Object screening method and device

Abstract

The embodiment of the invention provides an object screening method and device, and the method comprises the steps: determining a plurality of candidate objects, and obtaining the feature information of each candidate object under at least one feature dimension; determining a feature weight corresponding to each feature dimension according to the feature information of the plurality of candidate objects under each feature dimension; based on the feature weight of each feature dimension and the feature information of the target candidate object under each feature dimension, generating an object score corresponding to the target candidate object, the target candidate object being any one of the plurality of candidate objects; and screening a target object from the plurality of candidate objects according to the object score of each candidate object. The feature information of each candidate object under at least one feature dimension is obtained, the feature weight corresponding to each feature dimension is determined, and the object score of each candidate object is obtained, so that the accuracy and objectivity of object scoring can be improved, the diversity and comprehensiveness of candidate object evaluation can be enhanced, and the user experience is improved. And more efficient screening of the candidate objects is realized.

Description

Technical Field

[0001] The embodiments of the present specification relate to the field of computer technology, and more particularly to an object screening method and device. Background Art

[0002] Currently, screening objects, such as users and candidates, are usually done by professional business personnel who manually collect, count and evaluate object data offline. This not only takes a lot of time, but also fails to accumulate experience in object screening, resulting in object screening being highly dependent on professional business personnel, high labor costs and low efficiency. Therefore, a more efficient object screening method is urgently needed. Summary of the invention

[0003] In view of this, an embodiment of the present specification provides an object screening method. One or more embodiments of the present specification also relate to an object screening device, a computing device, a computer-readable storage medium and a computer program product to solve the technical defects existing in the prior art.

[0004] According to a first aspect of an embodiment of this specification, there is provided an object screening method, comprising: Determine multiple candidate objects, and obtain feature information of each candidate object in at least one feature dimension; Determine the feature weight corresponding to each feature dimension according to the feature information of multiple candidate objects in each feature dimension; Based on the feature weights of each feature dimension and the feature information of the target candidate object under each feature dimension, generating an object score corresponding to the target candidate object, wherein the target candidate object is any one of the multiple candidate objects; A target object is screened from multiple candidate objects according to the object score of each candidate object.

[0005] According to a second aspect of an embodiment of this specification, there is provided an object screening device, including: An acquisition module is configured to determine a plurality of candidate objects and acquire feature information of each candidate object in at least one feature dimension; A determination module is configured to determine a feature weight corresponding to each feature dimension according to feature information of multiple candidate objects in each feature dimension; A generating module is configured to generate an object score corresponding to a target candidate object based on a feature weight of each feature dimension and feature information of the target candidate object under each feature dimension, wherein the target candidate object is any one of the multiple candidate objects; The screening module is configured to screen a target object from a plurality of candidate objects according to an object score of each candidate object.

[0006] According to a third aspect of an embodiment of this specification, a computing device is provided, including: Memory and processor; The memory is used to store computer executable instructions, and the processor is used to execute the computer executable instructions. When the computer executable instructions are executed by the processor, the steps of the above-mentioned object screening method are implemented.

[0007] According to a fourth aspect of the embodiments of this specification, a computer-readable storage medium is provided, which stores computer-executable instructions, and when the instructions are executed by a processor, the steps of the above-mentioned object screening method are implemented.

[0008] According to a fifth aspect of the embodiments of this specification, a computer program product is provided, including a computer program / instruction, which implements the steps of the above-mentioned object screening method when executed by a processor.

[0009] An embodiment of the present specification realizes determining multiple candidate objects and obtaining feature information of each candidate object in at least one feature dimension; determining feature weights corresponding to each feature dimension based on the feature information of the multiple candidate objects in each feature dimension; generating an object score corresponding to the target candidate object based on the feature weights of each feature dimension and the feature information of the target candidate object in each feature dimension, wherein the target candidate object is any one of the multiple candidate objects; and screening the target object from the multiple candidate objects based on the object scores of each candidate object.

[0010] In this way, by obtaining the feature information of each candidate object in at least one feature dimension and determining the feature weight corresponding to each feature dimension, it is possible to obtain the object score of each candidate object based on the feature information in multiple feature dimensions and the weights corresponding to each feature dimension, thereby improving the accuracy and objectivity of the object score, while also enhancing the diversity and comprehensiveness of the evaluation of candidate objects and achieving more efficient screening of candidate objects. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] Figure 1 is a flow chart of an object screening method provided by one embodiment of this specification; Figure 2 is a process flow chart of an object screening method provided by an embodiment of the present specification; Figure 3 It is a schematic diagram of object information of an object screening method provided by an embodiment of this specification; Figure 4 It is a characteristic data schematic diagram of an object screening method provided by an embodiment of this specification; Figure 5 This is a schematic diagram of a display page of an object screening method provided by an embodiment of this specification; Figure 6 is a schematic diagram of the structure of an object screening device provided by an embodiment of this specification; Figure 7 It is a structural block diagram of a computing device provided by an embodiment of this specification. DETAILED DESCRIPTION

[0012] Many specific details are described in the following description to facilitate a full understanding of this specification. However, this specification can be implemented in many other ways than those described herein, and those skilled in the art can make similar generalizations without violating the connotation of this specification, so this specification is not limited to the specific implementation disclosed below.

[0013] The terms used in one or more embodiments of this specification are only for the purpose of describing specific embodiments, and are not intended to limit one or more embodiments of this specification. The singular forms of "a", "said" and "the" used in one or more embodiments of this specification and the appended claims are also intended to include plural forms, unless the context clearly indicates other meanings. It should also be understood that the term "and / or" used in one or more embodiments of this specification refers to and includes any or all possible combinations of one or more associated listed items.

