CN111861541B

CN111861541B - Method and device for determining cargo effect based on cargo video

Info

Publication number: CN111861541B
Application number: CN202010535741.5A
Authority: CN
Inventors: 王秋文; 李百川; 苏伟鹏
Original assignee: Youmi Technology Co ltd
Current assignee: Youmi Technology Co ltd
Priority date: 2020-06-12
Filing date: 2020-06-12
Publication date: 2024-04-19
Anticipated expiration: 2040-06-12
Also published as: CN111861541A

Abstract

The invention discloses a method and a device for determining a cargo effect based on cargo video, comprising the following steps: acquiring target data corresponding to all the video with goods of a certain target commodity in a certain target time period, wherein the target data comprises data with association relation with sales of the target commodity; calculating an allocation weight value of each cargo video according to the target data corresponding to each cargo video; and calculating a sales contribution value of each commodity-carrying video for the target commodity in the target time period according to the allocation weight value of each commodity-carrying video, the sum of the allocation weight values of all the commodity-carrying videos and the acquired sales volume data of the target commodity in the target time period. Therefore, the invention can accurately determine the contribution of each video on the commodity sales volume when a plurality of video bands sell the same commodity, namely accurately determine the video effect of each video, and is beneficial to providing an accurate reference basis for selecting proper video styles of the video bands and/or the video bands.

Description

Method and device for determining cargo effect based on cargo video

Technical Field

The invention relates to the technical field of Internet, in particular to a method and a device for determining a cargo effect based on cargo video.

Background

With the rapid development of the internet, internet users are increasing. In order to expand the audience range and influence of commodities, the marketing mode of commodities introduces video marketing based on internet besides traditional advertising marketing, for example: the advertiser can select a plurality of video bloggers to introduce and display commodities in a video recording or video live broadcasting mode, so that more people can be attracted to purchase the commodities, wherein videos which are released by the video bloggers and used for introducing and displaying the commodities can be also called as goods-carrying videos.

In practical application, for the same commodity, there are usually multiple video with goods distributed by multiple video bloggers, and the styles of the video with goods distributed by different video bloggers are various, and the sales contribution of the video with goods distributed by each video blogger is different. In order to gradually increase the cost performance of commodity video marketing, the commodity video with higher contribution to commodity sales needs to be determined from a plurality of commodity videos of a plurality of video bloggers. It can be seen that how to accurately determine the sales contribution value of each video-in-stock to the merchandise is important.

Disclosure of Invention

The invention aims to solve the technical problem of providing a method and a device for determining the effect of goods on the basis of the video of goods, which can accurately determine the sales contribution value of each video of goods on goods.

In order to solve the technical problem, a first aspect of the present invention discloses a method for determining a cargo effect based on a cargo video, the method comprising:

Acquiring target data corresponding to all the video with goods of a certain target commodity in a certain target time period, wherein the target data corresponding to the video with goods comprises data with association relation with sales volume of the target commodity in all the data corresponding to the video with goods;

Calculating an allocation weight value of each cargo video according to target data corresponding to each cargo video;

Calculating sales contribution value of each goods-carrying video for the target goods in the target time period according to the allocation weight value of each goods-carrying video, the sum of the allocation weight values of all the goods-carrying videos and the acquired sales data of the target goods in the target time period

The second aspect of the invention discloses a cargo effect determining device based on cargo video, which comprises:

The acquisition module is used for acquiring target data corresponding to all the commodity-carrying videos of a certain target commodity in a certain target time period, wherein the target data corresponding to the commodity-carrying videos comprises data with association relation with sales of the target commodity in all the data corresponding to the commodity-carrying videos;

the first calculation module is used for calculating an allocation weight value of each cargo video according to target data corresponding to each cargo video;

the second calculation module is used for calculating sales contribution values of each goods-carrying video for the target goods in the target time period according to the allocation weight value of each goods-carrying video, the sum of the allocation weight values of all the goods-carrying videos and the acquired sales data of the target goods in the target time period.

In a third aspect, the present invention discloses another apparatus for determining a cargo effect based on a cargo video, the apparatus comprising:

A memory storing executable program code;

A processor coupled to the memory;

The processor invokes the executable program code stored in the memory to perform some or all of the steps in the video-in-band effect determination method disclosed in the first aspect of the present invention.

A fourth aspect of the present invention discloses a computer storage medium storing computer instructions for performing part or all of the steps of the video-on-demand based effect determination method disclosed in the first aspect of the present invention when called.

Compared with the prior art, the embodiment of the invention has the following beneficial effects:

In the embodiment of the invention, the target data corresponding to all the goods-carrying videos of a certain target commodity in a certain target time period are acquired, wherein the target data corresponding to the goods-carrying videos comprises data which have an association relation with sales amount of the target commodity in all the data corresponding to the goods-carrying videos; calculating an allocation weight value of each cargo video according to the target data corresponding to each cargo video; and calculating a sales contribution value of each goods-carrying video for the target goods in the target time period according to the allocation weight value of each goods-carrying video, the sum of the allocation weight values of all the goods-carrying videos and the acquired sales data of the target goods in the target time period. Therefore, when the invention is implemented, the contribution of each video on the commodity sales can be accurately determined when a plurality of video bands sell the same commodity, namely the video effect of each video is accurately determined, and the invention is beneficial to providing accurate reference for selecting proper video styles of video with the video and/or the video with the video.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly described below, and it is apparent that the drawings in the following description are only some embodiments of the present invention, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.

FIG. 1 is a flow chart of a method for determining a cargo effect based on a cargo video according to an embodiment of the present invention;

FIG. 2 is a flow chart of another method for determining a cargo effect based on a cargo video according to an embodiment of the present invention;

FIG. 3 is a flowchart of a method for calculating sales contribution value of each of the video-in-band for a target commodity in a target time period according to an embodiment of the present invention;

FIG. 4 is a schematic diagram of a cargo effect determining device based on cargo video according to an embodiment of the present invention;

FIG. 5 is a schematic diagram of another cargo effect determination device based on cargo video according to an embodiment of the present invention;

FIG. 6 is a schematic diagram of a cargo effect determination device based on a cargo video according to an embodiment of the present invention;

Fig. 7 is a schematic structural diagram of a cargo effect determining apparatus based on a cargo video according to an embodiment of the present invention.

Detailed Description

In order that those skilled in the art will better understand the present invention, a technical solution in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in which it is apparent that the described embodiments are only some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.

The terms first, second and the like in the description and in the claims and in the above-described figures are used for distinguishing between different objects and not necessarily for describing a sequential or chronological order. Furthermore, the terms "comprise" and "have," as well as any variations thereof, are intended to cover a non-exclusive inclusion. For example, a process, method, apparatus, article, or article that comprises a list of steps or elements is not limited to only those listed but may optionally include other steps or elements not listed or inherent to such process, method, article, or article.

Reference herein to "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment may be included in at least one embodiment of the invention. The appearances of such phrases in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments. Those of skill in the art will explicitly and implicitly appreciate that the embodiments described herein may be combined with other embodiments.

The invention discloses a method and a device for determining a cargo effect based on cargo video, which can accurately determine the contribution of each cargo video to the sales volume of commodities when a plurality of cargo video are used for selling the same commodities, namely accurately determining the cargo effect of each cargo video, and are beneficial to providing accurate reference basis for selecting proper cargo bloggers and/or video styles of cargo video. The following will describe in detail.

Example 1

Referring to fig. 1, fig. 1 is a flowchart of a method for determining a cargo effect based on a cargo video according to an embodiment of the present invention. The method described in fig. 1 may be applied to a cargo effect determining device, such as a server, and the embodiment of the invention is not limited thereto. As shown in fig. 1, the cargo effect determining method based on the cargo video may include the following operations:

101. The goods effect determining device obtains target data corresponding to all goods video of a certain target commodity in a certain target time period.

In the embodiment of the invention, the target data corresponding to the video with goods comprises data which has an association relationship with the sales volume of the target commodity in all the data corresponding to the video with goods. Further, the target data corresponding to the video with goods comprise data which have an association relation with the sales amount of the target commodity and influence degree of the sales amount of the target commodity on the target commodity is larger than or equal to a preset degree threshold value, so that the accuracy and the efficiency of the allocation weight value of the video with goods calculated according to the target data of the video with goods can be improved. The certain target time period is any time period selected by the related personnel, and the certain target commodity is any commodity selected by the related personnel.

Still further optionally, the target data corresponding to the video-in-band may include at least one of fan data of a video blogger of the video-in-band, an accumulated viewing amount of the video-in-band in a target time period, and a peak viewing amount of the video-in-band in the target time period.

Still further optionally, all live video of the target commodity for the target time period includes all live video of the target commodity for the target time period and/or all non-live video of the target commodity for the target time period and the video content includes all live video of the target commodity.

Still further alternatively, the fan data of the video blogger with the goods video may include one or more combinations of the number of fans of the video blogger, the age level of the fans of the video blogger, the number of fans of different sexes in the fans of the video blogger, the number of fans of different areas in the fans of the video blogger, the number of fans of different social attributes in the fans of the video blogger, and the like, and the social attribute may be one of middle school students, college students, social workers, retired people, and the like. Preferably, the fan data of the video bloggers with the goods video at least comprises the number of the fan of the video bloggers.

