Content review method and apparatus, computer device, and storage medium

Abstract

The application relates to a content review method and device, computer equipment, a storage medium and a computer program product, and relates to the technical field of artificial intelligence. The method comprises the following steps: obtaining a first reader tag set obtained by pre-launching target content in a to-be-reviewed launch area; matching the first reader tag set with a reference reader tag set corresponding to the target content to obtain a matching result; in the case that the matching result meets a preset related condition, determining that the target content meets the review condition of the to-be-reviewed launch area, and pre-launching the target content into a launch area corresponding to a next level of the to-be-reviewed launch area; taking the launch area corresponding to the next level as the to-be-reviewed launch area, returning to the step of obtaining the first reader tag set obtained by pre-launching the target content in the to-be-reviewed launch area, and continuing until the level of the to-be-reviewed launch area reaches a target level. The method can improve the content review efficiency.

Description

Content moderation methods, devices, computer equipment and storage media Technical Field

[0001] This application relates to the field of artificial intelligence technology, and in particular to a content moderation method, apparatus, computer equipment, storage medium, and computer program product. Background Technology

[0002] In today's mobile internet era, information is fragmented, with a deluge of data flooding in. This has made everyone more discerning in their content selection, making content operation increasingly crucial within a comprehensive operational system. After content is produced, it needs to be published on content platforms. Current content platforms emphasize content production itself, and to ensure content quality meets standards, they require review before publication.

[0003] The existing content review method is mainly manual review. Before content is distributed to different regions, the reviewers in each region need to conduct a quality review of the content to ensure that it meets the quality requirements of that region.

[0004] However, given the massive amount of content awaiting publication that requires review, existing content review methods are extremely time-consuming, and the timeliness of approved content cannot be guaranteed. Therefore, existing content review methods are inefficient. Summary of the Invention

[0005] Therefore, it is necessary to provide a content moderation method, apparatus, computer equipment, computer-readable storage medium, and computer program product that can improve efficiency in addressing the aforementioned technical problems.

[0006] Firstly, this application provides a content moderation method. The method includes:

[0007] Obtain the first set of reader tags by pre-deploying the target content in the area awaiting review;

[0008] The first reader tag set is matched with the reference reader tag set corresponding to the target content to obtain a matching result; the reference reader tag set is determined according to the level of the region to be reviewed and distributed.

[0009] If the matching result meets the preset relevant conditions, it is determined that the target content meets the review conditions of the pending review and placement area, and the target content is pre-placed in the placement area corresponding to the next level of the pending review and placement area;

[0010] The next level's corresponding delivery area is designated as the delivery area to be reviewed. The process returns to the step of obtaining the first reader tag set obtained by pre-delivering the target content in the delivery area to be reviewed, until the level of the delivery area to be reviewed reaches the target level.

[0011] In one embodiment, matching the first reader tag set with the reference reader tag set corresponding to the target content to obtain a matching result includes:

[0012] If the level of the area to be reviewed is the primary level, the first reader tag set is matched with the historical reader tag set corresponding to the target content to obtain the matching result;

[0013] If the level of the area to be reviewed and distributed is not the primary level, the first reader tag set is matched with the second reader tag set obtained by pre-distributing the target content in the distribution area corresponding to the previous level of the area to be reviewed and distributed, and a matching result is obtained.

[0014] In one embodiment, the step of matching the first reader tag set with the second reader tag set obtained by pre-deploying the target content in the delivery area corresponding to the next higher level of the delivery area to be reviewed, and obtaining the matching result, includes:

[0015] In the second reader tag set obtained by pre-deploying the first reader tag set and the target content in the delivery area corresponding to the level above the delivery area to be reviewed, a first identical tag is determined.

[0016] Calculate the ratio of the number of the first identical tags to the number of reader tags in the first reader tag set to obtain the first matching value;

[0017] For each first identical tag, determine the first proportion of the first identical tag in the second reader tag set and the second proportion of the first identical tag in the first reader tag set, and calculate the ratio of the first proportion to the second proportion to obtain the second matching value;

[0018] The first matching value and the second matching value are used to form a matching result.

[0019] In one embodiment, matching the first reader tag set with the historical reader tag set corresponding to the target content to obtain a matching result includes:

[0020] For each historical submission of the target content, a second identical tag is determined from the first reader tag set and the historical reader tag subset corresponding to the historical submission content;

[0021] Calculate the ratio of the number of the second identical tags to the number of reader tags in the first reader tag set to obtain the third matching value corresponding to the historical submission content;

[0022] For each second identical tag, determine the third proportion of the second identical tag in the historical reader tag subset corresponding to the historical submission content and the fourth proportion of the second identical tag in the first reader tag set, and calculate the ratio of the third proportion to the fourth proportion to obtain the fourth matching value corresponding to the historical submission content.

[0023] The third and fourth matching values ​​corresponding to each of the aforementioned historical submissions are used to form the matching result.

[0024] In one embodiment, the process of determining the correspondence between the level and the deployment area includes:

[0025] For each delivery dimension included in the delivery area, the values ​​of each dimension under the delivery dimension are sorted according to a preset level index to obtain the sequence number of each dimension value under the delivery dimension.

[0026] For each delivery area, the level corresponding to the delivery area is determined based on the ordinal number of each dimension value included in the delivery area under each delivery dimension and the weight corresponding to each delivery dimension.

[0027] In one embodiment, the method further includes:

[0028] For each historical submission, determine whether the third matching value corresponding to the historical submission is greater than or equal to a preset tag quantity threshold, and determine whether the fourth matching value corresponding to the historical submission is greater than or equal to a preset tag proportion threshold.

