Content processing method and device, computer equipment and computer readable storage medium

Abstract

The embodiment of the invention discloses a content processing method and device, computer equipment and a computer readable storage medium. The method comprises the following steps: acquiring a first submitted content amount and a first released content amount of a content pool at target time, and a second submitted content amount and a second released content amount of the content pool at historical time corresponding to the target time; calculating a ratio between the first submission content and the second submission content to obtain a first adjustment proportion; calculating the ratio of the first released inner content to the second released inner content to obtain a second adjusting proportion; according to the first adjustment proportion and the second adjustment proportion, adjusting the content quantity difference between the target submitted content quantity and the target content total quantity of the content pool to obtain the adjusted content quantity difference; and determining a screening condition based on the adjusted content quantity difference and the content sorted according to the auditing index value in the content pool, and screening the to-be-processed submitted content from the to-be-processed content according to the screening condition, so that the quantity of the submitted content can be kept stable.

Description

Technical Field

[0001] The present application relates to the field of communication technology, and in particular to a content processing method, apparatus, computer equipment and computer-readable storage medium. Background Art

[0002] With the development of Internet technology and the rise of self-media, the content on social platforms has become more colorful. Some of the content on social platforms requires manual review. Since the amount of content posted on the platform every day or every week is not fixed, the amount of content that needs to be reviewed is unstable, and it is difficult to reasonably arrange reviewers to conduct the review, resulting in waste of manpower or insufficient manpower. Summary of the invention

[0003] The embodiments of the present application provide a content processing method, apparatus, computer device, and computer-readable storage medium, which can keep the amount of content submitted for review stable, so as to better arrange personnel for review and avoid waste of manpower or insufficient manpower.

[0004] A content processing method provided in an embodiment of the present application includes:

[0005] Acquire the first submitted content volume and the first released content volume of the content pool at the target time, and the second submitted content volume and the second released content volume at the historical time corresponding to the target time;

[0006] Calculating a ratio between the volume of the first content submitted for review and the volume of the second content submitted for review to obtain a first adjustment ratio;

[0007] Calculating a ratio between the first released content volume and the second released content volume to obtain a second adjustment ratio;

[0008] According to the first adjustment ratio and the second adjustment ratio, adjusting the content quantity difference between the target content quantity submitted for review and the target content quantity of the content pool to obtain an adjusted content quantity difference;

[0009] Determining a screening condition based on the adjusted content quantity difference and the content in the content pool sorted according to the audit index value;

[0010] The content to be processed and submitted for review is filtered out from the content to be processed according to the filtering conditions.

[0011] Accordingly, an embodiment of the present application further provides a content processing device, including:

[0012] An acquisition unit, configured to acquire a first submitted content amount and a first released content amount of a content pool at a target time, and a second submitted content amount and a second released content amount at a historical time corresponding to the target time;

[0013] a first calculation unit, configured to calculate a ratio between the amount of the first content submitted for review and the amount of the second content submitted for review, to obtain a first adjustment ratio;

[0014] a second calculation unit, configured to calculate a ratio between the first released content amount and the second released content amount to obtain a second adjustment ratio;

[0015] An adjusting unit, configured to adjust a content quantity difference between a target content quantity for review and a target content quantity of the content pool according to the first adjustment ratio and the second adjustment ratio, to obtain an adjusted content quantity difference;

[0016] A condition determination unit, configured to determine a screening condition based on the adjusted content quantity difference and the content in the content pool sorted according to the audit index value;

[0017] The screening unit is used to screen out the content to be processed from the content to be processed according to the screening conditions.

[0018] In one embodiment, the adjusted content quantity difference includes a first adjusted content quantity difference and a second adjusted content quantity difference, and the adjusting unit includes:

[0019] A quantity calculation subunit, used to calculate the content quantity difference between the target content quantity submitted for review and the total content quantity submitted for review;

[0020] The first adjustment subunit is configured to adjust the content quantity difference based on the first adjustment ratio to obtain the adjusted content quantity difference if the first adjustment ratio satisfies a first ratio condition and the second adjustment ratio does not satisfy a second ratio condition.

[0021] The second adjustment subunit is configured to adjust the content quantity difference based on the second adjustment ratio to obtain the adjusted content quantity difference if the first adjustment ratio does not satisfy the first ratio condition and the second adjustment ratio satisfies the second ratio condition.

[0022] In one embodiment, the adjustment unit further includes:

[0023] a third adjustment subunit, configured to perform a first adjustment process on the content quantity difference based on the first adjustment ratio to obtain a candidate adjusted content quantity difference if the first adjustment ratio satisfies the first ratio condition and the second adjustment ratio satisfies the second ratio condition;

[0024] The fourth adjustment subunit is configured to perform a second adjustment process on the candidate adjusted content quantity difference based on the second adjustment ratio to obtain the adjusted content quantity difference.

[0025] In one embodiment, the content pool includes a submitted content pool, a buffer content pool, and a released content pool; the target content volume is the total volume of content in the submitted content pool and the buffer content pool; each content in the content pool corresponds to a risk score; the audit index value includes the risk score; and the screening unit includes:

[0026] A first content determination subunit is configured to determine target content based on the content quantity difference and the content in the released content pool sorted based on the risk score;

[0027] A first condition generating subunit is used to generate a first screening condition according to the risk score corresponding to the target content if the total amount of the target content is less than the target amount for review;

[0028] The second condition generating subunit is used to generate a second screening condition according to the risk score corresponding to the target content if the total amount of the target content is greater than the target review amount.

[0029] In one embodiment, the screening unit further comprises:

[0030] a content screening subunit, configured to screen the content to be entered into the released content pool according to the first screening condition, and determine the first content to be entered into the released content pool and the second content to be entered into the buffer content pool;

[0031] The filtering subunit is configured to filter the content in the buffer content pool according to the second filtering condition.

[0032] In one embodiment, the content pool includes multimedia content, and the content processing device further includes:

[0033] A content acquisition unit, configured to acquire multimedia content, wherein the multimedia content includes at least two sub-contents in different modes;

[0034] An extraction unit, configured to extract features of the sub-contents in the at least two modes respectively, to obtain feature information of the sub-contents corresponding to the different modes;

[0035] A fusion unit, used for fusing sub-content feature information corresponding to different modalities to obtain fusion feature information of the multimedia content;

[0036] An identification unit is used to perform risk identification on the multimedia content according to the fused feature information to obtain a risk score of the multimedia content.

[0037] In one embodiment, the at least two modalities include a text modality, the text modality corresponds to at least one text sub-content, and the extraction unit includes:

[0038] A word set acquisition subunit is used to acquire a sensitive word set;

[0039] A detection subunit, configured to perform sensitive word detection on each text sub-content in the text modality according to the sensitive word set, and determine the position of the sensitive word in each text sub-content;

[0040] A marking subunit, used to mark the sensitive word position of each text sub-content according to the position, and obtain the sensitive word position information corresponding to each text sub-content;

[0041] The feature extraction subunit is used to extract text features from each text sub-content based on the sensitive word position information to obtain sub-content feature information of the text modality.

