is determined according to the candidate characteristic values, and the data distribution information is used to indicate fluctuation degrees of the candidate characteristic values of multiple candidate objects under the object attributes; In the case where the multiple object attributes include continuous characteristic attributes and discrete characteristic attributes, the object attributes belonging to the continuous characteristic attributes are sorted according to standard deviations of the data distribution information to obtain continuous sorting information, and the object attributes belonging to the discrete characteristic attributes are sorted according to information entropy of the data distribution information to obtain discrete sorting information; The discrete sorting information and the continuous sorting information are combined to obtain attribute sorting information; Target characteristic values of the multiple object attributes of a target object are obtained, the target characteristic values of the object attributes of the target object are sequentially expanded according to the attribute sorting information, and at least one target characteristic range of the object attributes is obtained; According to at least one target characteristic range and the target characteristic values that are not expanded, the candidate characteristic values of the corresponding object attributes are matched, and similar objects of the target object are determined from the multiple candidate objects according to a matching result. According to the attribute sorting information, reference attributes are sequentially determined from the object attributes of the target object, and the following steps are performed based on the reference attributes: When the reference attribute belongs to a continuous characteristic attribute, an attribute difference value is obtained according to maximum and minimum values of the candidate characteristic values of all the candidate objects under the reference attribute; a parameter deviation value is determined according to the attribute difference value and a preset reference coefficient; and a target characteristic value of the target object under the reference attribute is expanded to obtain a target characteristic range of the reference attribute; Or, When the reference attribute belongs to a discrete characteristic attribute, the candidate characteristic values of all the candidate objects under the reference attribute are classified to obtain various parameter groups; the parameter groups are sorted according to the number of the candidate characteristic values in the parameter groups to obtain parameter sorting information; a reference characteristic is determined from the candidate characteristic values of all the candidate objects under the reference attribute based on the parameter sorting information; and a target characteristic value of the target object under the reference attribute is expanded according to the reference characteristic to obtain a target characteristic range of the reference attribute. Before the data distribution information of the object attributes is determined according to the candidate characteristic values, the method further comprises:

2. The method of claim 1, wherein, Target characteristic values of the multiple object attributes of a target object are obtained, the target characteristic values are matched with the candidate characteristic values of the corresponding object attributes, and the candidate objects containing the candidate characteristic values matched with all the target characteristic values are determined as similar objects of the target object. ​ 3. The method of claim 1, wherein, The data distribution information of the object attribute is determined according to the candidate feature values, and the data distribution information comprises: When the object attribute belongs to a continuous feature attribute, standard deviation calculation processing is performed on the candidate feature values of the plurality of candidate objects under the object attribute, to obtain the data distribution information of the object attribute.

4. The method of claim 1, wherein, The data distribution information of the object attribute is determined according to the candidate feature values, and the data distribution information comprises: When the object attribute belongs to a discrete feature attribute, information entropy calculation processing is performed on the candidate feature values of the plurality of candidate objects under the object attribute, to obtain the data distribution information of the object attribute.

5. A similar object determination apparatus characterized by comprising: The apparatus comprises: a parameter acquisition module configured to acquire, for each candidate object, candidate feature values of a plurality of object attributes; a feature distribution module configured to determine data distribution information of the object attributes according to the candidate feature values, the data distribution information being used to indicate fluctuation degrees of the candidate feature values of the plurality of candidate objects under the object attributes; an attribute ranking module configured to, when the plurality of object attributes comprise continuous feature attributes and discrete feature attributes, rank the object attributes belonging to the continuous feature attributes according to standard deviations of the data distribution information, to obtain continuous ranking information, and rank the object attributes belonging to the discrete feature attributes according to information entropies of the data distribution information, to obtain discrete ranking information; and combine the discrete ranking information and the continuous ranking information, to obtain attribute ranking information; a feature expansion module configured to acquire target feature values of the plurality of object attributes of a target object, and expand the target feature values of the object attributes of the target object in sequence according to the attribute ranking information, to obtain at least one target feature range of the object attributes; The target feature values of the object attributes of the target object are expanded in sequence according to the attribute ranking information, to obtain at least one target feature range of the objects attributes, and the method comprises: reference attributes are determined from the object attributes of the target object in sequence according to the attribute ranking information, and one of the following steps is performed based on the reference attributes: When the reference attribute belongs to a continuous feature attribute, an attribute difference value is obtained according to maximum values and minimum values of the candidate feature values of all the candidate objects under the reference attribute; a parameter deviation value is determined according to the attribute difference value and a preset reference coefficient; and a target feature value of the target object under the reference attribute is expanded according to the parameter deviation value, to obtain a target feature range of the reference attribute. Or, When the reference attribute belongs to a discrete feature attribute, the candidate feature values of all the candidate objects under the reference attribute are classified, to obtain respective parameter groups; the respective parameter groups are ranked according to quantities of the candidate feature values in the parameter groups, to obtain parameter ranking information; a reference feature is determined from the candidate feature values of all the candidate objects under the reference attribute based on the parameter ranking information; and a target feature value of the target object under the reference attribute is expanded according to the reference feature, to obtain a target feature range of the reference attribute. The feature matching module is configured to match the candidate feature values of the corresponding object attribute with the target feature values in the target feature range and the target feature values not extended, and determine a similar object of the target object from the multiple candidate objects according to a matching result.

6. An electronic device, comprising: The electronic device includes a memory and a processor, the memory stores a computer program, and the processor implements the similar object determination method in any one of claims 1 to 4 when executing the computer program.

7. A computer-readable storage medium storing a computer program, wherein the computer program comprises the following steps of: receiving a request for a resource from a client; determining whether the client is authorized to access the resource; and if the client is authorized to access the resource, providing the resource to the client. The computer program is executed by the processor to implement the similar object determination method in any one of claims 1 to 4.

Images