[0029] If a third matching value corresponding to a historical submission is greater than or equal to a preset tag quantity threshold, and a fourth matching value corresponding to the historical submission is greater than or equal to a preset tag proportion threshold, then the matching result is determined to meet the preset relevant conditions.

[0030] Secondly, this application also provides a content moderation device. The device includes:

[0031] The acquisition module is used to acquire the first set of reader tags obtained by pre-deploying the target content in the area to be reviewed and approved;

[0032] The matching module is used to match the first reader tag set with the reference reader tag set corresponding to the target content to obtain a matching result; the reference reader tag set is determined according to the level of the region to be reviewed and distributed.

[0033] The pre-delivery module is used to determine that the target content meets the review conditions of the delivery area to be reviewed when the matching result meets the preset relevant conditions, and to pre-deliver the target content to the delivery area corresponding to the next level of the delivery area to be reviewed;

[0034] The update module is used to designate the distribution area corresponding to the next level as the distribution area to be reviewed, and return to the step of obtaining the first reader tag set obtained by pre-distributing the target content in the distribution area to be reviewed, until the level of the distribution area to be reviewed reaches the target level.

[0035] In one embodiment, the matching module is specifically used for:

[0036] If the level of the area to be reviewed is the primary level, the first reader tag set is matched with the historical reader tag set corresponding to the target content to obtain the matching result;

[0037] If the level of the area to be reviewed and distributed is not the primary level, the first reader tag set is matched with the second reader tag set obtained by pre-distributing the target content in the distribution area corresponding to the previous level of the area to be reviewed and distributed, and a matching result is obtained.

[0038] In one embodiment, the matching module is specifically used for:

[0039] In the second reader tag set obtained by pre-deploying the first reader tag set and the target content in the delivery area corresponding to the level above the delivery area to be reviewed, a first identical tag is determined.

[0040] Calculate the ratio of the number of the first identical tags to the number of reader tags in the first reader tag set to obtain the first matching value;

[0041] For each first identical tag, determine the first proportion of the first identical tag in the second reader tag set and the second proportion of the first identical tag in the first reader tag set, and calculate the ratio of the first proportion to the second proportion to obtain the second matching value;

[0042] The first matching value and the second matching value are used to form a matching result.

[0043] In one embodiment, the matching module is specifically used for:

[0044] For each historical submission of the target content, a second identical tag is determined from the first reader tag set and the historical reader tag subset corresponding to the historical submission content;

[0045] Calculate the ratio of the number of the second identical tags to the number of reader tags in the first reader tag set to obtain the third matching value corresponding to the historical submission content;

[0046] For each second identical tag, determine the third proportion of the second identical tag in the historical reader tag subset corresponding to the historical submission content and the fourth proportion of the second identical tag in the first reader tag set, and calculate the ratio of the third proportion to the fourth proportion to obtain the fourth matching value corresponding to the historical submission content.

[0047] The third and fourth matching values ​​corresponding to each of the aforementioned historical submissions are used to form the matching result.

[0048] In one embodiment, the device further includes:

[0049] The sorting module is used to sort the values ​​of each dimension under each delivery dimension according to a preset level index for each delivery dimension included in the delivery area, so as to obtain the sequence number of each dimension value under the delivery dimension.

[0050] The first determining module is used to determine the level corresponding to each delivery area based on the ordinal number of each dimension value included in the delivery area under each delivery dimension and the weight corresponding to each delivery dimension.

[0051] In one embodiment, the device further includes:

[0052] The judgment module is used to determine, for each historical submission, whether the third matching value corresponding to the historical submission is greater than or equal to a preset tag quantity threshold, and whether the fourth matching value corresponding to the historical submission is greater than or equal to a preset tag proportion threshold.

[0053] The second determining module is used to determine that the matching result satisfies preset relevant conditions if there is a third matching value corresponding to a historical submission that is greater than or equal to a preset tag quantity threshold, and a fourth matching value corresponding to the historical submission that is greater than or equal to a preset tag proportion threshold.

[0054] Thirdly, this application also provides a computer device. The computer device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the steps described in the first aspect.

[0055] Fourthly, this application also provides a computer-readable storage medium. The computer-readable storage medium stores a computer program thereon, which, when executed by a processor, performs the steps described in the first aspect.

[0056] Fifthly, this application also provides a computer program product. The computer program product includes a computer program that, when executed by a processor, implements the steps described in the first aspect.

[0057] The aforementioned content review method, apparatus, computer equipment, storage medium, and computer program product pre-divide different distribution areas and determine the corresponding levels for each area. Target content is pre-distributed to the distribution areas awaiting review, and a first set of reader tags is obtained from this pre-distribution. This first set of reader tags is then matched with a reference set of reader tags corresponding to the target content. The matching result is then determined to satisfy preset conditions to assess whether the target content meets the review criteria for the distribution areas awaiting review. If the target content meets these criteria, it is pre-distributed to the next level distribution area, which is then designated as the next level distribution area. This process is repeated until the level of the distribution areas reaches the target level, signifying that the target content has been successfully distributed across all distribution areas. In this way, when target content needs to be distributed to different areas, levels are defined based on the characteristics of each area. Distribution is then progressively and tentatively implemented according to the distribution level, continuously verifying quality and reducing distribution risk. This automated content review significantly reduces review time and improves efficiency compared to manual review. Attached Figure Description

[0058] Figure 1 is a flowchart illustrating a content moderation method in one embodiment;

[0059] Figure 2 is a flowchart illustrating the steps of matching the first reader tag set with the reference reader tag set corresponding to the target content in one embodiment.