102. The loading effect determining device calculates an allocation weight value of each loading video according to the target data corresponding to each loading video.

As an alternative embodiment, the apparatus for determining a cargo effect calculates an allocation weight value of each cargo video according to the target data corresponding to each cargo video, which may include:

The belt cargo effect determining device calculates the sum of all data in the target data corresponding to each belt cargo video to obtain a calculation result (also called as a bottom layer weight value) corresponding to each belt cargo video;

The goods effect determining device executes conversion operation on the calculation result corresponding to each goods video according to a predetermined conversion formula to obtain a conversion result corresponding to each goods video;

The goods effect determining device determines a conversion result corresponding to each goods video as an allocation weight value of the goods video; or determining a weight value correction parameter corresponding to each video with goods, and correcting a conversion result corresponding to the video with goods according to the weight value correction parameter corresponding to each video with goods to obtain an allocation weight value of the video with goods.

In this alternative embodiment, for example, assuming that the number of each of the video clips is i, the target data corresponding to the video clip i includes the number of vermicelli F _i, the cumulative viewing amount is Q _i, and the viewing amount peak value M _i, the calculation result corresponding to the video clip i is: k _i＝F_i+Q_i+M_i.

In this alternative embodiment, the apparatus for determining a cargo effect calculates the sum of all data in the target data corresponding to each cargo video, and before obtaining the calculation result corresponding to each cargo video, the apparatus for determining a cargo effect may further perform the following operations:

For any of the above-mentioned video-in-charge, the video-in-charge effect determining means performs data conversion on data that does not satisfy the calculation order of magnitude among the target data corresponding to the video-in-charge, so that the data that does not satisfy the calculation order of magnitude is converted into data that satisfies the calculation order of magnitude.

In the alternative embodiment, as the data in the target data corresponding to the video with goods is various, before the summation calculation is performed, data conversion is performed on the data which does not meet the calculation order of magnitude in the target data so as to obtain the data which meets the calculation order of magnitude, and further the subsequent summation calculation is performed, so that the rationality and the accuracy of the calculated allocation weight value can be improved.

In this optional embodiment, further optionally, the apparatus for determining a loading effect may determine a weight value correction parameter corresponding to each loading video, which may include:

The on-load effect determining device determines a weight value correction parameter corresponding to each on-load video according to at least one of the video type, the delivery channel and the interval duration from the initial release time to the starting time of the target time period of the on-load video.

Still further alternatively, the above conversion formula is:

wherein k _i' is a conversion result corresponding to the ith video, k _i is a calculation result corresponding to the ith video, and i is a positive integer.

The conversion result corresponding to each video with goods is corrected according to the weight value correction parameter corresponding to each video with goods, and a calculation formula corresponding to the apportioned weight value of the video with goods is obtained as follows:

Wherein k _i' is an allocated weight value of the ith video with goods obtained after the conversion result is corrected, and P _i is a weight value correction parameter corresponding to the ith video with goods.

For example, when determining the weight correction parameter according to the delivery channel (i.e., delivery platform) of the video-in-stock, P _i may use the average 1 ten thousand fan price as the weight correction parameter of the i-th video-in-stock.

Therefore, in this optional implementation manner, the conversion operation needs to be performed on the calculation result corresponding to each of the cargo videos through the conversion formula, so as to solve the problems of difference caused by extreme data, basically ineffective low-end flow and low loyalty of the flow stars, compress the obvious difference caused by data like vermicelli, and is beneficial to improving the middle waist effect. In addition, because the influence of some factors can cause the difference of the feelings of different audience groups on the target commodity, the influence weight (namely the weight value correction parameter) can be determined according to the influence factors, and then the conversion result is corrected or adjusted so as to improve the accuracy of the finally determined shared weight value of the video with goods.

103. The goods-carrying effect determining device calculates sales contribution values of each goods-carrying video for the target goods in the target time period according to the allocation weight value of each goods-carrying video, the sum of the allocation weight values of all the goods-carrying videos and the obtained sales data of the target goods in the target time period.

Optionally, the sales contribution value of each of the video-in-band for the target commodity in the target time period may be represented by a percentage, or may be represented by corresponding sales data, or may be represented by a sales contribution level, which is not limited by the embodiment of the present invention.

In yet another alternative embodiment, after performing the finishing step 103, the method may further comprise the operations of:

The goods effect determining device sorts the calculated sales contribution value of each goods video for the target goods in the target time period, and sends the sorting result to the marketing thrower of the target goods; and/or the number of the groups of groups,

The video effect determining device screens a plurality of video with sales contribution values which are greater than or equal to preset sales contribution values according to the calculated sales contribution values of each video with goods for the target goods in the target time period, and counts one or more combinations of the types of video bloggers of the video with goods, the video styles of the video with goods, the video types of the video with goods, the delivery channels of the video with goods and the like.

Therefore, the optional embodiment can also automatically sort sales contribution values and send the sales contribution values to marketing players after calculating the sales contribution values of each video with goods for the target goods in the target time period, so that the marketing players can know the effects of different videos with goods marketing in time; in addition, one or more of the types of video bloggers of the plurality of video with goods, the video styles of the plurality of video with goods, the video types of the plurality of video with goods, the delivery channels of the plurality of video with goods and the like can be counted, so that reasonable reference basis is provided for other marketing dispensers who want to market goods through the video with goods.

In yet another alternative embodiment, the method may further comprise the operations of:

and the goods effect determining device judges whether a plurality of goods videos with the same video blogger exist in the goods videos, and when the judgment result is yes, calculates the sum of sales contribution values of all the goods videos with the same video blogger, which are brought by the goods videos with the same video blogger, for the target goods in the target time period, and takes the sum as the sales contribution value of the video blogger, which is brought by the goods to the target goods in the target time period.

Still further, the method may further comprise the operations of:

After calculating sales contribution values of video bloggers with a plurality of goods video with the same video blogger for target goods in the target time period, calculating the ratio of the sales contribution values of the video bloggers for the target goods in the target time period to the quantity of the goods video distributed by the video blogger in all the goods video, and taking the ratio as the average sales contribution value of single goods video of the video bloggers for the target goods in the target time period.

Therefore, the optional embodiment can also intelligently calculate the sales contribution value of the video blogger for the target commodity in the target time period, and further can intelligently calculate the average sales contribution value of the single video with the commodity for the target commodity in the target time period, so that an effective data basis is provided for searching a proper video blogger for commodity marketing.

The video effect determining device divides all the video with goods into video groups corresponding to different video types according to the video types of all the video with goods;

And for the video group corresponding to any video type, the video effect determining device divides all the video in the video group into a plurality of sub-video groups with different video styles according to the video styles, calculates the average sales contribution value of each sub-video group for the target commodity in the target time period, and obtains the average sales contribution value of the video with different video styles for the target commodity.

Wherein the average sales contribution value of each sub-commodity video set for the target commodity in the target time period is equal to the sum of sales contribution values of all commodity videos in the sub-commodity video set for the target commodity in the target time period divided by the number of all commodity videos in the sub-commodity video set.

It can be seen that the optional embodiment can also intelligently calculate average sales contribution values of the video with different video styles in different video types for the target commodity in the target time period, so as to provide effective reference for selecting proper video types and video styles in commodity marketing.

It can be seen that when the method described by the embodiment of the invention is implemented, the contribution of each video on the sales volume of the commodity can be accurately determined when a plurality of video bands sell the same commodity, that is, the video effect of each video is accurately determined, so that an accurate reference basis is provided for selecting proper video styles of the video bands and/or video bands.

Example two

Referring to fig. 2, fig. 2 is a flowchart of another method for determining a cargo effect based on a cargo video according to an embodiment of the invention. The method described in fig. 2 may be applied to a cargo effect determining device, such as a server, and the embodiment of the invention is not limited thereto. As shown in fig. 2, the cargo effect determining method based on the cargo video may include the following operations:

201. The goods effect determining device obtains target data corresponding to all goods video of a certain target commodity in a certain target time period.

202. The loading effect determining device calculates an allocation weight value of each loading video according to the target data corresponding to each loading video.

203. The loading effect determining device performs a preprocessing operation on the assigned weight values of all the loaded videos.

204. The loading effect determining means updates the allocation weight value of each of the loaded videos after the preprocessing operation is performed to the allocation weight value of each of the loaded videos.

Wherein the sum of the assigned weight values of all the video-in-stock after the preprocessing operation is performed is equal to a predetermined fixed value, such as 1.

In the embodiment of the invention, in order to make the sum of the allocation weight values of each of the goods-carrying videos be 1 and to facilitate the use of the proportion value in the subsequent calculation of the leveling effect, the allocation weight values of all the goods-carrying videos can be directly subjected to softmax processing to obtain the final allocation weight value, and the final allocation weight value is as follows:

k_fi＝softmax([k"₁,k"₂,k"₃,...,k"_n])；

Wherein k _fi is the final assigned weight value of the ith video, n is the total number of the video, and i is less than or equal to n.