[0014] It should be understood that although the terms first, second, etc. may be used to describe various information in one or more embodiments of this specification, this information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of one or more embodiments of this specification, the first may also be referred to as the second, and similarly, the second may also be referred to as the first. Depending on the context, the word "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining".

[0015] In addition, it should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in one or more embodiments of this specification are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with the relevant laws, regulations and standards of the relevant countries and regions, and provide corresponding operation entrances for users to choose to authorize or refuse.

[0016] First, the terms involved in one or more embodiments of this specification are explained.

[0017] Expert: refers to an individual or team who can promote products on the Internet platform through multimedia forms such as live broadcasts, short videos, notes, articles, etc., such as Internet celebrities in daily life.

[0018] Information Entropy: It describes the uncertainty of possible events in an information source and the amount of information, and can reflect the probability distribution characteristics of each event in the information source.

[0019] Insurance product: It is provided by insurance companies and can pay a certain amount of insurance money to the insurer or beneficiary in accordance with the terms of the insurance contract when the insured encounters a specific risk accident or suffers property loss or personal injury due to a specific event, or when the insurance period expires and reaches the age, term and other conditions stipulated in the contract.

[0020] In this specification, an object screening method is provided. This specification also relates to an object screening device, a computing device, a computer-readable storage medium, and a computer program product, which are described in detail one by one in the following embodiments.

[0021] See also Figure 1 , Figure 1 A flow chart of an object screening method provided according to an embodiment of the present specification is shown, which specifically includes the following steps.

[0022] Step 102: Determine multiple candidate objects, and obtain feature information of each candidate object in at least one feature dimension.

[0023] In practical applications, multiple candidate objects may be determined, and feature information of each candidate object in at least one feature dimension may be obtained. Based on the feature information of each candidate object in at least one feature dimension, subsequent screening of the candidate objects may be performed.

[0024] The object screening method provided in the embodiments of this specification can be applied to the object screening platform of the business system. The business system is the merchant's system, and the object screening platform can be used to store, manage, screen and display the relevant information of the object.

[0025] Specifically, the object can be understood as the object that the merchant wants to cooperate with, which can be a user, a candidate, etc. For example, in the application scenario of product marketing, the object can be a talent. The candidate object can be understood as an object waiting to be screened.

[0026] In actual applications, if a merchant wants to establish cooperative relationships with a preset number of objects, he or she can first obtain multiple candidate objects through channels such as suppliers, and then screen out target objects with which cooperative relationships can be established from the multiple candidate objects.

[0027] Specifically, the feature dimension can represent the information type, that is, one information type can correspond to one feature dimension. The feature information can be understood as the feature value under the feature dimension.

[0028] In the actual implementation process, since the information types of an object may include multiple types, object screening based on object information under only one type of information is likely to result in inaccurate and subjective screening results. Therefore, according to one or more optional embodiments of the present specification, for each candidate object, object information of each candidate object under multiple types of information may be collected. Furthermore, object screening may be performed on each candidate object based on the object information of each candidate object under multiple types of information.

[0029] According to an optional embodiment of the present specification, determining multiple candidate objects may include: Get multiple candidate objects provided by a supplier.

[0030] Furthermore, on the basis of determining a plurality of candidate objects, each candidate object and object information of each candidate object may be uploaded to the object screening platform.

[0031] According to another optional embodiment of the present specification, determining multiple candidate objects may further include: A plurality of candidate objects are determined from the objects stored in the object screening platform.

[0032] Optionally, the object screening platform can receive candidate objects and object information uploaded by users. The user can be understood as a business person. Further, the object screening platform can store the object information in the form of a data table or file.

[0033] According to an optional embodiment of the present specification, obtaining feature information of each candidate object in at least one feature dimension may include: Feature extraction is performed on the object information to obtain feature information of each candidate object in at least one feature dimension.

[0034] Optionally, feature extraction may be performed on the object information of each candidate object through a model or a tool to obtain feature information of each candidate object in at least one feature dimension.

[0035] Optionally, the feature dimensions and feature information obtained through feature extraction may be recorded in a feature information table.

[0036] Exemplarily, the feature information table may include the object ID of the candidate object, each feature dimension, and feature information under each feature dimension.

[0037] Step 104: Determine the feature weight corresponding to each feature dimension according to the feature information of the plurality of candidate objects in each feature dimension.

[0038] In practical applications, after determining multiple candidate objects and obtaining feature information of each candidate object in at least one feature dimension, the feature weight corresponding to each feature dimension can be determined according to the feature information of the multiple candidate objects in each feature dimension.

[0039] Specifically, the feature weight can represent the importance of each feature dimension to the scoring calculation during the scoring calculation of the object; it can also represent the influence of each feature dimension on the object score.

[0040] According to one or more optional embodiments of the present specification, determining the feature weight corresponding to each feature dimension according to feature information of multiple candidate objects in each feature dimension may include the following steps: According to the feature information of multiple candidate objects in each feature dimension, determine the information entropy corresponding to each feature dimension; Based on the information entropy corresponding to each feature dimension, the feature weight corresponding to each feature dimension is obtained.

[0041] Specifically, information entropy can characterize the richness of information contained and the degree of influence on the scoring results. The larger the information entropy, the richer the information content of the feature dimension, and the greater the influence of the feature dimension on the scoring results.

