KOC user published content quality evaluation method and device, terminal and medium

Abstract

The invention discloses a KOC user published content quality evaluation method and device, a terminal and a medium. According to the scheme provided by the invention, when published content uploaded by a KOC user is received, content features of published content data are extracted, and the published content data are stored according to a published content database used for storing historical published content; and performing similarity comparison on the content features and historical content features stored in a published content database to obtain a content similarity comparison result corresponding to the published content data so as to determine whether the content published by the KOC user belongs to high-repetition low-quality content or not according to the content similarity comparison result. Therefore, the content quality evaluation result of the content published by the KOC user this time is obtained, the low-quality published content is efficiently recognized, and platform side excitation resources are prevented from being wasted.

Description

Technical Field

[0001] The present application relates to the field of information processing technology, and in particular to a method, device, terminal and medium for evaluating the quality of content published by a KOC user. Background Art

[0002] With the advancement of information technology such as mobile Internet, social networks represented by Weibo, WeChat, and Xiaohongshu have achieved rapid development. The value of social networks in the fields of information dissemination, advertising marketing, and brand reputation building has become increasingly significant.

[0003] KOC (Key Opinion Consumer) is a type of consumer who can influence consumption in their own social circles. They focus more on sharing real consumption experiences and influencing other consumers' purchasing decisions by sharing real usage experiences and personal opinions. Therefore, they become the focus of brand and platform training. In order to encourage KOC to continue to create, many platforms have formulated corresponding incentive mechanisms. In the actual operation of this incentive mechanism, some KOCs may frequently publish low-quality or repetitive content to increase the incentives they receive, which affects the fairness of incentive distribution and causes a waste of platform incentive resources. Summary of the invention

[0004] The present application provides a method, device, terminal and medium for evaluating the quality of content published by KOC users, which realizes the quality evaluation of KOC published content, so as to achieve the invention purpose of efficiently identifying low-quality published content and avoiding wasting platform incentive resources.

[0005] In order to solve the above technical problems, the first aspect of the present application provides a method for evaluating the quality of content published by KOC users, comprising:

[0006] Receiving published content data uploaded by a KOC user, and extracting content features of the published content data;

[0007] According to a preset published content database, the content feature is compared with the historical content feature stored in the published content database for similarity, so as to obtain a content similarity comparison result corresponding to the published content data;

[0008] The content quality evaluation result of the published content data is determined according to the content similarity comparison result and in combination with a preset correspondence between the content similarity comparison result and the content quality evaluation result.

[0009] Preferably, the published content data includes: text-based published content, picture-based published content, mixed text-and-picture published content, and video-based published content.

[0010] Preferably, when the published content data is a mixed text and picture type published content, extracting the content features of the published content data includes:

[0011] Identify and split the published content data into text content and image content respectively;

[0012] The text content and the picture content are respectively subjected to feature extraction using a pre-trained content feature extraction model to obtain corresponding text content features and picture content features.

[0013] Preferably, when the published content data is video-type published content, extracting content features of the published content data includes:

[0014] Splitting the published content data into video to obtain voice data and image frame data;

[0015] Performing text conversion processing on the voice data to obtain text content;

[0016] According to the image frame data, performing image frame extraction processing on the image frame data containing the target product to obtain picture content;

[0017] The text content and the picture content are respectively subjected to feature extraction using a pre-trained content feature extraction model to obtain corresponding text content features and picture content features.

[0018] Preferably, when the published content data is a mixed text and picture type published content or a video type published content, the content features are compared with the historical content features stored in the published content database according to the preset published content database to obtain the content similarity comparison result corresponding to the published content data, including:

[0019] Comparing the text content features and the picture content features with the historical text content features and the historical picture content features stored in the published content database to obtain text similarity and picture similarity;

[0020] A comprehensive similarity is calculated according to the text similarity and the image similarity, so as to determine a content similarity comparison result corresponding to the published content data according to the comprehensive similarity.

