CN110290400B

CN110290400B - Suspicious brushing amount video identification method, real playing amount estimation method and device

Info

Publication number: CN110290400B
Application number: CN201910690295.2A
Authority: CN
Inventors: 王中晴; 彭涛
Original assignee: Beijing QIYI Century Science and Technology Co Ltd
Current assignee: Beijing QIYI Century Science and Technology Co Ltd
Priority date: 2019-07-29
Filing date: 2019-07-29
Publication date: 2022-06-03
Anticipated expiration: 2039-07-29
Also published as: CN110290400A

Abstract

The embodiment of the invention provides a suspicious brushing amount video identification method, a real playing amount estimation method and a device, which are applied to the technical field of video detection. The identification method comprises the following steps: determining a target video to be identified and a target playing end, wherein the target playing end is a playing end for video identification; determining each judgment dimension according to when the suspicious brushing amount video is identified aiming at the target playing end; calculating judgment values of the target video under each judgment dimension in a preset time period; and identifying whether the target video is a suspicious brushing amount video or not based on the evaluation value under each evaluation dimension to obtain an identification result. Therefore, whether the video is the suspicious brushing volume video or not can be effectively identified through the scheme.

Description

Suspicious brushing amount video identification method, real playing amount estimation method and device

Technical Field

The invention relates to the technical field of video detection, in particular to a suspicious brushing amount video identification method, a real playing amount estimation method and a device.

Background

With the rapid development of the internet, video websites are gradually known and used by the public. With the improvement of the openness of the video website to the user, homemade dramas, homemade artists and the like in the video website become pets of all large video websites.

However, under the prosperous situation, a lot of gray zones exist, namely video brushing amount, which not only damages the benefit of the video website platform side, but also easily recommends the video brushing amount to more users due to the existence of the recommendation system, thereby undoubtedly damaging the knowledge right of the public. Wherein, the video brushing amount is as follows: in order to enable the video submitted to the video website by the video uploader to get higher attention and click number, the video uploader simulates the behavior of clicking the video by human beings by using a third-party mechanism.

Therefore, how to effectively identify whether the video is a suspicious video with a brushing amount, namely, whether the video has the suspicious brushing amount is an urgent problem to be solved.

Disclosure of Invention

The embodiment of the invention aims to provide a method and a device for identifying a suspicious brushing amount video, so as to effectively identify whether the video is the suspicious brushing amount video. In addition, the embodiment of the invention also provides a method and a device for estimating the real playing amount of the suspicious brushing amount video, so that the real playing amount of the video can be effectively estimated when the video is identified to be the suspicious brushing amount video. The specific technical scheme is as follows:

in a first aspect, an embodiment of the present invention provides a method for identifying a suspicious brushing volume video, including:

determining a target video to be identified and a target playing end; the target playing end is a playing end for video identification;

determining each judgment dimension according to when the suspicious brushing amount video is identified aiming at the target playing end; each judgment dimension is the dimension influenced when the video is brushed;

calculating judgment values of the target video under each judgment dimension in a preset time period;

and identifying whether the target video is a suspicious brushing amount video or not based on the evaluation value under each evaluation dimension to obtain an identification result.

Optionally, the step of determining each judgment dimension according to which the suspicious brushing video is identified for the target playing terminal includes:

acquiring a corresponding relation between each playing end and an evaluation dimension according to each playing end, wherein the evaluation dimension according to each playing end is the evaluation dimension according to when the suspicious brushing amount video is identified aiming at the playing end;

and acquiring each judgment dimension according to the suspicious brushing amount video identified by the target playing terminal from the obtained corresponding relation.

Optionally, each of the playing terminals includes: the system comprises a webpage end, a client end and an audio and video equipment end;

the corresponding relationship between each playing end and the evaluation dimension according to each playing end includes:

the web page side corresponds to one or more of the following judging dimensions:

the method comprises the steps that the playing amount of a newly-added user playing a video accounts for the ratio, the correlation between the playing amount of the video at each sub-period and the playing amount of a channel to which the video belongs at each sub-period, the correlation between the playing amount of the video at each playing end and the playing amount of the channel to which the video belongs at each playing end, the ratio of the playing amount of the video in the total playing amount of each playing end, the ratio of a login user in a user playing the video, the positive film playing rate of the video, and the ratio of a playing event meeting a preset playing time length in the playing event of the video; the newly added users are users meeting preset newly added conditions;

the client corresponds to one or more of the following evaluation dimensions:

the method comprises the steps that the playing volume proportion of a newly added user playing a video, the correlation between the playing volume of the video in each sub-period and the playing volume of a channel to which the video belongs in each sub-period, the proportion of users logging in the users playing the video, the feature film playing rate of the video and the proportion of playing events meeting preset playing duration in the playing events of the video are determined;

the video and audio equipment end corresponds to one or more of the following evaluation dimensions:

the relevance of the playing amount of the video in each sub-period and the playing amount of the channel to which the video belongs in each sub-period, the occupation ratio of login users in the users playing the video, the feature film playing rate of the video, and the occupation ratio of playing events meeting the preset playing time length in the playing events of the video.

Optionally, the step of identifying whether the target video is a suspicious brushing volume video based on the evaluation values in the evaluation dimensions to obtain an identification result includes:

judging whether the judgment value under each judgment dimension meets a preset brushing amount condition corresponding to the judgment dimension or not to obtain a judgment result aiming at each judgment dimension in the judgment dimensions; if the number of judgment results which show that the judgment results meet the corresponding preset brushing amount conditions in the obtained judgment results is larger than a preset number threshold value, determining that the target video is the suspicious brushing amount video, otherwise, determining that the target video is not the suspicious brushing amount video;

or,

performing weighted calculation on the evaluation values under each evaluation dimension to obtain a calculation result; and if the calculation result meets the preset brushing amount condition corresponding to the weighting calculation, determining that the target video is a suspicious brushing amount video, otherwise, determining that the target video is not the suspicious brushing amount video.