[0042] Optionally, the information entropy corresponding to each feature dimension may be calculated according to the feature information of multiple candidate objects in each feature dimension; and the feature weight corresponding to each feature dimension may be determined based on the information entropy corresponding to each feature dimension.

[0043] By applying this embodiment, by determining the information entropy corresponding to each feature dimension according to the feature information of multiple candidate objects in each feature dimension, it is possible to take into account the correlation between each feature dimension based on the information entropy, so that the feature weight corresponding to each feature dimension can be determined in combination with the correlation between each feature dimension, thereby avoiding the problems of subjective weighting and inconsistency, improving the accuracy of feature weights, and thus improving the accuracy of object scoring.

[0044] In an optional embodiment of the present specification, determining the information entropy corresponding to each feature dimension according to feature information of multiple candidate objects in each feature dimension may include the following steps: According to the feature information of the multiple candidate objects in each feature dimension, a feature array corresponding to the multiple candidate objects is obtained; According to the preset operation rules, based on the feature array, the information entropy corresponding to each feature dimension is obtained.

[0045] Specifically, the feature array can be understood as a two-dimensional array containing feature information of each candidate object in each feature dimension. The feature array can be a two-dimensional data of m*n specification, where m can represent the number of feature dimensions and n can represent the number of candidate objects.

[0046] Optionally, the feature array may be extracted from a feature information table.

[0047] Specifically, the preset operation rule is an operation rule for calculating the information entropy corresponding to each feature dimension, and may include an information entropy operation formula. The information entropy operation formula may be specifically referred to in the following formula (1).

[0048] Formula (1) Optionally, obtaining the information entropy corresponding to each feature dimension based on the feature array according to a preset operation rule may include: First, calculate the Part. Specifically, it can be calculated in a loop, and the loop length is the number of candidates. Represents the j-th element of the feature array.

[0049] Then the calculation result is compared with Multiply. Where n is the number of feature dimensions.

[0050] Through the above method, the information entropy corresponding to each feature dimension can be calculated; the information entropy corresponding to each feature dimension can be saved in the information entropy array.

[0051] By applying this embodiment, by obtaining feature arrays corresponding to multiple candidate objects according to feature information of multiple candidate objects in each feature dimension, it is possible to obtain information entropy corresponding to each feature dimension based on the feature array according to preset operation rules, thereby improving the efficiency of information entropy calculation of each feature dimension.

[0052] In an optional embodiment of the present specification, obtaining feature arrays corresponding to multiple candidate objects according to feature information of multiple candidate objects in each feature dimension may include the following steps: Determining a number of candidate objects and a number of feature dimensions of at least one feature dimension for the plurality of candidate objects; A feature array is obtained according to the number of candidate objects, the number of feature dimensions, and the feature information under each feature dimension.

[0053] Optionally, the feature array can be extracted from the feature information table. According to the number of candidate objects of multiple candidate objects in the feature information table and the number of feature dimensions of each feature dimension, feature data can be extracted from the feature information table and saved in an array to obtain the feature array.

[0054] By applying this embodiment, a feature array is obtained according to the number of candidate objects, the number of feature dimensions and the feature information under each feature dimension, which can facilitate the subsequent calculation of the information entropy formula based on the feature array and improve the efficiency of calculating information entropy for each feature dimension.

[0055] In an optional embodiment of the present specification, obtaining the feature weight corresponding to each feature dimension based on the information entropy corresponding to each feature dimension may include the following steps: Determine the initial weight of the information entropy corresponding to each feature dimension; The initial weights are normalized to obtain the feature weights corresponding to each feature dimension.

[0056] Specifically, the initial weight can be understood as an initial weight value obtained based on the information entropy corresponding to each feature dimension.

[0057] Optionally, determining the initial weight of the information entropy corresponding to each feature dimension may include: Based on the information entropy array, the initial weights of the information entropy corresponding to each feature dimension are calculated.

[0058] For example, assuming that the feature dimensions include 3 and the information entropy array corresponding to the feature dimensions is [0.5, 0.3, 0.7], the calculation process of the initial weight is: w1 = 1-.05=0.5 w2 = 1-0.3=0.7 w3 = 1-0.7=0.3 That is, the initial weights of the information entropy corresponding to each feature dimension are 0.5, 0.7, and 0.3, respectively.

[0059] Optionally, normalizing the initial weights to obtain feature weights corresponding to each feature dimension may include: According to the initial weights of the information entropy corresponding to each feature dimension, the sum of the initial weights is obtained; Based on the proportion of each initial weight to the total initial weight, the feature weight corresponding to each feature dimension is obtained.

[0060] For example, following the above example, after calculating that the initial weights of the information entropy corresponding to each feature dimension are 0.5, 0.7, and 0.3 respectively, the sum of the initial weights can be calculated based on the initial weights of the information entropy corresponding to each feature dimension: sum = 0.5+0.7+0.3=1.5; further, the feature weights corresponding to each feature dimension can be calculated separately: nor_w1 = 0.5 / 1.5=0.33 nor_w2 = 0.7 / 1.5=0.47 nor_w3 = 0.3 / 1.5=0.2 In this way, the feature weights corresponding to each feature dimension can be calculated, and the feature weights corresponding to each feature dimension can be saved in an array to obtain a feature weight array [0.33, 0.47, 0.2].

[0061] By applying this embodiment, the initial weights of the information entropy corresponding to each feature dimension are determined; the initial weights are normalized to obtain the feature weights corresponding to each feature dimension, which can ensure that the sum of the feature weights is 1, ensure the rationality of the weight distribution in the subsequent weighting process, improve the accuracy and consistency of the feature weights, and thus improve the accuracy of the information entropy calculation.