205. The goods-carrying effect determining device calculates sales contribution values of each goods-carrying video for the target goods in the target time period according to the allocation weight value of each goods-carrying video, the sum of the allocation weight values of all the goods-carrying videos and the obtained sales data of the target goods in the target time period.

In the embodiment of the present invention, for other detailed descriptions of step 201, step 202 and step 205, please refer to the detailed descriptions of step 101-step 103 in the first embodiment, and the detailed descriptions of the embodiment of the present invention are omitted.

It can be seen that when the method described by the embodiment of the invention is implemented, the contribution of each video on the sales volume of the commodity can be accurately determined when a plurality of video bands sell the same commodity, that is, the video effect of each video is accurately determined, so that an accurate reference basis is provided for selecting proper video styles of the video bands and/or video bands. In addition, the preliminarily calculated split weight values can also be softmax processed to facilitate the use of the scale values in the subsequent calculation of the flattening effect (i.e., sales contribution value). And for the commodity video with the video type being the live broadcast type, the marketing effect brought by each commodity video with the commodity video for each new commodity video with the commodity video and the new commodity can be calculated by the method disclosed by the embodiment of the invention.

In an optional embodiment, in step 205, a specific process of calculating a sales contribution value of each of the video with goods in the target time period according to the allocation weight value of each of the video with goods, the sum of the allocation weight values of all of the video with goods, and the obtained sales data of the target goods in the target time period may be as shown in fig. 3, and fig. 3 is a schematic flow diagram of a method for calculating the sales contribution value of each of the video with goods in the target time period according to an embodiment of the present invention. As shown in fig. 3, the calculation method of sales contribution value of each of the video-in-stock for the target commodity in the target time period may include the following operations:

301. The goods-carrying effect determining device calculates sales volume increasing data of sub-time periods corresponding to the time intervals of the target time period from the starting time of the target time period according to the obtained sales volume data of the target time period from the starting time of the target time period at the time intervals of the target time period, the sales volume data of the starting time of the target time period and the sales volume data of the ending time of the target time period.

The sales volume increasing data of each sub-time period is equal to the sales volume data of the ending time of the sub-time period minus the sales volume data of the starting time of the sub-time period, and further, the calculated sales volume increasing data can be rounded up or rounded down. The overlapping of two sub-periods refers to the time crossing of the sub-periods, and does not include the overlapping of the ending time of one sub-period and the starting time of the next sub-period. The sub-time period corresponding to each target interval time from the starting time of the target time period is specifically a time period composed of the starting time of the target time period, each target time and the ending time of the target time period, namely, the starting time and the first target time compose a sub-time period, the second target time and the third target time compose a sub-time period, and so on until the last target time and the ending time of the target time period compose the sub-time period.

For example, assuming that the target time period is 09:00-09:10 at 31 am in 05 month 2020 and the target interval duration is 1min, the tape effect determining apparatus needs to acquire sales volume data of 09:00, sales volume data of 09:01, sales volume data of 09:02, sales volume data of 09:03, sales volume data of 09:04, sales volume data of 09:05, sales volume data of 09:06, sales volume data of 09:07, sales volume data of 09:08, sales volume data of 09:09, and sales volume data of 09:10, and sub-time periods corresponding to the target interval duration from the start time of the target time period are 09:00-09:01、09:01-09:02、09:02-09:03、09:03-09:04、09:04-09:05、09:05-09:06、09:06-09:07、09:07-09:08、09:08-09:09 and 09:09-09:10, respectively, wherein sales volume increase data of the sub-time period of 09:04-09:05 is the sales volume data of 09:05 minus the sales volume data of 09:05.

302. The goods-carrying effect determining device calculates sub-sales contribution values of each goods-carrying video for the target goods in each sub-time period according to the allocation weight value of each goods-carrying video, the sum of the allocation weight values of all the goods-carrying videos and calculated sales quantity increasing data of the target goods in each sub-time period.

In the embodiment of the invention, the sub-sales contribution value brought by each commodity-carrying video in each sub-time period for the target commodity is equal to the ratio of the sales volume increment data of the sub-time period multiplied by the sum of the allocation weight value of the commodity-carrying video and the allocation weight value of all the commodity-carrying videos. The calculation formula of the sub-sales contribution value brought by the ith commodity video for the target commodity in the jth sub-time period is as follows:

wherein D _ij is a sub-sales contribution value brought by the ith commodity video in the jth sub-time period for the target commodity, and S _j is a sales volume increment of the target commodity in the jth sub-time period.

303. The carry effect determining means calculates an accumulated value of sub-sales contribution values for the target commodity for each sub-period of time for each carry video as a sales contribution value for the target commodity for the carry video in the target period of time.

The calculation formula of the sub-sales contribution value brought by the ith commodity video for the target commodity in the target time period is as follows:

D_i＝D_i1+D_i2+...+D_ij+...D_im；

wherein m is the number or the number of sub-time periods included in the target time period.

Therefore, in the alternative embodiment, the sales contribution value of each of the video-in-band for the target commodity in the target time period can be calculated by calculating the accumulated value of the sub-sales contribution value of each of the video-in-band for the target commodity in each of the sub-time periods, so that the accuracy of the calculated sales contribution value of the video-in-band for the target commodity in the target time period is improved, and the shorter the target interval time is, the more accurate the calculated sales contribution value is.

Alternatively, the above step 302 may be replaced by the following steps, namely:

The goods effect determining device calculates sub-sales contribution values of each goods video for the target goods in each sub-time period according to the allocation weight value of each goods video, the sum of the allocation weight values of all the goods video, calculated sales quantity increasing data of the target goods in each sub-time period and the occurrence condition of the target goods in each sub-time period.

In practical application, which of the two embodiments is specifically adopted to calculate the sub-sales contribution value of each of the video with goods for the target goods in each sub-time period may be determined according to the target interval duration. If the target interval duration is shorter, the occurrence condition of the target commodity in each video with goods in each sub-time period does not need to be considered, namely, the first mode is used; if the target interval duration is longer, the occurrence of the target commodity in each video with goods in each sub-time period needs to be considered, namely, the second mode is used for calculation.

It can be seen that the optional embodiment can also consider the appearance of the target commodity in each sub-time period when calculating the sub-sales contribution value of each video with commodity for the target commodity in each sub-time period, which is beneficial to further improving the accuracy of the calculated sub-sales contribution value of each video with commodity for the target commodity in each sub-time period.

In another alternative embodiment, the method may further comprise the following operations, before performing step 205 described above:

the method comprises the steps that a cargo effect determining device obtains sales volume data of target commodities at each target time of every target interval duration from the starting time of a target time period;

the goods carrying effect determining device judges whether empty data time when sales volume data is empty exists in all target time according to the acquired sales volume data of all target time;

when judging that the empty data time exists in all the target time, triggering and executing the operation of calculating the sales contribution value of each video with goods for the target goods in the target time period according to the allocation weight value of each video with goods, the sum of the allocation weight values of all the videos with goods and the acquired sales data of the target goods in the target time period by the aid of the goods effect determining device.

Taking the target time period of 09:00-09:10 and the target interval duration of 1min as an example for illustration, if sales data of the starting time 09:00, the target time 09:01, the target time 09:02, the target time 09:03, the target time 09:04, the target time 09:05, the target time 09:06, the target time 09:07, the target time 09:08, the target time 09:09, and the target time 09:10 are all obtained, it can be determined that no empty data time exists in the starting time and the target time.

In this alternative embodiment, in the case of acquiring sales volume data through a crawler technology, there may be a problem of data loss in the middle, and before calculation, it is ensured that no empty time data with empty sales volume data exists in all target time points, so as to improve the accuracy of the calculated sales contribution value of each video with goods for the target commodity in the target time period.

In this further alternative embodiment, the method may further comprise the operations of:

When judging that at least one empty data moment exists in all the target moments, the goods-carrying effect determining device supplements sales volume data of each empty data moment and triggers and executes the operation of calculating sales contribution values of each goods-carrying video for the target goods in the target time period according to the allocation weight value of each goods-carrying video, the sum of the allocation weight values of all the goods-carrying videos and the acquired sales volume data of the target goods in the target time period.

Taking the target time period of 09:00-09:10 and the target interval duration of 1min as an example for illustration, if sales volume data of the target time period of 09:02, the target time period of 09:04, the target time period of 09:05, the target time period of 09:06, the target time period of 09:07, the target time period of 09:08 and the target time period of 09:09 are not obtained, the time periods are all referred to as empty data time periods, and when at least one empty data time period exists in all the target time periods, the cargo carrying effect determining device needs to supplement corresponding sales volume data for the empty data time periods first, and then performs subsequent calculation operations.

Therefore, in this optional embodiment, before calculating the sales contribution value of each of the video with goods for the target commodity in the target time period, whether the target time has the empty data time is determined, if so, the sales volume data of the empty data time needs to be supplemented, which is beneficial to improving the calculation accuracy and the calculation efficiency.

In this alternative embodiment, still further alternatively, the method may further comprise, prior to supplementing sales volume data for each empty data time instance, the operations of:

The cargo effect determining device determines all first type empty data time points from all empty data time points, wherein the first type empty data time points are not adjacent to any empty data time point except the first type empty data time points; or the tape effect determining means determines all the second type of null data instants from among all the null data instants, the second type of null data instants being adjacent to at least one null data instant other than the second type of null data instant.