In a second aspect, an embodiment of the present invention provides a method for estimating a true playing amount of a suspicious brushing amount video, including:

when the target video is identified as the suspicious brushing volume video according to the identification method provided by the first aspect, counting the sum pca _ other _ sum _ bvv of the playing volume of the target video in other playing ends except the target playing end within a preset time period used by the identification method; the target playing end is a playing end for which suspicious brushing amount video identification aims;

counting the playing amount pca-bvv of the target video in the target playing end within the preset time period;

counting the sum all _ sum _ bvv of the playing amount of the target video after data cleaning of each playing end in the preset time period; wherein the data cleaning is the processing of a user for removing the brushing amount;

estimating the real playing amount pca _ bvv _ predict of the target video in the target playing end in the preset time based on a preset formula by using the calculated pca _ other _ sum _ bvv, pca _ bvv and all _ sum _ bvv;

wherein the predetermined formula includes:

the pca _ ratio is pca _ bvv/all _ sum _ bvv.

In a third aspect, an embodiment of the present invention provides an apparatus for identifying a suspicious brushing volume video, including:

the first determining unit is used for determining a target video to be identified and a target playing end; the target playing end is a playing end for video identification;

the second determining unit is used for determining each judgment dimension according to which the suspicious brushing amount video is identified aiming at the target playing terminal; each judgment dimension is the dimension influenced when the video is brushed;

the calculating unit is used for calculating judgment values of the target video under each judgment dimension in a preset time period;

and the identification unit is used for identifying whether the target video is a suspicious brushing amount video or not based on the judgment values under all the judgment dimensions to obtain an identification result.

Optionally, the second determining unit includes:

a corresponding relation determining subunit, configured to obtain a corresponding relation between each playing end and an evaluation dimension according to which each playing end depends, where the evaluation dimension according to each playing end is an evaluation dimension according to which the suspicious brushing amount video is identified for the playing end;

and the dimension acquiring subunit is used for acquiring each judgment dimension according to the suspicious brushing amount video identified by the target playing terminal from the obtained corresponding relation.

the web page side corresponds to one or more of the following evaluation dimensions:

the client corresponds to one or more of the following evaluation dimensions:

the correlation between the playing amount of the video in each sub-period and the playing amount of the channel to which the video belongs in each sub-period, the occupation ratio of users logging in the users playing the video, the feature film playing rate of the video, and the occupation ratio of playing events meeting the preset playing time length in the playing events of the video.

Optionally, the identification unit comprises:

the first identification subunit is used for judging whether the judgment value under each judgment dimension meets a preset brushing amount condition corresponding to the judgment dimension or not according to each judgment dimension in each judgment dimension to obtain a judgment result; if the number of judgment results which indicate that the corresponding preset brushing amount conditions are met in the obtained judgment results is larger than a preset number threshold value, determining that the target video is a suspicious brushing amount video, otherwise, determining that the target video is not the suspicious brushing amount video;

or,

the second identification subunit is used for performing weighted calculation on the judgment values under each judgment dimension to obtain a calculation result; and if the calculation result meets the preset brushing amount condition corresponding to the weighting calculation, determining that the target video is a suspicious brushing amount video, otherwise, determining that the target video is not the suspicious brushing amount video.

In a fourth aspect, an embodiment of the present invention provides an estimation apparatus for a true playing amount of a suspicious brushing amount video, including:

a first statistical unit, configured to, when the target video is identified as a suspicious brushing volume video according to the identification method provided in the first aspect, perform statistics on a sum pca _ other _ sum _ bvv of playing volumes of the target video in other playing ends except a target playing end within a predetermined time period used by the identification method; the target playing end is a playing end for which suspicious brushing amount video identification aims;

a second counting unit, configured to count a playing amount pca _ bvv of the target video at the target playing end within the predetermined time period;

a third counting unit, configured to count a sum all _ sum _ bvv of playing amounts of the target video after data cleaning at each playing end in the predetermined period; wherein the data cleaning is the processing of a user for removing the brushing amount;

a calculating unit, configured to estimate, based on a predetermined formula, a real playing amount pca _ bvv _ predict of the target video in the target playing end in the predetermined time by using the calculated pca _ other _ sum _ bvv, pca _ bvv, and all _ sum _ bvv;

wherein the predetermined formula comprises:

the pca _ ratio is pca _ bvv/all _ sum _ bvv.

In a fifth aspect, an embodiment of the present invention provides a server, including a processor, a communication interface, a memory, and a communication bus, where the processor and the communication interface complete communication between the memory and the processor through the communication bus;

a memory for storing a computer program;

a processor, configured to implement the steps of the identification method provided in the first aspect when executing the program stored in the memory.

In a sixth aspect, an embodiment of the present invention provides a server, including a processor, a communication interface, a memory, and a communication bus, where the processor and the communication interface complete communication between the memory and the processor through the communication bus;

a memory for storing a computer program;

and the processor is used for realizing the steps of the estimation method provided by the second aspect when executing the program stored in the memory.

In a seventh aspect, the present invention further provides a computer-readable storage medium, in which a computer program is stored, and the computer program, when executed by a processor, implements the steps of the identification method provided in the first aspect.

In an eighth aspect, the present invention further provides a computer-readable storage medium, in which a computer program is stored, and the computer program, when executed by a processor, implements the steps of the estimation method provided in the second aspect.

In a ninth aspect, an embodiment of the present invention further provides a computer program product containing instructions, which when run on a computer, causes the computer to perform the steps of the identification method provided in the first aspect.

In a tenth aspect, embodiments of the present invention further provide a computer program product containing instructions, which when run on a computer, cause the computer to perform the steps of the estimation method provided in the second aspect.

