CN116127131A

CN116127131A - Video generation method, device, equipment and storage medium

Info

Publication number: CN116127131A
Application number: CN202211543311.3A
Authority: CN
Inventors: 侯志强
Original assignee: Beijing Baidu Netcom Science and Technology Co Ltd
Current assignee: Beijing Baidu Netcom Science and Technology Co Ltd
Priority date: 2022-12-02
Filing date: 2022-12-02
Publication date: 2023-05-16

Abstract

The disclosure provides a video generation method, a device, equipment and a storage medium, relates to the technical field of computers, and particularly relates to a computer vision technology and a natural language processing technology. The specific implementation scheme is as follows: acquiring a target search statement searched by a user; carrying out semantic analysis on the target search statement, and judging whether the target search statement is matched with the element to be issued of the target search statement; the element to be distributed comprises a target commodity word corresponding to the target search statement, at least one target parameter item of the target commodity word and a parameter value of each target parameter item; and generating a target video for indicating the element to be distributed under the condition of matching the element to be distributed.

Description

Video generation method, device, equipment and storage medium

Technical Field

The present disclosure relates to the field of computer technologies, and in particular, to a computer vision technology and a natural language processing technology, and in particular, to a video generating method, apparatus, device, and storage medium.

Background

In the B2B (business to business) purchasing platform, merchants can market through issuing videos so as to reflect the business conditions, product details and product use effects of the merchants through the videos.

Disclosure of Invention

The disclosure provides a video generation method, a device, equipment and a storage medium, which are used for reducing the cost of video production in a B2B purchasing platform and improving the conversion of video in the B2B purchasing platform.

According to an aspect of the present disclosure, there is provided a video generating method including: acquiring a target search statement searched by a user;

carrying out semantic analysis on the target search statement, and judging whether the target search statement is matched with the element to be issued of the target search statement; the element to be distributed comprises a target commodity word corresponding to the target search statement, at least one target parameter item of the target commodity word and a parameter value of each target parameter item;

and generating a target video for indicating the element to be distributed under the condition of matching the element to be distributed.

According to another aspect of the present disclosure, there is provided a video generating apparatus including an acquisition unit, a judgment unit, and a generation unit; an acquisition unit for acquiring a target search statement searched by a user; the judging unit is used for carrying out semantic analysis on the target search statement and judging whether the element to be issued of the target search statement is matched or not; the element to be distributed comprises a target commodity word corresponding to the target search statement, at least one target parameter item of the target commodity word and a parameter value of each target parameter item; and the generation unit is used for generating a target video for indicating the element to be distributed under the condition of matching the element to be distributed.

According to another aspect of the present disclosure, there is provided an electronic device including: at least one processor; and a memory communicatively coupled to the at least one processor; wherein the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the video generation method provided by the present disclosure.

According to another aspect of the present disclosure, there is provided a non-transitory computer-readable storage medium storing computer instructions for causing a computer to perform the video generation method provided by the present disclosure.

According to another aspect of the present disclosure, there is provided a computer program product comprising a computer program which, when executed by a processor, implements the video generation method provided by the present disclosure.

According to the technical scheme provided by the disclosure, the target commodity words, the target parameter items and the parameter values which are required to be searched by the user can be determined based on the target search statement searched by the user, and the target video is generated according to the determined target commodity words, the target parameter items and the parameter values. Therefore, the video generation method can help the merchant to generate videos according to the requirements of users, and can reduce the cost of video production of the merchant while helping the merchant to improve video conversion.

It should be understood that the description in this section is not intended to identify key or critical features of the embodiments of the disclosure, nor is it intended to be used to limit the scope of the disclosure. Other features of the present disclosure will become apparent from the following specification.

Drawings

The drawings are for a better understanding of the present solution and are not to be construed as limiting the present disclosure. Wherein:

FIG. 1 is a schematic diagram of a video generation method according to an embodiment of the present disclosure;

FIG. 2 is a schematic diagram of one video generation method shown in accordance with an embodiment of the present disclosure;

FIG. 3 is a schematic diagram of a video generation system shown in accordance with an embodiment of the present disclosure;

FIG. 4 is a flow diagram illustrating a video generation method according to an embodiment of the present disclosure;

FIG. 5 is a flow diagram illustrating a video generation method according to an embodiment of the present disclosure;

FIG. 6 is a flow diagram illustrating a video generation method according to an embodiment of the present disclosure;

FIG. 7 is a flow diagram illustrating a video generation method according to an embodiment of the present disclosure;

FIG. 8 is a flow diagram illustrating a video generation method according to an embodiment of the present disclosure;

FIG. 9 is a flow diagram illustrating a video generation method according to an embodiment of the present disclosure;

FIG. 10 is a flow diagram illustrating a video generation method according to an embodiment of the present disclosure;

FIG. 11 is a flow diagram illustrating a video generation method according to an embodiment of the present disclosure;

FIG. 12 is a schematic diagram of a search results page shown in accordance with an embodiment of the present disclosure;

fig. 13 is a block diagram of a video generating apparatus for implementing a video generating method of an embodiment of the present disclosure;

fig. 14 is a block diagram of an electronic device for implementing a video generation method of an embodiment of the present disclosure.

Detailed Description

Exemplary embodiments of the present disclosure are described below in conjunction with the accompanying drawings, which include various details of the embodiments of the present disclosure to facilitate understanding, and should be considered as merely exemplary. Accordingly, one of ordinary skill in the art will recognize that various changes and modifications of the embodiments described herein can be made without departing from the scope and spirit of the present disclosure. Also, descriptions of well-known functions and constructions are omitted in the following description for clarity and conciseness.

In the technical scheme of the disclosure, the related processes of collecting, storing, using, processing, transmitting, providing, disclosing and the like of the personal information of the user accord with the regulations of related laws and regulations, and the public order colloquial is not violated.

First, the inventive principles of the embodiments of the present disclosure are described:

at present, a B2B purchasing platform is a platform for exchanging products, services and information between enterprises (merchants) and enterprises (users) through the internet, and in the B2B purchasing platform, the merchants can issue videos to conduct marketing so as to reflect business conditions, product details and product use effects of the merchants through the videos.

However, since the B2B purchasing platform is different from the common short video software, the video in the common short video software starts for the C-terminal user scene, and the video content is mostly based on freshness and interestingness, and cannot be applied to the B2B purchasing platform. The commodity in the B2B purchasing platform is generally mass production applied to industry, the commodity parameter items are more, the parameter value specificity is higher, and based on the commodity parameter items, a series of problems exist in the process of video production by merchants. For example, merchants do not know what video titles, content, or covers to make. Even if a merchant produces video, the merchant does not know how to produce quality video, so that the produced video is difficult to attract flow, and the conversion is low. If a merchant seeks a professional team to make video for it, this results in high costs.

Based on the above technical problems, as shown in fig. 1, the present disclosure considers that the search behavior value of a search sentence can be determined based on the portrait characteristics of a user, the number of searches of the search sentence, and the interactive behavior of the user to the search result of the search sentence, and at the same time, the associated value of the search sentence is determined based on the number, quality, number, and rank of associated videos of the search result, and the production value of the search sentence is determined based on the search behavior value and the associated value. Further, a target search term having the highest production value may be determined from the plurality of search terms based on the production value. As shown in fig. 2, after determining the target search sentence, it is determined whether the commodity word, the parameter item, and the parameter value corresponding to the target search sentence can be matched. If the matching is possible, generating a text to be issued according to the commodity word, the parameter item and the parameter value. Further, a target video for indicating the text to be distributed can be generated according to the text to be distributed. According to the technical scheme, the requirements of a user when using a product or consulting the product, such as information of commodity words, parameter items, parameter values and the like, and the video is generated based on the determined information of the commodity words, the parameter items, the parameter values and the like, so that an automatic video production method is provided for a merchant, the cost of the merchant is reduced, and meanwhile, the requirements of the user are held, the video can be produced in a targeted manner, the produced video can meet the requirements of the user, the flow of the merchant can be improved, and the conversion of the video is improved.