[0021] Preferably, the method for constructing the published content database includes:

[0022] Acquire historical published content data, perform feature extraction on the historical published content data, and obtain feature vectors of each of the historical published content data;

[0023] Dividing each of the feature vectors into a plurality of feature vector segments, clustering the feature vector segments of each segment to obtain a plurality of vector segment categories;

[0024] According to the corresponding relationship between each of the vector segment categories and the feature vector, index feature information is generated, and the index feature information is stored in a preset database to obtain a published content database.

[0025] Preferably, the comparing the content features with the historical content features stored in the published content database to obtain the content similarity comparison result corresponding to the published content data specifically includes:

[0026] Divide the content feature into a plurality of content feature vector segments, and compare the similarity between the content feature vector segments and the index feature information in the published content database to obtain similar index feature information corresponding to each content feature vector segment, so as to determine the target feature vector most similar to the content feature from each of the similar index feature information;

[0027] According to the similarity comparison result between the content feature and the target feature vector, a content similarity comparison result corresponding to the published content data is determined.

[0028] At the same time, the second aspect of the present application provides a device for evaluating the quality of content published by a KOC user, comprising:

[0029] A content feature extraction unit, configured to receive the published content data uploaded by the KOC user and extract content features of the published content data;

[0030] A content similarity comparison unit is used to compare the content features with the historical content features stored in the published content database according to a preset published content database, and obtain a content similarity comparison result corresponding to the published content data;

[0031] The content quality evaluation unit is used to determine the content quality evaluation result of the published content data according to the content similarity comparison result and the preset corresponding relationship between the content similarity comparison result and the content quality evaluation result.

[0032] A third aspect of the present application provides a KOC user published content quality assessment terminal, including: a memory and a processor;

[0033] The memory is used to store program codes, and the program codes correspond to a method for evaluating the quality of content published by a KOC user as provided in the first aspect of the present application;

[0034] The processor is used for reading and executing the program code.

[0035] The fourth aspect of the present application provides a computer-readable storage medium, in which a program code corresponding to a method for evaluating the quality of content published by a KOC user as provided in the first aspect of the present application is stored, and the program code is used to be read and executed by a processor to implement a method for evaluating the quality of content published by a KOC user as provided in the first aspect of the present application.

[0036] It can be seen from the above technical solutions that this application has the following advantages:

[0037] The solution provided by the present application extracts content features of the published content data when receiving the published content uploaded by the KOC user, and compares the content features with the historical content features stored in the published content database for storing historical published content, thereby obtaining a content similarity comparison result corresponding to the published content data, and determining whether the content published by the KOC user is highly repetitive and low-quality content based on the content similarity comparison result, thereby obtaining a content quality evaluation result of the content published by the KOC user this time, thereby achieving efficient identification of low-quality published content and avoiding wasting platform incentive resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0039] Figure 1 A flowchart of an embodiment of a method for evaluating the quality of content published by a KOC user provided in this application.

[0040] Figure 2 A schematic diagram of a structure of an embodiment of a device for evaluating the quality of content published by a KOC user provided in the present application.

[0041] Figure 3 This is a schematic diagram of the structure of an embodiment of a KOC user published content quality assessment terminal provided in the present application. DETAILED DESCRIPTION

[0042] In the actual operation of this incentive mechanism, some KOCs may frequently publish low-quality or repetitive content, or slightly modify the published content, and publish essentially the same articles multiple times in order to obtain more incentives, in order to increase the chances of obtaining incentives and obtain more incentive resources. For users, the existence of a large amount of similar content will reduce the diversity of platform content, affect users' browsing experience, and may even cause users to question the authenticity of product content and the platform's image; for the platform, it will not be able to achieve the preset promotion effect, and may also issue incentives for essentially the same content multiple times, resulting in a waste of platform resources.