In the embodiment of the invention, the video brushing amount is considered to be a single-end brushing amount behavior, so that a target playing end for suspicious video brushing amount identification is determined while a target video to be identified is determined; further, determining each judgment dimension according to which the suspicious video brushing amount is identified aiming at the target playing end, wherein each judgment dimension is the dimension influenced by the video brushing amount; calculating the judgment value of the target video under each judgment dimension within a preset time period; and finally, identifying whether the target video is a suspicious brushing amount video or not based on the judgment values under each judgment dimension to obtain an identification result. Therefore, whether the video is the suspicious brushing volume video or not can be effectively identified through the scheme.

In addition, when the identification method provided by the embodiment of the invention is used for identifying that the target video is the suspicious brushing volume video, the real playing volume of the target video can be effectively estimated by the real playing volume estimation method provided by the embodiment of the invention, so that the playing condition of the target video can be more comprehensively known.

Of course, not all of the advantages described above need to be achieved at the same time in the practice of any one product or method of the invention.

Drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.

Fig. 1 is a flowchart of a method for identifying a suspicious volume-brushing video according to an embodiment of the present invention;

fig. 2 is a flowchart of a method for estimating a real playing amount of a suspicious brushing amount video according to an embodiment of the present invention;

fig. 3 is a schematic structural diagram of an apparatus for identifying a suspicious video according to an embodiment of the present invention;

fig. 4 is a schematic structural diagram of a device for estimating a real playing amount of a suspicious brushing amount video according to an embodiment of the present invention;

fig. 5 is a schematic structural diagram of a server according to an embodiment of the present invention;

fig. 6 is a schematic structural diagram of a server according to an embodiment of the present invention.

Detailed Description

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

In order to effectively identify whether the video is a suspicious volume-brushing video, namely whether the video has the suspicious volume-brushing video, the embodiment of the invention provides an identification method of the suspicious volume-brushing video.

The execution subject of the method for identifying the suspicious video brushing amount provided by the embodiment of the invention can be an identification device of the suspicious video brushing amount, and the device can be operated in a server of a video website. In addition, the target video to be identified in the embodiment of the present invention may be a single video, and certainly, may also be a video album in which a plurality of video files exist, for example: a video album that carries videos, or, alternatively, a television show album. In a specific application, a single video or a video album can be selected as a target video to be identified according to actual requirements.

As shown in fig. 1, a method for identifying a suspicious brushing volume video according to an embodiment of the present invention may include the following steps:

s101, determining a target video to be identified and a target playing end;

the target playing end is a playing end for video identification.

When a trigger condition for identification of a suspicious brushing volume video is satisfied, a target video to be identified may be determined. And considering that the video brushing amount is generally a single-end brushing amount behavior, the target playing end for suspicious brushing amount video identification can be determined while the target video to be identified is determined. And then, after the target video and the target playing end are determined, the subsequent flow of the identification method is executed. Wherein the trigger condition may detect an identification instruction, or reach a predetermined point in time, etc.

The determination of the target video to be recognized is as follows: and determining the video identification of the target video to be recognized. Similarly, the target playing end for determining the suspicious brushing amount video identification is: and determining the end identification of the target playing end for which the suspicious brushing amount video identification is aimed. And, it can be determined by manual or system-specified manner which videos are the target videos to be identified; and determining a target playing end for which the suspicious brushing amount video identification aims by a manual specification or system specification mode.

It is understood that, in a specific application, the playing end capable of playing the video may include: the system comprises a webpage end, a client end and a video equipment end. The webpage end is a webpage which runs in a browser of the electronic equipment and can play videos; the client is an application program installed in the electronic device and capable of playing videos; the video and audio device end is a hardware device capable of providing a video and audio playing function, such as a television box, a screen projection device, and the like, or an application program running in a device with a video and audio playing function, such as an intelligent television, an intelligent projection device, and the like. In addition, in a specific application, according to different types of electronic devices, the web page may include: PC (Personal Computer) web page side and mobile web page side, and the client may include: a PC client and a mobile client. The PC webpage end is a webpage operated in a browser of the PC; the mobile webpage end is a webpage operated in a browser of the mobile terminal; the PC client is an application installed in the PC; the mobile client is an application installed in the mobile terminal.

Based on the above description about the playing end capable of playing video, each playing end targeted by the embodiment of the present invention may include: the determined target playing end can be a webpage end, a client end or an audio-video equipment end. Of course, for more detailed identification, each playing end targeted by the embodiment of the present invention may include a PC web page end, a mobile web page end, a PC client, a mobile client, and an audio-visual device end, and then, the determined target playing end may be the PC web page end, the mobile web page end, the PC client, the mobile client, or the audio-visual device end.

S102, determining each judgment dimension according to which the suspicious brushing amount video is identified aiming at the target playing terminal;

because different playing ends have different influences when the video is subjected to the amount brushing, after the target playing end is determined, in order to realize the video identification aiming at the target playing end, the judgment dimensions according to which the suspicious amount brushing video is identified aiming at the target playing end can be determined, and each judgment dimension is the dimension influenced when the video is subjected to the amount brushing at the target playing end.

It can be understood that there are various specific implementation manners of each judgment dimension according to when determining that the suspicious brushing video is identified for the target playing terminal.

For example, in an implementation manner, the step of determining each judgment dimension according to which the suspicious brushed video is identified for the target playing end may include:

In this specific implementation manner, the corresponding relationship between each playing end and the evaluation dimension according to each playing end may be set based on an empirical value or statistical analysis. Of course, each evaluation dimension according to which the suspicious brushing amount video is determined to be identified for the target playing end based on the corresponding relationship is merely an example, and should not be construed as a limitation to the embodiment of the present invention. For example: it is also reasonable to manually set each judgment dimension according to when the suspicious brushing amount video is identified for the target playing terminal when the target video to be identified is set.

For clarity of the scheme and clarity of the layout, the corresponding relationship between each playing end and the evaluation dimension according to each playing end is exemplarily described later.