Next, description is made for an application scenario of the embodiment of the present disclosure:

fig. 3 is a schematic diagram of a structure of a video generating system for implementing a video generating method according to an embodiment of the present disclosure, and referring to fig. 3, the video generating system 10 includes an electronic device 11 and a video generating apparatus 12. The electronic device 11 is connected to the video generating apparatus 12, and the electronic device 11 and the video generating apparatus 12 may be connected by a wired system or may be connected by a wireless system, which is not limited in the embodiment of the present disclosure.

The electronic device 11 may be a server for providing searching and retrieving services for users in the B2B model. For example, the electronic device 11 may acquire a search term that the user searches for in a history period and transmit the search term to the video generating apparatus.

The video generating apparatus 12 is configured to determine, after receiving the search term sent by the electronic device 11, whether a preset product information mapping table exists in a target product word corresponding to the search term, at least one target parameter item of the target product word, and a parameter value of each target parameter item.

Further, after determining that the target commodity word, the at least one target parameter item, and the numerical value corresponding to the search statement exist in the commodity information mapping table, the video generating apparatus 12 is further configured to instruct the target video of the above information, and send the target video to the electronic device.

Accordingly, after receiving the target video, the electronic device 11 may display the target video in the search result corresponding to the search statement based on the search statement, so as to reduce the cost of the merchant and improve the flow of the merchant and the conversion of the target video.

The electronic device and the video generating apparatus may be physical machines, for example: desktop computers, also known as desktop computers or desktops (desktops), tablet computers, notebook computers, ultra-mobile personal computer (UMPC), netbooks, personal digital assistants (personal digital assistant, PDA), and other terminal devices. Meanwhile, the electronic device and the video generating apparatus may be servers, or may be a server group including a plurality of servers.

The electronic device 11 and the video generating apparatus 12 may be independent devices or may be integrated into the same device, which is not particularly limited in this disclosure.

When the electronic device 11 and the video generating apparatus 12 are integrated in the same device, the communication between the electronic device 11 and the video generating apparatus 12 is performed by communication between internal modules of the device. In this case, the communication flow therebetween is the same as "in the case where the electronic device 11 and the video generating apparatus 12 are independent of each other".

In the following embodiments provided in the present disclosure, the present disclosure is described taking an example in which the electronic device 11 and the video generating apparatus 12 are provided independently of each other.

In a possible implementation manner, the video generating method provided by the embodiment of the present disclosure may be performed by the electronic device 11, may be performed by the video generating apparatus 12 inside or outside the electronic device 11, or may be performed by other similar devices. Hereinafter, an example will be described in which the video generating apparatus 12 executes a video generating method.

Fig. 4 is a flow chart illustrating a video generation method according to an embodiment of the present disclosure, as shown in fig. 4, the method including the steps of:

s201, the video generating device acquires a target search statement searched by a user.

As one possible implementation manner, the video generating apparatus acquires a search sentence input by the user in the B2B purchase platform, and takes the search sentence as a target search sentence.

It should be noted that, the target search statement may be a search statement that the user searches in the historical time period, or may be a search statement that the user searches in real time.

For example, the target search statement may be a weak purchase search statement such as "how power of the steamed stuffed bun machine is set", or "how noisy the steamed stuffed bun machine is".

In some embodiments, the search terms to which the present disclosure relates refer to weak purchase search terms in the B2B purchase platform.

The weak purchase search statement indicates that the user has the requirement of inquiring the content, and keywords such as purchase, price or quotation are not included in the weak purchase search statement.

It should be noted that, the B2B model generally includes a strong purchase search term and a weak purchase search term.

The strong purchase search statement indicates that the user has strong purchase intention, and the strong purchase search statement contains the requirement (such as keywords of purchase, price, quotation, etc.) that the user wants to use for purchase. For example, the strong purchase search statement may be "how the steamed stuffed bun machine is in a market," or "the price of the steamed stuffed bun machine.

In the practical application process, the video generating device inputs the search statement into a pre-trained search statement classification model after acquiring the search statement searched by the user so as to determine that the search statement is a strong purchase search statement or a weak purchase search statement.

If the search term is a strong purchase search term, the video generating apparatus discards the search term. If the search term is a weak purchase search term, the video generating apparatus performs the subsequent step using the weak purchase search term as a target search term.

S202, the video generating device performs semantic analysis on the target search statement and judges whether the target search statement is matched with the element to be distributed.

The element to be distributed comprises a target commodity word corresponding to the target search statement, at least one target parameter item of the target commodity word and a parameter value of each target parameter item.

As one possible implementation manner, the video generating device performs semantic analysis of a natural language processing technology on the target search statement, determines whether a main word and an auxiliary word exist in the target search statement, and determines whether a target commodity word corresponding to the main word exists in a preset commodity information mapping table and whether at least one target parameter item corresponding to the auxiliary word and a parameter value of each parameter item exist in the commodity information mapping table under the condition that the main word and the auxiliary word exist in the target search statement.

Further, in the case that the target commodity word, at least one target parameter item, and parameter values of the respective parameter items exist in the commodity information mapping table, the video generating apparatus determines the element to be issued that matches the target search statement.

It should be noted that, the main word may be generally understood as a main word of the user in the target search sentence, generally, a commodity word, and the adverbs are generally required contents of the main word by the user. The commodity information mapping table comprises different commodity words, different parameter items of each commodity word and parameter values of each parameter item.

For example, in the target search statement "how power of the steamed stuffed bun machine is set", the steamed stuffed bun machine is a main word, and how power is set as an adverb. For another example, in the target search sentence "how noisy the steamed stuffed bun machine is," the steamed stuffed bun machine is the main word, and how noisy the steamed stuffed bun machine is the auxiliary word.

The specific implementation of this step may refer to the following description of the embodiments of the present disclosure, and will not be described herein.

S203, the video generating device generates target video for indicating the element to be distributed when the target video is matched with the element to be distributed.

As one possible implementation manner, the video generating device obtains the element to be distributed and generates the target video based on the content in the element to be distributed when the target search statement is matched with the element to be distributed.

As another possible implementation manner, the video generating apparatus may determine a partial target parameter item from at least one target parameter item after the element to be distributed is later, and generate the target video based on the target commodity word, the partial target parameter item, and the parameter value of the partial target parameter item.

For a specific implementation of this step, reference may be made to the following description of the embodiments of the present disclosure, which is not repeated here.

Subsequently, the video generation video sends the target video to the electronic device, so that in the case that other users search for the same target search statement, the electronic device can present the search result page including the target video to the user.

It can be appreciated that, according to the technical scheme provided by the embodiment of the disclosure, the target commodity word, the target parameter item and the parameter value which the user wants to search can be determined based on the target search statement searched by the user, and the target video is generated according to the determined target commodity word, the target parameter item and the parameter value. Therefore, the video generation method can help the merchant to generate videos according to the requirements of users, and can reduce the cost of video production of the merchant while helping the merchant to improve video conversion.

In some embodiments, since in the B2B purchasing platform, a plurality of users search for different search sentences, if video is generated based on each search sentence, the cost is high, and low-value video is also flooded, so in order to generate high-traffic and high-conversion target video, as shown in fig. 5, in the video generating method provided in the embodiment of the disclosure, S201 is specifically implemented as follows:

S301, the video generating device acquires a plurality of search sentences.

Wherein the plurality of search terms includes search terms searched by different users.

As one possible implementation, the video generating apparatus obtains, from the electronic device, a plurality of search sentences searched by different users over a historical period of time.