[0043] The embodiments of the present application provide a method, device, terminal and medium for evaluating the quality of content published by KOC users, which realizes the quality evaluation of KOC published content, so as to achieve the invention purpose of efficiently identifying low-quality published content and avoiding wasting platform incentive resources.

[0044] In order to make the purpose, features, and advantages of the invention of this application more obvious and easy to understand, the technical solutions in the embodiments of this application will be clearly and completely described below in conjunction with the drawings in the embodiments of this application. Obviously, the embodiments described below are only part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0045] First, a detailed description of an embodiment of a method for evaluating the quality of content published by a KOC user provided by the present application is as follows:

[0046] See also Figure 1 , the present application provides a method for evaluating the quality of content published by a KOC user, comprising:

[0047] Step 101: receiving published content data uploaded by a KOC user and extracting content features of the published content data;

[0048] Step 102: Based on the preset published content database, the content features are compared with the historical content features stored in the published content database to obtain a content similarity comparison result corresponding to the published content data;

[0049] Step 103: Determine the content quality evaluation result of the published content data according to the content similarity comparison result and the preset corresponding relationship between the content similarity comparison result and the content quality evaluation result.

[0050] It should be noted that in the scenario where KOC users publish content, KOC users will first upload the prepared published content data to the publishing platform. When the publishing platform receives the published content data, it first performs feature extraction processing on the published content data to obtain content features of the published content data. Then, based on the preset published content database, the content features are compared with the historical content features stored in the published content database for similarity to obtain content similarity comparison results corresponding to the published content data. The content similarity comparison results can be used to determine the similarity comparison results between the current published content data and the historical published content stored in the published content database. Then, based on the correspondence between the preset content similarity comparison results and the content quality evaluation results, the content quality evaluation results of the current published content data are determined. For example, when the content similarity comparison results show that the similarity comparison results between the current published content data and the historical published content data are lower than the preset threshold, the content quality evaluation result of the published content data can be determined to be qualified. Otherwise, the content quality evaluation result of the published content data is determined to be unqualified. The solution provided by this embodiment can efficiently identify the purpose of publishing low-quality content, so that the platform operation and maintenance party can adjust the proportion of incentive resources for low-quality published content according to the content quality evaluation results, so as to achieve more reasonable and accurate incentive resource distribution and avoid waste of incentive resources. At the same time, it also helps to improve the content quality and diversity of platform content, thereby improving the overall KOC operation effect and user browsing experience.

[0051] On the basis of the above basic embodiment, the method for evaluating the quality of KOC user published content provided in this embodiment may further include the following technical contents:

[0052] The published content data includes: text-based published content, picture-based published content, mixed text and picture-based published content, and video-based published content.

[0053] For the above types of published content data, the content features of the extracted published content data corresponding to each data type are as follows:

[0054] It should be noted that, for published content with a single content feature, such as copywriting-type published content and picture-type published content, conventional natural language processing models and image feature recognition models can be used to obtain corresponding text content features or picture content features.

[0055] When the published content data is a mixed text and image type published content, the content features of the published content data extracted in step 101 include:

[0056] Identify and split the published content data into text content and image content respectively;

[0057] Through the pre-trained content feature extraction model, feature extraction is performed on the text content and the image content respectively to obtain corresponding text content features and image content features.

[0058] Furthermore, when the published content data is video-type published content, the content features of the published content data extracted in step 101 include:

[0059] Splitting the published content data into video to obtain voice data and image frame data;

[0060] Perform text conversion on the voice data to obtain text content;

[0061] According to the image frame data, image frame extraction processing is performed on the image frame data containing the target product to obtain the image content;

[0062] Through the pre-trained content feature extraction model, feature extraction is performed on the text content and the image content respectively to obtain corresponding text content features and image content features.