S103, calculating judgment values of the target video in each judgment dimension within a preset time period;

after each judgment dimension is determined, the judgment value of the target video under each judgment dimension in a preset time period can be calculated according to a calculation mode or a statistical mode matched with each judgment dimension.

In addition, it can be understood that the video play log of the target video and/or the play click log of each user may be obtained, and then, the evaluation value of the target video in each evaluation dimension within a predetermined period of time is calculated by using log information in the video play log and/or the play click log of each user. In specific application, a kafka open source stream processing platform can be used for collecting a play click log of a user, the play click log is led into a Hadoop big data platform, and HIVE is used as a data warehouse for data storage.

It is understood that the duration of the predetermined time period may be set according to actual requirements, for example: a day, a week, a month, etc. Also, the characterization of the predetermined period of time may be various. For example, in one implementation, a starting point in time and duration may be given, such as: the starting time point was 2019, 5, 14, 00:00 and the duration was one week. In yet another implementation, a start time point and an end time point may be given, for example: the starting time point is 5/month and 14/day 00:00 in 2019, and the ending time point is 5/month and 21/day 00:00 in 2019.

And S104, identifying whether the target video is a suspicious brushing amount video or not based on the judgment values under each judgment dimension to obtain an identification result.

After the evaluation values under each evaluation dimension are obtained, whether the target video is the suspicious brushing amount video or not can be identified through a preset analysis mode based on the evaluation values under each evaluation dimension, and an identification result is obtained. In addition, when the target video is determined to be the suspicious brushing amount video, the suspicious brushing amount of the target video is indicated.

It can be understood that, there are various specific implementation manners for identifying whether the target video is a suspicious brushing amount video based on the evaluation values in each evaluation dimension.

For example, in an implementation manner, the step of identifying whether the target video is a suspicious brushing volume video based on the evaluation values in each evaluation dimension to obtain an identification result may include:

aiming at each judgment dimension in each judgment dimension, judging whether a judgment value under the judgment dimension meets a preset brushing amount condition corresponding to the judgment dimension to obtain a judgment result; and if the number of the judgment results which show that the judgment results meet the corresponding preset brushing amount conditions in the obtained judgment results is larger than a preset number threshold value, determining that the target video is the suspicious brushing amount video, otherwise, determining that the target video is not the suspicious brushing amount video.

In the implementation mode, each judgment dimension corresponds to a preset brushing condition, so that whether the judgment value under the judgment dimension meets the preset brushing condition or not can be judged according to each judgment dimension to obtain a judgment result; and then determining whether the target video is a suspicious brushing amount video or not based on the number of judgment results meeting the corresponding preset brushing amount condition in each judgment result.

It can be understood that the predetermined brushing amount condition corresponding to each evaluation dimension may be set according to the corresponding evaluation dimension, and then exemplified by combining the specific evaluation dimension. In addition, the predetermined number of thresholds may be set according to practical situations, for example: the predetermined number threshold may be a value less than the total number of the respective evaluation dimensions or may be a value equal to the total number of the respective evaluation dimensions.

In another implementation manner, the identifying whether the target video is a suspicious brushing amount video based on the evaluation values in each evaluation dimension to obtain an identification result may include:

carrying out weighted calculation on the evaluation values under each evaluation dimension to obtain a calculation result; and if the calculation result meets the preset brushing amount condition corresponding to the weighting calculation, determining that the target video is the suspicious brushing amount video, otherwise, determining that the target video is not the suspicious brushing amount video.

Since different playback ends may have different evaluation dimensions, the corresponding weight for each evaluation dimension may be set for each playback end.

In the embodiment of the invention, the video brushing amount is considered to be a single-end brushing amount behavior, so that a target playing end for suspicious video brushing amount identification is determined while a target video to be identified is determined; further, determining each judgment dimension according to when the suspicious video brushing amount is identified aiming at the target playing end, wherein each dimension is the dimension influenced when the video brushing amount is identified; calculating the judgment value of the target video under each judgment dimension in a preset time period; and finally, identifying whether the target video is a suspicious brushing amount video or not based on the judgment values under each judgment dimension to obtain an identification result. Therefore, whether the video is the suspicious brushing volume video or not can be effectively identified through the scheme.

For clarity of the scheme and clarity of layout, the following exemplary description relates to the correspondence between each playback end and the evaluation dimension according to each playback end.

Illustratively, in one implementation, each of the broadcast ends includes: the system comprises a webpage end, a client end and an audio and video equipment end;

the corresponding relationship between each playing terminal and the evaluation dimension according to each playing terminal includes:

the method comprises the steps of playing video, wherein the ratio of playing amount of a newly added user playing video, the correlation between the playing amount of the video at each sub-period and the playing amount of a channel to which the video belongs at each sub-period, the correlation between the playing amount of the video at each playing end and the playing amount of the channel to which the video belongs at each playing end, the ratio of the playing amount of the video at the total playing amount of each playing end, the ratio of users logging in the users playing the video, the positive film playing rate of the video and the ratio of playing events meeting a preset playing time length in the playing events of the video are determined; wherein, the new user is a user meeting the preset new conditions;

the clients correspond to one or more of the following judgment dimensions:

the video and audio equipment side corresponds to one or more of the following evaluation dimensions:

Illustratively, in another implementation manner, each of the playing terminals includes: the system comprises a PC webpage end, a mobile webpage end, a PC client end, a mobile client end and an audio-visual equipment end;

the PC webpage side and the mobile webpage side correspond to one or more of the following evaluation dimensions:

the method comprises the steps that the playing amount of a newly-added user playing a video accounts for the ratio, the correlation between the playing amount of the video at each sub-period and the playing amount of a channel to which the video belongs at each sub-period, the correlation between the playing amount of the video at each playing end and the playing amount of the channel to which the video belongs at each playing end, the ratio of the playing amount of the video in the total playing amount of each playing end, the ratio of a login user in a user playing the video, the positive film playing rate of the video, and the ratio of a playing event meeting a preset playing time length in the playing event of the video; wherein, the new user is a user meeting the preset new conditions;

both the PC client and the mobile client correspond to one or more of the following evaluation dimensions:

the correlation between the playing amount of the video in each sub-period and the playing amount of the channel to which the video belongs in each sub-period, the occupation ratio of login users among users playing the video, the feature film playing rate of the video, and the occupation ratio of playing events meeting the playing time length in the playing events of the video.