[0042] Correspondingly, an embodiment of the present application also provides a computer device, including a memory and a processor; the memory stores a computer program, and the processor is used to run the computer program in the memory to execute any content processing method provided in the embodiment of the present application.

[0043] Correspondingly, an embodiment of the present application further provides a computer-readable storage medium, which is used to store a computer program, and the computer program is loaded by a processor to execute any content processing method provided in the embodiment of the present application.

[0044] The embodiment of the present application obtains the first amount of content submitted for review and the first amount of content released from the content pool at a target time, and the second amount of content submitted for review and the second amount of content released at a historical time corresponding to the target time; calculates the ratio between the first amount of content submitted for review and the second amount of content submitted for review to obtain a first adjustment ratio; calculates the ratio between the first amount of content released and the second amount of content released to obtain a second adjustment ratio; adjusts the content quantity difference between the target amount of content submitted for review and the target total amount of content in the content pool according to the first adjustment ratio and the second adjustment ratio to obtain an adjusted content quantity difference; determines a screening condition based on the adjusted content quantity difference and the contents in the content pool sorted according to the review index value, and screens out the content submitted for review to be processed from the contents to be processed according to the screening condition.

[0045] In the embodiment of the present application, screening conditions are generated based on the difference between the number of target contents in the content pool and the target amount of content to be submitted for review, as well as based on the ratio of the number of content at the target time and the historical time. New content can be screened based on the screening conditions so that the number of target contents in the content pool is dynamically kept close to the target amount of content to be submitted for review, that is, the number of content to be submitted for review is kept stable, so that personnel can be better arranged for review and the problem of waste of manpower or insufficient manpower can be avoided. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0047] Figure 1 is a flow chart of a content processing method provided by an embodiment of the present application;

[0048] Figure 2 It is a schematic diagram of the model structure provided in the embodiment of the present application;

[0049] Figure 3 is a schematic diagram of the network structure of the fusion network provided in the embodiment of the present application;

[0050] Figure 4 is a schematic diagram of a content risk identification device provided in an embodiment of the present application;

[0051] Figure 5 It is a schematic diagram of the structure of a computer device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0052] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of this application.

[0053] The embodiments of the present application provide a content processing method, apparatus, computer device and computer readable storage medium. The content processing apparatus can be integrated in a computer device, which can be a server or a terminal.

[0054] The terminal may include a mobile phone, a wearable smart device, a tablet computer, a laptop computer, a personal computer (PC), and a vehicle-mounted computer.

[0055] Among them, the server can be an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN, as well as big data and artificial intelligence platforms.

[0056] It should be noted that the description order of the following embodiments is not intended to limit the preferred order of the embodiments.

[0057] This embodiment will be described from the perspective of a content processing device. The content processing device may be integrated into a computer device, which may be a server or a terminal.

[0058] The present application provides a content processing method, such as Figure 1 As shown, the specific process of the content processing method can be as follows:

[0059] 101. Obtain a first amount of content submitted for review and a first amount of content released in a content pool at a target time, and a second amount of content submitted for review and a second amount of content released at a historical time corresponding to the target time.

[0060] Among them, the target time may include a time period corresponding to a preset time length in the past, such as the past 2 hours, the past 1 hour, the past 30 minutes, or the Nth hour of today, etc. The historical time corresponding to the target time may be the specific moment of the target time on the current date, the time corresponding to the past date, for example, the past 2 hours was 9:00-11:00 today, then the historical time may be 9:00-11:00 yesterday, or 9:00-11:00 the day before yesterday, etc., 9:00-11:00 of other dates before today.

[0061] Among them, the content pool may include a submitted content pool, a buffer content pool and a released content pool. Among them, the content in the submitted content pool will be submitted for review first, and the content in the buffer content pool will determine whether to submit for review based on whether the amount of submitted content meets the requirements. If the amount of content in the submitted content pool is insufficient, it will be supplemented by the content in the buffer content pool, and the content in the released content pool can be pushed to users.

[0062] The amount of content submitted for review may be the amount of content in the content pool submitted for review, and the amount of content released may be the amount of content in the content pool released.

[0063] The first amount of content submitted for review may be the number of contents in the content pool at the target time, the second amount of content submitted for review may be the number of contents in the content pool at a historical time, the first amount of content released may be the number of contents in the released content pool at the target time, and the second amount of content released may be the number of contents in the released content pool at a historical time.

[0064] The second amount of content submitted for review may be an average amount of content submitted for review in at least one historical period, and the second amount of content released may be an average amount of content released in at least one historical period.

[0065] It can be understood that the target time is relative to the current time. If the processing time unit is one hour, and the current time is between 8:00-9:00, then the target time can be the past hour, that is, 7:00-8:00; if the current time is 9:00-10:00, then the target time can be the past hour, that is, 8:00-9:00. Therefore, the embodiment of the present application can dynamically adjust the amount of content in the content pool so that the target content amount in the content pool is dynamically kept close to the target content amount for review.

[0066] 102. Calculate the ratio between the amount of the first content submitted for review and the amount of the second content submitted for review to obtain a first adjustment ratio.

[0067] The ratio between the first amount of content submitted for review and the second amount of content submitted for review is calculated to obtain a first adjustment ratio. By comparing the amount of content submitted for review at the target time and the amount of content submitted for review at the historical time, it can be determined whether the amount of content submitted for review at the target time is too much or too little, so as to adjust the amount submitted for review.

[0068] Exemplarily, the first adjustment ratio Yraito=the number of contents in the content pool submitted for review at the Nth hour yesterday / the number of contents in the content pool submitted for review at the Nth hour today.

[0069] 103. Calculate the ratio between the first released content volume and the second released content volume to obtain a second adjustment ratio.

[0070] The ratio between the first amount of released content and the second amount of released content is calculated to obtain the second adjustment ratio. By comparing the amount of content submitted for review at the target time and the historical time, it can be determined whether the amount of released content at the target time is too much or too little, so as to adjust the amount submitted for review.

[0071] Exemplarily, the first adjustment ratio Fraito=the number of contents released from the content pool at the Nth hour yesterday / the number of contents released from the content pool at the Nth hour today.

[0072] 104. According to the first adjustment ratio and the second adjustment ratio, the content quantity difference between the target content quantity for review and the target total content quantity of the content pool is adjusted to obtain an adjusted content quantity difference.

[0073] The target amount of content to be submitted for review may be the amount of content to be submitted for review within a specified time period set according to actual review needs or review manpower, such as the amount of content to be submitted for review in one day, or the amount of content to be submitted for review in one week, etc.

[0074] Among them, the content pool can include a submitted content pool, a buffer content pool and a released content pool. Among them, the content in the submitted content pool will be submitted for review first, and the content in the buffer content pool will determine whether to submit for review based on whether the amount of submitted content meets the requirements. The content in the released content pool can be pushed to users.

[0075] The content pool can include content acquired during a preset time period in the past. For example, if the target amount of content to be submitted for review is one day's amount of content to be submitted for review, the content pool can include content acquired during the past 24 hours, and the content pool can be updated every hour. When content enters the content pool in the last hour, the oldest hour of content in the content pool will be removed.