[0060] Figure 3 is a flowchart illustrating the steps of matching the first reader tag set with the second reader tag set obtained by pre-deploying the target content in the delivery area corresponding to the next higher level of the delivery area to be reviewed in one embodiment.

[0061] Figure 4 is a flowchart illustrating the steps of matching the first reader tag set with the historical reader tag set corresponding to the target content in one embodiment.

[0062] Figure 5 is a flowchart illustrating the process of determining the correspondence between grade and deployment area in one embodiment;

[0063] Figure 6 is a flowchart illustrating the content moderation method in another embodiment;

[0064] Figure 7 is a structural block diagram of a content moderation device in one embodiment;

[0065] Figure 8 is an internal structure diagram of a computer device in one embodiment. Detailed Implementation

[0066] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0067] In one embodiment, as shown in Figure 1, a content moderation method is provided. This embodiment illustrates the method by applying it to a terminal. It is understood that this method can also be applied to a server, and further to a system including both a terminal and a server, and is implemented through interaction between the terminal and the server. The terminal can be, but is not limited to, various personal computers, laptops, smartphones, tablets, IoT devices, and portable wearable devices. IoT devices can include smart speakers, smart TVs, smart air conditioners, smart in-vehicle devices, etc. Portable wearable devices can include smartwatches, smart bracelets, head-mounted devices, etc. The server can be implemented using a standalone server or a server cluster consisting of multiple servers. In this embodiment, the method includes the following steps:

[0068] Step 101: Obtain the first set of reader tags obtained by pre-deploying the target content in the area to be reviewed.

[0069] In this embodiment, the target content is content to be published and intended for distribution to different distribution areas. The target content can be an article or a video. Different distribution areas can be different content platforms or different distribution ranges within the same content platform. For example, different distribution areas could be content platform A and content platform B. Alternatively, different distribution areas could be distribution range c and distribution range d of content platform A. Distribution range c may not overlap with distribution range d, or it may overlap with distribution range d, and distribution range c may also belong to distribution range d. The distribution area to be reviewed is the distribution area where the target content is pre-distributed. The first reader tag set includes multiple reader tags obtained from the pre-distribution of the target content in the distribution area to be reviewed. Reader tags are tags of users who browse or watch the target content. Reader tags are used to reflect the characteristics of the target content's readers. For example, reader tags can be reader gender tags, reader age tags, and tags representing areas of interest to the reader.

[0070] The terminal pre-delivers the target content to the designated delivery area awaiting review. Then, the terminal obtains the first set of reader tags obtained from the pre-delivery of the target content to the designated delivery area.

[0071] In one example, the terminal presets a pre-deployment duration. The terminal obtains the first set of reader tags obtained during the pre-deployment period when the target content is pre-deployed in the region awaiting review. The pre-deployment duration is the length of time the target content is pre-deployed. For example, the pre-deployment duration could be 7 days.

[0072] Step 102: Match the first reader tag set with the reference reader tag set corresponding to the target content to obtain the matching result.

[0073] The reference reader tag set is determined based on the level of the region to be reviewed and distributed.

[0074] In this embodiment, the reference reader tag set includes multiple reference reader tags. The reference reader tag set reflects the characteristics of previous readers of the target content. The matching result indicates the similarity between the first reader tag set and the reference reader tag set corresponding to the target content.

[0075] The terminal matches the reader tags in the first reader tag set with the reader tags in the reference reader tag set corresponding to the target content to obtain the matching result.

[0076] Step 103: If the matching result meets the preset relevant conditions, determine that the target content meets the review conditions of the pending review and placement area, and pre-place the target content in the placement area corresponding to the next level of the pending review and placement area.

[0077] In this embodiment, relevant conditions are used to measure the similarity between the first reader tag set and the reference reader tag set. These relevant conditions may include relevant thresholds. The correspondence between the delivery area and the level is predetermined. The terminal delivers the target content sequentially to the delivery area corresponding to each level, according to the level.

[0078] The terminal pre-determines the corresponding level for each delivery area. If the matching results meet preset conditions, the terminal determines that the target content meets the review criteria for the delivery area to be reviewed. Then, the terminal determines the level of that delivery area to be reviewed, i.e., the current level. Finally, the terminal pre-delivers the target content to the delivery area corresponding to the next lower level of the delivery area to be reviewed.

[0079] Step 104: Take the distribution area corresponding to the next level as the distribution area to be reviewed, and return to the step of obtaining the first reader tag set obtained by pre-distributing the target content in the distribution area to be reviewed, until the level of the distribution area to be reviewed reaches the target level.

[0080] In this embodiment, the target level can be the highest level. The delivery area corresponding to the target level can be the most important delivery area or the delivery area with the largest delivery range.

[0081] The terminal designates the delivery area corresponding to the next lower level as the delivery area to be reviewed. Then, the terminal determines the level of the delivery area to be reviewed. Then, the terminal returns to the step of obtaining the first set of reader tags obtained by pre-delivering the target content in the delivery area to be reviewed, until the level of the delivery area to be reviewed reaches the target level.