[0062] Step 106: Generate an object score corresponding to the target candidate object based on the feature weight of each feature dimension and the feature information of the target candidate object under each feature dimension, wherein the target candidate object is any one of the multiple candidate objects.

[0063] In practical applications, based on determining the feature weight corresponding to each feature dimension according to the feature information of multiple candidate objects in each feature dimension, an object score corresponding to the target candidate object can be generated based on the feature weight of each feature dimension and the feature information of the target candidate object in each feature dimension.

[0064] Specifically, the target candidate object is any candidate object among the multiple candidate objects. The object score can represent the quality or recommendation degree of the candidate object. The higher the object score, the higher the quality of the object and the more recommended it is to establish a cooperative relationship with it. The object score can be a score value, or it can be data such as star rating and grade.

[0065] According to one or more optional embodiments of the present specification, generating an object score corresponding to a target candidate object based on a feature weight of each feature dimension and feature information of the target candidate object under each feature dimension may include the following steps: Determine a target feature weight corresponding to a target feature dimension and target feature information of a target candidate object under the target feature dimension, wherein the target feature dimension is any one of the feature dimensions; Multiply the target feature weight and the target feature information to generate the score of the target candidate under the target feature dimension; According to the scores of the target candidate objects in each feature dimension, the object scores corresponding to the target candidate objects are obtained.

[0066] Specifically, the target feature dimension is any one of the feature dimensions.

[0067] In practical applications, when the feature weights corresponding to each feature dimension are calculated through information entropy, the object score of each candidate object can be calculated respectively for each candidate object.

[0068] This embodiment takes any one of multiple candidate objects as an example to further explain the calculation process of the object score.

[0069] Since the target candidate object has feature information under each feature dimension, the score of the target candidate object under each feature dimension can be calculated separately. The score under each feature dimension is calculated based on the feature weight corresponding to the feature dimension and the feature value under the feature dimension. Further, on the basis of calculating the score of the target candidate object under each feature dimension, each score can be added to obtain the comprehensive score of the target candidate object, that is, the object score corresponding to the target candidate object.

[0070] By applying this embodiment, the score of the target candidate object under the target feature dimension is generated by multiplying the target feature weight and the target feature information; the object score corresponding to the target candidate object is obtained according to the score of the target candidate object under each feature dimension, and the object score of each candidate object can be obtained based on the feature weight corresponding to the information entropy, thereby improving the efficiency and accuracy of the object score calculation.

[0071] According to one or more optional embodiments of the present specification, after generating an object score corresponding to the target candidate object based on the feature weight of each feature dimension and the feature information of the target candidate object under each feature dimension, the following steps may also be included: Associating the object score of each candidate object with each candidate object in the object information library; Each candidate object and the corresponding object score are displayed on the display page.

[0072] Specifically, the object information library can be understood as a database corresponding to the object screening platform, which is used to store data such as object information and object scores of each object.

[0073] In practical applications, when the object score of each candidate object is obtained, the object score can be associated with each candidate object in the object information library, that is, the object score can be associated with the corresponding candidate object. Furthermore, each candidate object and the corresponding object score can be displayed on the display page, so that it is convenient for business personnel to screen each candidate object based on the object score. The candidate objects can also be arranged and displayed in order according to the object score from high to low or from low to high on the display page.

[0074] By applying this embodiment, the object score of each candidate object is associated with each candidate object in the object information library; each candidate object and the corresponding object score are displayed in the display page, which can support business personnel to select target objects with greater intention of cooperation from multiple candidate objects online according to the scores from high to low, thereby facilitating object screening of candidate objects and helping to improve object screening efficiency.

[0075] Step 108: Filter the target object from the multiple candidate objects according to the object score of each candidate object.

[0076] In actual applications, when the object score corresponding to each candidate object is obtained, the target object can be screened from multiple candidate objects according to the object score of each candidate object.

[0077] Specifically, the target object can be understood as the object recommended for establishing a cooperative relationship, or it can be understood as the object with which the merchant is more interested in cooperating.

[0078] Optionally, the target object can be obtained by screening from multiple candidate objects according to the object score.

[0079] Optionally, according to a preset number of target objects, a preset number of objects with top scores among the candidate objects may be determined as target objects according to the object scores from high to low.

[0080] By screening the target object from multiple candidate objects according to the object score of each candidate object, it is possible to automatically screen the objects and improve the object screening efficiency.

[0081] In the actual implementation process, the object screening method provided in this embodiment can be used to construct an object screening model, and the object screening model can include an object screening algorithm. Specifically, the implementation steps of the object screening algorithm can include: Determine multiple candidate objects, and obtain feature information of each candidate object in at least one feature dimension; Determine the feature weight corresponding to each feature dimension according to the feature information of multiple candidate objects in each feature dimension; Based on the feature weights of each feature dimension and the feature information of the target candidate object in each feature dimension, an object score corresponding to the target candidate object is generated, wherein the target candidate object is any one of the multiple candidate objects.

[0082] Among them, determining the feature weight corresponding to each feature dimension according to the feature information of multiple candidate objects in each feature dimension can be implemented as follows: first, calculating the information entropy corresponding to each feature dimension according to the feature information of multiple candidate objects in each feature dimension; and then obtaining the feature weight corresponding to each feature dimension based on the information entropy corresponding to each feature dimension.

[0083] In practical applications, an object screening model can be constructed by pre-configuring feature dimensions, feature weight calculation methods, etc.