Taking the target time period of 09:00-09:10 and the target interval duration of 1min as an example for illustration, if sales data of the target time period of 09:02, the target time period of 09:04, the target time period of 09:05, the target time period of 09:06, the target time period of 09:07, the target time period of 09:08 and the target time period of 09:09 are not acquired, all the time periods are referred to as null data time periods, wherein the target time period of 09:02 is not adjacent to any other null data time period, the first null data time period is classified, and the target time period of 09:04, the target time period of 09:05, the target time period of 09:06, the target time period of 09:07, the target time period of 09:08 and the target time period of 09:09 are adjacent to at least one null data time period, so that the first null data time period can be classified as the second null data time period.

Wherein the loading effect determining means supplements sales data for each empty data time, may include:

the tape effect determining device supplements sales data at each empty data time according to a data supplementing mode matched with the category of each empty data time.

It will be seen that in this alternative embodiment, after determining the category for each empty data instance, sales data may be supplemented in accordance with the category-matched data supplementation.

The category of the null data time is used to specifically indicate whether the null data time belongs to the first category of null data time or the second category of null data time.

In this another alternative embodiment, as an alternative embodiment, the loading effect determining apparatus supplements sales volume data of each empty data time according to a data supplementing manner matched with a category of each empty data time, and may include:

For any first type of empty data time, the cargo-carrying effect determining device sequentially obtains first sales volume data of each preceding non-empty data time in a first preset number of preceding non-empty data time and second sales volume data of each following non-empty data time in a second preset number of following non-empty data time after the first type of empty data time from all target time in order of short and long time from the first type of empty data time;

the tape effect determining means supplements sales volume data at the first type of empty data time based on first sales volume data at each preceding non-empty data time and second sales volume data at each following non-empty data time.

In this alternative embodiment, the first preset number may be equal to 1 and the second preset number may be equal to 1. The loading effect determining device supplements sales volume data of first class empty data time according to first sales volume data of each preceding non-empty data time and second sales volume data of each following non-empty data time, and may include:

The goods carrying effect determining device calculates the average value of the second sales volume data at the later non-empty data time and the first sales volume data at the earlier non-empty data time;

When the average value is an integer, the cargo effect determining device supplements sales volume data of the first class of empty data time as the average value;

When the average value is a non-integer, the sales volume data at the first type of empty data time is supplemented by the tape effect determining device as an up-rounding value or a down-rounding value of the average value.

After calculating the average value, the loading effect determining apparatus may directly supplement the sales volume data at the first type of empty data time to the average value, without determining whether the average value is an integer.

Therefore, when the sales volume data is supplemented for the first kind of empty data moment, the alternative embodiment can directly supplement the sales volume data for the first kind of empty data moment according to the average value of the sales volume data of the adjacent preceding non-empty data moment and the adjacent following non-empty data moment, so that the error is smaller, and the data supplementing speed is high.

The cargo effect determining device divides all second-class empty data time into a plurality of time sets, any two second-class empty data time points included in the same time set are directly adjacent or indirectly adjacent through at least one second-class empty data time point, and any two second-class empty data time points in different time sets are not directly adjacent and are not indirectly adjacent through the second-class empty data time points;

for any time set, the cargo effect determining device determines the target number of sub-time periods included in the time set, the minimum second-class empty data time and the maximum second-class empty data time in the time set, and obtains first sales volume data of the prior non-empty data time closest to the minimum second-class empty data time and second sales volume data of the subsequent non-empty data time closest to the maximum second-class empty data time from all the target time, wherein the target number is equal to the number of the second-class empty data time included in the time set plus 1;

the tape effect determining means calculates a difference between the second sales volume data and the first sales volume data, and supplements sales volume data at each second class empty data time included in the time set based on the second sales volume data, the first sales volume data, the target quantity, and the difference.

The sub-time period included in the time set consists of a sub-time period formed by all directly adjacent second-class empty data time instants in the time set, a sub-time period formed by a previous non-empty data time instant and a minimum second-class empty data time instant in the time set, and a sub-time period formed by a later non-empty data time instant and a maximum second-class empty data time instant in the time set.

It should be noted that indirect adjacency refers to indirect adjacency through one or more second type null data moments, and if any two second type null data moments are not indirectly adjacency, it means that the two second type null data moments cannot be indirectly adjacency through one or more second type null data moments. Taking the example of the target time period being 9:00-09:10 and the target interval duration being 1min as an illustration, if sales data of the target time points 09:02, 09:03, 09:05, 09:06, 09:07, 09:08 and 09:09 are not obtained, these time points are all referred to as null data time points and all belong to the second type of null data time points. And all the second type of null data time can be divided into two time sets, wherein the first time set comprises a target time 09:02 and a target time 09:03, the target time comprised by the first time set is directly adjacent, the second time set comprises a target time 09:05, a target time 09:06, a target time 09:07, a target time 09:08 and a target time 09:09, any two target times comprised by the second time set are directly adjacent or indirectly adjacent, and as the first time set and the second time set are separated by a non-null data time 09:04, any one target time in the first time set is not directly adjacent or indirectly adjacent to any one target time in the second time set.

In this another alternative embodiment, as an alternative embodiment, the loading effect determining apparatus supplements sales volume data of each second type of empty data time included in the time set according to the second sales volume data, the first sales volume data, the target quantity, and the difference value, and may include:

And the goods carrying effect determining device is used for obtaining sales volume data of each second type of empty data moment in the moment set according to the second sales volume data and the first sales volume data and according to average value and average value to each second type of empty data moment included in the moment set.

The sub-time periods included in each time set are sequentially composed of a preceding non-null data time moment closest to the minimum second-class null data time moment of the time set, each second-class null data time moment and a following non-null data time moment closest to the maximum second-class null data time moment of the time set, and each sub-time period has no intersection or overlapping of other times except for the coincidence of one time end point, and the time length of each sub-time period is equal to the target interval duration.

Specifically, the alternative embodiment first calculates a sales volume increment average value of each sub-time period in the time set, where the sales volume increment average value is equal to a difference between sales volume data at a second time point (i.e., a last non-empty data time point closest to a maximum second empty data time point of the time set) and sales volume data at a first time point (i.e., a previous non-empty data time point closest to the minimum second empty data time point of the time set) divided by a number of sub-time periods included in the time set. In practical application, because there is no less than one sales volume increment, the calculated sales volume increment average value needs to be estimated as an integer, that is, the sales volume increment average value of sub-time periods corresponding to each second type of empty data time instant except the maximum second type of empty data time instant (or the minimum second type of empty data time instant) in the time instant set is sequentially estimated as an integer in a rounding manner, then the sub-time periods formed by the maximum second type of empty data time instant (or the minimum second type of empty data time instant) and the second time instant (or the first time instant) in the time instant set bear errors so as to ensure that the sales volume increment and the difference value between the sales volume data of the second time instant and the sales volume data of the first time instant keep consistent, or, the average value of sales volume increment of sub-time periods corresponding to each second type of empty data time except the maximum second type of empty data time (or the minimum second type of empty data time) in the time set can be estimated as an integer in turn according to a mode of cross rounding up and down, and then errors are accepted by sub-time periods formed by the maximum second type of empty data time (or the minimum second type of empty data time) and the second time point (or the first time point) in the time set, so that the difference value between sales volume increment and sales volume data of the second time point and sales volume data of the first time period is kept consistent.

For example, a certain target time period or a certain time set of the target time period of 02 month 01 of 2020 is taken as an example, wherein the starting time is 00:01, the ending time is 00:06, and the time interval duration is 1min, and the number obtained by the crawler technology is as follows:

{ time: 2020-02-01:01, sales: 1};

{ time: 2020-02-01:04, sales: 8};

{ time: 2020-02-01 00:05, sales: 9};

{ time: 2020-02-01:06, sales: 12}.

The idea of carrying out data supplementation by adopting the mean value flattening and rounding mode is as follows:

Determining that the blank data time is 00:02 and 00:03, and determining that the two time points are 00:01 and 00:04 respectively, wherein the corresponding sub-time periods are 00:01-00:02, 00:02-00:03 and 00:03-00:04 respectively, namely corresponding to 3 sub-time periods;

Calculating the sales volume increment of 7 in the 00:01-00:04 time period according to the sales volume of 00:01 and the sales volume of 00:04, increasing by 2.33 sales volumes per minute according to the average value, wherein the sales volume is not required to be less than one, the monitoring granularity is the minute, the increased sales volume is required to be estimated as an integer according to a rounding rule, the increment of 3 sub-time periods is [2, 2 and 2], and the calculated sales volume increment in the 00:01-00:04 time period is 6, which is less than the actual sales volume increment by 1, and the difference is required to be supplemented in a certain sub-time period; according to the concept of inertia economy, the sales increase trend should steadily or continuously decrease and increase, and the middle sudden decrease or sudden increase is not reasonable, so the difference value is selected to be compensated in the last sub-time period, namely, the increment of 3 sub-time periods is [2, 2 and 3];

after determining the increment of 3 sub-time periods as [2, 3], the empty data time requiring the sales volume data supplementation can be subjected to data supplementation, and the supplementation result is as follows:

{ time: 2020-02-01:02, sales: 3},

{ Time 2020-02-01:03, sales: 5}.