[0076] The target total amount of content may be the sum of the content included in the submitted content pool and the content included in the buffer content pool.

[0077] Since the content in the review content pool and the buffer content pool can all be submitted for review, by comparing the target review content volume with the target total content volume, it can be determined whether the current review content volume is appropriate.

[0078] Calculate the difference between the target amount of content to be submitted for review and the target total amount of content to obtain the content quantity difference. If the target amount of content to be submitted for review is greater than the target total amount of content, the content quantity difference indicates that the amount of content to be submitted for review needs to be reduced. If the target amount of content to be submitted for review is less than the target total amount of content, the content quantity difference indicates that the amount of content to be submitted for review needs to be increased.

[0079] Then, the content quantity difference is dynamically adjusted according to the first adjustment ratio and the second adjustment ratio, so that the target total content amount of the submitted content pool and the buffer content pool gradually approaches the target submitted content amount.

[0080] The first adjustment ratio indicates the difference between the amount of content submitted for review at the target time and the historical event, so the content quantity difference can be adjusted when the first adjustment ratio and the second adjustment ratio meet the conditions. That is, in one embodiment, the adjusted content quantity difference includes the first adjusted content quantity difference and the second adjusted content quantity difference. The step of "adjusting the content quantity difference according to the content quantity difference between the amount of content submitted for review and the total amount of content in the content pool, the first adjustment ratio and the second adjustment ratio to obtain the adjusted content quantity difference" may include:

[0081] Calculate the difference between the amount of content submitted for review and the total amount of content submitted for review;

[0082] If the first adjustment ratio satisfies the first ratio condition and the second adjustment ratio does not satisfy the second ratio condition, the content quantity difference is adjusted based on the first adjustment ratio to obtain an adjusted content quantity difference.

[0083] If the first adjustment ratio does not satisfy the first ratio condition, and the second adjustment ratio satisfies the second ratio condition, the content quantity difference is adjusted based on the second adjustment ratio to obtain an adjusted content quantity difference.

[0084] The first ratio condition may include that the first adjustment ratio is greater than a preset ratio threshold, and the second ratio condition may include that the second adjustment ratio is less than a second preset ratio threshold.

[0085] For example, if the first adjustment ratio is greater than a preset ratio threshold, and the second adjustment ratio is greater than a second preset ratio threshold, the content quantity difference is adjusted based on the first adjustment ratio to obtain the adjusted content quantity difference. If the first adjustment ratio is less than the preset ratio threshold and the second adjustment ratio is less than the second preset ratio threshold, the content quantity difference is adjusted based on the second adjustment ratio to obtain the adjusted content quantity difference.

[0086] For example, if the target submission amount is 60,000, the total target content in the submission content pool and the buffer content pool is 40,000, and the content quantity difference is 20,000, assuming that the first ratio threshold and the second ratio threshold are both 0.9, if Yrati>0.9 and Frati>0.9, then 20,000 / Yratio is used as the adjusted content quantity difference, if Fraito<0.9 and Yraito<0.9, then 20,000*Fraito is used as the adjusted content quantity difference.

[0087] If both the first adjustment ratio and the second adjustment ratio satisfy the adjustment, the content quantity difference can be adjusted in combination with the first adjustment ratio and the second adjustment ratio. That is, in one embodiment, the step of "calculating the content quantity difference between the content quantity submitted for review and the total content submitted for review" and the content processing method provided in the embodiment of the present application can also include:

[0088] If the first adjustment ratio satisfies the first ratio condition, and the second adjustment ratio satisfies the second ratio condition, performing a first adjustment on the content quantity difference based on the first adjustment ratio to obtain a candidate adjusted content quantity difference;

[0089] A second adjustment is performed on the candidate adjusted content quantity difference based on the second adjustment ratio to obtain the adjusted content quantity difference.

[0090] For example, if the first adjustment ratio is greater than a preset ratio threshold, and the second adjustment ratio is less than a second preset ratio threshold, a first adjustment is performed on the content quantity difference based on the first adjustment ratio to obtain a candidate adjusted content quantity difference; and a second adjustment is performed on the candidate adjusted content quantity difference based on the second adjustment ratio to obtain an adjusted content quantity difference.

[0091] For example, if the target review volume is 60,000, the total target content in the review content pool and the buffer content pool is 40,000, and the content quantity difference is 20,000, assuming that the first ratio threshold and the second ratio threshold are both 0.9, if Yrati>0.9 and Fraito<0.9, then 20,000 / Yratio*Fraito is used as the adjusted content quantity difference.

[0092] 105. Determine the screening condition based on the difference in the number of adjusted contents and the contents in the content pool sorted according to the audit index value.

[0093] 106. Filter out the submitted content to be processed from the pending content based on the screening conditions.

[0094] For example, if the target review quantity is greater than the target total content, it means that the amount of content to be reviewed needs to be increased. The content in the released content pool can be sorted from large to small according to the risk score of the content. The risk score can indicate the risk level of the content. The higher the risk score, the more risky the content is. The lower the risk score, the lower the risk of the content. The target content corresponding to the adjusted content quantity difference can be determined based on the adjusted content quantity difference and the content sorting in the released content pool, and the screening conditions can be generated based on the risk score of the target content. That is, in one embodiment, the content pool includes a review content pool, a buffer content pool, and a released content pool. The target content quantity is the total amount of content in the review content pool and the buffer content pool. Each content in the content pool has a corresponding risk score. The step of "determining the screening conditions based on the content quantity difference and the content sorted according to the review index value in the content pool" can include:

[0095] Determine the target content based on the difference in content quantity and the content in the released content pool sorted by the audit index value;

[0096] If the total amount of target content is less than the target review amount, the first screening condition is generated according to the risk score corresponding to the target content;

[0097] If the total amount of target content is greater than the target review amount, a second screening condition is generated based on the risk score corresponding to the target content.

[0098] If the target review quantity is greater than the total target content quantity, the target content (target content) corresponding to the adjusted content quantity difference can be determined based on the adjusted content quantity difference and the content ranking in the released content pool, and the screening conditions can be generated based on the risk score of the target content.

[0099] For example, assuming that the difference in the number of adjusted contents is 12,000, the content ranked 12,000 in the released content pool is the target content, and the risk score of the target content is used as a standard to screen the new content (i.e., the content to be processed). For example, content with a risk score greater than that of the target content is moved to the buffer content pool, and content with a risk score less than that of the target content is stored in the released content pool.

[0100] Therefore, the first screening condition generated according to the risk score of the target content is that if the risk score of the content to be processed is greater than the risk score of the target content, the content is moved to the buffer content pool, and if the risk score of the content is less than the risk score of the target content, the content is stored in the released content pool.

[0101] Optionally, a fixed risk score can be set to determine whether the content enters the review content pool or the released content pool. If the content is greater than the fixed risk score, it will enter the review content pool; if the content is less than or equal to the fixed risk score, it will enter the released content pool. The first screening condition obtained is used to determine whether the content less than or equal to the fixed risk score will enter the released content pool. For example, content with a risk score greater than the fixed risk score will enter the review content pool; for content with a risk score less than or equal to the fixed risk score, if the content has a risk score greater than the target content, it will be moved to the buffer content pool; if the content has a risk score less than the target content, it will be stored in the released content pool.