[0082] In the above content review method, different distribution areas are pre-divided, and the corresponding levels for each distribution area are determined. Target content is pre-distributed to the distribution areas to be reviewed, and the first set of reader tags obtained from the pre-distribution of the target content in the distribution areas to be reviewed is obtained. Then, the first set of reader tags is matched with the reference set of reader tags corresponding to the target content to obtain the matching result. The matching result is then judged to see if it meets the preset relevant conditions to determine whether the target content meets the review conditions of the distribution areas to be reviewed. If the target content meets the review conditions of the distribution areas to be reviewed, the target content is pre-distributed to the distribution area corresponding to the next level of the distribution area to be reviewed, and the distribution area corresponding to the next level is designated as the distribution area to be reviewed. The above process is repeated until the level of the distribution area to be reviewed reaches the target level, that is, the target content has been distributed in each distribution area. In this way, when the target content needs to be distributed to different distribution areas, the level is divided according to the characteristics of different distribution areas, and the distribution is carried out gradually and tentatively according to the distribution level, continuously verifying the quality and reducing the distribution risk. It can automatically realize content review, which significantly reduces the review time and improves the efficiency of content review compared to manual review. Moreover, this content moderation method is an independent and comprehensive content moderation system that can be applied to various content platforms, thus improving the universality of content moderation.

[0083] In one embodiment, as shown in Figure 2, the specific process of matching the first reader tag set with the reference reader tag set corresponding to the target content to obtain the matching result includes the following steps:

[0084] Step 201: If the level of the area to be reviewed and distributed is the primary level, match the first reader tag set with the historical reader tag set corresponding to the target content to obtain the matching result.

[0085] In this embodiment, the primary level is the lowest level at which the target content is permitted to be placed. The primary level can be the lowest level or a higher level. For example, the levels include 3 levels, with the importance of the placement areas corresponding to levels 1 to 3 increasing sequentially. The primary level can be level 1 or level 2. The historical reader tag set is the collection of reader tags for the historical submissions of the user who contributed the target content.

[0086] When the level of the area to be reviewed and distributed is the primary level, the terminal will match the reader tags in the first reader tag set with the reader tags in the historical reader tag set corresponding to the target content to obtain the matching result.

[0087] Step 202: If the level of the area to be reviewed is not the primary level, match the first reader tag set with the second reader tag set obtained by pre-deploying the target content in the area corresponding to the next higher level of the area to be reviewed, and obtain the matching result.

[0088] In this embodiment of the application, the second reader tag set is a set of reader tags obtained by pre-deploying the target content in the delivery area corresponding to the next higher level of the delivery area to be reviewed.

[0089] If the level of the area to be reviewed and distributed is not the primary level, the terminal will match the reader tags in the first reader tag set with the reader tags in the second reader tag set obtained by pre-distributing the target content in the distribution area corresponding to the next higher level of the area to be reviewed and distributed, and obtain the matching result.

[0090] In the above content review method, the reference reader tag set is determined based on the level of the region to be reviewed. If the region is at the basic level, the first reader tag set is matched with the historical reader tag set corresponding to the target content to obtain a matching result. If the region is not at the basic level, the first reader tag set is matched with the second reader tag set obtained from pre-deployment of the target content in the region corresponding to the previous level of the region to be reviewed, to obtain a matching result. This approach, using different reference reader tag sets and matching methods for different levels of regions to be reviewed, better reflects the actual situation of content review and improves its accuracy.

[0091] In one embodiment, as shown in Figure 3, the specific process of matching the first reader tag set with the second reader tag set obtained by pre-deploying the target content in the delivery area corresponding to the next higher level of the delivery area to be reviewed and delivered, and obtaining the matching result, includes the following steps:

[0092] Step 301: In the second reader tag set obtained by pre-deploying the first reader tag set and the target content in the next higher level of the pending deployment area, determine the first identical tag.

[0093] In this embodiment, the terminal obtains a second set of reader tags obtained by pre-delivering the target content in the delivery area corresponding to the level above the delivery area to be reviewed. Then, in the first set of reader tags and the second set of reader tags obtained by pre-delivering the target content in the delivery area corresponding to the level above the delivery area to be reviewed, the terminal determines the same tag, namely the first same tag.

[0094] Step 302: Calculate the ratio of the number of first identical tags to the number of reader tags in the first reader tag set to obtain the first matching value.

[0095] In this embodiment, the first matching value represents the similarity in quantity between the tags in the first reader tag set and the tags in the second reader tag set. The terminal calculates the ratio of the number of identical tags to the number of reader tags in the first reader tag set to obtain the first matching value.

[0096] Step 303: For each first identical tag, determine the first proportion of the first identical tag in the second reader tag set and the second proportion of the first identical tag in the first reader tag set, and calculate the ratio of the first proportion to the second proportion to obtain the second matching value.

[0097] In this embodiment, the second matching value represents the proportional similarity between tags in the first reader tag set and the second reader tag set. For each first identical tag, the terminal determines a first proportion of the first identical tag in the second reader tag set and a second proportion of the first identical tag in the first reader tag set. Then, the terminal divides the first proportion by the second proportion to obtain the ratio of the first proportion to the second proportion. The terminal then uses the ratio of the first proportion to the second proportion as the second matching value.

[0098] Step 304: Combine the first matching value and the second matching value to form the matching result.

[0099] In this embodiment of the application, the terminal uses the first matching value and the second matching value to form a matching result.

[0100] In the aforementioned content review method, a first matching tag is determined from the first reader tag set and the second reader tag set obtained by pre-deploying the target content in the distribution area corresponding to the next higher level of the distribution area to be reviewed. The ratio of the number of first matching tags to the number of reader tags in the first reader tag set is calculated to obtain a first matching value. Then, for each first matching tag, a first proportion of that first matching tag in the second reader tag set and a second proportion of that first matching tag in the first reader tag set are determined, and the ratio of the first proportion to the second proportion is calculated to obtain a second matching value. The first matching value and the second matching value constitute the matching result. In this way, when the level of the distribution area to be reviewed is not the primary level, by calculating the first matching value and the second matching value, the similarity of tags in the first reader tag set and the second reader tag set can be comprehensively evaluated from the number and proportion of matching tags. This is more in line with the actual situation of content review and can improve the comprehensiveness and accuracy of content review.