[0084] Through the object screening model, the object score of each candidate object can be obtained. Specifically, it can be implemented as follows: The feature information of multiple candidate objects in at least one feature dimension is input into the object screening model to obtain the object score corresponding to each candidate object output by the object screening model.

[0085] In the actual implementation process, on the basis of obtaining the object screening model, in order to continuously improve the accuracy and reliability of the model, the object screening model can also be subjected to feature precipitation and model iteration.

[0086] According to one or more optional embodiments of the present specification, after screening the target object from multiple candidate objects according to the object score of each candidate object, the following steps may also be included: Collect the content reading volume corresponding to the published content of the target object and determine the object ranking of the target object; According to the content reading volume and object ranking, determine whether each feature dimension and the feature weight corresponding to each feature dimension meet the preset adjustment conditions; If so, each feature dimension is analyzed, and each feature dimension and / or a feature weight corresponding to each feature dimension is adjusted based on the analysis result.

[0087] Specifically, the published content can be understood as the multimedia content published by the target object for the product to be promoted, which can specifically include videos, articles, notes, etc. The content reading volume can be understood as the number of times the published content is read on the Internet platform.

[0088] In actual applications, after screening and obtaining the target object, the merchant can establish a cooperative relationship with the target object, and the target object can publish multimedia content on the Internet platform for the product to be promoted to achieve the promotion of the product to be promoted.

[0089] Optionally, the content reading volume may be the total reading volume of each published content published by the target object for the product to be promoted.

[0090] For example, the target object can publish multimedia content for the product to be promoted in different forms such as short videos, articles, notes, etc. on multiple different Internet platforms. When collecting content readings, the content readings corresponding to the target object's published content on each platform can be collected according to the content ID of the published content, and then the total content readings, that is, the content readings of the target object, can be obtained based on the content readings of each published content.

[0091] Optionally, in the process of calculating the total content reading volume, the content reading volume of each published content may be weighted according to different multimedia content or different Internet platforms.

[0092] Specifically, the object ranking can be understood as the order of the object score of the target object among the objects. The preset adjustment condition may include: the difference between the content reading volume and the object ranking is greater than a preset threshold.

[0093] In actual applications, for the target objects that have been cooperated with, business personnel can collect the reading volume of the content actually delivered based on the content ID; compare the reading volume with the object ranking recorded by the target object in the object screening platform; if the difference between the reading volume and the object ranking is too large, it means that there is a certain error in the output result of the object screening model, and feature analysis is needed, and based on the analysis results, the weights of each feature dimension should be adjusted, or more features should be added to iterate the model.

[0094] This embodiment is applied to collect the content reading volume corresponding to the published content of the target object and determine the object ranking of the target object; based on the content reading volume and the object ranking, when each feature dimension and the feature weight corresponding to each feature dimension meet the preset adjustment conditions, analyze each feature dimension, and adjust each feature dimension and / or the feature weight corresponding to each feature dimension based on the analysis result. In the actual application process of the model, the model can be continuously adjusted according to the actual delivery effect of the object, thereby realizing the feature precipitation and iteration of the model, making the output result of the model more precise and the object scoring more accurate.

[0095] In practical applications, on the basis of obtaining the object screening model, it is also possible to verify whether the score given by the model has reference significance, that is, whether the model is useful.

[0096] According to an optional embodiment of the present specification, the verification model may include: Determine a historical target object, and obtain the reading volume corresponding to the published content of the historical target object, and characteristic information of the historical target object in at least one characteristic dimension; Based on the object screening model, obtaining an object score for each historical target object according to feature information of the historical target object in at least one feature dimension; The object screening model is verified based on the reading volume corresponding to the published content of the historical target objects and the object score of each historical target object.

[0097] In practical applications, the objects that have been manually selected in the past can be determined. Since the objects that have been manually selected in the past have published corresponding multimedia content such as product promotion videos and articles, the corresponding reading volume of the multimedia content can be collected. At this time, the model can calculate the score of each object based on the feature information of each historical target object, and then sort the object scores from high to low, and determine whether the order of the object scores from high to low matches the order of the reading volume from high to low, so as to verify the model.

[0098] According to another optional embodiment of the present specification, the verification model may further include: Determine a plurality of candidate objects to be processed, and obtain feature information of each candidate object to be processed in at least one feature dimension; Based on the object screening model, obtaining an object score for each candidate object to be processed according to feature information of each candidate object to be processed in at least one feature dimension; Screening a target object to be processed from a plurality of candidate objects to be processed according to the object score of each candidate object to be processed; The reading volume of the published content of the target object to be processed is collected, and the object screening model is verified according to the reading volume of the published content of the target object to be processed.

[0099] Specifically, the candidate objects to be processed can be understood as objects that are not selected by the business personnel.

[0100] In actual applications, you can test content that is about to be promoted in the future. For example, if you need to select 100 objects from a batch of objects in the future, you can reserve a preset number of names for the model to filter. The model filters a preset number of objects from the candidate objects that the business personnel have not selected. The preset number of objects filtered by the model can be actually put on the market to test the effect, that is, the reading volume, so as to verify the model.

[0101] An object screening method provided by an embodiment of the present specification realizes determining multiple candidate objects and obtaining feature information of each candidate object in at least one feature dimension; determining feature weights corresponding to each feature dimension based on the feature information of the multiple candidate objects in each feature dimension; generating an object score corresponding to the target candidate object based on the feature weights of each feature dimension and the feature information of the target candidate object in each feature dimension, wherein the target candidate object is any one of the multiple candidate objects; and screening the target object from the multiple candidate objects based on the object scores of each candidate object.