[0102] Therefore, the first screening condition can be that for the target content to be processed that is to enter the released content pool, if the risk score of the target content to be processed is greater than the risk score of the target content, the content is moved to the buffer content pool, and if the risk score of the target content is less than the risk score of the target content, the content is stored in the released content pool.

[0103] If the total amount of content is greater than the target amount for review, it means that the amount of content for review needs to be reduced. The second filter condition can be generated based on the content sorting in the buffer content pool.

[0104] For example, the contents in the buffer content pool may be sorted from small to large according to the risk scores of the contents, and the target content (second target content) corresponding to the adjusted content quantity difference may be determined based on the adjusted content quantity difference and the content sorting in the buffer content pool, and filtering conditions may be generated based on the risk scores of the second target content.

[0105] For example, assuming that the difference in the number of adjusted contents is 12,000, the content ranked 12,000 in the released content pool is the target content, and the risk score of the target content is used as a standard to filter the contents of the buffer content pool. For example, content with a risk score smaller than the target content is moved to the released content pool; content with a risk score greater than the target content is retained in the buffer content pool.

[0106] Therefore, the second screening condition generated according to the risk score of the second target content is that, for the content in the buffer content pool, the content with a risk score smaller than the target content is moved to the released content pool; and the content with a risk score larger than the target content is retained in the buffer content pool.

[0107] Optionally, a fixed risk score can be set to determine whether the content enters the review content pool or the released content pool. If the risk score is greater than the fixed risk score, the content enters the review content pool; if the risk score is less than or equal to the fixed risk score, the content enters the released content pool. The obtained screening condition is used to filter the content in the released buffer pool to reduce the amount of content in the buffer pool.

[0108] The content in the review content pool is submitted to the reviewer for review. If the content in the review content pool is insufficient, it is supplemented by the content in the buffer content pool.

[0109] When adding the content to be processed into the content pool, the content that first entered the pool within a preset time period may also be removed from the content pool.

[0110] For example, assuming that the processing time unit is one hour, the content pool stores the content within the past 24 hours, namely the content within the past 24 hours, the past 23 hours, and so on, within the past hour. The content within the current hour is screened and then enters the content pool, and the content within the past 24 hours of the content pool for review is removed.

[0111] The content stored in the content pool may be multimedia content, and the corresponding risk score may be determined by risk identification. That is, in one embodiment, the content pool includes multimedia content, and the content processing method provided in the embodiment of the present application may further include:

[0112] S201, acquiring multimedia content, where the multimedia content includes at least two sub-contents in different modes;

[0113] S202, extracting features of sub-contents in at least two modes respectively to obtain sub-content feature information corresponding to different modes;

[0114] S203, fusing sub-content feature information corresponding to different modalities to obtain fused feature information of multimedia content;

[0115] S204: Identify the risks of the multimedia content according to the fused feature information to obtain a risk score of the multimedia content.

[0116] Step S201 may be: the multimedia content may be content including multiple media forms, for example, may include sub-contents of at least two different modes among text, image, audio and video.

[0117] For example, multimedia content may be obtained from a database to identify whether the multimedia content is illegal or in violation of regulations, and determine risk information of the multimedia content. For details of the specific implementation process, see the description of the relevant content of steps S202 to S204.

[0118] S202 may specifically be: sub-contents of different modalities may be subjected to feature extraction through different neural networks; sub-contents of image modality may be subjected to image feature extraction through a Resnet network to obtain sub-content feature information corresponding to the image modality; sub-contents of text modality may be subjected to feature extraction through a Bert network to obtain sub-content feature information corresponding to the text modality; and sub-contents of audio modality may be subjected to feature extraction through a Transformer network to obtain sub-content feature information corresponding to the audio modality.

[0119] Feature extraction can also be performed through other neural networks, which is not limited here.

[0120] In one embodiment, the multimedia content includes a sub-content belonging to a text mode, which may be referred to as a text sub-content. Sensitive words in the text sub-content may be marked according to a sensitive word set, so that when extracting features from the text sub-content, more attention may be paid to the sensitive words in the text sub-content, thereby improving the recognition effect of risk information. That is, in the embodiment, at least two modes include a text mode, and the text mode corresponds to at least one text sub-content. Step S202 may include:

[0121] Get the sensitive word set;

[0122] Perform sensitive word detection on each text sub-content in the text mode according to the sensitive word set, and determine the position of the sensitive word in each text sub-content;

[0123] Mark the sensitive word position of each text sub-content according to the position, and obtain the sensitive word position information corresponding to each text sub-content;

[0124] Based on the sensitive word location information, text features are extracted for each text sub-content to obtain sub-content feature information of the text modality.

[0125] The sensitive word set may be a pre-set word set, which includes sensitive words related to content risks.

[0126] The text mode may have at least one text sub-content, for example, it may include a title, a body, and text recognized from a video or an image by optical character recognition (OCR) technology.

[0127] Sensitive words are searched from each text sub-content according to the sensitive word set, the position of the sensitive words in the sensitive word set in each text sub-content is determined, and then the position of the sensitive words is marked to obtain the sensitive word position information corresponding to each text content.

[0128] In one embodiment, each text unit in the text sub-content corresponds to an identifier, which can indicate whether the text unit at that position is a keyword. For example, 0 indicates that the text unit at that position is not a keyword, and 1 indicates that the text unit at that position is a keyword. The sensitive word position information can be obtained based on the identifier corresponding to each text unit in the text content.

[0129] Assume that there is a text sub-content titled "Which is more bitter, bitter melon or American coffee?", in which bitter melon is a sensitive word in the sensitive word set. Then the sensitive word position information corresponding to the text sub-unit can be 11000000, where 1 indicates that the text unit at this position is a sensitive word, and 0 indicates that the sensitive word at this position is a non-sensitive word. The first text unit and the second text unit in the text sub-content are determined to be sensitive words based on the sensitive word position information.

[0130] Each text sub-content and the corresponding sensitive word position information can be input into a neural network for feature extraction to obtain text feature information of each text sub-content, and the text feature information of each text sub-content can be used as sub-content feature information of the text modality; optionally, the text feature information of each text sub-content can also be subjected to feature fusion processing to obtain sub-content feature information of the sub-content of the text modality.

[0131] In one embodiment, the Figure 2 The fusion network in the model structure shown in the figure fuses each text feature information to obtain the sub-content feature information of the text modality. The network structure of the fusion network can be entered into Figure 3 As shown, the feature fusion layer is used to fuse at least two text feature information to obtain initial sub-content features, and then the global pooling layer performs pooling on the initial sub-content features, performs linear transformation through the fully connected layer, maps the input information to a preset numerical range through the ReLU layer, processes it, performs linear transformation again through the fully connected layer, maps the input information to a preset numerical range through the Sigmod layer, and finally scales the size of the text feature information to an appropriate size for calculation through the scaling layer.