[0063] It should be noted that when the published content data is a copywriting type of published content or a picture type of published content, the text content features or picture content features of the published content data can be compared with the historical content features stored in the published content database for similarity, and the text similarity or picture similarity obtained by comparison is used as the content similarity comparison result corresponding to the published content data. Taking the case where the published content data is a picture type of published content as an example, it can be determined based on the content similarity comparison result whether there is target data similar to the picture of the current published content data in the collected historical published content data. If the number of similar pictures reaches the preset picture number threshold, it can be determined that the current published content data has a high degree of similarity with the collected target data.

[0064] When the published content data is a mixed text and picture type published content or a video type published content, in step 102, based on a preset published content database, the content features are compared with the historical content features stored in the published content database for similarity, and the content similarity comparison results corresponding to the published content data are obtained, including:

[0065] Compare the text content features and the image content features with the historical text content features and the historical image content features stored in the published content database to obtain text similarity and image similarity;

[0066] The comprehensive similarity is calculated according to the text similarity and the image similarity, so as to determine the content similarity comparison result corresponding to the published content data according to the comprehensive similarity.

[0067] It should be noted that if the published content data is a mixed text and image publication content or a video publication content that contains multiple content features, the corresponding text content features and image content features can be obtained respectively based on the method mentioned in the above embodiment, and then, based on different types of features, similarity comparison is performed with the historical text content features and historical image content features stored in the published content database to obtain text similarity and image similarity. Based on the text similarity and image similarity, a comprehensive similarity is calculated, and the comprehensive similarity is used as the basis for determining the content similarity comparison result corresponding to the published content data. Among them, regarding the calculation method of the comprehensive similarity, this embodiment preferably adopts the weighted sum value of the text similarity and the image similarity as the comprehensive similarity, but in some embodiments, the ordinary sum value, ordinary product value or weighted product value of the text similarity and the image similarity can also be used as the comprehensive similarity.

[0068] The publishing content database mentioned in this embodiment is constructed in the following manner:

[0069] Acquire historical published content data, perform feature extraction on the historical published content data, and obtain feature vectors of each historical published content data;

[0070] Each eigenvector is divided into multiple eigenvector segments, and the eigenvector segments of each segment are clustered to obtain several vector segment categories;

[0071] According to the corresponding relationship between each vector segment category and the feature vector, index feature information is generated, and the index feature information is stored in a preset database to obtain a published content database.

[0072] Further, based on the above-mentioned method of constructing the published content database, the content features are compared with the historical content features stored in the published content database in step 102 of this embodiment to obtain the content similarity comparison result corresponding to the published content data, which specifically includes:

[0073] Divide the content feature into multiple content feature vector segments, compare the content feature vector segments with the index feature information in the published content database according to the similarity, obtain the similar index feature information corresponding to each content feature vector segment, and determine the target feature vector that is most similar to the content feature from each similar index feature information;

[0074] According to the similarity comparison result between the content feature and the target feature vector, the content similarity comparison result corresponding to the published content data is determined.

[0075] It should be noted that, regarding the construction of the published content database, first, all the published content uploaded by KOC users on the platform, that is, the historical published content data, is obtained, and then the historical published content data is deduplicated, damaged data is removed, and other data cleaning is performed to obtain standard data. The deep learning basic network model structure Resnet structure is adopted, and the pre-trained model is first used to extract feature vectors from standard pictures and text data. Then, the loss function adopts triplet (Triplet Loss) to train the feature extraction model, so as to use the feature extraction model to extract features from the historical published content data, and obtain feature vectors of each historical published content data, including: text content feature vector and picture content feature vector, so as to save these feature vectors into a preset database, thereby forming the published content database mentioned in this embodiment.