[0102] In this way, by obtaining the feature information of each candidate object in at least one feature dimension and determining the feature weight corresponding to each feature dimension, it is possible to obtain the object score of each candidate object based on the feature information in multiple feature dimensions and the weights corresponding to each feature dimension, thereby improving the accuracy and objectivity of the object score, while also enhancing the diversity and comprehensiveness of the evaluation of candidate objects and achieving more efficient screening of candidate objects.

[0103] In actual applications, influencers refer to individuals or teams that promote products on Internet platforms through multimedia forms such as live broadcasts, short videos, notes, and articles. Through the influencer's fan base and influence, it is possible to increase the exposure of brands and products, increase product sales, and thus promote commercial conversion. Efficient screening of influencers can help merchants better identify and select partners, which is conducive to expanding traffic and promoting product sales.

[0104] The following combination Figure 2-Figure 5 , taking the application of the object screening method provided in this specification in the expert selection scheme as an example, the object screening method is further described. Figure 2 A processing flow chart of an object screening method provided by an embodiment of the present specification is shown, which specifically includes the following steps.

[0105] Step 202: Obtain a list of talents provided by the supplier, and upload the list of talents to the talent database of the platform.

[0106] Specifically, the expert list may include expert information of each expert under multiple information types.

[0107] See also Figure 3 , Figure 3 FIG. 1 is a schematic diagram showing object information of an object screening method provided by an embodiment of the present specification. Figure 3 As shown, the expert list may include various types of expert information, and each column in the table may correspond to one type of expert information. The platform may store this expert information in the expert library for subsequent expert screening.

[0108] Step 204: extracting features from the expert information in the expert list to obtain feature information of each candidate expert in multiple feature dimensions.

[0109] See also Figure 4 , Figure 4 FIG. 1 is a schematic diagram showing characteristic data of an object screening method provided by an embodiment of the present specification. Figure 4As shown, assuming that the feature dimensions used to calculate information entropy include the number of fans (kol_fans_cnt), the number of likes and collections (kol_like_cnt) and the active fan ratio (activate_fans_rate), the feature information under the number of fans, the number of likes and collections and the active fan ratio can be extracted based on the information in the expert list; and a feature information table is obtained based on the expert id (kol_id), the number of fans (kol_fans_cnt), the number of likes and collections (kol_like_cnt) and the active fan ratio (activate_fans_rate); based on the feature information of each kol_id in the feature information table under kol_fans_cnt, kol_like_cnt and activate_fans_rate, a feature array for calculating information entropy is obtained.

[0110] Step 206: Based on the feature information of each candidate expert in multiple feature dimensions, calculate the information entropy corresponding to each feature dimension respectively.

[0111] Optionally, the information entropy calculated for each feature dimension can be saved in an array.

[0112] Step 208: Obtain the feature weight corresponding to each feature dimension according to the information entropy corresponding to each feature dimension.

[0113] Optionally, for the information entropy corresponding to each feature dimension, the feature weight corresponding to each feature dimension calculated may also be saved in an array.

[0114] Step 210: Multiply the feature information by the corresponding feature weight to obtain the comprehensive score of the expert.

[0115] Step 212: The comprehensive score is associated with the expert database through the expert ID and displayed on the display page.

[0116] See also Figure 5 , Figure 5 A schematic diagram of a display page of an object screening method provided by an embodiment of this specification is shown. Figure 5 As shown, the display page can display the expert ID, expert nickname, expert rating, number of fans, number of likes and collections, and the increase in fans in the past 30 days in a table. Through these data, business personnel can clearly know the various information of the expert and the score and rating of each expert. The display page can also include interactive controls, which can trigger the experts to be sorted and displayed in order based on the score; interactive controls can also be used to automatically select a preset number of target experts based on the score and display them on the page.

[0117] By applying this embodiment, feature extraction is performed on expert information to obtain feature information of each expert in multiple feature dimensions, and the information entropy corresponding to each feature dimension is determined based on the feature information of each expert in multiple feature dimensions. Based on the information entropy, the correlation between each feature dimension can be taken into account, so that the feature weight corresponding to each feature dimension can be determined in combination with the correlation between each feature dimension, thereby avoiding the problems of subjective weighting and inconsistency, improving the accuracy of feature weights, thereby improving the accuracy of scoring, and realizing automatic and efficient expert selection.

[0118] Corresponding to the above method embodiment, this specification also provides an object screening device embodiment, Figure 6 FIG. 2 shows a schematic diagram of the structure of an object screening device provided by an embodiment of the present specification. Figure 6 As shown, the device comprises: The acquisition module 602 is configured to determine a plurality of candidate objects and acquire feature information of each candidate object in at least one feature dimension.

[0119] Determination module 604: is configured to determine the feature weight corresponding to each feature dimension according to the feature information of multiple candidate objects in each feature dimension.

[0120] Generation module 606: configured to generate an object score corresponding to a target candidate object based on the feature weights of each feature dimension and the feature information of the target candidate object under each feature dimension, wherein the target candidate object is any one of the multiple candidate objects.

[0121] Screening module 608: configured to screen a target object from a plurality of candidate objects according to the object score of each candidate object.

[0122] Optionally, the determination module 604 is further configured to: According to the feature information of multiple candidate objects in each feature dimension, determine the information entropy corresponding to each feature dimension; Based on the information entropy corresponding to each feature dimension, the feature weight corresponding to each feature dimension is obtained.