[0132] The feature fusion processing is performed on the text feature information of at least two text sub-contents of the text modality, and the feature fusion processing can be performed on at least two text feature information by means of concatenate or add; corresponding weights can be set for different text sub-contents. For example, the OCR text of the title and the cover image of the multimedia content is usually the first text sub-content that users look at, so compared with other multimedia content, it is more urgent to filter the multimedia content containing risky content in the title or cover image. The text sub-content weight can be assigned according to the display position of the text sub-content in the multimedia content. That is, in one embodiment, the step of "based on the sensitive word position information, extracting text features for each text sub-content to obtain sub-content feature information of the text modality" can specifically include:

[0133] Based on the sensitive word location information, text feature extraction is performed on each text sub-content to obtain text feature information of each text sub-content;

[0134] Determining the content weight of each text sub-content according to the display position of each text sub-content in the multimedia content;

[0135] According to the content weight of each text sub-content, the text feature information of each text sub-content is weighted and fused to obtain the sub-content feature information of the text modality.

[0136] For example, a correspondence between display position and content weight can be preset, and the content weight of each text sub-content can be determined according to the display position of each text sub-content in the multimedia content. For example, the content weight of the OCR text of the title and cover image can be set to 1, and the content weight of the text can be set to 0.8, and so on.

[0137] Based on the sensitive word location information, text feature extraction is performed on each text sub-content to obtain text feature information of each text sub-content. Then, according to the content weight of each text sub-content, weighted fusion processing is performed on the text feature information of each text sub-content to obtain sub-content feature information of the text modality.

[0138] According to the sub-content feature information corresponding to different modes of multimedia content, the risk information of the multimedia content can be identified. The risk information may include the risk level of the multimedia content, and may also include the risk type of the multimedia content, such as the violence type and the pornography type, etc. Therefore, the sensitive word set may include sensitive word subsets corresponding to different risk types. Different sensitive word subsets may be marked in different ways. For example, different identifiers may be used to indicate that sensitive words belong to different sensitive word subsets. That is, in one embodiment, the sensitive word set includes sensitive word subsets corresponding to different risk types. The step of "marking the sensitive word position of each text sub-content according to the position to obtain the sensitive word position information corresponding to each text sub-content" may include:

[0139] For each sensitive word subset, mark the sensitive words in each text sub-content that match the sensitive word subset according to the position and the marking method corresponding to the sensitive word subset;

[0140] Based on the tags corresponding to each sensitive word subset in the text sub-content, the sensitive word position information of each text sub-content is obtained.

[0141] For example, sensitive words belonging to different sensitive word subsets in the text sub-content can be marked by different identifiers. Assuming that the sensitive word set contains three sensitive word subsets, the sensitive words in the sensitive word set in the text sub-content can be marked with 1, 2, and 3 respectively.

[0142] For each sensitive word subset, the sensitive words matching the sensitive word subset in each text sub-content are marked according to the position and the marking method corresponding to the sensitive word subset, and the sensitive word position information of each text sub-content is obtained based on the mark corresponding to each sensitive word subset in the text sub-content. Using different methods to mark different sensitive word subsets can make the neural network pay more attention to the sensitive words in the text sub-content and the risk type to which the sensitive words belong, which can improve the accuracy and speed of determining the risk information of multimedia content.

[0143] Exemplarily, a text sub-content is "Which is more bitter, bitter melon or American style", where bitter melon is a sensitive word in sensitive word subset A, and American style is a sensitive word in sensitive word subset B. The sensitive words in sensitive word subset A are marked with 1, and those in sensitive word subset B are marked with 2. Then the sensitive word position information corresponding to the text sub-unit can be 11022000, where 0 indicates that the sensitive word at this position is a non-sensitive word, 1 indicates that the text unit at this position is a sensitive word in sensitive word subset A, and 2 indicates that the text unit at this position is a sensitive word in sensitive word subset B. According to the sensitive word position information, it is determined that the sensitive words of the first text unit and the second text unit in the text sub-content belong to sensitive word set A, and the sensitive words of the fourth text unit and the fifth text unit belong to sensitive word set B.

[0144] The multimedia content may include sub-content of text modality and may also include sub-content of image modality. The sub-content of image modality may be referred to as image sub-content. By performing feature extraction on the image sub-content, image feature information of each image sub-content is used as sub-content feature information corresponding to the image modality. Alternatively, the image feature information of each image sub-content may be fused to obtain sub-content feature information. That is, in one embodiment, at least two modalities include an image modality, and the image modality corresponds to at least one image sub-content. Step S202 may include:

[0145] Extracting image features from at least one image sub-content respectively to obtain image feature information of each image sub-content;

[0146] The image feature information of each image sub-content is fused to obtain the sub-content feature information corresponding to the image modality.

[0147] Among them, an image sub-content can be an image, a complete video, or multiple video frames extracted from a video, for example, it can include a cover image of multimedia content, a cover frame and first frame of a video, etc.

[0148] For example, image features may be extracted from at least one image sub-content under the image modality to obtain image feature information of each image sub-content, and the image feature information of each image sub-content may be fused to obtain sub-content feature information corresponding to the image modality.

[0149] Optionally, weights can be set for different image sub-contents to perform weighted fusion processing on the image feature information of different image sub-contents to obtain the sub-content feature information corresponding to the image modality. For example, the content weights can be set according to the display order of the image sub-contents. The content weights of the cover image and the cover frame or the first frame of the video can be set to 1, and the content weights of other image sub-contents can be set to 0.7, etc.

[0150] In one embodiment, the Figure 2 The fusion network in the model structure shown performs fusion processing on each image feature information to obtain sub-content feature information of the image modality.

[0151] S203 may specifically be: the sub-content feature information corresponding to each modality may be fused to obtain fused feature information of the multimedia content.

[0152] In one embodiment, the Figure 2 The fusion network in the model structure shown fuses the sub-content feature information corresponding to each modality to obtain fused feature information of the multimedia content.

[0153] In one embodiment, different weights may be set for sub-contents of different modalities to perform weighted fusion processing, that is, at least two modalities include a text modality and an image modality, and step S203 may include:

[0154] Obtain the number of text units contained in the sub-content of the text mode in the multimedia content, and obtain the number of images contained in the sub-content of the image mode;

[0155] Determining a first weight of sub-content feature information in a text modality and a second weight of sub-content feature information in an image modality according to a quantitative relationship between the number of text units and the number of images;

[0156] Based on the first weight and the second weight, weighted fusion processing is performed on the sub-content feature information corresponding to different modalities to obtain fused feature information.

[0157] For example, the number of text units contained in the text sub-content in the multimedia content is obtained, and the number of images in the image sub-content is obtained; the ratio between the number of text units and the number of images is calculated, and the ratio is equivalent to the number of text units corresponding to one image.

[0158] If the ratio is greater than the first preset threshold and less than the second preset threshold, it means that the amount of information contained in the image sub-content is equivalent to the amount of information contained in the text sub-content. In this case, the first weight of the sub-content feature information of the text modality and the second weight of the sub-content feature information of the image modality can be set equal. For example, the first weight and the second weight are both 1.