[0076] In order to further improve the efficiency of subsequent feature vector matching, this embodiment may further perform the following processing on the above feature vectors when constructing the published content database:

[0077] If there are N features, the feature dimension is D, which can be represented as N*D in the feature library. Each feature vector is divided into M segments, and then the image feature library is clustered according to the features of each segment. The number of categories is C. According to the category center, the corresponding feature vector is found, and each feature vector corresponds to the nearest category. This category is used as the index number of each segment of the corresponding feature vector. That is, a feature vector has only M dimensions from the original D dimensions to the index number representation. In this way, the feature library is stored according to this calculation method to form a publishing content database based on index features.

[0078] More specifically, for the newly uploaded content data to be evaluated by KOC users, the features of the published content are processed in the same way as when the above-mentioned published content database was constructed to generate M feature vector segments, which are then compared with the historical features in the published content database. The category center in the first segment of the feature is first found. Here, only C distances need to be calculated to find the most similar feature segment. Similarly, the category centers of 2, 3, ...M segments are calculated respectively. Finally, the index values ​​of multiple segments are combined to obtain the retrieved feature vector. The calculation method here greatly improves efficiency, and brute force calculation of millions of feature distances is no longer used. The retrieval response speed is improved, and the evaluation efficiency in the face of large-scale published content data comparison scenarios is improved.

[0079] The above is a detailed description of an embodiment of a method for evaluating the quality of content published by a KOC user provided by the present application. The following is a detailed description of an embodiment of a device for evaluating the quality of content published by a KOC user provided by the present application.

[0080] See also Figure 2 The present application provides a KOC user published content quality assessment device embodiment, comprising:

[0081] The content feature extraction unit 201 is used to receive the published content data uploaded by the KOC user and extract the content features of the published content data;

[0082] The content similarity comparison unit 202 is used to compare the content features with the historical content features stored in the published content database according to the preset published content database, and obtain the content similarity comparison result corresponding to the published content data;

[0083] The content quality evaluation unit 203 is used to determine the content quality evaluation result of the published content data according to the content similarity comparison result and the preset corresponding relationship between the content similarity comparison result and the content quality evaluation result.

[0084] Furthermore, the present application also provides a detailed description of a KOC user published content quality assessment terminal and a computer-readable storage medium embodiment.

[0085] See also Figure 3 , the present application provides a KOC user published content quality evaluation terminal embodiment, the terminal implementation types include but are not limited to: personal computers, industrial computers, servers and embedded intelligent devices, the terminal mainly includes: a memory 33 and a processor 31, the memory 33 and the processor 31 can be connected via a communication bus 34;

[0086] The memory is used to store program codes, and the program codes correspond to a method for evaluating the quality of content published by a KOC user provided in the above embodiment;

[0087] The processor is used to read and execute program code.

[0088] In a fourth aspect, the present application provides a computer-readable storage medium, in which a program code corresponding to a method for evaluating the quality of content published by a KOC user as provided in the above embodiment is stored, and the program code is used to be read and executed by a processor to implement a method for evaluating the quality of content published by a KOC user as provided in the above embodiment.

[0089] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the terminals, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0090] In the several embodiments provided in the present application, it should be understood that the disclosed terminals, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of units is only a logical function division, and there may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0091] The terms "first", "second", "third", "fourth", etc. (if any) in the specification of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present application described herein can, for example, be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0092] It should be understood that in the present application, "at least one (item)" means one or more, and "plurality" means two or more. "And / or" is used to describe the association relationship of associated objects, indicating that three relationships may exist. For example, "A and / or B" can mean: only A exists, only B exists, and A and B exist at the same time, where A and B can be singular or plural. The character " / " generally indicates that the objects associated before and after are in an "or" relationship. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, at least one of a, b or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, c can be single or multiple.

[0093] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0094] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0095] If the integrated unit is implemented in the form of 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 the present invention, 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. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0096] As described above, the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for evaluating the quality of content published by KOC users, characterized in that: include: Receiving published content data uploaded by a KOC user, and extracting content features of the published content data; According to a preset published content database, the content feature is compared with the historical content feature stored in the published content database for similarity, so as to obtain a content similarity comparison result corresponding to the published content data; The content quality evaluation result of the published content data is determined according to the content similarity comparison result and in combination with a preset correspondence between the content similarity comparison result and the content quality evaluation result.