In this another alternative embodiment, as another alternative embodiment, the tape effect determining apparatus supplements sales volume data of each second type of empty data time included in the time set according to the second sales volume data, the first sales volume data, the target quantity, and the difference value, and may include:

Dividing the determined mother time set into a front time set and a rear time set corresponding to the mother time set according to a dichotomy, calculating the sales volume increment of the front time set and the sales volume increment of the rear time set corresponding to the mother time set according to the sales volume increment of the mother time set and the allocation proportion corresponding to the mother time set, caching the sales volume increment of the sub time sets of which the number of sub time periods included in the front time set and the rear time set corresponding to the mother time set meets the preset number condition, and judging whether the sub time sets of which the number of sub time periods included in the front time set and the rear time set does not meet the preset number condition exist or not;

When the judgment result is negative, the goods carrying effect determining device supplements sales volume data of each second type of empty data moment included in the moment set according to the second sales volume data, the first sales volume data and sales volume increment of all cached time moment sets, wherein the number of the time periods included in the second sales volume data meets the preset number condition;

And when the judgment result is yes, the cargo effect determining device determines a sub-time set, of which the number of sub-time periods included in the previous time set and the next time set corresponding to the mother time set does not meet the preset number condition, as a new mother time set, repeatedly executes the operation of dividing the determined mother time set into the previous time set and the next time set corresponding to the mother time set according to the bisection method, calculating the sales volume increment of the previous time set and the sales volume increment of the next time set corresponding to the mother time set according to the sales volume increment of the mother time set and the allocation proportion corresponding to the mother time set, caching the sales volume increment of the sub-time set, of which the number of sub-time periods included in the previous time set and the next time set corresponding to the mother time set meets the preset number condition, and judging whether the number of sub-time periods included in the previous time set and the next time set corresponding to the mother time set does not meet the preset number condition.

The initial determined mother time set is the time set, and the sales volume increment of the time set is equal to the difference value.

In this alternative embodiment, the number of sub-time periods included in the set of mother time instances may be considered when dividing the set of mother time instances into a preceding set of time instances and a following set of time instances corresponding to the set of mother time instances. If the number is even, the number of the sub-time periods included in the previous time set is consistent with the number of the sub-time periods included in the later time set, namely the mother time set is equally divided into two; if the number is odd, the number of sub-time periods included in the previous time instance set may be 1 or 1 less than the number of sub-time periods included in the subsequent time instance set. The sales volume increment corresponding to the master time set is equal to the sales volume data of the maximum time of the sub time period included in the master time set minus the sales volume data of the minimum time of the sub time period included in the master time set, wherein the maximum time of the sub time period included in the master time set is the adjacent non-empty data time of the maximum time included in the master time set, and the minimum time of the sub time period included in the master time set is the adjacent non-empty data time of the minimum time included in the master time set.

For example, if the null data time included in the master time set is 00:01, 00:02, 00:03, and 00:04, the number of sub-time periods included in the master time set is 5, and is 00:00-00:01, 00:01-00:02, 00:02-00:03, 00:03-00:04, and 00:04-00:05, respectively, so the sub-time period included in the previous time set corresponding to the master time set may be 00:00-00:01, 00:01-00:02, the sub-time period included in the next time set corresponding to the master time set may be 00:02-00:03, 00:03-00:04, and 00:04-00:05, or the sub-time period included in the previous time set corresponding to the master time set may be 00:00-00:01, 00:01-00:02, and 00:02-00:03, and the sub-00:03:03.

For example, if the null data time included in the master time set is 00:01, 00:02, and 00:03, the number of sub-time periods included in the master time set is 4, and is 00:00-00:01, 00:01-00:02, 00:02-00:03, and 00:03-00:04, respectively, so that the sub-time period included in the previous time set corresponding to the master time set may be 00:00-00:01, and 00:01-00:02, and the sub-time period included in the later time set corresponding to the master time set may be 00:02-00:03, and 00:03-00:04.

In this alternative embodiment, the allocation proportion corresponding to the master time set includes a first allocation proportion of a preceding time set and a second allocation proportion of a subsequent time set, where the first allocation proportion of the preceding time set is equal to a number of sub-periods included in the preceding time set to a total number of sub-periods included in the master time set, the second allocation proportion of the subsequent time set is equal to a number of sub-periods included in the subsequent time set to a total number of sub-periods included in the master time set, and the sales volume increment of the preceding time set corresponding to the master time set is determined according to a product of the first allocation proportion and the sales volume increment of the master time set, and the sales volume increment of the subsequent time set corresponding to the master time set is determined according to a product of the second allocation proportion and the sales volume increment of the master time set. When the product of one of the allocation ratios and the sales volume increment of the mother time set is not an integer, the product of the allocation ratio and the sales volume increment of the mother time set needs to be rounded in one of up, down and rounding modes, after the product of the allocation ratio and the sales volume increment of the mother time set is rounded to obtain the corresponding sales volume increment, then a substitute value of the product of the other allocation ratio and the sales volume increment of the mother time set can be obtained according to the difference between the sales volume increment of the mother time set and the sales volume increment obtained after rounding, and the substitute value is an integer.

Therefore, the alternative implementation mode can realize the supplementation of the sales volume data at the moment of empty data in a dichotomy mode, compared with a mean value sharing mode, the method can reduce the occurrence of data storm or storm drop in the last error receiving sub-time period, can enable the supplemented sales volume data to accord with the increasing trend of the sales volume data, and improves the rationality and reliability of the supplemented sales volume data. The advantages of the implementation of the dichotomy over the implementation of the mean value sharing will be described below by way of an example.

The data obtained from the crawler data is described as [ (2020-02-01:00, 0) ] (2020-02-02:00, 15129). The empty data time when the data needs to be replenished is the whole minute time of the time period 2020-02-01:01-2020-02-01-23:59. If the sales volume increment of the sub-time period included in 2020-02-01:00-2020-02-01-23:59 is 10 according to the rounding mode of combining mean value allocation with rounding, the sales volume increment of 2020-02-01:59-2020-02-02 00:00 is in a surge 719; if the method is implemented by the dichotomy, the method is divided into two time periods of [ 2020-02-01:00, 2020-02-01:12:00) and [ 2020-02-01:12:00, 2020-02-02:00) firstly, and the sales volume increment of [ 2020-02-01:00, 2020-02-01:12:00) is as follows according to the average value average and the last interval filling difference value principle: 7564, continuing to apportion to know that the sales volume increment of each sub-time period tends to be stable.

In yet another alternative embodiment, the loading effect determining apparatus may perform the following operations before supplementing sales volume data at each empty data time:

the goods effect determining device performs data fitting operation on sales volume data of all non-empty data moments of all target moments except all empty data moments, and obtains a sales volume curve corresponding to the target time period, wherein the sales volume curve is used for representing the corresponding relation between the moments and the sales volume data in the target time period.

At this time, the shipment effect determination means supplements sales volume data for each empty data time, and may include:

The cargo effect determining device substitutes each empty data moment into the sales volume curve in turn to obtain sales volume data of each empty data moment.

It can be seen that the further alternative embodiment can also supplement sales volume data at each empty data moment in a curve fitting manner, which is beneficial to improving the matching degree of the sales volume data after supplement and the sales volume increasing trend, and can improve the supplementation efficiency of the sales volume data at the empty data moment compared with a bisection method and a mean average.

Example III

Referring to fig. 4, fig. 4 is a schematic structural diagram of a cargo effect determining device based on cargo video according to an embodiment of the present invention. As shown in fig. 4, the cargo effect determining apparatus based on cargo video may include:

The obtaining module 401 is configured to obtain target data corresponding to all the video with goods of a certain target commodity in a certain target time period, where the target data corresponding to the video with goods includes data, which has an association relationship with sales of the target commodity, in all the data corresponding to the video with goods.

A first calculation module 402 is configured to calculate an allocation weight value of each of the video in charge according to the target data corresponding to each of the video in charge.

The second calculating module 403 is configured to calculate a sales contribution value of each of the video with goods for the target goods in the target time period according to the allocated weight value of each of the video with goods, the sum of the allocated weight values of all of the video with goods, and the obtained sales volume data of the target goods in the target time period.

Optionally, the target data corresponding to the video with goods includes at least one of vermicelli data of a video blogger of the video with goods, cumulative viewing amount of the video with goods in the target time period, and viewing amount peak value of the video with goods in the target time period.

Optionally, all the live video of the target commodity in the target time period includes all live video of the target commodity in the target time period and/or all the non-live video of the target commodity in the target time period.

It can be seen that implementing the pickup effect determining apparatus described in fig. 4 can accurately determine the contribution of each pickup video to the sales volume of the same commodity when a plurality of pickup videos sell the same commodity, that is, accurately determine the pickup effect of each pickup video, which is beneficial to providing an accurate reference for selecting a proper pickup blog and/or video style of the pickup video.

In an alternative embodiment, as shown in fig. 5, the first computing module 402 may include:

The calculating submodule 4021 is configured to calculate a sum of all data in the target data corresponding to each of the video with goods, and obtain a calculation result corresponding to each of the video with goods.