For example, the video generating apparatus acquires a plurality of search sentences searched by different users in the past 30 days.

It should be noted that the plurality of search terms are weak purchase search terms.

S302, the video generating device determines the production value of each search statement.

Wherein the production value of a search term is positively correlated with the search behavior value, and/or the associated value, of a search term. The search behavior value of one search statement is used for representing the execution condition of the user to execute interactive behavior on the search result of one search statement. The associated value of a search term is used to characterize the distribution of search results of a search term.

As one possible implementation, the video generating apparatus determines, for each of a plurality of search sentences, a search behavior value of each search sentence according to execution conditions of an interactive behavior performed on each search sentence by a different user, and determines a production value of each search sentence based on the determined search behavior values.

It should be noted that the interaction behavior may specifically include non-click, single click, multiple click, and conversion behavior.

As one example, the video generating apparatus may directly determine the determined search behavior value as the production value of the search term.

As another possible implementation manner, the video generating apparatus may also determine a search behavior value of each search term according to the distribution condition of different associated contents in the search result of each search term, and determine the production value of each search term according to the determined associated value.

The publishing condition may specifically include the number of associated content, the quality of each associated content, the number of associated content authors, and the rank of the associated content authors in the search result.

As one example, the video generating apparatus may directly determine the determined association value as the production value of the search term.

As yet another possible implementation manner, the video generating apparatus may further determine a generation value of each search term according to the search behavior value and the association value of the search term after determining the search behavior value and the association value.

In this step, the implementation manner of determining the search behavior value and the associated value of each search term and the implementation manner of determining the generation value of each search term according to the search behavior value and the associated value of each search term may refer to the subsequent description of the embodiments of the present disclosure, which is not described herein.

S303, the video generating device determines a target search statement from a plurality of search statements according to the production value of each search statement.

The production value of the target search statement is larger than a preset threshold value.

As one possible implementation, the video generating apparatus determines, after determining the production value of each search term, a search term having a production value greater than a preset threshold as the target search term.

In some cases, in order to determine the preset threshold, the video generating apparatus may further sort the plurality of search sentences based on the size of the production value, and take the first N search sentences as the target search sentences, where N is greater than or equal to 1, and in this case, the preset threshold is the production value of the nth search sentence.

It can be appreciated that according to the technical scheme provided by the embodiment of the disclosure, the production value of each search statement can be determined based on the search behavior value and the association value of each search statement, so that the target search statement is determined from a plurality of search statements searched by different users based on the production value of the search statement, the obtained target search statement can reflect the common requirements of most users, further the conversion of the subsequent generation of target videos can be improved, and the flow of merchants is improved.

In some embodiments, in order to determine the search behavior value of each search term, as shown in fig. 6, the video generating method provided by the embodiments of the present disclosure further includes the following steps:

s401, the video generating device acquires the searching times of each searching sentence searched by different users in a preset time period, the user information of the different users and the interactive behavior data.

Wherein the user information includes industries and professions of different users. The interactive behavior data is used for indicating the times that the search results of the search statement are executed by different users for different interactive behaviors.

As a possible implementation manner, the video generating device may directly obtain, from the electronic device, the number of searches each search term is searched by a different user.

As another possible implementation manner, after acquiring the plurality of search sentences, the video generating apparatus may further semantically classify the plurality of search sentences to acquire the number of searches of each search sentence, so as to obtain a plurality of candidate search sentence clusters. Each candidate search statement cluster comprises a plurality of candidate search statements, and the plurality of candidate search statements in one candidate search statement cluster have the same semantic meaning.

In this case, for one search term, the video generating apparatus determines the number of searches for the search term based on the number of searches for a plurality of candidate search terms in a candidate search term cluster in which the search term is located.

Specifically, the number of searches of one search term may be a mean or a sum of the number of searches of the corresponding plurality of candidate search terms.

By way of example, the number of searches per search term may be as shown in Table 1 below:

table 1 number of searches for search sentences

As shown in table 1, for the search term "how to use the steamed stuffed bun machine", the plurality of candidate search terms corresponding to the search term are "how to use the steamed stuffed bun machine", "how to operate the steamed stuffed bun machine", and "common operation method of the steamed stuffed bun machine", wherein the sum of the search times of the candidate search terms is 220, and the search times of the search term "use the steamed stuffed bun machine" is 220.

In some embodiments, the video generating apparatus may also determine a candidate search term having the largest number of searches from each candidate search term cluster, and compare the production value of the candidate search term having the largest number of searches. Thus, it is unnecessary to calculate the production value for other candidate search terms among the candidate search terms, and the calculation resources can be reduced.

As shown in table 1, among the plurality of candidate search sentences, the number of searches of the "steamed stuffed bun machine use" is largest, and the video generating apparatus determines only the production value of the search sentence "steamed stuffed bun machine use" as the data source of the subsequent determination target search sentence.

In some embodiments, the interactive behavior data includes a number of clicks, and a number of conversion operations.

Wherein the number of non-clicks includes a number of times the user has not performed a click operation in a search results page including search results of the search statement. The single click times include the times that the user has clicked one time on the search result in the above-described search result page and has not performed any operation on the one-click landing page. The number of times of clicking is the number of times that the user performs clicking operations and does not perform conversion operations in the search result page and/or the landing page. The conversion operation times are times that the user performs multiple clicking operations in the search result page or the landing page and performs conversion behaviors.

The transformation behavior comprises the behavior of initiating operations such as attention, sharing, consultation, purchase and the like in the page.

S402, the video generating device determines the search behavior value of each search statement according to the search times, the user information and the interaction behavior data.

As a possible implementation, the video generating apparatus determines the user value of each search term according to the user information.

The user value is used for indicating the correlation degree between the user information and the target commodity word.

And the video generating device further carries out weighting processing on the non-clicked times, the single-clicked times, the multi-clicked times and the conversion operation times to obtain the interactive behavior value of each search statement.

The weighting process may be a weighted sum or a weighted average. In the case where the above-described weighting process is a weighted average, the interactive behavior value of each search term satisfies the following expression:

wherein H is ₁ Representing the interactive behavior value of the search statement, h ₁ For the number of times of non-clicking, weight of the number of times of non-clicking, h ₂ For the number of single clicks, b is the weight of the number of single clicks, h ₃ For the number of clicks, c is the weight of the number of clicks, h ₄ For the number of conversion operations, d is a weight of the number of conversion operations, and e is a first preset coefficient.

In some cases, the conversion operation number of times is greater than the multiple click number of times, the multiple click number of times is greater than the single click number of times, the single click number of times is greater than the non-click number of times, and the first preset coefficient may be the same as the conversion operation number of times.

For example, the conversion operation may have a weight of 10, the multiple clicks may have a weight of 3, the single click may have a weight of 1, and the no click may have a weight of 0.

It can be understood that the conversion operation frequency weight is set to be greater than the multi-click frequency weight, the multi-click frequency weight is set to be greater than the single-click frequency weight, and the single-click frequency weight is set to be greater than the non-click frequency weight, so that the influence of the interaction behavior of the user on the search result on the interaction behavior value is considered, and the interaction behavior value obtained by the weighting processing can be more accurate.

Further, the video generating apparatus determines a search behavior value of each search term based on the number of clicks, the user value, and the interactive behavior value.

Wherein, the search behavior value is positively correlated with the click times, the user value and the interaction behavior value.

In one case, the search behavior value of the search statement satisfies the following expression:

H＝H ₃ (1+H ₁ )+H ₂

wherein H represents the search behavior value of the search statement, H ₁ Representing interactive behavior value of search statement, H ₂ Representing the user value of the search statement, H ₃ Representing the number of searches of the search term.