[0101] In one embodiment, as shown in Figure 4, the specific process of matching the first reader tag set with the historical reader tag set corresponding to the target content to obtain the matching result includes the following steps:

[0102] Step 401: For each historical submission of the target content, determine the second identical tag in the first reader tag set and the historical reader tag subset corresponding to the historical submission.

[0103] In this embodiment, the historical submissions refer to the content successfully submitted by the user who submitted the target content in various delivery areas. The historical reader tag subset is a set of historical reader tags corresponding to a historical submission. The historical reader tag set includes the historical reader tag subset corresponding to each historical submission.

[0104] For each historical submission of the target content, the terminal determines the same tag, i.e., the second same tag, in the first reader tag set and the historical reader tag subset corresponding to the historical submission.

[0105] Step 402: Calculate the ratio of the number of second identical tags to the number of reader tags in the first reader tag set to obtain the third matching value corresponding to the historical submission content.

[0106] In this embodiment, the third matching value represents the numerical similarity between the first reader tag set and the tags in the historical reader tag subset corresponding to a historical submission. The terminal calculates the ratio of the number of identical tags to the number of reader tags in the first reader tag set to obtain the third matching value corresponding to the historical submission.

[0107] Step 403: For each second identical tag, determine the third proportion of the second identical tag in the historical reader tag subset corresponding to the historical submission content and the fourth proportion of the second identical tag in the first reader tag set, and calculate the ratio of the third proportion to the fourth proportion to obtain the fourth matching value corresponding to the historical submission content.

[0108] In this embodiment, the fourth matching value represents the proportional similarity between the first reader tag set and the tags in a subset of historical reader tags corresponding to a historical submission. For each second identical tag, the terminal determines a third proportion of the second identical tag in the subset of historical reader tags corresponding to the historical submission and a fourth proportion of the second identical tag in the first reader tag set. Then, the terminal calculates the ratio of the third proportion to the fourth proportion to obtain the fourth matching value corresponding to the historical submission.

[0109] Step 404: Combine the third and fourth matching values ​​corresponding to each historical submission to form the matching result.

[0110] In this embodiment of the application, the terminal uses the third and fourth matching values ​​corresponding to each historical submission to form a matching result.

[0111] In one example, for each historical submission, the terminal uses the third and fourth matching values ​​corresponding to that historical submission to form a sub-matching result. Then, the terminal combines the sub-matching results of each historical submission to form the final matching result.

[0112] In the above content review method, for each historical submission of the target content, a second identical tag is determined in the first reader tag set and the corresponding historical reader tag subset. The ratio of the number of second identical tags to the number of reader tags in the first reader tag set is calculated to obtain the third matching value corresponding to the historical submission. Then, for each second identical tag, the third proportion of the second identical tag in the corresponding historical reader tag subset and the fourth proportion of the second identical tag in the first reader tag set are determined, and the ratio of the third proportion to the fourth proportion is calculated to obtain the fourth matching value corresponding to the historical submission. The third matching value and the fourth matching value corresponding to each historical submission are combined to form the matching result. In this way, when the level of the area to be reviewed is at the primary level, not only is the similarity between the first reader tag set and the historical reader tag set comprehensively evaluated based on the number and proportion of identical tags by calculating the third and fourth matching values ​​corresponding to each historical submission, which is more in line with the actual situation of content review, but the similarity between the first reader tag set and the tags in the historical reader tag subsets corresponding to each historical submission is also evaluated separately. This is more in line with the actual situation that the submitting user of the target content may publish content in multiple technical fields, which can further improve the comprehensiveness and accuracy of content review.

[0113] In one embodiment, as shown in Figure 5, the process of determining the correspondence between grade and delivery area includes the following steps:

[0114] Step 501: For each delivery dimension included in the delivery area, sort the values ​​of each dimension under the delivery dimension according to the preset level index to obtain the sequence number of each dimension value under the delivery dimension.

[0115] In this embodiment, the targeting dimension serves as the basis for dividing the targeting area. Targeting dimensions include, but are not limited to, targeting location and targeting range. Dimension values ​​are specific values ​​for each targeting dimension. For example, if the targeting dimension is the targeting range, the values ​​for each dimension are County A, City B, and Province C, respectively. The ranking metric is an indicator used to classify rankings and represent the importance of each dimension value. The ranking metric can be the number of monthly active users (MAU).

[0116] For each delivery dimension included in the delivery area, the terminal sorts the values ​​of each dimension under that delivery dimension according to the preset level indicators, and obtains the sequence number of each dimension value under that delivery dimension.

[0117] In one example, the terminal presets a time limit for each level. For each delivery dimension included in the delivery area, the terminal collects the level indicator values ​​corresponding to each dimension value under that delivery dimension at predetermined time intervals. Then, the terminal sorts the dimension values ​​under that delivery dimension according to the preset level indicators, obtaining the sequence number of each dimension value under that delivery dimension. The time limit for level determination is the interval between each two level determinations. For example, the time limit for level determination can be one month. The time limit for level determination can be a fixed value or can be adjusted at any time according to actual conditions. For example, the time limit for level determination can be one month.

[0118] Step 502: For each delivery area, determine the level corresponding to the delivery area based on the ordinal number of each dimension value included in the delivery area under each delivery dimension and the weight corresponding to each delivery dimension.