For clarity of the scheme, the following detailed description is presented for each evaluation dimension:

(1) the play amount of the newly added user playing the video is in proportion:

since the video volume is usually a mass play behavior in a short time, and the users of the mass play behavior are usually new registered users or new visitors, the play volume of the new user playing the video can be used as a judgment dimension. Moreover, the higher the play volume ratio of the newly added user playing the video, the higher the dubbing of the video brushing volume.

Wherein, the new user is a user meeting the preset new conditions. For example, the predetermined newly added condition may be: no play action has occurred for a specified duration before the predetermined period. For example: the preset time period is from 5/month 10/day 00 in 2019 to 5/month 10/day 24/00 in 2019, and the preset newly increased conditions can be as follows: no play activity occurred in one month between 5 and 10 months 00:00 in 2019 and 24:00 in 5 and 10 months in 2019. Of course, the predetermined addition condition is not limited to the example given above.

The play volume ratio of the newly added user playing the video is specifically as follows: for the pointed playing end, the ratio of the playing quantity generated by the newly added user in the total playing quantity of the video to the total playing quantity of the video. By analyzing the video playing logs of the video and/or the playing click logs of each user, the playing amount ratio of the newly added users playing the video can be calculated.

The preset brushing amount condition corresponding to the playing amount ratio of the newly added user playing the video can be as follows: greater than a predetermined first proportional threshold. The first proportional threshold may be set according to actual conditions, for example: the first proportional threshold may be 70%, 75%, 80%, etc.

For the judgment dimension that the playing amount of the newly added user playing the video is in proportion to, in the calculation preset period, the judgment value of the target video in each judgment dimension may specifically be:

and calculating a specific ratio of the playing quantity of the newly added user playing the target video to the ratio in a preset time period.

(2) The correlation between the playing amount of the video in each sub-period and the playing amount of the channel to which the video belongs in each sub-period is as follows:

for a large data environment, the time-interval playing amount of each video and the channel where the video is located is positively correlated, and if the positive correlation between the playing amount of the video in each sub-interval and the playing amount of the channel where the video belongs in each sub-interval is weaker, the higher the doubtability that the video belongs to the brushing volume video is. Wherein, the correlation can be generally characterized by a correlation coefficient, the value of the correlation coefficient is between [ -1,1], and the smaller the value, the more suspicious the video is represented.

Optionally, the correlation between the playing amount of the video in each sub-period and the playing amount of the channel to which the video belongs in each sub-period includes:

the Pearson correlation coefficient between the playing amount of the video in each sub-period and the playing amount of the channel to which the video belongs in each sub-period.

For example, if the predetermined time period is a certain day, the playing amount of the video in each hour and the playing amount of the channel to which the video belongs in each sub-period may be counted, and the pearson correlation coefficient between the playing amount of the video in each hour and the playing amount of the channel to which the video belongs in each hour is calculated.

It is understood that, in statistics, the Pearson correlation coefficient (Pearson product-moment correlation coefficient, abbreviated to PPMCC or PCCs) is used to measure the correlation (linear correlation) between two variables X and Y, and the value thereof is between-1 and 1. Wherein the pearson correlation coefficient between two variables is defined as the quotient of the covariance and the standard deviation between the two variables. In this embodiment, for the evaluation dimension, the variable X and the variable Y are respectively: the amount of video played in each hour, and the amount of video played in each sub-period on the channel to which the video belongs.

Of course, in another implementation manner, the playing amount of the video in each sub-period may be fitted to a curve, and the playing amount of the channel to which the video belongs in each sub-period may be fitted to a curve, and then, the correlation between the two curves is calculated. The smaller the correlation, the more suspicious the video is indicative of belonging to a brush-size video.

For the evaluation dimension, the corresponding predetermined brushing amount condition may be: less than a predetermined first coefficient threshold. The first coefficient threshold may be set according to actual conditions, for example: the first coefficient threshold may be-0.5, -0.6, -0.7, -0.8, etc.

For the judgment dimension, in the calculating the judgment value of the target video in each judgment dimension in the predetermined time period, specifically, the judgment value may be:

and calculating a specific value of the correlation between the playing amount of the target video in each sub-period and the playing amount of the channel to which the video belongs in each sub-period in a preset time period.

(3) The correlation between the playing amount of the video at each playing end and the playing amount of the channel to which the video belongs at each playing end is as follows:

for a big data environment, the positive correlation between the video and the sub-platform playing amount of the channel where the video is located is positive, and if the positive correlation between the video and the sub-platform playing amount of the channel where the video is located is weaker, the higher the suspiciousness that the video belongs to the flash video is. Wherein, the correlation can be generally characterized by a correlation coefficient, the value of the correlation coefficient is between [ -1,1], and the smaller the value, the more suspicious the video is represented.

The correlation between the playing amount of the video at each playing end and the playing amount of the channel to which the video belongs at each playing end comprises the following steps:

the Pearson correlation coefficient between the playing amount of the video at each playing end and the playing amount of the channel to which the video belongs at each playing end.

For example, if the playing end where the video is located is the above five playing ends: the system comprises a PC client, a PC webpage end, a mobile client, a mobile webpage end and an audio-visual equipment end, and therefore the Pearson correlation coefficient of the playing amount of the video at each playing end and the playing amount of the channel to which the video belongs at each playing end can be counted. In this embodiment, for the evaluation dimension, when calculating the pearson correlation coefficient, the variable X and the variable Y are respectively: the playing amount of the video at each playing end and the playing amount of the channel to which the video belongs at each playing end.