It can be appreciated that, according to the technical scheme provided by the embodiment of the disclosure, based on the industry and occupation of the user and the interaction behavior of the user and the search result of the search statement, the determined search behavior value can be more accurate, so that the production value of each search statement is more accurate, and finally, the determined target search statement can be ensured to be more fit with the requirement of the user, and the generated target video can meet the requirement of more users.

Meanwhile, the number of non-clicks, the number of single clicks, the number of multiple clicks and the number of conversion operations are weighted, so that the dimension for determining the interactive behavior value is wider, the interactive behavior value is more accurate, and the determined search behavior value is more accurate.

In some embodiments, in order to determine the user value of each search term, as shown in fig. 7, in S402 provided in the embodiment of the present disclosure, "determining the search behavior value of each search term according to the number of searches, the user information, and the interaction behavior data", may be specifically implemented as follows:

s501, the video generating device determines a first correlation degree and a second correlation degree.

The first relevance is the duty ratio of users in the same industry with the target commodity word in different users. The second relativity is the duty ratio of occupations of different users to preset occupations, and the conversion rate corresponding to the preset occupations is larger than the preset conversion rate.

As one possible implementation manner, the video generating device determines that, among search users of the search sentence, a search word corresponding to the search sentence is a number of users in the same industry, and determines a first relevance according to a ratio of the number of users in the same industry to a total number of search users.

For example, the total number of search users for the search term "steamed stuffed bun machine in use" is 1000, wherein the number of users in the same industry as the commercial word "steamed stuffed bun machine" is 800, and the first relevance is 0.8.

The video generating device determines that the ratio of the number of users with the user occupation being a preset occupation to the total number of searching users in the searching users of the searching sentences is a second relativity.

For example, the preset occupation may be a purchasing occupation, and the total number of searching users of the search sentence "steamed stuffed bun machine use" is 1000, wherein the number of users of the occupation purchasing is 400, and the second relatedness is 0.4.

S502, the video generating device determines the user value based on the first correlation degree and the second correlation degree.

Wherein the user value is positively correlated with the first correlation and the second phase Guan Du.

As one possible implementation, the user value of the search term satisfies the following expression:

H ₂ ＝f(h ₅ +h ₆ )

wherein H is ₂ Representing the user value of the search statement, h ₅ For the first correlation, h ₅ And f is a second preset coefficient for the second correlation degree. The second preset coefficient may be the same as the first preset coefficient, for example.

It can be appreciated that by adopting the technical scheme, the relevance between the user and the target commodity word can be indicated more based on the proportion of the user in the same industry with the target commodity word in different users and the proportion of the preset professional user in different users, so that the determined user value can reflect the requirement of the user more, and the accuracy of the search behavior value can be ensured.

In some embodiments, the search results of each search term include a plurality of associated content, and to be able to determine the associated value of each search term, as shown in fig. 8, determining the associated value of each search term in embodiments of the present disclosure may be implemented as follows:

s601, the video generating apparatus acquires the number of the plurality of associated contents, the content quality score of each associated content, the number of the plurality of associated content authors, and the author rank score of each author.

Wherein the content quality score is determined based on a rating operation of the user, and the author rating score is used for indicating popularity of the author.

For example, after searching for "steamed stuffed bun machine use," the user includes 50 videos in the search results, involving 35 authors, such that the number of associated content authors is 50 and the number of associated content authors is 35.

S602, the video generating device determines the value of the associated content of each search statement according to the quantity of the plurality of associated contents and the content quality score of each associated content.

Wherein the associated content value is positively correlated with the number of the plurality of associated content and the content quality score.

As a possible implementation manner, the video generating apparatus performs weighting processing on the number of the plurality of associated contents and the content quality score of each associated content, so as to obtain the associated content value of each search term.

S603, the video generating device determines the associated author value of each search statement according to the number of the plurality of associated content authors and the author grade score of each author.

Wherein the associated author value is positively correlated with the number of the plurality of associated content authors and the author rank score.

As a possible implementation manner, the video generating apparatus performs weighting processing on the number of the plurality of associated content authors and the author rank score of each author, to obtain an associated author value of each search sentence.

S604, the video generating device determines the association value of each search statement according to the association content value and the association author value of each search statement.

Wherein the associated value is positively correlated with the associated content value and the associated author value.

As a possible implementation manner, the video generating device performs weighting processing on the associated content value and the associated author value of each search statement to obtain the associated value of each search statement.

It should be noted that, the weighting processing related to the embodiment of the present disclosure may refer to the description in embodiment S402 of the present disclosure, and will not be described in detail herein.

It can be understood that the relevance value determined by the multi-dimension can fully reflect the competition condition of the search sentence according to the number of the plurality of associated contents, the content quality score of each associated content, the number of the plurality of associated content authors and the author grade score of each author, so that the determined production value is more accurate, and further the conversion of the subsequent target video can be improved.

In some embodiments, in order to determine whether the target search statement can be matched to the element to be published, as shown in fig. 9, S202 provided by the embodiment of the disclosure may be implemented as follows:

S701, the video generating device performs semantic analysis on the target search sentence to obtain a main word and an adverb in the target search sentence.

As a possible implementation manner, the video generating apparatus performs semantic analysis on the target search sentence, determines a target semantic structure of the target search sentence from the plurality of semantic structures, and determines the main word and the adverbs from the target search sentence according to the target semantic structure.

Specifically, the video generating apparatus stores a plurality of semantic structures in advance, and table 2 exemplarily shows the plurality of semantic structures:

TABLE 2 multiple semantic structures

The main word is a main word in the search sentence, and is usually a commodity word. The adverbs are the core appeal of the user in the search statement, typically parameters, performance, etc.

Specifically, the video generating apparatus parses the main word and the adverb from the target search sentence according to the positions of the main word and the adverb in the target semantic structure after determining the target semantic structure.

Table 3 shows an example of parsing a main word and an adverb from a target search term.

TABLE 3 resolution results of target search statement

The adverbs can be divided into direct adverbs and indirect adverbs, wherein the direct adverbs comprise parameter items of commodities, and the parameter items directly indicate the requirements of users, such as power, horsepower, efficiency and the like in table 3. Indirect adverbs do not include parameter items for the merchandise and more indicate the needs of the user, such as yellow leaves, usage class, etc. in table 3.

In some cases, the video generating apparatus analyzes the target search sentence, and in the process of determining whether the target search sentence includes the main word and the adverb, it may determine whether the main word exists first, then determine whether the adverb exists, or determine whether the adverb exists first, then determine whether the main word exists, or determine whether the main word exists and whether the adverb exists simultaneously.

In the case where the video generating apparatus determines that the main word and the adverb do not exist in the target search sentence at the same time, the target search sentence is discarded.

In the case where the video generating apparatus determines that the adverb exists in the target search term and that the adverb does not exist, the video generating apparatus may further determine the subject or the target commodity word that the user desires to search for, for example, "what medicine is used for cucumber Miao Shezi" in the above search term, and in the case where the adverb does not exist, the video generating apparatus may still determine the medicine name corresponding to the search term of the user based on the adverb and use the medicine name as the subject.

It can be understood that, according to the target semantic structures in the plurality of semantic structures, the main word and the adverbs in the target search sentence are determined, so that the determined main word and the determined adverbs are more accurate, are more fit and are more suitable for understanding the requirements of users, further, the target commodity words and the target parameter items which are determined later are more accurate, and the conversion of the target video can be further improved.

S702, the video generating device judges whether a target commodity word corresponding to the main word exists in a preset commodity information mapping table according to the main word.

The commodity information mapping table comprises a plurality of commodity words, parameter items of each commodity word and parameter values of each parameter item.

As one possible implementation manner, the video generating apparatus queries whether a target commodity word corresponding to the main word exists from a plurality of commodity words included in the commodity information mapping table according to the main word.