[0119] In this embodiment, for each delivery region, the terminal, based on the ordinal numbers of the dimensions included in the delivery region under each delivery dimension, queries the delivery dimension level corresponding to each dimension value included in the delivery region in a preset mapping relationship between delivery dimension ordinal numbers and delivery dimension levels. Then, the terminal calculates the delivery value corresponding to the delivery region based on the delivery dimension level corresponding to each dimension value and the weight corresponding to each delivery dimension. Finally, the terminal queries the level corresponding to the delivery region based on the delivery value corresponding to the delivery region in a preset mapping relationship between delivery value and delivery level. Here, the weight corresponding to the delivery dimension is used to represent the importance of that delivery dimension. The delivery value is used to represent the importance or review difficulty of the delivery region.

[0120] In one embodiment, the ranking metric is monthly active users (MAU), and the targeting dimensions are targeting location and targeting area. Both targeting location and targeting area have a weight of 50%, and the ranking is determined over a one-month period. Every month, the terminal's data platform ranks targeting locations based on MAU. Higher rankings result in higher rankings, with a total of 5 levels. Higher rankings indicate greater importance for the targeting location, with each level representing a 20% increase. Simultaneously, every month, the terminal's data platform also ranks targeting areas based on MAU. Higher rankings result in higher rankings, with a total of 5 levels. Higher rankings indicate greater importance for the targeting location, with each level representing a 20% increase. For each targeting area, the terminal determines its ranking based on the ordinal number of each dimension value within that targeting area and the corresponding weight of each dimension. For example, targeting location E might rank in the top 10% in MAU, resulting in a ranking of 5, while targeting area F might rank in the top 50% in MAU, resulting in a ranking of 3. Then, the terminal calculates the weighted delivery value corresponding to the delivery area, which can be expressed as: 5*50% + 3*50% = 4. Then, based on the delivery value corresponding to the delivery area, the terminal queries the mapping relationship between delivery value and delivery level to find that the corresponding level for the delivery area is 4.

[0121] In the aforementioned content review method, for each delivery dimension included in the delivery area, the values ​​of each dimension under that delivery dimension are sorted according to preset level indicators to obtain the ordinal number of each dimension value under that delivery dimension. Then, for each delivery area, the level corresponding to that delivery area is determined based on the ordinal number of each dimension value under each delivery dimension and the corresponding weight of each delivery dimension. In this way, at fixed intervals or when there are significant changes in the level indicators, the level corresponding to each delivery area can be determined based on the sorting of each dimension value under each delivery dimension obtained according to the level indicators and the weight of each delivery dimension. This not only accurately determines the level of each delivery area but also allows for timely and accurate adjustment of the determined level. It enables timely and accurate determination of the difficulty of content review in different delivery areas, which is more in line with the actual situation and can further improve the comprehensiveness and accuracy of content review.

[0122] In one embodiment, as shown in Figure 6, the content moderation method further includes the following steps:

[0123] Step 601: For each historical submission, determine whether the third matching value corresponding to the historical submission is greater than or equal to the preset tag quantity threshold, and determine whether the fourth matching value corresponding to the historical submission is greater than or equal to the preset tag proportion threshold.

[0124] In this embodiment, the tag quantity threshold (also referred to as the first tag quantity threshold for ease of distinction) is used to measure whether the first reader tag set is substantially similar in quantity to the tags in the historical reader tag subset corresponding to a historical submission. The tag ratio threshold (also referred to as the first tag ratio threshold for ease of distinction) is used to measure whether the first reader tag set is substantially similar in ratio to the tags in the historical reader tag subset corresponding to a historical submission. For each historical submission, the terminal determines whether the third matching value corresponding to the historical submission is greater than or equal to the preset first tag quantity threshold. Simultaneously, the terminal determines whether the fourth matching value corresponding to the historical submission is greater than or equal to the preset first tag ratio threshold.

[0125] Step 602: If there is a third matching value corresponding to a historical submission that is greater than or equal to a preset tag quantity threshold, and the fourth matching value corresponding to the historical submission is greater than or equal to a preset tag proportion threshold, then the matching result is determined to meet the preset relevant conditions.

[0126] In this embodiment of the application, if there is a third matching value corresponding to a historical submission that is greater than or equal to a preset threshold for the number of first tags, and a fourth matching value corresponding to the historical submission that is greater than or equal to a preset threshold for the proportion of first tags, then the terminal determines that the matching result meets the preset relevant conditions.

[0127] In the aforementioned content review method, for each historical submission, it is determined whether the third matching value corresponding to that historical submission is greater than or equal to a preset tag quantity threshold, and whether the fourth matching value corresponding to that historical submission is greater than or equal to a preset tag proportion threshold. If both the third matching value and the fourth matching value of a historical submission are greater than or equal to the preset tag quantity threshold and the preset tag proportion threshold, then the matching result is determined to meet the preset conditions. Thus, when the level of the region to be reviewed is at the primary level, by comparing the third matching value of each historical submission with the tag quantity threshold and the fourth matching value with the tag proportion threshold, if any historical submission exhibits a similarity in both the number and proportion of identical tags to the tags in the subset of historical reader tags corresponding to the historical submission, then the target content is determined to meet the primary level review requirements. This approach better reflects the reality that the submitter of the target content may publish content in multiple technical fields, further improving the comprehensiveness and accuracy of content review.

[0128] In one embodiment, when the content review area is not at the primary level, the content review method further includes the following steps: determining whether a first matching value is greater than or equal to a preset second tag quantity threshold, and determining whether the second matching value is greater than or equal to a preset second tag proportion threshold; if the first matching value is greater than or equal to the preset second tag quantity threshold and the second matching value is greater than or equal to the preset second tag proportion threshold, then the matching result is determined to meet preset relevant conditions. The second tag quantity threshold is used to measure whether the number of tags in the first reader tag set and the second reader tag subset are substantially similar. The second tag proportion threshold is used to measure whether the proportion of tags in the first reader tag set and the second reader tag subset is substantially similar.