The conversion submodule 4022 is configured to perform a conversion operation on the calculation result corresponding to each of the video with goods according to a predetermined conversion formula, so as to obtain a conversion result corresponding to each of the video with goods.

The determining submodule 4023 is configured to determine a conversion result corresponding to each of the video with goods as an allocated weight value of the video with goods, or determine a weight value correction parameter corresponding to each of the video with goods, and correct the conversion result corresponding to the video with goods according to the weight value correction parameter corresponding to each of the video with goods, so as to obtain the allocated weight value of the video with goods.

In this optional embodiment, further optionally, the specific manner in which the determining submodule 4023 determines the weight value correction parameter corresponding to each of the video under load is:

and determining a weight value correction parameter corresponding to each video with goods according to at least one of the video type, the delivery channel and the interval duration from the initial release time to the starting time of the target time period of each video with goods.

In another alternative embodiment, the above conversion formula is:

Therefore, the implementation of the cargo effect determining apparatus described in fig. 5 also needs to perform a conversion operation on the calculation result corresponding to each cargo video through a conversion formula, so as to process the problems of difference caused by extreme data, basically no effect on low-end flow and low loyalty of flow stars, compress the obvious difference caused by data like vermicelli, and is beneficial to improving the middle waist effect. In addition, because the influence of some factors can cause the difference of the feelings of different audience groups on the target commodity, the influence weight (namely the weight value correction parameter) can be determined according to the influence factors, and then the conversion result is corrected or adjusted so as to improve the accuracy of the finally determined shared weight value of the video with goods.

In yet another alternative embodiment, as shown in fig. 5, the apparatus may further include:

the preprocessing module 404 is configured to perform a preprocessing operation on the assigned weight values of all the video under charge after the first computing module 402 computes the assigned weight value of each video under charge according to the target data corresponding to each video under charge.

An updating module 405, configured to update the assigned weight value of each of the video with goods after the preprocessing operation is performed to the assigned weight value of each of the video with goods.

Wherein the sum of the assigned weight values of all the video-in-stock after the preprocessing operation is performed is equal to a predetermined fixed value, e.g. 1.

In this alternative embodiment, in order to make the sum of the allocation weight values of each of the cargo videos be 1 and to facilitate the subsequent calculation of the leveling effect, the allocation weight values of all the cargo videos may be directly subjected to softmax processing, so as to obtain a final allocation weight value, where the final allocation weight value is:

k_fi＝softmax([k"₁,k"₂,k"₃,...,k"_n])；

Wherein k _fi is the final allocation weight value of the ith video, n is the total number of all the videos, and i is less than or equal to n.

It can be seen that this alternative embodiment is also capable of softmax processing the initially calculated split weight values in order to facilitate the use of the scaling values in the subsequent calculation of the split effect (i.e. sales contribution value). And for the video type is live video, the marketing effect brought by each video with goods for the goods marketed by the video with goods can be calculated for each new video with goods and new goods by the method disclosed by the optional embodiment.

In yet another alternative embodiment, the second calculating module 403 calculates the sales contribution value of each of the video with goods for the target time period according to the allocated weight value of each of the video with goods, the sum of the allocated weight values of all of the video with goods, and the obtained sales data of the target goods for the target time period, which is specifically as follows:

calculating sales volume increase data of sub-time periods corresponding to the target time periods from the starting moment of the target time period every other target interval duration according to the acquired sales volume data of the target commodity at the target time periods from the starting moment of the target time period every other target interval duration, the starting moment of the target time period and the sales volume data of the ending moment of the target time period, wherein any two sub-time periods are not overlapped;

Calculating a sub-sales contribution value of each commodity-carrying video for the target commodity in each sub-time period according to the allocation weight value of each commodity-carrying video, the sum of the allocation weight values of all the commodity-carrying videos and the calculated sales quantity increase data of the target commodity in each sub-time period; or calculating sub-sales contribution values of each commodity-carrying video for the target commodity in each sub-time period according to the allocation weight value of each commodity-carrying video, the sum of the allocation weight values of all the commodity-carrying videos, calculated sales quantity increase data of the target commodity in each sub-time period and the occurrence condition of the target commodity in each sub-time period;

and calculating an accumulated value of sub-sales contribution values of each commodity-carrying video for the target commodity in each sub-time period, and taking the accumulated value as the sales contribution value of the commodity-carrying video for the target commodity in the target time period.

In this alternative embodiment, the sub-sales contribution value for each sub-period of time for the target commodity for each of the video offerings is equal to the ratio of the sales volume increment data for that sub-period of time multiplied by the sum of the assigned weight value for that video offerings and the assigned weight value for all video offerings. The calculation formula of the sub-sales contribution value brought by the ith commodity video for the target commodity in the jth sub-time period is as follows:

D_i＝D_i1+D_i2+...+D_ij+...D_im；

It can be seen that this alternative embodiment is further capable of calculating, by calculating the accumulated value of the sub-sales contribution value of each of the video-in-band for the target commodity in each sub-time period, the sales contribution value of the video-in-band for the target commodity in the target time period, so that the accuracy of the calculated sales contribution value of the video-in-band for the target commodity in the target time period is improved, and the shorter the above-mentioned target interval duration is, the more accurate the calculated sales contribution value is.

In yet another alternative embodiment, the obtaining module 401 may be further configured to obtain sales data of the target commodity at each target time interval of a target duration from a starting time of the target time period, and as shown in fig. 5, the apparatus may further include:

The judging module 406 is configured to judge whether there is an empty data time when the sales volume data is empty in all the target time according to the obtained sales volume data of all the target time, and when it is judged that there is no empty data time in all the target time, trigger the second calculating module 403 to perform an operation of calculating a sales contribution value of each of the video with goods for the target goods in the target time period according to the allocation weight value of each of the video with goods, the sum of the allocation weight values of all the video with goods, and the obtained sales volume data of the target goods in the target time period.

It can be seen that this alternative embodiment can also ensure that there is no empty time data with empty sales data in all the target time before calculation, so as to improve the accuracy of the calculated sales contribution value of each of the video-in-band for the target commodity in the target time period.

and the supplementing module 407 is configured to supplement sales data of each empty data time when the judging module 406 judges that at least one empty data time exists in all the target time, and trigger the second calculating module 403 to perform the above-mentioned operations according to the allocation weight value of each of the video with goods, the sum of the allocation weight values of all the video with goods, and the acquired sales data of the target goods in the target time period, and calculate sales contribution value of each of the video with goods for the target goods in the target time period.

Therefore, the optional embodiment can judge whether the empty data moment exists in the target moment before calculating the sales contribution value of each goods-carrying video for the target goods in the target time period, and if the empty data moment exists, the sales volume data of the empty data moment needs to be supplemented, so that the calculation accuracy and the calculation efficiency are improved.

A determining module 408, configured to determine all first type null data moments from all null data moments, where the first type null data moment is not adjacent to any null data moment except the first type null data moment; or determining all second-class null data instants from all null data instants, the second-class null data instants being adjacent to at least one null data instant other than the second-class null data instant.

The specific way for the supplementing module 407 to supplement sales data at each empty data time is:

And supplementing sales data of each empty data moment according to a data supplementing mode matched with the category of each empty data moment.

It will be appreciated that implementing this alternative embodiment enables sales data to be replenished in accordance with the category-matched credit data replenishment mode after the category for each empty data time is determined.

Optionally, the supplementing module 407 supplements sales data of each empty data moment according to a data supplementing mode matched with the category of each empty data moment, which may include:

for any first type of null data time, sequentially acquiring first sales volume data of each preceding non-null data time in a first preset number of preceding non-null data time and second sales volume data of each following non-null data time in a second preset number of following non-null data time in the first type of null data time from all target time according to the sequence of short and long time lengths from the first type of null data time;

and supplementing sales data of the first type of null data moment according to the first sales data of each preceding non-null data moment and the second sales data of each following non-null data moment.

Further optionally, the first preset number is equal to 1 and the second preset number is equal to 1;

The specific way for the supplementing module 407 to supplement sales volume data of the first kind of null data time according to the first sales volume data of each preceding non-null data time and the second sales volume data of each following non-null data time is:

Calculating an average value of the second sales volume data at the later non-empty data time and the first sales volume data at the earlier non-empty data time;

When the average value is an integer, supplementing sales volume data of the first type of empty data time as the average value;

and when the average value is a non-integer, supplementing sales data at the moment of the first type of empty data as an upward rounding value or a downward rounding value of the average value.

It can be seen that, in this alternative embodiment, when sales volume data is further added to the first type of null data, sales volume data at the first type of null data can be directly added according to the average value of sales volume data at the adjacent preceding non-null data time and the adjacent following non-null data time, so that errors are small, and data adding efficiency is high.