For example, the target search sentence includes a main word steamed stuffed bun machine, and the video generating device can directly determine that the target commodity word exists as the steamed stuffed bun machine from a plurality of commodity word steamed stuffed bun machines and small steamed stuffed bun machines in the commodity information mapping table.

As another possible implementation manner, the video generating device normalizes the main word, matches the main word with the existing commodity word, and queries whether the commodity word exists in the plurality of commodity words from the commodity information mapping table according to the commodity word.

It should be noted that, the normalization is generally determined based on a relevance model, and is used for determining according to the relevance between texts. For example, the target search statement includes a full-automatic steamed stuffed bun machine, but commodity words related to the steamed stuffed bun machine in the commodity information mapping table are the steamed stuffed bun machine and the small steamed stuffed bun machine. The video generating device can determine that the full-automatic steamed stuffed bun machine is normalized to be a small steamed stuffed bun machine through the correlation model, and then determine that the target commodity word corresponding to the target search statement is the small steamed stuffed bun machine.

S703, the video generating device judges whether at least one target parameter item and the parameter value of each target parameter item exist in the commodity information mapping table according to the adverbs when the target commodity word exists in the commodity information mapping table.

As one possible implementation manner, the video generating apparatus first determines whether the adverbs in the target search statement are direct adverbs or indirect adverbs, and determines whether at least one target parameter item and a parameter value of each target parameter item exist in the merchandise information mapping table according to a determination result.

Specifically, in the case that the adverbs are direct adverbs, that is, the adverbs of the target search statement include parameter items, the video generating device determines, according to the parameter items in the adverbs, whether at least one target parameter item exists in the commodity information mapping table, and determines, in the case that at least one target parameter item exists, a parameter value of the at least one parameter item from the commodity information mapping table.

It should be noted that, in this step, it is determined whether the parameter item in the direct adverb corresponds to the target parameter item, and specifically, it may be determined through normalization.

For example, the adverbs "size" and "size" are normalized to "size". For the implementation process of normalization, reference may be made to the description in S702 in the embodiment of the disclosure, which is not described herein.

In another case, in the case where the adverb is an indirect adverb, that is, the adverb does not include a parameter item, the video generating apparatus determines whether or not there is a parameter value corresponding to the adverb in the commodity information mapping table according to the adverb. Further, in the case where the parameter value corresponding to the adverb exists in the commodity information mapping table, the video generating apparatus determines at least one target parameter item based on the parameter value corresponding to the adverb.

It should be noted that, because the indirect adverbs often include parameter values of effect types and scene types, for example, aiming at a target search statement "how the noise suddenly increases during the use of the steamed stuffed bun machine returns", the commodity information mapping table includes notes "the notes of the steamed stuffed bun machine include: the water needs to be added at a constant speed, otherwise, the running noise of the machine becomes large. Therefore, the adverbs in the target search statement are matched, so that the parameter items with high noise can be matched, and further the parameter items with high noise can be matched to the target parameter items as notice matters.

It can be understood that, since the adverbs can be divided into direct adverbs and indirect adverbs, according to different situations, different judging modes are adopted, and the parameter values of at least one target parameter item and each target parameter item can be comprehensively and accurately matched according to the adverbs.

S704, the video generating device determines that the target parameter item is matched with the element to be distributed under the condition that at least one target parameter item and the parameter value of each target parameter item exist in the commodity information mapping table.

It can be appreciated that by judging whether the main word and the adverb exist in the target search statement, the target commodity word can be matched based on the main word, at least one target parameter item and a parameter value can be matched based on the adverb, and a specific implementation mode for matching the element to be issued is provided.

In the actual application process, in order to determine whether a target video corresponding to a target search statement can be generated, as shown in fig. 10, an embodiment of the disclosure may further be implemented in the following manner:

s801, the video generating device acquires a target search statement.

S802, the video generating device judges whether a main word exists in the target search statement.

S803, the video generating device normalizes the main word to obtain the commodity word when the main word exists in the target search sentence.

S804, the video generating device judges whether the target commodity words corresponding to the normalized commodity words exist in the commodity information mapping table.

S805, the video generating device determines that the target video cannot be generated and discards the target search statement when the target commodity word does not exist in the commodity information mapping table.

S806, the video generating device judges whether indirect adverbs exist in the target search statement or not under the condition that the target search statement does not include the main word or includes the main word and the main word can be matched with the target commodity word.

S807, the video generating device inquires whether a parameter value exists in the commodity information mapping table according to the indirect adverbs when the indirect adverbs exist in the target search statement, and determines at least one target parameter item according to the parameter value.

S808, under the condition that the direct adverbs exist in the target search statement, the video generating device normalizes the parameter items in the direct adverbs and inquires whether at least one target parameter item and the parameter value of each target parameter item exist in the commodity information mapping table.

S809, the video generating device determines that the target video cannot be generated and discards the target search statement under the condition that at least one target parameter item and the parameter value of each target parameter item do not exist in the commodity information mapping table.

S810, the video generating device generates a target video according to the target commodity word, the target parameter item and the parameter value of each target parameter item when at least one target parameter item and the parameter value of each target parameter item exist in the commodity information mapping table.

In some embodiments, in order to be able to generate a target video, as shown in fig. 11, S203 provided by the embodiments of the present disclosure may be implemented as follows:

and S901, the video generating device determines a text to be distributed for expressing the element to be distributed from a preset industrial knowledge graph according to the element to be distributed.

The industrial knowledge graph comprises different commodities, different parameter items and expression modes of parameter values.

As one possible implementation manner, the video generating device queries the text to be distributed in the expression manner corresponding to the element to be distributed from the industrial knowledge graph.

As another possible implementation manner, the video generating device determines a target parameter item with a degree of correlation with the user being greater than a preset degree of correlation from at least one target parameter item of the element to be distributed, and a corresponding parameter value and a target commodity word, and determines a text to be distributed from the industrial knowledge graph.

For example, the different parameter items include a commodity professional parameter, a commodity name, a brand, a function advantage, a scene field, an industrial band, a usage mode, a price, and the like, and the video generating apparatus may determine a target parameter item having a high degree of relevance to the user, such as a function advantage, a scene field, a price, and the like, from the plurality of parameter items as a source for determining the text to be distributed. In practical application, the plurality of target parameter items can be cleaned, normalized and researched to extract the target parameter item with the user relevance greater than the preset relevance from the plurality of target parameter items.

Table 4 shows various expressions of commodity expertise parameters in an industrial knowledge graph.

TABLE 4 expression of commodity-specific parameters in Industrial knowledge graph

The above-mentioned [ parameter item ] ranges from [ parameter value 1] (minimum value) to [ parameter value 2] (maximum value), and applies to the boundary between the maximum value and the minimum value, and if the parameter value is > 10, such expression is not applied. The parameter items are mostly larger than the parameter value 1 (more than 80%), but a small part of the parameter items are the parameter values 2 and 3 (less than 10%), and are suitable for calculating the average value by taking the parameter values as a range, and not calculating the average value if the parameter values are not bordered around the parameter value range. The average value of the parameter items is about 1 (average value), and common parameter items are 2, 3 and 4 (random example), which are suitable for calculating the average value by taking the parameter values as the range, and not calculating the average value if the parameter values have no boundary around the range.

Table 5 shows various expressions of commodity names in the industrial knowledge graph.

TABLE 5 expression of commodity names in Industrial knowledge graph

Table 6 shows various expressions of commercial word brands in the industrial knowledge graph.

TABLE 6 expression of commercial word brands in Industrial knowledge graph

/>

In table 6, [ parameter value 4] is the brand of the video merchant, and if there are a plurality of brands, brand 2 with the largest number of commodities is selected, and [ scene ] is the application scene of the commodity word.

Table 7 shows various expression patterns in the industrial knowledge graph for the functional advantage of commercial words.