2. A method for evaluating the quality of content published by KOC users according to claim 1, characterized in that: The published content data includes: text-based published content, picture-based published content, mixed text and picture-based published content, and video-based published content.

3. A method for evaluating the quality of content published by KOC users according to claim 2, characterized in that: When the published content data is a mixed text and picture type published content, extracting the content features of the published content data includes: Identify and split the published content data into text content and image content respectively; The text content and the picture content are respectively subjected to feature extraction using a pre-trained content feature extraction model to obtain corresponding text content features and picture content features.

4. A method for evaluating the quality of content published by KOC users according to claim 2, characterized in that: When the published content data is video-type published content, extracting content features of the published content data includes: Splitting the published content data into video to obtain voice data and image frame data; Performing text conversion processing on the voice data to obtain text content; According to the image frame data, performing image frame extraction processing on the image frame data containing the target product to obtain picture content; The text content and the picture content are respectively subjected to feature extraction using a pre-trained content feature extraction model to obtain corresponding text content features and picture content features.

5. A method for evaluating the quality of content published by KOC users according to claim 3 or 4, characterized in that: When the published content data is a mixed text and picture type published content or a video type published content, the content features are compared with the historical content features stored in the published content database according to the preset published content database, and the content similarity comparison result corresponding to the published content data is obtained, including: Comparing the text content features and the picture content features with the historical text content features and the historical picture content features stored in the published content database to obtain text similarity and picture similarity; A comprehensive similarity is calculated according to the text similarity and the image similarity, so as to determine a content similarity comparison result corresponding to the published content data according to the comprehensive similarity.

6. A method for evaluating the quality of content published by KOC users according to claim 1, characterized in that: The method for constructing the published content database includes: Acquire historical published content data, perform feature extraction on the historical published content data, and obtain feature vectors of each of the historical published content data; Dividing each of the feature vectors into a plurality of feature vector segments, clustering the feature vector segments of each segment to obtain a plurality of vector segment categories; According to the corresponding relationship between each of the vector segment categories and the feature vector, index feature information is generated, and the index feature information is stored in a preset database to obtain a published content database.

7. A method for evaluating the quality of content published by KOC users according to claim 6, characterized in that: The comparing the content feature with the historical content feature stored in the published content database to obtain the content similarity comparison result corresponding to the published content data specifically includes: Divide the content feature into a plurality of content feature vector segments, and compare the similarity between the content feature vector segments and the index feature information in the published content database to obtain similar index feature information corresponding to each content feature vector segment, so as to determine the target feature vector most similar to the content feature from each of the similar index feature information; According to the similarity comparison result between the content feature and the target feature vector, a content similarity comparison result corresponding to the published content data is determined.

8. A device for evaluating the quality of content published by a KOC user, characterized in that: include: A content feature extraction unit, configured to receive the published content data uploaded by the KOC user and extract content features of the published content data; A content similarity comparison unit is used to compare the content features with the historical content features stored in the published content database according to a preset published content database, and obtain a content similarity comparison result corresponding to the published content data; The content quality evaluation unit is used to determine the content quality evaluation result of the published content data according to the content similarity comparison result and the preset corresponding relationship between the content similarity comparison result and the content quality evaluation result.

9. A KOC user published content quality assessment terminal, characterized in that: include: Memory and processor; The memory is used to store program codes, and the program codes correspond to a method for evaluating the quality of content published by a KOC user according to any one of claims 1 to 7; The processor is used for reading and executing the program code.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a program code corresponding to the method for evaluating the quality of content published by a KOC user as claimed in any one of claims 1 to 7, and the program code is used to be read and executed by a processor to implement the method for evaluating the quality of content published by a KOC user as claimed in any one of claims 1 to 7.

Images