[0159] If the ratio is greater than the second preset threshold, indicating that the amount of information contained in the image sub-content is less than the amount of information contained in the text sub-content, the first weight of the sub-content feature information in the text modality can be set greater than the second weight of the sub-content feature information in the image modality.

[0160] If the ratio is less than the first preset threshold, indicating that the image sub-content contains more information than the text sub-content, the first weight of the sub-content feature information in the text mode may be set smaller than the second weight of the sub-content feature information in the image mode.

[0161] Based on the first weight and the second weight, weighted fusion processing is performed on the sub-content feature information corresponding to different modalities to obtain fused feature information.

[0162] S204 may specifically identify the content risk of the multimedia content according to the fused feature information to obtain the risk information of the multimedia content.

[0163] Among them, the risk information may include a risk score or risk level of the multimedia content. After determining the risk score or risk level of the multimedia content, the multimedia content may be filtered according to the risk score or risk level to avoid recommending risky multimedia content to users.

[0164] Optionally, multimedia content that requires manual review can also be determined based on the risk score or risk level. For example, multimedia content with a higher risk score or risk level can be directly filtered, and multimedia content with a low risk score or risk level can be directly recommended to users, reducing the amount of content that requires manual review. For multimedia content with a medium risk score or risk level, manual review can be performed to determine whether it can be recommended to users.

[0165] Optionally, the risk information may further include the risk type of the multimedia content, and the multimedia content to be filtered may be determined by sorting according to the risk type and risk level. That is, in one embodiment, there are multiple multimedia contents, and the risk information includes the risk type and risk level. After step S204, the following may be further included:

[0166] Classify and process each multimedia content according to its risk type, and obtain multimedia content under each risk type;

[0167] The multimedia content under each risk type is filtered according to the risk level to obtain the multimedia content to be reviewed.

[0168] For example, each multimedia content can be classified according to its risk type to obtain the multimedia content of each risk type; the multimedia content of each risk type can be filtered according to the risk level to obtain the multimedia content to be reviewed for each risk type; the multimedia content to be reviewed for each risk type can be pushed to different reviewers for review, thereby improving the review efficiency.

[0169] As can be seen from the above, the embodiment of the present application obtains the first amount of content submitted for review and the first amount of content released from the content pool at the target time, and the second amount of content submitted for review and the second amount of content released at the historical time corresponding to the target time; calculates the ratio between the first amount of content submitted for review and the second amount of content submitted for review to obtain a first adjustment ratio; calculates the ratio between the first amount of content released and the second amount of content released to obtain a second adjustment ratio; adjusts the content quantity difference between the target amount of content submitted for review and the target total amount of content in the content pool according to the first adjustment ratio and the second adjustment ratio to obtain an adjusted content quantity difference; determines a screening condition based on the adjusted content quantity difference and the content in the content pool sorted according to the review index value, and screens out the content submitted for review to be processed from the content to be processed according to the screening condition.

[0170] In the embodiment of the present application, screening conditions are generated based on the difference between the number of target contents in the content pool and the target amount of content to be submitted for review, as well as based on the ratio of the number of content at the target time and the historical time. New content can be screened based on the screening conditions so that the number of target contents in the content pool is dynamically kept close to the target amount of content to be submitted for review, that is, the number of content to be submitted for review is kept stable, so that personnel can be better arranged for review and the problem of waste of manpower or insufficient manpower can be avoided.

[0171] In order to better implement the content processing method provided in the embodiment of the present application, a content processing device is also provided in one embodiment. The meanings of the terms are the same as those in the above content processing method, and the specific implementation details can refer to the description in the method embodiment.

[0172] The content processing device may be integrated into a computer device, such as Figure 4 As shown, the content processing device may include: an acquisition unit 301, a first calculation unit 302, a second calculation unit 303, an adjustment unit 304, a condition determination unit 305 and a screening unit 306, which are specifically as follows:

[0173] (1) An acquisition unit 301 is used to acquire a first amount of submitted content and a first amount of released content in a content pool at a target time, and a second amount of submitted content and a second amount of released content at a historical time corresponding to the target time.

[0174] (2) A first calculation unit 302 is used to calculate the ratio between the first amount of content submitted for review and the second amount of content submitted for review to obtain a first adjustment ratio.

[0175] (3) A second calculation unit 303, used to calculate the ratio between the first released content amount and the second released content amount to obtain a second adjustment ratio.

[0176] (4) An adjustment unit 304, configured to adjust the content quantity difference between the target content quantity for review and the target total content quantity of the content pool according to the first adjustment ratio and the second adjustment ratio, to obtain an adjusted content quantity difference.

[0177] In one embodiment, the adjusted content quantity difference includes a first adjusted content quantity difference and a second adjusted content quantity difference, and the adjusting unit includes:

[0178] A quantity calculation subunit is used to calculate the content quantity difference between the target content quantity submitted for review and the total content quantity submitted for review;

[0179] The first adjustment subunit is configured to adjust the content quantity difference based on the first adjustment ratio to obtain an adjusted content quantity difference if the first adjustment ratio satisfies a first ratio condition and the second adjustment ratio does not satisfy a second ratio condition.

[0180] The second adjustment subunit is configured to adjust the content quantity difference based on the second adjustment ratio to obtain an adjusted content quantity difference if the first adjustment ratio does not satisfy the first ratio condition and the second adjustment ratio satisfies the second ratio condition.

[0181] In one embodiment, the adjustment unit further includes:

[0182] a third adjustment subunit, configured to perform a first adjustment on the content quantity difference based on the first adjustment ratio to obtain a candidate adjusted content quantity difference if the first adjustment ratio satisfies a first ratio condition and the second adjustment ratio satisfies a second ratio condition;

[0183] The fourth adjustment subunit is configured to perform a second adjustment on the candidate adjusted content quantity difference based on a second adjustment ratio to obtain an adjusted content quantity difference.

[0184] (5) a condition determination unit 305, configured to determine a screening condition based on the adjusted content quantity difference and the content in the content pool sorted according to the review index value; in one embodiment, the content pool includes a submitted content pool, a buffer content pool, and a released content pool; the target content quantity is the total quantity of content in the submitted content pool and the buffer content pool; each content in the content pool has a corresponding risk score; and the screening unit includes:

[0185] A first content determination subunit is used to determine target content based on the content quantity difference and the content in the released content pool sorted according to the review index value;

[0186] The first condition generating subunit is used to generate a first screening condition according to the risk score corresponding to the target content if the total amount of target content is less than the target amount for review.

[0187] The second condition generating subunit is used to generate a second screening condition according to the risk score corresponding to the second target content if the total amount of target content is greater than the target submission amount.

[0188] (6) A screening unit 306, which screens out the submitted content to be processed from the content to be processed according to the screening conditions.

[0189] In one embodiment, the screening unit further comprises:

[0190] A condition determination unit, used to determine a screening condition based on the adjusted content quantity difference and the content in the content pool sorted according to the audit index value;

[0191] The screening unit is used to screen out the submitted content to be processed from the content to be processed according to the screening conditions.