Of course, in another implementation manner, the playing amount of the video at each playing end may be fitted to a curve, and the playing amount of the channel to which the video belongs at each playing end may be fitted to a curve, and then, the correlation between the two curves is calculated. The smaller the correlation, the more suspicious the video is indicative of belonging to a brush-size video.

For the evaluation dimension, the corresponding predetermined brushing amount condition may be: less than a predetermined second coefficient threshold. The second coefficient threshold may be set according to actual conditions, for example: the second coefficient threshold may be-0.5, -0.6, -0.7, -0.8, etc.

and calculating a specific value of the correlation between the playing amount of the target video at each playing end and the playing amount of the channel to which the video belongs at each playing end in a preset time period.

(4) The ratio of the playing amount of the video in the total playing amount of each playing end is as follows:

through big data analysis, the higher the ratio of the playing amount of the video at a certain playing end to the total playing amount of each playing end, the higher the dubious of the brushing amount of the video at the end.

For the evaluation dimension, the corresponding predetermined brushing amount condition may be: greater than a predetermined second proportional threshold. The second proportional threshold may be set according to actual conditions, for example: the second proportion threshold may be 50%, 55%, 60%, 65%, 70%, etc.

and calculating the ratio of the playing amount of the target video at the target playing end to the total playing amount of each playing end in a preset time period.

(5) The duty ratio of the logged-in users in the users playing the videos is as follows:

through big data analysis, the ratio of the logged users is relatively high for a normal high-play episode. If a video has a high playing amount, but the login user proportion is low, the video is indicated to have suspicious behaviors of brushing the amount. Therefore, the duty ratio of the logged-in user among the users playing the video is taken as a judgment dimension.

For the evaluation dimension, the corresponding predetermined brushing amount condition may be: less than a predetermined third proportional threshold. The third proportional threshold may be set according to actual conditions, for example: the third proportion threshold may be 10%, 20%, 30%, 40%, etc.

and calculating the ratio of the duty ratio of the login user in the users playing the target video aiming at the target playing end in a preset time period.

(6) The positive film playing rate of the video;

wherein, the positive film playing rate of the video is as follows: ratio of positive viewing volume to playing volume.

As can be seen from the big data analysis, a video with a high playing amount exists, which represents that the video should have high heat and interest, and the user should watch the content of the video instead of closing the video by clicking, and if the ratio of the watching amount to the playing amount is too low, it also represents that the video may have a brushing amount behavior. The judgment dimension can be applied to the identification of videos into which advertisements are divided.

For the evaluation dimension, the corresponding predetermined brushing amount condition may be: less than a predetermined fourth proportional threshold. The fourth proportional threshold may be set according to actual conditions, for example: the fourth proportion threshold may be 10%, 20%, 30%, 40%, etc.

and calculating the specific value of the feature film playing rate of the target video aiming at the target playing end in a preset time period.

(7) The occupation ratio of the playing events meeting the preset playing time length in the playing events of the video is determined;

as can be known from big data analysis, in the brushing amount video, in order to obtain the click number, the percentage of the playing time with a short playing time is relatively large, a predetermined playing time may be set, such as 1 minute, 2 minutes, 3 minutes, 4 minutes, 5 minutes, and the like, and the percentage of the playing events satisfying the predetermined playing time in the playing events of the video is calculated.

For the evaluation dimension, the corresponding predetermined brushing amount condition may be: greater than a predetermined fifth proportional threshold. The fifth proportional threshold may be set according to actual conditions, for example: the fifth scaling threshold may be 50%, 60%, 65%, 70%, etc. It is understood that one predetermined play-out time period may be set, or a plurality of predetermined play-out time periods may be set. When a plurality of preset playing time lengths are set, the corresponding preset brushing amount conditions may be: the ratio mean is greater than the predetermined sixth ratio threshold, but is not limited thereto. For the judgment dimension, in the calculating the judgment value of the target video in each judgment dimension in the predetermined time period, specifically, the judgment value may be:

and calculating the ratio of the playing events meeting the preset playing time length in the playing events of the target video aiming at the target playing end in the preset time period.

In addition, when the video is identified to belong to the suspicious video, the real playing amount of the video is generally required to be known, so that the playing condition of the video can be more comprehensively known. Based on the requirement, on the basis of the steps S101 to S104, the embodiment of the present invention further provides a method for estimating a real playing amount of a suspicious brushing amount video. As shown in fig. 2, the method for estimating the real playing amount of the suspicious brushing amount video according to the embodiment of the present invention may further include the following steps:

s201, when the target video is identified as the suspicious brushing amount video according to the identification method provided by the embodiment of the invention, counting the sum pca _ other _ sum _ bvv of the playing amount of the target video in other playing ends except the target playing end in a preset time period used by the identification method; the target playing end is a playing end for which suspicious brushing amount video identification aims;

s202, counting the playing amount pca _ bvv of the target video in the target playing end within the preset time period;

s203, counting the sum all _ sum _ bvv of the playing amount of the target video after data cleaning of each playing end in the preset time period;

wherein the data cleaning is the processing of the user for removing the brushing amount. The user who removes the brush amount may adopt any filtering method of the user who removes the brush amount, which is not limited herein.

S204, estimating the real playing amount pca _ bvv _ predict of the target video in the target playing end in the preset time based on a preset formula by using the calculated pca _ other _ sum _ bvv, pca _ bvv and all _ sum _ bvv;

wherein the predetermined formula may include:

the pca _ ratio is pca _ bvv/all _ sum _ bvv.

It is understood that the video play logs of the target video and/or the play click logs of the users can be analyzed to obtain the statistics in S201 to S203.