In yet another alternative embodiment, the supplementing module 407 supplements sales data at each empty data time according to a data supplementing manner matched with the category of each empty data time, which is specifically:

Dividing all second-class empty data time instants into a plurality of time instant sets, wherein any two second-class empty data time instants included in the same time instant set are directly adjacent or indirectly adjacent through at least one second-class empty data time instant, and any two second-class empty data time instants in different time instant sets are not directly adjacent and are not indirectly adjacent through the second-class empty data time instant;

For any time set, determining the target number of sub-time periods included in the time set, the minimum second-class empty data time and the maximum second-class empty data time in the time set, and acquiring first sales volume data of a preceding non-empty data time closest to the minimum second-class empty data time and second sales volume data of a following non-empty data time closest to the maximum second-class empty data time from all target time;

Calculating the difference value between the second sales volume data and the first sales volume data, and supplementing sales volume data of each second class of empty data moment included in the moment set according to the second sales volume data, the first sales volume data, the target quantity and the difference value;

The sub-time period included in the time set consists of a sub-time period formed by all two directly adjacent second-class empty data time points in the time set, a sub-time period formed by the previous non-empty data time point and the minimum second-class empty data time point, and a sub-time period formed by the latter non-empty data time point and the maximum second-class empty data time point.

In this alternative embodiment, the supplementing module 407 supplements the sales volume data of each second type of empty data time included in the time set according to the second sales volume data, the first sales volume data, the target quantity, and the difference value in a specific manner that:

determining the sales volume increment of each sub-time period included in the time set according to the result of dividing the difference value by the target quantity;

And supplementing sales volume data of each second class of empty data moment included in the moment set according to the sales volume increment of each sub-time period included in the second sales volume data, the first sales volume data and the moment set.

Therefore, the optional embodiment also provides a mean average sales volume data supplementing mode, which can supplement sales volume data corresponding to empty data time, and is further beneficial to subsequent calculation of sales contribution values.

In yet another alternative embodiment, the supplementing module 407 supplements sales volume data of each second type of empty data time included in the time set according to the second sales volume data, the first sales volume data, the target quantity, and the difference value in a specific manner that:

Dividing the determined mother time set into a front time set and a rear time set corresponding to the mother time set according to a dichotomy, calculating the sales volume increment of the front time set and the sales volume increment of the rear time set corresponding to the mother time set according to the sales volume increment of the mother time set and the allocation proportion corresponding to the mother time set, caching the sales volume increment of the sub time sets of which the number of sub time periods included in the front time set and the rear time set corresponding to the mother time set meets the preset number condition, and judging whether the sub time sets of which the number of sub time periods included in the front time set and the rear time set corresponding to the mother time set does not meet the preset number condition exist or not;

When the judgment result is negative, supplementing sales volume data of each second class of empty data moment included in the moment set according to sales volume increment of all the moment sets, wherein the sales volume increment of all the moment sets is met by the quantity of the second sales volume data, the first sales volume data and the cached quantity of the included sub time periods meeting the preset quantity condition;

And when the judgment result is yes, determining a sub-time set, which is included in the previous time set and the next time set and corresponds to the mother time set, of which the number of sub-time periods does not meet the preset number condition as a new mother time set, repeatedly executing the operation of dividing the determined mother time set into the previous time set and the next time set, which are corresponding to the mother time set, according to the bisection method, calculating the sales volume increment of the previous time set and the sales volume increment of the next time set, which are corresponding to the mother time set, according to the sales volume increment of the mother time set and the allocation proportion, and caching the sales volume increment of the sub-time set, which is included in the previous time set and the next time set, of which the number of sub-time periods included in the previous time set and the next time set meets the preset number condition, and judging whether the sub-time set, which is included in the previous time set and the next time set, which does not meet the preset number condition, exists.

The number of the included sub-time periods meets the preset number condition, namely the number of the included sub-time periods is 1, and the number of the included sub-time periods does not meet the preset number condition, namely the number of the included sub-time periods is larger than 1.

The initial determined mother time set is a time set, and the sales increment of the time set is equal to the difference value.

Therefore, the optional embodiment can realize the supplementation of the data at the moment of empty data in a dichotomy mode, and compared with a mean value sharing mode, the method can reduce the occurrence of data storm or storm falling in the last error receiving sub-time period, can enable the supplemented sales volume data to accord with the increasing trend of the sales volume data, and improves the rationality and reliability of the supplemented sales volume data.

In yet another alternative embodiment, as shown in fig. 6, the apparatus may further include:

the data fitting module 409 is configured to perform a data fitting operation on sales volume data of all non-null data moments remaining except for all null data moments in all target moments, so as to obtain a sales volume curve corresponding to the target time period, where the sales volume curve is used to represent a correspondence between moments and sales volume data in the target time period.

In this alternative embodiment, the specific manner in which the replenishment module 407 supplements sales volume data for each empty data time instant is:

And substituting each empty data moment into the sales volume curve in turn to obtain sales volume data of each empty data moment.

Therefore, the alternative embodiment can also supplement sales volume data of each empty data moment in a curve fitting mode, which is beneficial to improving the matching degree of the supplemented sales volume data and sales volume increasing trend, and compared with a dichotomy and average value average, the supplementation efficiency of the sales volume data of the empty data moment can be improved.

Example IV

Referring to fig. 7, fig. 7 is a schematic structural diagram of a cargo effect determining apparatus based on cargo video according to another embodiment of the present invention. As shown in fig. 7, the apparatus may include:

a memory 501 in which executable program codes are stored;

a processor 502 coupled to the memory 501;

the processor 502 invokes executable program code stored in the memory 405 to perform the steps in the video-in-band effect determination method disclosed in the first or second embodiment of the present invention.

Example five

The embodiment of the invention discloses a computer storage medium which stores computer instructions for executing the steps in the method for determining the effect of a video-on-a-belt disclosed in the first embodiment or the second embodiment of the invention when the computer instructions are called.

The apparatus embodiments described above are merely illustrative, wherein the modules illustrated as separate components may or may not be physically separate, and the components shown as modules may or may not be physical, i.e., may be located in one place, or may be distributed over a plurality of network modules. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art will understand and implement the present invention without undue burden.

From the above detailed description of the embodiments, it will be apparent to those skilled in the art that the embodiments may be implemented by means of software plus necessary general hardware platforms, or of course by means of hardware. Based on such understanding, the foregoing technical solutions may be embodied essentially or in part in the form of a software product that may be stored in a computer-readable storage medium including read-only memory (ROM), random access memory (Random Access Memory, RAM), programmable read-only memory (Programmable Read-only memory, PROM), erasable programmable read-only memory (Erasable Programmable Read Only Memory, EPROM), one-time programmable read-only memory (OTPROM), electrically erasable programmable read-only memory (EEPROM), compact disc read-only memory (Compact Disc Read-only memory, CD-ROM) or other optical disc memory, magnetic disc memory, tape memory, or any other medium that can be used for computer-readable carrying or storing data.

Finally, it should be noted that: the embodiment of the invention discloses a method and a device for determining a cargo effect based on cargo video, which are disclosed as preferred embodiments of the invention, and are only used for illustrating the technical scheme of the invention, but not limiting the technical scheme; although the invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art will understand that; the technical scheme recorded in the various embodiments can be modified or part of technical features in the technical scheme can be replaced equivalently; such modifications and substitutions do not depart from the spirit and scope of the corresponding technical solutions.

Claims

1. A method for determining a cargo effect based on a cargo video, the method comprising:

Calculating a sales contribution value of each goods-carrying video for the target commodity in the target time period according to the allocation weight value of each goods-carrying video, the sum of the allocation weight values of all the goods-carrying videos and the acquired sales data of the target commodity in the target time period;

the calculating the allocation weight value of each video according to the target data corresponding to each video comprises the following steps:

calculating the sum of all data in the target data corresponding to each cargo video to obtain a calculation result corresponding to each cargo video;

performing conversion operation on the calculation result corresponding to each video with goods according to a predetermined conversion formula to obtain a conversion result corresponding to each video with goods;

determining a conversion result corresponding to each cargo video as an allocation weight value of the cargo video; or alternatively

And determining a weight value correction parameter corresponding to each video with goods, and correcting a conversion result corresponding to the video with goods according to the weight value correction parameter corresponding to each video with goods to obtain an allocation weight value of the video with goods.

2. The method for determining a video-on-demand effect according to claim 1, wherein the target data corresponding to the video-on-demand includes at least one of vermicelli data of a video blogger of the video-on-demand, an accumulated viewing amount of the video-on-demand in the target period, and a viewing amount peak of the video-on-demand in the target period;

all live video of the target commodity within the target time period includes all live video of the target commodity within the target time period and/or all non-live video of the target commodity is viewed within the target time period and video content includes all non-live video of the target commodity.

3. The method for determining a cargo effect based on a cargo video according to claim 1, wherein said determining a weight value correction parameter corresponding to each of said cargo videos comprises:

4. A method of determining a video-on-hold effect according to claim 1 or 3, wherein the conversion formula is:

5. The method of claim 4, wherein after calculating the assigned weight value for each of the video-in-band based on the target data corresponding to each of the video-in-band, the method further comprises:

performing preprocessing operation on the assigned weight values of all the cargo videos;

Updating the assigned weight value of each of the video-in-stock after the preprocessing operation is executed to be the assigned weight value of each of the video-in-stock;

wherein the sum of the assigned weight values of all the video-in-stock after the preprocessing operation is performed is equal to a predetermined fixed value.