[0192] In one embodiment, the content pool includes multimedia content, and the content processing device further includes:

[0193] A content acquisition unit, configured to acquire multimedia content, wherein the multimedia content includes at least two sub-contents in different modes;

[0194] An extraction unit, configured to extract features of sub-contents in at least two modes respectively, to obtain feature information of sub-contents corresponding to different modes;

[0195] A fusion unit, used to fuse the sub-content feature information corresponding to different modalities to obtain fused feature information of the multimedia content;

[0196] The identification unit is used to identify the risks of multimedia content according to the fused feature information to obtain the risk score of the multimedia content.

[0197] In one embodiment, the at least two modalities include a text modality, the text modality corresponds to at least one text sub-content, and the extraction unit includes:

[0198] A word set acquisition subunit is used to acquire a sensitive word set;

[0199] A detection subunit, used to perform sensitive word detection on each text sub-content in the text mode according to the sensitive word set, and determine the position of the sensitive word in each text sub-content;

[0200] The marking subunit is used to mark the sensitive word position of each text sub-content according to the position, and obtain the sensitive word position information corresponding to each text sub-content;

[0201] The feature extraction subunit is used to extract text features from each text sub-content based on the sensitive word position information to obtain sub-content feature information of the text modality.

[0202] As can be seen from the above, the content processing device of the embodiment of the present application obtains the first amount of content submitted for review and the first amount of content released in the content pool at the target time, and the second amount of content submitted for review and the second amount of content released at the historical time corresponding to the target time through the acquisition unit 301; the first calculation unit 302 calculates the ratio between the first amount of content submitted for review and the second amount of content submitted for review, and obtains the first adjustment ratio; the second calculation unit 303 calculates the ratio between the first amount of content released and the second amount of content released, and obtains the second adjustment ratio; the adjustment unit 304 adjusts the content quantity difference between the target amount of content submitted for review and the target total amount of content in the content pool according to the first adjustment ratio and the second adjustment ratio, and obtains the adjusted content quantity difference; the condition determination unit 305 determines the screening condition based on the adjusted content quantity difference and the content in the content pool sorted according to the review index value, and the screening unit 306 screens out the content submitted for review to be processed from the content to be processed according to the screening condition.

[0203] In the embodiment of the present application, screening conditions are generated based on the difference between the number of target contents in the content pool and the target amount of content to be submitted for review, as well as based on the ratio of the number of content at the target time and the historical time. New content can be screened based on the screening conditions so that the number of target contents in the content pool is dynamically kept close to the target amount of content to be submitted for review, that is, the number of content to be submitted for review is kept stable, so that personnel can be better arranged for review and the problem of waste of manpower or insufficient manpower can be avoided.

[0204] The present application also provides a computer device, which may be a terminal or a server. Figure 5 As shown, it shows a schematic diagram of the structure of the computer device involved in the embodiment of the present application, specifically:

[0205] The computer device may include components such as a processor 1001 with one or more processing cores, a memory 1002 with one or more computer-readable storage media, a power supply 1003, and an input unit 1004. Those skilled in the art will appreciate that Figure 5 The computer device structure shown in the figure does not constitute a limitation on the computer device, and may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently. Among them:

[0206] The processor 1001 is the control center of the computer device. It uses various interfaces and lines to connect various parts of the entire computer device. By running or executing software programs and / or modules stored in the memory 1002 and calling data stored in the memory 1002, it executes various functions of the computer device and processes data, thereby monitoring the computer device as a whole. Optionally, the processor 1001 may include one or more processing cores; preferably, the processor 1001 may integrate an application processor and a modem processor, wherein the application processor mainly processes the operating system, user interface and computer programs, etc., and the modem processor mainly processes wireless communications. It is understandable that the above-mentioned modem processor may not be integrated into the processor 1001.

[0207] The memory 1002 can be used to store software programs and modules. The processor 1001 executes various functional applications and data processing by running the software programs and modules stored in the memory 1002. The memory 1002 may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system, a computer program required for at least one function (such as a sound playback function, an image playback function, etc.), etc.; the data storage area may store data created according to the use of the computer device, etc. In addition, the memory 1002 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other volatile solid-state storage devices. Accordingly, the memory 1002 may also include a memory controller to provide the processor 1001 with access to the memory 1002.

[0208] The computer device also includes a power supply 1003 for supplying power to various components. Preferably, the power supply 1003 can be logically connected to the processor 1001 through a power management system, so as to manage charging, discharging, and power consumption through the power management system. The power supply 1003 can also include any components such as one or more DC or AC power supplies, recharging systems, power failure detection circuits, power converters or inverters, and power status indicators.

[0209] The computer device may further include an input unit 1004, which may be used to receive input digital or character information and generate keyboard, mouse, joystick, optical or trackball signal input related to user settings and function control.

[0210] Although not shown, the computer device may further include a display unit, etc., which will not be described in detail herein. Specifically, in this embodiment, the processor 1001 in the computer device will load the executable files corresponding to the processes of one or more computer programs into the memory 1002 according to the following instructions, and the processor 1001 will run the computer programs stored in the memory 1002, thereby realizing various functions, as follows:

[0211] Obtaining the first submitted content volume and the first released content volume of the content pool at the target time, and the second submitted content volume and the second released content volume at the historical time corresponding to the target time;

[0212] Calculate the ratio between the amount of the first content submitted for review and the amount of the second content submitted for review to obtain a first adjustment ratio;

[0213] Calculate the ratio between the first released content volume and the second released content volume to obtain a second adjustment ratio;

[0214] According to the first adjustment ratio and the second adjustment ratio, adjusting the content quantity difference between the target content quantity submitted for review and the target content quantity of the content pool to obtain an adjusted content quantity difference;

[0215] Determine the screening criteria based on the difference in the number of adjusted contents and the contents in the content pool sorted by the audit indicator values;

[0216] Filter out the submitted content to be processed from the pending content according to the screening conditions.

[0217] The specific implementation of the above operations can be found in the previous embodiments and will not be described in detail here.

[0218] As can be seen from the above, the computer device of the embodiment of the present application can obtain the first amount of content submitted for review and the first amount of content released in the content pool at the target time, and the second amount of content submitted for review and the second amount of content released at the historical time corresponding to the target time; calculate the ratio between the first amount of content submitted for review and the second amount of content submitted for review to obtain a first adjustment ratio; calculate the ratio between the first amount of content released and the second amount of content released to obtain a second adjustment ratio; adjust the content quantity difference between the target amount of content submitted for review and the target total amount of content in the content pool according to the first adjustment ratio and the second adjustment ratio to obtain an adjusted content quantity difference; determine a screening condition based on the adjusted content quantity difference and the content in the content pool sorted according to the review index value; and screen out the content submitted for review to be processed from the content to be processed according to the screening condition.