[0129] In one embodiment, before obtaining the first set of reader tags for pre-deployment of target content in the pending review and distribution area, the method further includes the following steps: if the level of the pending review and distribution area is the primary level, obtain the historical submission level of the target content; if there is a historical submission level greater than or equal to the primary level, then pre-deploy the target content to the pending review and distribution area. Here, the historical submission level refers to the level at which the submitting user of the target content previously successfully submitted to various distribution areas. Thus, before pre-deployment to the distribution area corresponding to the primary level, it is first determined whether the historical submission level of the target content is greater than or equal to the primary level, and then the target content with a historical submission level greater than or equal to the primary level is pre-deployed to the distribution area corresponding to the primary level. This adds an extra layer of protection before the start of all pre-deployment for content review, further improving the accuracy of content review.

[0130] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0131] Based on the same inventive concept, this application also provides a content moderation apparatus for implementing the content moderation method described above. The solution provided by this apparatus is similar to the implementation scheme described in the above method; therefore, the specific limitations in one or more content moderation apparatus embodiments provided below can be found in the limitations of the content moderation method described above, and will not be repeated here.

[0132] In one embodiment, as shown in FIG7, a content review device 700 is provided, including: an acquisition module 710, a matching module 720, a pre-deployment module 730, and an update module 740, wherein:

[0133] Module 710 is used to obtain the first set of reader tags obtained by pre-deploying the target content in the area to be reviewed and approved.

[0134] The matching module 720 is used to match the first reader tag set with the reference reader tag set corresponding to the target content to obtain a matching result; the reference reader tag set is determined according to the level of the area to be reviewed and distributed.

[0135] The pre-delivery module 730 is used to determine that the target content meets the review conditions of the delivery area to be reviewed when the matching result meets the preset relevant conditions, and to pre-deliver the target content to the delivery area corresponding to the next level of the delivery area to be reviewed;

[0136] The update module 740 is used to take the delivery area corresponding to the next level as the delivery area to be reviewed, and return to the step of obtaining the first reader tag set obtained by pre-delivering the target content in the delivery area to be reviewed, until the level of the delivery area to be reviewed reaches the target level.

[0137] Optionally, the matching module 720 is specifically used for:

[0138] If the level of the area to be reviewed is the primary level, the first reader tag set is matched with the historical reader tag set corresponding to the target content to obtain the matching result;

[0139] If the level of the area to be reviewed and distributed is not the primary level, the first reader tag set is matched with the second reader tag set obtained by pre-distributing the target content in the distribution area corresponding to the previous level of the area to be reviewed and distributed, and a matching result is obtained.

[0140] Optionally, the matching module 720 is specifically used for:

[0141] In the second reader tag set obtained by pre-deploying the first reader tag set and the target content in the delivery area corresponding to the level above the delivery area to be reviewed, a first identical tag is determined.

[0142] Calculate the ratio of the number of the first identical tags to the number of reader tags in the first reader tag set to obtain the first matching value;

[0143] For each first identical tag, determine the first proportion of the first identical tag in the second reader tag set and the second proportion of the first identical tag in the first reader tag set, and calculate the ratio of the first proportion to the second proportion to obtain the second matching value;

[0144] The first matching value and the second matching value are used to form a matching result.

[0145] Optionally, the matching module 720 is specifically used for:

[0146] For each historical submission of the target content, a second identical tag is determined from the first reader tag set and the historical reader tag subset corresponding to the historical submission content;

[0147] Calculate the ratio of the number of the second identical tags to the number of reader tags in the first reader tag set to obtain the third matching value corresponding to the historical submission content;

[0148] For each second identical tag, determine the third proportion of the second identical tag in the historical reader tag subset corresponding to the historical submission content and the fourth proportion of the second identical tag in the first reader tag set, and calculate the ratio of the third proportion to the fourth proportion to obtain the fourth matching value corresponding to the historical submission content.

[0149] The third and fourth matching values ​​corresponding to each of the aforementioned historical submissions are used to form the matching result.

[0150] Optionally, the device 700 further includes:

[0151] The sorting module is used to sort the values ​​of each dimension under each delivery dimension according to a preset level index for each delivery dimension included in the delivery area, so as to obtain the sequence number of each dimension value under the delivery dimension.

[0152] The first determining module is used to determine the level corresponding to each delivery area based on the ordinal number of each dimension value included in the delivery area under each delivery dimension and the weight corresponding to each delivery dimension.

[0153] Optionally, the device 700 further includes:

[0154] The judgment module is used to determine, for each historical submission, whether the third matching value corresponding to the historical submission is greater than or equal to a preset tag quantity threshold, and whether the fourth matching value corresponding to the historical submission is greater than or equal to a preset tag proportion threshold.

[0155] The second determining module is used to determine that the matching result satisfies preset relevant conditions if there is a third matching value corresponding to a historical submission that is greater than or equal to a preset tag quantity threshold, and a fourth matching value corresponding to the historical submission that is greater than or equal to a preset tag proportion threshold.

[0156] Each module in the aforementioned content moderation device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in the processor of a computer device in hardware form or independent of it, or stored in the memory of the computer device in software form, so that the processor can call and execute the corresponding operations of each module.