6. The method according to any one of claims 1,2,3 and 5, wherein calculating a sales contribution value of each of the video in the target period based on an allocation weight value of each of the video, a sum of allocation weight values of all the video, and acquired sales volume data of the target commodity in the target period, comprises:

Calculating sales volume increase data of sub-time periods corresponding to the target interval duration from the starting moment of the target time period according to the acquired sales volume data of the target commodity at the target moment of the target interval duration from the starting moment of the target time period, the sales volume data of the starting moment and the sales volume data of the ending moment of the target time period, wherein any two sub-time periods are not overlapped;

calculating a sub-sales contribution value of each commodity-carrying video in each sub-time period according to the allocation weight value of each commodity-carrying video, the sum of the allocation weight values of all the commodity-carrying videos and the calculated sales quantity increase data of the target commodity in each sub-time period; or calculating a sub-sales contribution value of each goods-carrying video for the target goods in each sub-time period according to the allocation weight value of each goods-carrying video, the sum of the allocation weight values of all the goods-carrying videos, calculated sales quantity increase data of the target goods in each sub-time period and the occurrence condition of the target goods in each sub-time period of each goods-carrying video;

and calculating an accumulated value of sub-sales contribution values of each carried video for the target commodity in each sub-time period, and taking the accumulated value as the sales contribution value of the carried video for the target commodity in the target time period.

7. The method for determining a video-on-demand effect according to claim 6, wherein before calculating a sales contribution value of each of the video-on-demand for the target commodity in the target period based on an allocation weight value of each of the video-on-demand, a sum of allocation weight values of all the video-on-demand, and the acquired sales volume data of the target commodity in the target period, the method further comprises:

Acquiring sales volume data of the target commodity at each target time of every target interval duration from the starting time of the target time period;

Judging whether empty data time when sales volume data are empty exists in all the target time according to the acquired sales volume data of all the target time;

And when judging that the empty data time does not exist in all the target time, triggering and executing the operation of calculating sales contribution values of each video with goods for the target goods in the target time period according to the allocation weight value of each video with goods, the sum of the allocation weight values of all the video with goods and the acquired sales data of the target goods in the target time period.

8. The method of cargo effect determination based on cargo video of claim 7 further comprising:

And when judging that at least one empty data moment exists in all the target moments, supplementing sales volume data of each empty data moment, and triggering and executing the operation of calculating sales contribution values of each cargo video for the target commodity in the target time period according to the allocation weight value of each cargo video, the sum of the allocation weight values of all the cargo videos and the acquired sales volume data of the target commodity in the target time period.

9. The method for determining a video-on-demand effect based on a video-on-demand of claim 8, wherein prior to said supplementing sales volume data for each of said empty data moments, said method further comprises:

Determining all first-type null data time instants from all the null data time instants, wherein the first-type null data time instants are not adjacent to any null data time instant except the first-type null data time instant; or alternatively

Determining all second-class null data moments from all the null data moments, wherein the second-class null data moments are adjacent to at least one null data moment except the second-class null data moment;

Wherein said supplementing sales volume data for each of said null data moments comprises:

10. The method for determining a cargo effect based on a cargo video according to claim 9, wherein said supplementing sales volume data at each of said empty data time according to a data supplementing method matched with a category of each of said empty data time comprises:

For any first type of null data time, sequentially acquiring first sales volume data of each of a first preset number of preceding non-null data time before the first type of null data time and second sales volume data of each of a second preset number of following non-null data time after the first type of null data time from all target time according to the sequence of short and long time lengths from the first type of null data time;

And supplementing sales volume data of the first type of null data moment according to the first sales volume data of each preceding non-null data moment and the second sales volume data of each following non-null data moment.

11. The method of claim 10, wherein the first preset number is equal to 1 and the second preset number is equal to 1;

wherein the supplementing sales volume data of the first type of empty data time according to the first sales volume data of each preceding non-empty data time and the second sales volume data of each following non-empty data time comprises:

calculating the average value of the second sales volume data at the later non-empty data moment and the first sales volume data at the earlier non-empty data moment;

when the average value is an integer, supplementing sales volume data of the first type of empty data moment to the average value;

and when the average value is a non-integer, supplementing sales volume data of the first type of empty data time to be an upward rounding value or a downward rounding value of the average value.

12. The method for determining a cargo effect based on a cargo video according to claim 9, wherein said supplementing sales volume data at each of said empty data time according to a data supplementing method matched with a category of each of said empty data time comprises:

For any time set, determining a target number of sub-time periods included in the time set, a minimum second-class empty data time and a maximum second-class empty data time in the time set, and acquiring first sales volume data of a preceding non-empty data time closest to the minimum second-class empty data time and second sales volume data of a following non-empty data time closest to the maximum second-class empty data time from all the target times, wherein the target number is equal to the number of second-class empty data times included in the time set plus 1;

calculating a difference value between the second sales volume data and the first sales volume data, and supplementing sales volume data of each second class of empty data moment included in the moment set according to the second sales volume data, the first sales volume data, the target quantity and the difference value;

The sub-time period included in the time set is composed of a sub-time period formed by all directly adjacent two second-class empty data time instants in the time set, a sub-time period formed by the preceding non-empty data time instant and the minimum second-class empty data time instant, and a sub-time period formed by the following non-empty data time instant and the maximum second-class empty data time instant.

13. The method of claim 12, wherein supplementing sales volume data for each second type of empty data time instance included in the time instance set based on the second sales volume data, the first sales volume data, the target quantity, and the difference value comprises:

Determining sales volume increment of each sub-time period included in the time set according to the result of dividing the difference value by the target quantity;

And supplementing sales volume data of each second class of empty data moment included in the moment set according to the second sales volume data, the first sales volume data and sales volume increment of each sub-time period included in the moment set.

14. The method of claim 12, wherein supplementing sales volume data for each second type of empty data time instance included in the time instance set based on the second sales volume data, the first sales volume data, the target quantity, and the difference value comprises:

Dividing the determined mother time set into a front time set and a rear time set corresponding to the mother time set according to a dichotomy, calculating the sales volume increment of the front time set and the sales volume increment of the rear time set corresponding to the mother time set according to the sales volume increment of the mother time set and the allocation proportion corresponding to the mother time set, caching the sales volume increment of the sub-time set of which the number of sub-time periods included in the front time set and the rear time set corresponding to the mother time set meets the preset number condition, and judging whether the sub-time set of which the number of sub-time periods included in the front time set and the rear time set corresponding to the mother time set does not meet the preset number condition exists or not;

When the judgment result is negative, supplementing sales volume data of each second type of empty data moment included in the moment set according to the second sales volume data, the first sales volume data and the sales volume increment of all cached time sets, wherein the number of the time periods included in the second sales volume data meets the preset number condition;

When the judgment result is yes, determining a sub-time set, which is corresponding to the mother time set and is included in the back time set, of sub-time periods not meeting the preset number condition as a new mother time set, repeatedly executing the operation of dividing the determined mother time set into the front time set and the back time set corresponding to the mother time set according to a bisection method, calculating the sales volume increment of the front time set and the sales volume increment of the back time set corresponding to the mother time set according to the sales volume increment of the mother time set and the allocation proportion corresponding to the mother time set, caching the sales volume increment of the sub-time set, which is included in the front time set and the back time set corresponding to the mother time set, of the sub-time set, and judging whether the sub-time set, which is not meeting the preset number condition, exists in the front time set and the back time set corresponding to the mother time set;

15. The method for determining a video-on-demand effect based on a video-on-demand of claim 8, wherein prior to said supplementing sales volume data for each of said empty data moments, said method further comprises:

Performing data fitting operation on sales volume data of all non-empty data moments of all the target moments except all the empty data moments to obtain a sales volume curve corresponding to the target time period, wherein the sales volume curve is used for representing the corresponding relation between the moments and the sales volume data in the target time period;

substituting each empty data moment into the sales volume curve in turn to obtain sales volume data of each empty data moment.

16. A cargo effect determination device based on cargo video, the device comprising:

The second calculation module is used for calculating a sales contribution value of each goods-carrying video for the target goods in the target time period according to the allocation weight value of each goods-carrying video, the sum of the allocation weight values of all the goods-carrying videos and the acquired sales data of the target goods in the target time period;

Wherein the first computing module comprises:

The calculation sub-module is used for calculating the sum of all data in the target data corresponding to each video with goods to obtain a calculation result corresponding to each video with goods;

The conversion sub-module is used for executing conversion operation on the calculation result corresponding to each video with goods according to a predetermined conversion formula to obtain a conversion result corresponding to each video with goods;

The determining submodule is used for determining the conversion result corresponding to each video with goods as an allocation weight value of the video with goods; or determining a weight value correction parameter corresponding to each video with goods, and correcting a conversion result corresponding to the video with goods according to the weight value correction parameter corresponding to each video with goods to obtain an allocation weight value of the video with goods.

17. A cargo effect determination device based on cargo video, the device comprising:

A memory storing executable program code;

A processor coupled to the memory;

The processor invokes the executable program code stored in the memory to perform the video-in-band effect determination method of any of claims 1-15.

18. A computer storage medium storing computer instructions which, when invoked, are operable to perform the video-on-hold effect determination method of any one of claims 1 to 15.