[0123] Optionally, the determination module 604 is further configured to: According to the feature information of the multiple candidate objects in each feature dimension, a feature array corresponding to the multiple candidate objects is obtained; According to the preset operation rules, based on the feature array, the information entropy corresponding to each feature dimension is obtained.

[0124] Optionally, the determination module 604 is further configured to: Determining a number of candidate objects and a number of feature dimensions of at least one feature dimension for the plurality of candidate objects; A feature array is obtained according to the number of candidate objects, the number of feature dimensions, and the feature information under each feature dimension.

[0125] Optionally, the determination module 604 is further configured to: Determine the initial weight of the information entropy corresponding to each feature dimension; The initial weights are normalized to obtain the feature weights corresponding to each feature dimension.

[0126] Optionally, the generating module 606 is further configured to: Determine a target feature weight corresponding to a target feature dimension and target feature information of a target candidate object under the target feature dimension, wherein the target feature dimension is any one of the feature dimensions; Multiply the target feature weight and the target feature information to generate the score of the target candidate under the target feature dimension; According to the scores of the target candidate objects in each feature dimension, the object scores corresponding to the target candidate objects are obtained.

[0127] Optionally, the object screening device further includes a display module configured to: Associating the object score of each candidate object with each candidate object in the object information library; Each candidate object and the corresponding object score are displayed on the display page.

[0128] Optionally, the object screening device further includes an analysis module configured to: Collect the content reading volume corresponding to the published content of the target object and determine the object ranking of the target object; According to the content reading volume and object ranking, determine whether each feature dimension and the feature weight corresponding to each feature dimension meet the preset adjustment conditions; If so, each feature dimension is analyzed, and each feature dimension and / or a feature weight corresponding to each feature dimension is adjusted based on the analysis result.

[0129] By applying this embodiment, by obtaining feature information of each candidate object in at least one feature dimension and determining the feature weight corresponding to each feature dimension, it is possible to obtain an object score for each candidate object based on the feature information in multiple feature dimensions and the weights corresponding to each feature dimension, thereby improving the accuracy and objectivity of the object score, while also enhancing the diversity and comprehensiveness of the evaluation of the candidate objects and achieving more efficient screening of the candidate objects.

[0130] The above is a schematic scheme of an object screening device of this embodiment. It should be noted that the technical scheme of the object screening device and the technical scheme of the object screening method described above belong to the same concept, and the details not described in detail in the technical scheme of the object screening device can be referred to the description of the technical scheme of the object screening method described above.

[0131] Figure 7 The block diagram of a computing device 700 according to an embodiment of the present specification is shown. The components of the computing device 700 include but are not limited to a memory 710 and a processor 720. The processor 720 is connected to the memory 710 via a bus 730, and the database 750 is used to store data.

[0132] The computing device 700 also includes an access device 740 that enables the computing device 700 to communicate via one or more networks 760. Examples of these networks include a public switched telephone network (PSTN), a local area network (LAN), a wide area network (WAN), a personal area network (PAN), or a combination of communication networks such as the Internet. The access device 740 may include one or more of any type of network interface (e.g., a network interface card (NIC)) of wired or wireless, such as an IEEE 802.11 wireless local area network (WLAN) wireless interface, a world-wide interoperability for microwave access (Wi-MAX) interface, an Ethernet interface, a universal serial bus (USB) interface, a cellular network interface, a Bluetooth interface, and a near field communication (NFC).

[0133] In one embodiment of the present specification, the above components of the computing device 700 and Figure 7 Other components not shown in the figure may also be connected to each other, for example, via a bus. It should be understood that Figure 7 The computing device structure block diagram shown is only for the purpose of illustration, and is not intended to limit the scope of this specification. Those skilled in the art can add or replace other components as needed.

[0134] The computing device 700 may be any type of stationary or mobile computing device, including a mobile computer or mobile computing device (e.g., a tablet computer, a personal digital assistant, a laptop computer, a notebook computer, a netbook, etc.), a mobile phone (e.g., a smart phone), a wearable computing device (e.g., a smart watch, smart glasses, etc.), or other types of mobile devices, or a stationary computing device such as a desktop computer or a personal computer (PC). The computing device 700 may also be a mobile or stationary server.

[0135] The processor 720 is used to execute the following computer executable instructions, which implement the steps of the above-mentioned object screening method when executed by the processor.

[0136] The above is a schematic scheme of a computing device of this embodiment. It should be noted that the technical scheme of the computing device and the technical scheme of the above object screening method belong to the same concept, and the details not described in detail in the technical scheme of the computing device can be referred to the description of the technical scheme of the above object screening method.

[0137] An embodiment of the present specification further provides a computer-readable storage medium storing computer-executable instructions, which can implement the steps of the above-mentioned object screening method when executed by a processor.

[0138] The above is a schematic scheme of a computer-readable storage medium of this embodiment. It should be noted that the technical scheme of the storage medium and the technical scheme of the object screening method described above are of the same concept, and the details not described in detail in the technical scheme of the storage medium can be referred to the description of the technical scheme of the object screening method described above.

[0139] An embodiment of the present specification also provides a computer program product, including a computer program / instruction, which implements the steps of the above-mentioned object screening method when executed by a processor.

[0140] The above is a schematic scheme of a computer program product of this embodiment. It should be noted that the technical scheme of the computer program product and the technical scheme of the above object screening method belong to the same concept, and the details not described in detail in the technical scheme of the computer program product can be referred to the description of the technical scheme of the above object screening method.