[0157] In one embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram is shown in Figure 8. The computer device includes a processor, memory, input / output interface, communication interface, display unit, and input device. The processor, memory, and input / output interface are connected via a system bus, and the communication interface, display unit, and input device are also connected to the system bus via the input / output interface. The processor of the computer device provides computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The input / output interface of the computer device is used for exchanging information between the processor and external devices. The communication interface of the computer device is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, NFC (Near Field Communication), or other technologies. When the computer program is executed by the processor, it implements a content moderation method. The display unit of the computer device is used to form a visually visible image and may be a display screen, a projection device, or a virtual reality imaging device. The display screen can be an LCD screen or an e-ink screen. The input device of the computer device can be a touch layer covering the display screen, or buttons, trackballs, or touchpads set on the casing of the computer device, or external keyboards, touchpads, or mice, etc.

[0158] Those skilled in the art will understand that the structure shown in Figure 8 is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or may combine certain components, or may have different component arrangements.

[0159] In one embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments.

[0160] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps in the above method embodiments.

[0161] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above method embodiments.

[0162] 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, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data shall comply with the relevant laws, regulations and standards of the relevant countries and regions.

[0163] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0164] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0165] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A content moderation method, characterized in that, The method includes: obtaining a first reader tag set obtained by pre-deploying target content in a pending review area; matching the first reader tag set with a reference reader tag set corresponding to the target content to obtain a matching result; the reference reader tag set is determined according to the level of the pending review area; if the matching result meets preset relevant conditions, determining that the target content meets the review conditions of the pending review area, and pre-deploying the target content to the next level of the pending review area; using the next level of the pending review area as the pending review area, returning to the step of obtaining the first reader tag set obtained by pre-deploying target content in the pending review area, until the level of the pending review area reaches the target level; when the level of the pending review area is... In the case of the initial level, the step of matching the first reader tag set with the reference reader tag set corresponding to the target content to obtain a matching result includes: determining a first identical tag in the second reader tag set obtained by pre-deploying the first reader tag set and the target content in the deployment area corresponding to the previous level of the deployment area to be reviewed; calculating the ratio of the number of the first identical tags to the number of reader tags in the first reader tag set to obtain a first matching value; for each first identical tag, determining a first proportion of the first identical tags in the second reader tag set and a second proportion of the first identical tags in the first reader tag set, and calculating the ratio of the first proportion to the second proportion to obtain a second matching value; and combining the first matching value and the second matching value to form a matching result.

2. The method according to claim 1, characterized in that, When the level of the area to be reviewed is the primary level, the step of matching the first reader tag set with the reference reader tag set corresponding to the target content to obtain the matching result includes: matching the first reader tag set with the historical reader tag set corresponding to the target content to obtain the matching result.

3. The method according to claim 2, characterized in that, The step of matching the first reader tag set with the historical reader tag set corresponding to the target content to obtain a matching result includes: for each historical submission of the target content, determining a second identical tag in the first reader tag set and the historical reader tag subset corresponding to the historical submission; calculating the ratio of the number of the second identical tags to the number of reader tags in the first reader tag set to obtain a third matching value corresponding to the historical submission; for each second identical tag, determining a third proportion of the second identical tags in the historical reader tag subset corresponding to the historical submission and a fourth proportion of the second identical tags in the first reader tag set, and calculating the ratio of the third proportion to the fourth proportion to obtain a fourth matching value corresponding to the historical submission; and combining the third matching value and the fourth matching value corresponding to each historical submission to form a matching result.

4. The method according to claim 1, characterized in that, The process of determining the correspondence between the level and the delivery area includes: for each delivery dimension included in the delivery area, sorting the values ​​of each dimension under the delivery dimension according to the preset level index to obtain the sequence number of each dimension value under the delivery dimension; for each delivery area, determining the level corresponding to the delivery area according to the sequence number of each dimension value included in the delivery area under each delivery dimension and the weight corresponding to each delivery dimension.

5. The method according to claim 3, characterized in that, The method further includes: for each historical submission, determining whether the third matching value corresponding to the historical submission is greater than or equal to a preset tag quantity threshold, and determining whether the fourth matching value corresponding to the historical submission is greater than or equal to a preset tag proportion threshold; if there exists a historical submission with a third matching value greater than or equal to the preset tag quantity threshold and a historical submission with a fourth matching value greater than or equal to the preset tag proportion threshold, then the matching result is determined to meet preset relevant conditions.

6. A content moderation device, characterized in that, The device includes: an acquisition module, configured to acquire a first reader tag set obtained by pre-deploying target content in a pending review and distribution area; a matching module, configured to match the first reader tag set with a reference reader tag set corresponding to the target content to obtain a matching result; the reference reader tag set is determined according to the level of the pending review and distribution area; a pre-deployment module, configured to determine that the target content meets the review conditions of the pending review and distribution area when the matching result meets preset relevant conditions, and pre-deploy the target content to the distribution area corresponding to the next level of the pending review and distribution area; and an update module, configured to set the distribution area corresponding to the next level as the pending review and distribution area, return to the step of acquiring the first reader tag set obtained by pre-deploying target content in the pending review and distribution area, until the pending review and distribution area is updated. If the review and approval of the distribution area reaches the target level, and the distribution area to be reviewed is at the primary level, the matching module is specifically used to: determine a first identical tag in the second reader tag set obtained by pre-distributing the first reader tag set and the distribution area corresponding to the next higher level of the distribution area to be reviewed; calculate the ratio of the number of the first identical tags to the number of reader tags in the first reader tag set to obtain a first matching value; for each first identical tag, determine a first proportion of the first identical tags in the second reader tag set and a second proportion of the first identical tags in the first reader tag set, and calculate the ratio of the first proportion to the second proportion to obtain a second matching value; and combine the first matching value and the second matching value to form a matching result.

7. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 5.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 5.

9. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 5.

Images