TABLE 7 expression of the functional advantage of commercial words in industrial knowledge graph

/>

Table 8 shows various expressions in the industrial knowledge graph for the field of commercial word scenes.

TABLE 8 expression of Commodity word scene field in Industrial knowledge graph

Table 9 shows various expression patterns in the industrial knowledge graph with respect to the commercial word industrial zone.

TABLE 9 expression of commercial word industry bands in industrial knowledge graph

Table 10 shows various expressions of the commercial word usage patterns in the industrial knowledge graph.

TABLE 10 expression of the use of Commodity words in Industrial knowledge graph

Table 11 shows various expressions of commodity price in the industrial knowledge graph.

TABLE 11 expression of commodity word prices in Industrial knowledge graph

/>

/>

The expression (1) of price in table 11 has the following characteristics: 1. price characteristics: average prices of commodity words corresponding to skus are compared at annual granularity. The comparison is stable: the commodity words correspond to an average price of + -10% of the minimum stock keeping unit (stock keeping unit, SKU). Rising gradually in steady state: the commodity word corresponds to the average price of sku + [10%,30% ]. Gradually descending in steady state: commodity words correspond to the average price of sku- [10%,30% ]. The rising is obvious: commodity words correspond to the average price of sku + [30% ], + ]. The drop is obvious: the commodity word corresponds to the average price of sku- [30%, ++ infinity A kind of electronic device. 2. Unit description: price units in text need to be unified and concrete and apply unified data for computation, e.g., (1) 30 yuan/bucket cannot be used because the bucket cannot compute a specific capacity. If one barrel is lL, 30 yuan/1L can be changed, otherwise, the dirty data (2) is regarded as not allowing 450 yuan/large tertiary, because the large tertiary is obviously not a specification unit, the dirty data (3) is regarded as 850 yuan/ton, 1 yuan/kg can be uniformly changed into ton/kg, normalization is carried out to a unit with a large duty ratio, and the dirty data (3) and the data description are regarded as if the dirty data (2) cannot be normalized: dirty data such as a face, obvious abnormal data and non-data should be removed from the data, for example, (1) the data is obviously abnormal: data were 10 times higher than the highest data and 10 times lower than the lowest data (2) average price: after removing the dirty data, the average price (3) of unity is the highest/lowest price: after removing the dirty data, the current highest/low price (4) price per unit is displayed: [ data+Unit ]4, usage mode: the paragraphs with different brands, parameters and region prices can be used only one or a plurality of paragraphs.

The expression (2) of price in table 11 has the following characteristics: 1. the main supply is as follows: the commodity words are main commodity words, and the top 5 main commodity words are selected at most. 2. Percent = the merchant offer position (from high to low)/total offered merchant number.

S902, the video generating device generates a target video according to the text to be published.

It should be noted that the text to be distributed may be located in the video content of the target video, or may be located in a tag, a cover, or a video introduction of the target video.

It can be understood that the corresponding target expression mode is queried from the industrial knowledge graph, the expression mode which is more in line with the user semantics can be provided for the target video, and the generated target video can improve the user experience, video conversion and merchant flow.

In some embodiments, after the video generating device generates the target video, a search results page is also generated for the search results of the target search statement.

As shown in fig. 12, the search results page includes a video play window, store information, and conversion components.

The video playing window can be positioned at the left side of the search result page, and when the page slides to the complete video playing window to be exposed, the target video is automatically played, the playing is stopped, and the repeated playing is not performed.

The first line element of the video playing window comprises a video title, playing times and release time, wherein the release time is the release time of the video on the B2B purchasing platform for the video uploaded by a merchant, and the release time is the time of a user searching a target search statement for the target video generated by the embodiment of the present disclosure. The bottom of the video playing window also comprises a video brief introduction, and the video brief introduction can be 1-2 lines of text.

For example, taking the epoxy brand as the target search term, the video profile may be: the video is provided by a B2B purchasing platform merchant, the video content takes the brand of the epoxy resin as a theme, the brand ranking of the epoxy resin is introduced, and meanwhile, the information such as important parameters of the epoxy resin, the use mode of the epoxy resin, quotation of the epoxy resin and the like is displayed, so that the problem of the epoxy resin industry is solved.

Store information may be located on the right side of the search results page for displaying corresponding information of the store, such as identification (logo, name) of the store, store authentication information (in-field merchant, authenticity authentication, age of B2B purchasing platform), video quantity, source quantity. After clicking the store information, the user can jump to the store front page.

The conversion component may be located on the right side of the video playback window, including a telephone dialing control, a directional query control, and a full network query control.

The telephone dialing control is used for telephone communication between the user and the merchant. The directional query control is used by a user to initiate a query message to a merchant. The full-network inquiry control is used for a user to initiate inquiry messages to all merchants related to the target commodity words in the B2B purchasing platform. The query message is automatically filled with the target commodity word.

Meanwhile, the conversion component also displays a prompt text, for example, the prompt text of the telephone dialing control may be: clicking to obtain the base price. The prompt text displayed in the directional inquiry control can be low in price for communication, or after submission, the merchant will send a customer service representative to provide special services for you. The prompt text of the full web query control may be: does not find the desired source? Say your purchasing demand, will match for your intelligence. The bottom of the conversion component is also provided with a policy reminder, for example, the policy reminder may be: clicking free to get low price representatives agrees to user service agreement, privacy policy.

In some embodiments, as shown in fig. 12, in the case that the recall video is less than the preset number, the search result page further includes a popular video recommendation area including a plurality of recommended videos (recommended video 1, recommended video 2, and recommended video 3 are exemplarily shown in fig. 12, and more or less recommended videos may exist in the actual application process), and a title, a shop name, a call dialing control, a video cover, and a video duration of each recommended video.

It should be noted that the recommendation policy of the recommended video is determined based on the correlation with the target video. For example, recommendations are limited to follow the same merchandise word.

In some embodiments, a purchase control is also included in the search results page. When the user clicks the purchase control, the search results page automatically slides up to the top of the page and the conversion component is highlighted.

Fig. 13 is a block diagram showing a structure of a video generating apparatus according to an embodiment of the present disclosure, which is applied to an electronic device in a video generating system. Referring to fig. 13, the apparatus 1000 includes an acquisition unit 1001, a judgment unit 1002, and a generation unit 1003.

An obtaining unit 1001 is configured to obtain a target search term searched by a user.

A judging unit 1002, configured to perform semantic analysis on the target search statement, and judge whether the element to be issued of the target search statement is matched. The element to be distributed comprises a target commodity word corresponding to the target search statement, at least one target parameter item of the target commodity word and a parameter value of each target parameter item.

A generating unit 1003, configured to generate a target video for indicating an element to be distributed, in a case of matching to the element to be distributed.

As shown in fig. 13, the video generating apparatus 1000 provided in the embodiment of the present disclosure is specifically configured to:

a plurality of search terms is obtained. The plurality of search terms includes search terms that are searched by different users.

The production value of each search term is determined. The production value of a search term is positively correlated with the search behavior value, and/or the associated value, of a search term. The search behavior value of one search statement is used for representing the execution condition of the user to execute interactive behavior on the search result of one search statement. The associated value of a search term is used to characterize the distribution of search results of a search term.

A target search term is determined from the plurality of search terms based on the production value of each search term. The production value of the target search statement is greater than a preset threshold.

As shown in fig. 13, the video generating apparatus 1000 provided by the embodiment of the present disclosure further includes a determining unit 1004.

An obtaining unit 1001, configured to obtain the number of searches of each search sentence by different users in a preset period, user information of different users, and interaction behavior data. The user information includes industries and professions of different users. The interactive behavior data is used for indicating the times that the search results of the search statement are executed by different users for different interactive behaviors.

The determining unit 1004 is configured to determine a search behavior value of each search term according to the search times, the user information, and the interaction behavior data.