In the embodiment, when the video is identified to belong to the suspicious brushing volume video, the real playing volume of the video can be effectively estimated, so that the playing condition of the target video can be more comprehensively known.

Corresponding to the embodiment of the method for identifying the suspicious brushing volume video, the embodiment of the invention also provides a device for identifying the suspicious brushing volume video. As shown in fig. 3, an apparatus for identifying a suspicious brushing amount video according to an embodiment of the present invention may include:

a first determining unit 310, configured to determine a target video and a target playing end to be identified; the target playing end is a playing end for video identification;

a second determining unit 320, configured to determine each evaluation dimension according to which the suspicious brushing video is identified for the target playing end; each judgment dimension is the dimension influenced when the video is brushed;

a calculating unit 330, configured to calculate evaluation values of the target video in each evaluation dimension within a predetermined time period;

the identifying unit 340 is configured to identify whether the target video is a suspicious brushing amount video based on the evaluation values in the evaluation dimensions, so as to obtain an identification result.

In the embodiment of the invention, the video brushing amount is considered to be a single-end brushing amount behavior, so that a target playing end for suspicious video brushing amount identification is determined while a target video to be identified is determined; further, determining each judgment dimension according to which the suspicious brushing amount video is identified aiming at the target playing end, and calculating the judgment value of the target video under each judgment dimension in a preset time period; and finally, identifying whether the target video is a suspicious brushing amount video or not based on the judgment values under each judgment dimension to obtain an identification result. Therefore, whether the video is the suspicious brushing volume video or not can be effectively identified through the scheme.

Optionally, the second determining unit 320 may include:

and the dimension acquisition subunit is used for acquiring each judgment dimension according to which the suspicious brushing amount video is identified aiming at the target playing end from the obtained corresponding relation.

the method comprises the steps that the playing amount of a newly-added user playing a video accounts for the ratio, the correlation between the playing amount of the video at each sub-period and the playing amount of a channel to which the video belongs at each sub-period, the correlation between the playing amount of the video at each playing end and the playing amount of the channel to which the video belongs at each playing end, the ratio of the playing amount of the video in the total playing amount of each playing end, the ratio of a login user in a user playing the video, the positive film playing rate of the video, and the ratio of a playing event meeting a preset playing time length in the playing event of the video; the newly added user is a user meeting a preset newly added condition;

the client corresponds to one or more of the following evaluation dimensions:

the video and audio equipment end corresponds to one or more of the following judgment dimensions:

Optionally, the identifying unit 340 may include:

the first identification subunit is used for judging whether the judgment value under each judgment dimension meets a preset brushing amount condition corresponding to the judgment dimension or not according to each judgment dimension in each judgment dimension to obtain a judgment result; if the number of judgment results which show that the judgment results meet the corresponding preset brushing amount conditions in the obtained judgment results is larger than a preset number threshold value, determining that the target video is the suspicious brushing amount video, otherwise, determining that the target video is not the suspicious brushing amount video;

or,

Corresponding to the method for estimating the real playing amount of the suspicious brushing amount video, the embodiment of the invention also provides a device for estimating the real playing amount of the suspicious brushing amount video. As shown in fig. 4, the device for estimating the actual playing amount of a suspicious brushing amount video according to an embodiment of the present invention may include:

a first statistical unit 410, configured to, when the target video is identified as a suspicious brushing amount video according to the identification method provided in the embodiment of the present invention, perform statistics on a sum pca _ other _ sum _ bvv of playing amounts of the target video in other playing ends except for the target playing end within a predetermined time period used by the identification method; the target playing end is a playing end for which suspicious brushing amount video identification aims;

a second counting unit 420, configured to count a playing amount pca _ bvv of the target video at the target playing end within the predetermined time period;

a third counting unit 430, configured to count a sum all _ sum _ bvv of playing amounts of the target video after data cleaning at each playing end in the predetermined period; wherein the data cleaning is the processing of a user for removing the brushing amount;

a calculating unit 440, configured to estimate, by using the calculated pca _ other _ sum _ bvv, pca _ bvv, and all _ sum _ bvv, a real playing amount pca _ bvv _ predict of the target video in the target playing end within the predetermined time based on a predetermined formula;

wherein the predetermined formula comprises:

the pca _ ratio is pca _ bvv/all _ sum _ bvv.

In addition, compared with the above-mentioned suspicious brushing video identification method embodiment, the embodiment of the present invention further provides a server, as shown in fig. 5, including a processor 501, a communication interface 502, a memory 503 and a communication bus 504, where the processor 501, the communication interface 502 and the memory 503 complete mutual communication through the communication bus 504,

a memory 503 for storing a computer program;

the processor 501 is configured to implement the steps of the method for identifying a suspicious brushing video according to the embodiment of the present invention when executing the program stored in the memory 503.

In addition, compared with the above-mentioned embodiment of the method for estimating the actual playing amount of the suspicious brushing amount video, the embodiment of the present invention further provides a server, as shown in fig. 6, including a processor 601, a communication interface 602, a memory 603, and a communication bus 604, where the processor 601, the communication interface 602, and the memory 603 complete communication with each other through the communication bus 604,

a memory 603 for storing a computer program;

the processor 601 is configured to implement the steps of the method for estimating the real playing amount of the suspicious brushing amount video according to the embodiment of the present invention when executing the program stored in the memory 603.

The communication bus mentioned in the above server may be a Peripheral Component Interconnect (PCI) bus, an Extended Industry Standard Architecture (EISA) bus, or the like. The communication bus may be divided into an address bus, a data bus, a control bus, etc. For ease of illustration, only one thick line is shown, but this does not mean that there is only one bus or one type of bus.

The communication interface is used for communication between the server and other devices.

The Memory may include a Random Access Memory (RAM) or a Non-Volatile Memory (NVM), such as at least one disk Memory. Optionally, the memory may also be at least one memory device located remotely from the processor.