[0219] In the embodiment of the present application, screening conditions are generated based on the difference between the number of target contents in the content pool and the target amount of content to be submitted for review, as well as based on the ratio of the number of content at the target time and the historical time. New content can be screened based on the screening conditions so that the number of target contents in the content pool is dynamically kept close to the target amount of content to be submitted for review, that is, the number of content to be submitted for review is kept stable, so that personnel can be better arranged for review and the problem of waste of manpower or insufficient manpower can be avoided.

[0220] According to one aspect of the present application, a computer program product or a computer program is provided, the computer program product or the computer program including computer instructions, the computer instructions being stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device executes the methods provided in various optional implementations in the above-mentioned embodiments.

[0221] Those skilled in the art will appreciate that all or part of the steps in the various methods of the above embodiments may be completed by a computer program, or by controlling related hardware through a computer program. The computer program may be stored in a computer-readable storage medium and loaded and executed by a processor.

[0222] To this end, an embodiment of the present application provides a computer-readable storage medium, in which a computer program is stored. The computer program can be loaded by a processor to execute any content processing method provided in the embodiment of the present application.

[0223] The specific implementation of the above operations can be found in the previous embodiments, which will not be described in detail here.

[0224] The computer-readable storage medium may include: a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, etc.

[0225] Since the computer program stored in the computer-readable storage medium can execute any content processing method provided in the embodiments of the present application, the beneficial effects that can be achieved by any content processing method provided in the embodiments of the present application can be achieved. Please refer to the previous embodiments for details and will not be repeated here.

[0226] The above is a detailed introduction to a content processing method, device, computer equipment and computer-readable storage medium provided in the embodiments of the present application. Specific examples are used in this article to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method of the present application and its core idea; at the same time, for technical personnel in this field, according to the ideas of the present application, there will be changes in the specific implementation methods and application scopes. In summary, the content of this specification should not be understood as a limitation on the present application.

Claims

1. A content processing method, It is characterized in that include: Acquire the first submitted content volume and the first released content volume of the content pool at the target time, and the second submitted content volume and the second released content volume at the historical time corresponding to the target time; Calculating a ratio between the volume of the first content submitted for review and the volume of the second content submitted for review to obtain a first adjustment ratio; Calculating a ratio between the first released content volume and the second released content volume to obtain a second adjustment ratio; According to the first adjustment ratio and the second adjustment ratio, adjusting the content quantity difference between the target content quantity submitted for review and the target content quantity of the content pool to obtain an adjusted content quantity difference; Determining a screening condition based on the adjusted content quantity difference and the content in the content pool sorted according to the audit index value; The content to be processed and submitted for review is filtered out from the content to be processed according to the filtering conditions.

2. The method according to claim 1, It is characterized in that The adjusted content quantity difference includes a first adjusted content quantity difference and a second adjusted content quantity difference. The content quantity difference between the target content quantity submitted for review and the total content quantity of the content pool is adjusted according to the first adjustment ratio and the second adjustment ratio to obtain the adjusted content quantity difference, including: Calculate the difference between the target content quantity and the total content quantity; If the first adjustment ratio satisfies a first ratio condition and the second adjustment ratio does not satisfy a second ratio condition, the content quantity difference is adjusted based on the first adjustment ratio to obtain the adjusted content quantity difference. If the first adjustment ratio does not satisfy the first ratio condition, and the second adjustment ratio satisfies the second ratio condition, the content quantity difference is adjusted based on the second adjustment ratio to obtain the adjusted content quantity difference.

3. The method according to claim 2, It is characterized in that The method further comprises: If the first adjustment ratio satisfies the first ratio condition, and the second adjustment ratio satisfies the second ratio condition, performing a first adjustment process on the content quantity difference based on the first adjustment ratio to obtain a candidate adjusted content quantity difference; A second adjustment process is performed on the candidate adjusted content quantity difference based on the second adjustment ratio to obtain the adjusted content quantity difference.

4. The method according to claim 2, It is characterized in that The content pool includes a submitted content pool, a buffer content pool, and a released content pool; the target content amount is the total amount of content in the submitted content pool and the buffer content pool; each content in the content pool corresponds to a risk score; the audit index value includes the risk score; and the screening condition is determined based on the content quantity difference and the content in the content pool sorted according to the audit index value, including: Determine target content based on the content quantity difference and the content in the released content pool sorted based on the risk score; If the total amount of the target content is less than the target review amount, generating a first screening condition according to the risk score corresponding to the target content; If the total amount of the target content is greater than the target review amount, a second screening condition is generated according to the risk score corresponding to the target content.

5. The method according to claim 4, It is characterized in that The step of filtering out the content to be processed from the content to be processed according to the filtering condition includes: Screening the content to be entered into the released content pool according to the first screening condition, and determining the first content to be entered into the released content pool and the second content to be entered into the buffer content pool; The content in the buffer content pool is filtered according to the second screening condition.

6. The method according to claim 4, It is characterized in that The content pool includes multimedia content, and the method further includes: Acquire multimedia content, where the multimedia content includes at least two sub-contents in different modalities; Extracting features of the sub-contents in the at least two modes respectively to obtain feature information of the sub-contents corresponding to the different modes; Fusing the sub-content feature information corresponding to different modalities to obtain fused feature information of the multimedia content; The risk of the multimedia content is identified according to the fused feature information to obtain a risk score of the multimedia content.

7. The method according to claim 6, It is characterized in that The at least two modes include a text mode, and the text mode corresponds to at least one text sub-content. The feature extraction of the sub-contents under the at least two modes is performed to obtain sub-content feature information corresponding to different modes, including: Get the sensitive word set; Performing sensitive word detection on each text sub-content in the text mode according to the sensitive word set, and determining the position of the sensitive word in each text sub-content; Marking the sensitive word position of each text sub-content according to the position to obtain sensitive word position information corresponding to each text sub-content; Based on the sensitive word position information, text feature extraction is performed on each text sub-content to obtain sub-content feature information of the text modality.

8. A content processing device, It is characterized in that include: An acquisition unit, configured to acquire a first submitted content amount and a first released content amount of a content pool at a target time, and a second submitted content amount and a second released content amount at a historical time corresponding to the target time; a first calculation unit, configured to calculate a ratio between the amount of the first content submitted for review and the amount of the second content submitted for review, to obtain a first adjustment ratio; a second calculation unit, configured to calculate a ratio between the first released content amount and the second released content amount to obtain a second adjustment ratio; an adjusting unit, configured to adjust the difference between the target content quantity submitted for review and the content quantity in the content pool according to the first adjustment ratio and the second adjustment ratio, so as to obtain an adjusted content quantity difference; A condition determination unit, configured to determine a screening condition based on the adjusted content quantity difference and the content in the content pool sorted according to the audit index value; The screening unit is used to screen out the content to be processed from the content to be processed according to the screening conditions.

9. A computer device, It is characterized in that It comprises a memory and a processor; the memory stores a computer program, and the processor is used to run the computer program in the memory to execute the content processing method according to any one of claims 1 to 7.

10. A computer-readable storage medium, It is characterized in that The computer-readable storage medium is used to store a computer program, and the computer program is loaded by a processor to execute the content processing method according to any one of claims 1 to 7.

Images