As shown in fig. 13, the video generating apparatus 1000 provided in the embodiment of the present disclosure includes an unticked number of times, a single click number of times, a multiple click number of times, and a conversion operation number of times.

The determining unit 1004 is specifically configured to:

and determining the user value of each search statement according to the user information. The user value is used for indicating the correlation degree of the user information and the target commodity word.

And weighting the number of non-clicks, the number of single clicks, the number of multiple clicks and the number of conversion operations to obtain the interactive behavior value of each search statement.

The search behavior value of each search term is determined based on the number of clicks, the user value, and the interaction behavior value. The search behavior value is positively correlated with the number of clicks, the user value, and the interaction behavior value.

a first correlation and a second correlation are determined. The first relevance is the duty ratio of users in the same industry as the target commodity word in different users. The second relativity is the duty ratio of occupations of different users to preset occupations, and the conversion rate corresponding to the preset occupations is larger than the preset conversion rate.

The user value is determined based on the first correlation and the second correlation. The user value is positively correlated with the first correlation and the second correlation Guan Du.

As shown in fig. 13, the video generating apparatus 1000 provided in the embodiment of the present disclosure has a weight of the number of conversion operations greater than a weight of the number of clicks, the weight of the number of clicks greater than a weight of a single click, and the weight of the single click greater than a weight of the number of clicks not.

As shown in fig. 13, the video generating apparatus 1000 provided in the embodiment of the present disclosure, wherein the search result of each search sentence includes a plurality of associated contents.

The video generating apparatus 1000 further includes a determining unit 1004.

The obtaining unit 1001 is further configured to obtain a number of a plurality of associated contents, a content quality score of each associated content, a number of authors of the plurality of associated contents, and an author rank score of each author. The content quality score is determined based on the user's rating operation, and the author rating score is used to indicate the popularity of the author.

The determining unit 1004 is further configured to determine an associated content value of each search term according to the number of the plurality of associated contents and the content quality score of each associated content. The associated content value is positively correlated with the number of the plurality of associated content and the content quality score.

The determining unit 1004 is further configured to determine an associated author value of each search statement according to the number of the plurality of associated content authors and the author rank score of each author. The associated author value is positively correlated with the number of authors of the plurality of associated content, and the author rank score.

The determining unit 1004 is further configured to determine an associated value of each search term according to the associated content value and the associated author value of each search term. The associated value is positively correlated with the associated content value and the associated author value.

and carrying out semantic analysis on the target search statement to obtain a main word and an adverb in the target search statement.

And judging whether the target commodity word corresponding to the main word exists in a preset commodity information mapping table according to the main word. The commodity information mapping table comprises a plurality of commodity words, parameter items of each commodity word and parameter values of each parameter item.

And under the condition that the target commodity word exists in the commodity information mapping table, judging whether at least one target parameter item and the parameter value of each target parameter item exist in the commodity information mapping table according to the adverbs.

And under the condition that at least one target parameter item and parameter values of the target parameter items exist in the commodity information mapping table, determining that the commodity information mapping table is matched with the element to be distributed.

and carrying out semantic analysis on the target search statement, and determining a target semantic structure of the target search statement from the plurality of semantic structures.

And determining the main word and the adverbs from the target search statement according to the target semantic structure.

and judging whether at least one target parameter item exists in the commodity information mapping table according to the parameter items in the adverbs when the adverbs comprise the parameter items, and determining the parameter value of the at least one parameter item from the commodity information mapping table when the at least one target parameter item exists.

And judging whether a parameter value corresponding to the adverb exists in the commodity information mapping table according to the adverb under the condition that the adverb does not comprise the parameter item.

And determining at least one target parameter item based on the parameter value corresponding to the adverb when the parameter value corresponding to the adverb exists in the commodity information mapping table.

according to the elements to be distributed, determining a text to be distributed for expressing the elements to be distributed from a preset industrial knowledge graph, and generating a target video according to the text to be distributed. The industrial knowledge graph comprises different commodities, different parameter items and expression modes of parameter values.

According to an embodiment of the present disclosure, the present disclosure also provides an electronic device including at least one processor. And a memory communicatively coupled to the at least one processor. Wherein the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the video generation method provided by the present disclosure.

According to an embodiment of the present disclosure, the present disclosure also provides a non-transitory computer-readable storage medium storing computer instructions for causing the computer to perform the video generation method provided by the present disclosure.

According to an embodiment of the present disclosure, the present disclosure also provides a computer program product comprising a computer program which, when executed by a processor, implements the video generation method provided by the present disclosure.

FIG. 14 shows a schematic block diagram of an example electronic device that may be used to implement embodiments of the present disclosure. Electronic devices are intended to represent various forms of digital computers, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. The electronic device may also represent various forms of mobile devices, such as personal digital processing, cellular telephones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions, are meant to be exemplary only, and are not meant to limit implementations of the disclosure described and/or claimed herein.

As shown in fig. 14, the electronic device 1100 includes a computing unit 1101 that can perform various appropriate actions and processes according to a computer program stored in a Read Only Memory (ROM) 1102 or a computer program loaded from a storage unit 1108 into a Random Access Memory (RAM) 1103. In the RAM 1103, various programs and data required for the operation of the device 1100 can also be stored. The computing unit 1101, ROM 1102, and RAM 1103 are connected to each other by a bus 1104. An input/output (I/O) interface 1105 is also connected to bus 1104.

Various components in device 1100 are connected to I/O interface 1105, including: an input unit 1106, such as a keyboard, mouse, etc. The output unit 1107 is, for example, various types of displays, speakers, and the like. Storage unit 1108, such as a magnetic disk, optical disk, and the like. And a communication unit 1109 such as a network card, modem, wireless communication transceiver, or the like. The communication unit 1109 allows the device 1100 to exchange information/data with other devices through a computer network such as the internet and/or various telecommunication networks.

The computing unit 1101 may be a variety of general purpose and/or special purpose processing components having processing and computing capabilities. Some examples of computing unit 1101 include, but are not limited to, a Central Processing Unit (CPU), a Graphics Processing Unit (GPU), various specialized Artificial Intelligence (AI) computing chips, various computing units running machine learning model algorithms, a Digital Signal Processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 1101 performs the respective methods and processes described above, such as a video generation method. For example, in some embodiments, the video generation method may be implemented as a computer software program tangibly embodied on a machine-readable medium, such as storage unit 1108. In some embodiments, some or all of the computer programs may be loaded and/or installed onto device 1100 via ROM 1102 and/or communication unit 1109. When a computer program is loaded into the RAM 1103 and executed by the computing unit 1101, one or more steps of the video generation method described above may be performed. Alternatively, in other embodiments, the computing unit 1101 may be configured to perform the video generation method by any other suitable means (e.g., by means of firmware).

Various implementations of the systems and techniques described here above may be implemented in digital electronic circuitry, integrated circuit systems, field Programmable Gate Arrays (FPGAs), application Specific Integrated Circuits (ASICs), application Specific Standard Products (ASSPs), systems On Chip (SOCs), complex Programmable Logic Devices (CPLDs), computer hardware, firmware, software, and/or combinations thereof. These various embodiments may include: implemented in one or more computer programs, the one or more computer programs may be executed and/or interpreted on a programmable system including at least one programmable processor, which may be a special purpose or general-purpose programmable processor, that may receive data and instructions from, and transmit data and instructions to, a storage system, at least one input device, and at least one output device.

Program code for carrying out methods of the present disclosure may be written in any combination of one or more programming languages. These program code may be provided to a processor or controller of a general purpose computer, special purpose computer, or other programmable video generating device such that the program code, when executed by the processor or controller, causes the functions/operations specified in the flowchart and/or block diagram to be implemented. The program code may execute entirely on the machine, partly on the machine, as a stand-alone software package, partly on the machine and partly on a remote machine or entirely on the remote machine or server.