The Processor may be a general-purpose Processor, including a Central Processing Unit (CPU), a Network Processor (NP), and the like; but also Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Field Programmable Gate Arrays (FPGAs) or other Programmable logic devices, discrete Gate or transistor logic devices, discrete hardware components.

In yet another embodiment of the present invention, a computer-readable storage medium is further provided, in which a computer program is stored, and the computer program, when executed by a processor, implements the steps of the above-mentioned method for identifying suspicious brushing volume videos.

In another embodiment of the present invention, a computer-readable storage medium is further provided, in which a computer program is stored, and the computer program, when executed by a processor, implements the steps of the above-mentioned method for estimating the real playing amount of the suspicious brushing amount video.

In yet another embodiment of the present invention, a computer program product containing instructions is also provided, which when run on a computer causes the computer to execute the method for identifying suspicious brush videos of the above embodiments.

In another embodiment of the present invention, a computer program product containing instructions is provided, which when run on a computer, causes the computer to execute the method for estimating the real playing amount of a suspicious brushing video in the above embodiment.

In the above embodiments, all or part of the implementation may be realized by software, hardware, firmware, or any combination thereof. When implemented in software, may be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When loaded and executed on a computer, cause the processes or functions described in accordance with the embodiments of the invention to occur, in whole or in part. The computer may be a general purpose computer, a special purpose computer, a network of computers, or other programmable device. The computer instructions may be stored in a computer readable storage medium or transmitted from one computer readable storage medium to another, for example, from one website site, computer, server, or data center to another website site, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, Digital Subscriber Line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device, such as a server, a data center, etc., that incorporates one or more of the available media. The usable medium may be a magnetic medium (e.g., floppy Disk, hard Disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., Solid State Disk (SSD)), among others.

It is noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.

All the embodiments in the present specification are described in a related manner, and the same and similar parts among the embodiments may be referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, for the embodiments of the apparatus, the device, the storage medium, and the computer program product, since they are substantially similar to the method embodiments, the description is relatively simple, and for the relevant points, reference may be made to the partial description of the method embodiments.

The above description is only for the preferred embodiment of the present invention, and is not intended to limit the scope of the present invention. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention shall fall within the protection scope of the present invention.

Claims

1. A method for identifying suspicious brushing amount videos is characterized by comprising the following steps:

simultaneously determining a target video to be identified and a target playing end; the target playing end is a playing end which is aimed at by the suspicious brushing volume video identification;

2. The method according to claim 1, wherein the step of determining each judgment dimension according to which the suspicious brushed video is identified for the target player comprises:

and acquiring each judgment dimension according to which the suspicious brushing amount video is identified aiming at the target playing end from the obtained corresponding relation.

3. The method of claim 2, wherein each of the broadcasting terminals comprises: the system comprises a webpage end, a client end and an audio and video equipment end;

the client corresponds to one or more of the following evaluation dimensions:

4. The method according to any one of claims 1 to 3, wherein the step of identifying whether the target video is a suspicious brushing volume video based on the evaluation values in the respective evaluation dimensions to obtain an identification result comprises:

judging whether the judgment value under each judgment dimension meets a preset brushing amount condition corresponding to the judgment dimension or not to obtain a judgment result aiming at each judgment dimension in the judgment dimensions; if the number of judgment results which indicate that the corresponding preset brushing amount conditions are met in the obtained judgment results is larger than a preset number threshold value, determining that the target video is a suspicious brushing amount video, otherwise, determining that the target video is not the suspicious brushing amount video;

or,

5. A method for estimating the real playing amount of a suspicious brushing amount video is characterized by comprising the following steps:

when a target video is identified as a suspicious brushing amount video according to the identification method of any one of claims 1 to 4, counting the sum pca _ other _ sum _ bvv of the playing amounts of the target video in other playing ends except a target playing end within a predetermined time period utilized by the identification method; the target playing end is a playing end for which suspicious brushing amount video identification aims;

counting the playing amount pca _ bvv of the target video in the target playing end within the preset time period;

wherein the predetermined formula comprises:

the pca _ ratio is pca _ bvv/all _ sum _ bvv.

6. The utility model provides an identification means of suspicious brush volume video which characterized in that includes:

the first determining unit is used for simultaneously determining a target video to be identified and a target playing end; the target playing end is a playing end for which suspicious brushing amount video identification aims;

7. The apparatus according to claim 6, wherein the second determining unit comprises:

8. The apparatus of claim 7, wherein each of the broadcasting terminals comprises: the system comprises a webpage end, a client end and an audio and video equipment end;

the client corresponds to one or more of the following evaluation dimensions:

9. The apparatus according to any one of claims 6-8, wherein the identification unit comprises:

or,

10. A real playing amount estimation device for suspicious brushing amount videos is characterized by comprising:

a first statistical unit, configured to, when the target video is identified as a suspicious brushing volume video according to the identification method of any one of claims 1 to 4, count a sum pca _ other _ sum _ bvv of playing volumes of the target video in other playing ends except a target playing end within a predetermined period utilized by the identification method; the target playing end is a playing end for which suspicious brushing amount video identification aims;

a second statistical unit, configured to count a play amount pca _ bvv of the target video at the target playback end within the predetermined time period; wherein, the data cleaning is the processing of the user for removing the brushing amount;

a third counting unit, configured to count a sum all _ sum _ bvv of playing amounts of the target video after data cleaning at each playing end in the predetermined period;

wherein the predetermined formula comprises:

the pca _ ratio is pca _ bvv/all _ sum _ bvv.

11. A server is characterized by comprising a processor, a communication interface, a memory and a communication bus, wherein the processor and the communication interface are used for realizing the communication between the processor and the memory through the communication bus;

a memory for storing a computer program;

a processor for implementing the method steps of any one of claims 1 to 5 when executing a program stored in the memory.