[0141] The above is a description of a specific embodiment of the specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0142] The computer instructions include computer program codes, which may be in source code form, object code form, executable files or some intermediate forms, etc. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal and software distribution medium, etc. It should be noted that the content contained in the computer-readable medium may be appropriately increased or decreased according to the requirements of patent practice. For example, in some regions, according to patent practice, computer-readable media do not include electric carrier signals and telecommunication signals.

[0143] It should be noted that, for the above-mentioned method embodiments, for the sake of simplicity of description, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the embodiments of this specification are not limited by the order of the actions described, because according to the embodiments of this specification, some steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the embodiments of this specification.

[0144] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0145] The preferred embodiments of this specification disclosed above are only used to help explain this specification. The optional embodiments do not describe all the details in detail, nor do they limit the invention to only the specific implementation methods described. Obviously, many modifications and changes can be made according to the content of the embodiments of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the embodiments of this specification, so that technicians in the relevant technical field can well understand and use this specification. This specification is only limited by the claims and their full scope and equivalents.

Claims

1. A method for object screening, characterized in that: include: Determine multiple candidate objects, and obtain feature information of each candidate object in at least one feature dimension; Determining, according to the feature information of the multiple candidate objects in each feature dimension, a feature weight corresponding to each feature dimension; Based on the feature weights of the feature dimensions and the feature information of the target candidate object under the feature dimensions, generating an object score corresponding to the target candidate object, wherein the target candidate object is any one of the multiple candidate objects; A target object is screened from the plurality of candidate objects according to the object score of each candidate object.

2. The method according to claim 1, characterized in that The determining, according to the feature information of the plurality of candidate objects in each feature dimension, the feature weight corresponding to each feature dimension comprises: Determine, according to the feature information of the plurality of candidate objects in each feature dimension, the information entropy corresponding to each feature dimension; Based on the information entropy corresponding to each feature dimension, the feature weight corresponding to each feature dimension is obtained.

3. The method according to claim 2, characterized in that The determining, according to the feature information of the plurality of candidate objects in each feature dimension, the information entropy corresponding to each feature dimension respectively includes: Obtaining feature arrays corresponding to the multiple candidate objects according to feature information of the multiple candidate objects in each feature dimension; According to preset operation rules, based on the feature array, the information entropy corresponding to each feature dimension is obtained.

4. The method according to claim 3, characterized in that The obtaining, according to the feature information of the multiple candidate objects in each feature dimension, feature arrays corresponding to the multiple candidate objects includes: Determining the number of candidate objects of the plurality of candidate objects and the number of feature dimensions of the at least one feature dimension; The feature array is obtained according to the number of candidate objects, the number of feature dimensions and the feature information under each feature dimension.

5. The method according to any one of claims 2 to 4, characterized in that: The obtaining, based on the information entropy corresponding to each feature dimension, the feature weight corresponding to each feature dimension, includes: Determine the initial weights of the information entropy corresponding to each of the feature dimensions; The initial weights are normalized to obtain feature weights corresponding to the feature dimensions.

6. The method according to claim 1, characterized in that The generating an object score corresponding to the target candidate object based on the feature weights of the feature dimensions and the feature information of the target candidate object under the feature dimensions includes: Determine a target feature weight corresponding to a target feature dimension, and target feature information of the target candidate object under the target feature dimension, wherein the target feature dimension is any one of the feature dimensions; Multiplying the target feature weight and the target feature information to generate a score of the target candidate object under the target feature dimension; According to the scores of the target candidate objects in the various feature dimensions, an object score corresponding to the target candidate objects is obtained.

7. The method according to claim 1, characterized in that After generating the object score corresponding to the target candidate object based on the feature weights of the feature dimensions and the feature information of the target candidate object in the feature dimensions, the method further includes: Associating the object score of each candidate object with each candidate object in the object information library; The candidate objects and the corresponding object scores are displayed on the display page.

8. The method according to claim 1, characterized in that After screening the target object from the plurality of candidate objects according to the object score of each candidate object, the method further includes: Collecting the content reading volume corresponding to the published content of the target object, and determining the object ranking of the target object; According to the content reading volume and the object ranking, determining whether each feature dimension and the feature weight corresponding to each feature dimension meet a preset adjustment condition; If so, the feature dimensions are analyzed, and the feature dimensions and / or the feature weights corresponding to the feature dimensions are adjusted based on the analysis results.

9. An object screening device, characterized in that: include: An acquisition module is configured to determine a plurality of candidate objects and acquire feature information of each candidate object in at least one feature dimension; A determination module is configured to determine a feature weight corresponding to each feature dimension according to feature information of the plurality of candidate objects in each feature dimension; A generating module configured to generate an object score corresponding to the target candidate object based on the feature weights of the feature dimensions and the feature information of the target candidate object under the feature dimensions, wherein the target candidate object is any one of the multiple candidate objects; The screening module is configured to screen a target object from the plurality of candidate objects according to an object score of each candidate object.

10. A computing device, characterized in that: include: Memory and processor; The memory is used to store computer programs / instructions, and the processor is used to execute the computer programs / instructions. When the computer programs / instructions are executed by the processor, the steps of the object screening method according to any one of claims 1 to 8 are implemented.

11. A computer-readable storage medium, characterized in that: It stores a computer program / instruction, which, when executed by a processor, implements the steps of the object screening method according to any one of claims 1 to 8.

12. A computer program product, characterized in that The invention comprises a computer program / instruction, which, when executed by a processor, implements the steps of the object screening method according to any one of claims 1 to 8.

Images