In the context of this disclosure, a machine-readable medium may be a tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. The machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

To provide for interaction with a user, the systems and techniques described here can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to a user. And a keyboard and pointing device (e.g., a mouse or trackball) by which a user can provide input to the computer. Other kinds of devices may also be used to provide for interaction with a user. For example, feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback). And input from the user may be received in any form, including acoustic input, speech input, or tactile input.

The systems and techniques described here can be implemented in a computing system that includes a background component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front-end component (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such background, middleware, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: local Area Networks (LANs), wide Area Networks (WANs), and the internet.

The computer system may include a client and a server. The client and server are typically remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. The server may be a cloud server, a server of a distributed system, or a server incorporating a blockchain.

It should be appreciated that various forms of the flows shown above may be used to reorder, add, or delete steps. For example, the steps recited in the present disclosure may be performed in parallel or sequentially or in a different order, provided that the desired results of the technical solutions of the present disclosure are achieved, and are not limited herein.

The above detailed description should not be taken as limiting the scope of the present disclosure. It will be apparent to those skilled in the art that various modifications, combinations, sub-combinations and alternatives are possible, depending on design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present disclosure are intended to be included within the scope of the present disclosure.

Claims

1. A video generation method, comprising:

acquiring a target search statement searched by a user;

carrying out semantic analysis on the target search statement, and judging whether the target search statement is matched with the element to be issued; the element to be distributed comprises a target commodity word corresponding to the target search statement, at least one target parameter item of the target commodity word and a parameter value of each target parameter item;

And generating a target video for indicating the element to be distributed under the condition that the element to be distributed is matched.

2. The video generating method according to claim 1, wherein the acquiring the target search statement searched by the user includes:

acquiring a plurality of search sentences; the plurality of search terms includes search terms searched by different users;

determining a production value of each search term; the production value of one search term is positively correlated with the search behavior value, and/or the associated value, of said one search term; the search behavior value of the one search statement is used for representing the execution condition of executing interaction behavior on the search result of the one search statement by a user; the association value of the one search statement is used for representing the release condition of the search result of the one search statement;

determining the target search statement from the plurality of search statements according to the production value of each search statement; and the production value of the target search statement is greater than a preset threshold.

3. The video generation method of claim 2, the method further comprising:

acquiring the searching times of each searching sentence searched by different users in a preset time period, the user information of the different users and the interactive behavior data; the user information comprises industries and professions of the different users; the interactive behavior data is used for indicating the times that the search results of the search statement are executed by different users for different interactive behaviors;

And determining the search behavior value of each search statement according to the search times, the user information and the interaction behavior data.

4. The video generation method of claim 3, wherein the interactive behavior data includes a number of non-clicks, a number of single clicks, a number of clicks, and a number of conversion operations;

the determining the search behavior value of each search sentence according to the search times, the user information and the interactive behavior data includes:

determining the user value of each search statement according to the user information; the user value is used for indicating the correlation degree between the user information and the target commodity word;

weighting the non-clicked times, the single-clicked times, the multi-clicked times and the conversion operation times to obtain the interactive behavior value of each search statement;

determining a search behavior value of each search statement based on the number of clicks, the user value, and the interaction behavior value; the search behavior value is positively correlated with the number of clicks, the user value, and the interaction behavior value.

5. The video generation method of claim 4, wherein the determining the user value of each search term from the user information comprises:

determining a first correlation degree and a second correlation degree; the first relevance is the duty ratio of users in the same industry with the target commodity word in the different users; the second relevance is the duty ratio of occupation of preset occupation users in the different users, and the conversion rate corresponding to the preset occupation is larger than the preset conversion rate;

determining the user value based on the first relevance and the second relevance; the user value is positively correlated with the first correlation and the second phase Guan Du.

6. The video generation method according to claim 4 or 5, wherein the conversion operation number is greater in weight than the multi-click number, the multi-click number is greater in weight than the single-click number, and the single-click number is greater in weight than the non-click number.

7. The video generation method of any of claims 2-6, wherein the search results of each search statement include a plurality of associated content;

the method further comprises the steps of:

Acquiring the number of the plurality of associated contents, the content quality score of each associated content, the number of authors of the plurality of associated content, and the author grade score of each author; the content quality score is determined based on a rating operation of the user, and the author grade score is used for indicating popularity of authors;

determining the associated content value of each search statement according to the number of the plurality of associated contents and the content quality score of each associated content; the associated content value is positively correlated with the number of the plurality of associated content and the content quality score;

determining the associated author value of each search statement according to the number of the plurality of associated content authors and the author grade score of each author; the associated author value positively correlates with the number of the plurality of associated content authors and the author rank score;

determining the associated value of each search statement according to the associated content value of each search statement and the associated author value; the associated value is positively correlated with the associated content value and the associated author value.

8. The video generation method according to any one of claims 1-7, wherein the performing semantic analysis on the target search statement to determine whether the element to be published of the target search statement is matched, includes:

Carrying out semantic analysis on the target search statement to obtain a main word and an adverb in the target search statement;

judging whether a target commodity word corresponding to the main word exists in a preset commodity information mapping table according to the main word; the commodity information mapping table comprises a plurality of commodity words, parameter items of each commodity word and parameter values of each parameter item;

judging whether at least one target parameter item and the parameter value of each target parameter item exist in the commodity information mapping table according to the adverbs under the condition that the target commodity word exists in the commodity information mapping table;

and under the condition that the at least one target parameter item and the parameter values of the target parameter items exist in the commodity information mapping table, determining that the commodity information mapping table is matched with the element to be distributed.

9. The video generating method according to claim 8, wherein the performing semantic analysis on the target search sentence to obtain a main word and an adverb in the target search sentence includes:

carrying out semantic analysis on the target search statement, and determining a target semantic structure of the target search statement from a plurality of semantic structures;

And determining the main word and the adverb from the target search statement according to the target semantic structure.

10. The video generating method according to claim 8 or 9, wherein the determining whether the at least one target parameter item and the parameter value of each of the target parameter items exist in the commodity information mapping table according to the adverbs includes:

judging whether at least one target parameter item exists in the commodity information mapping table according to the parameter items in the adverbs when the adverbs comprise the parameter items, and determining the parameter value of the at least one parameter item from the commodity information mapping table when the at least one target parameter item exists;

judging whether a parameter value corresponding to the adverb exists in the commodity information mapping table according to the adverb under the condition that the adverb does not include the parameter item;

and under the condition that the parameter value corresponding to the adverb exists in the commodity information mapping table, determining the at least one target parameter item based on the parameter value corresponding to the adverb.

11. The video generation method according to any one of claims 1 to 10, wherein the generating a target video indicating the element to be distributed includes:

Determining a text to be distributed for expressing the element to be distributed from a preset industrial knowledge graph according to the element to be distributed, and generating the target video according to the text to be distributed; the industrial knowledge graph comprises different commodities, different parameter items and expression modes of parameter values.

12. A video generating device comprises an acquisition unit, a judging unit and a generating unit;

the acquisition unit is used for acquiring target search sentences searched by a user;

the judging unit is used for carrying out semantic analysis on the target search statement and judging whether the target search statement is matched with the element to be issued; the element to be distributed comprises a target commodity word corresponding to the target search statement, at least one target parameter item of the target commodity word and a parameter value of each target parameter item;

the generating unit is used for generating a target video for indicating the element to be distributed under the condition that the element to be distributed is matched.

13. An electronic device, comprising:

at least one processor; and

a memory communicatively coupled to the at least one processor; wherein,,

the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-11.

14. A non-transitory computer readable storage medium storing computer instructions for causing the computer to perform the method of any one of claims 1-11.

15. A computer program product comprising a computer program which, when executed by a processor, implements the method according to any of claims 1-11.