CN114219586A - Shopping recommendation method, device, equipment and storage medium based on video - Google Patents
Shopping recommendation method, device, equipment and storage medium based on video Download PDFInfo
- Publication number
- CN114219586A CN114219586A CN202210040493.6A CN202210040493A CN114219586A CN 114219586 A CN114219586 A CN 114219586A CN 202210040493 A CN202210040493 A CN 202210040493A CN 114219586 A CN114219586 A CN 114219586A
- Authority
- CN
- China
- Prior art keywords
- video
- product
- target
- videos
- user
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 68
- 230000000694 effects Effects 0.000 claims abstract description 17
- 238000012216 screening Methods 0.000 claims description 27
- 230000006399 behavior Effects 0.000 claims description 6
- 238000013480 data collection Methods 0.000 claims description 5
- 230000009193 crawling Effects 0.000 claims description 4
- 238000012544 monitoring process Methods 0.000 claims description 3
- 238000004364 calculation method Methods 0.000 claims description 2
- 238000005457 optimization Methods 0.000 abstract description 3
- 241000282326 Felis catus Species 0.000 description 11
- 230000008569 process Effects 0.000 description 11
- 230000006870 function Effects 0.000 description 9
- 238000006243 chemical reaction Methods 0.000 description 8
- 238000004458 analytical method Methods 0.000 description 7
- 238000010586 diagram Methods 0.000 description 6
- 238000000605 extraction Methods 0.000 description 5
- 238000005065 mining Methods 0.000 description 5
- 238000012552 review Methods 0.000 description 5
- 230000007547 defect Effects 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 230000003252 repetitive effect Effects 0.000 description 3
- 238000013507 mapping Methods 0.000 description 2
- 230000001360 synchronised effect Effects 0.000 description 2
- 238000009825 accumulation Methods 0.000 description 1
- 230000001413 cellular effect Effects 0.000 description 1
- 238000004590 computer program Methods 0.000 description 1
- 230000008451 emotion Effects 0.000 description 1
- 230000008676 import Effects 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0631—Item recommendations
Landscapes
- Business, Economics & Management (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Marketing (AREA)
- Strategic Management (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The application provides a shopping recommendation method, device, equipment and storage medium based on videos, wherein the method comprises the following steps: collecting video activity data of a target user within a preset time period; acquiring the weight of each data index of related videos of the same kind of products and a score corresponding to an interval to which the index value of the data index belongs; calculating the label interest index of the first commodity label corresponding to the similar product according to the weight of each data index and the corresponding score; if the tag interest index is larger than or equal to a first threshold, a first target video is collected, the priority recommendation level of the first target video is determined according to the tag interest indexes of different products, and the first target video is recommended according to the priority recommendation level. According to the method and the device, the product which the user is interested in is predicted through the behavior of the user browsing the video, and the video corresponding to the product which the user is interested in is recommended to the user, so that the optimization of the matching mode of the current shopping requirement is realized, and the potential requirement of the user is effectively mined.
Description
Technical Field
The present application relates to the field of computer technologies, and in particular, to a method, an apparatus, a device, and a storage medium for video-based shopping recommendation.
Background
The market potential of video shopping is huge, but how to accurately detect consumption demands in the process of refreshing videos of users is very critical and is also an industrial problem. The demand recommendation mode in the prior art mainly comprises live shopping recommendation, shopping cart link placement below a video and similar commodity recommendation below a related shopping page. However, not every video is embedded with a product link while the user is browsing the videos, resulting in no links or few links being accessible for the items that the user wishes to purchase. Due to the fact that the default or selectivity of the purchasing channel is small, and the video platform is different from a professional shopping platform, similar commodities cannot be searched directly or quickly, the consumption requirements of users cannot be met completely, and a large part of markets are missed.
Disclosure of Invention
The method aims to solve the technical problem that commodity links cannot be provided for users in time in the existing video shopping technology, so that part of markets are missed. The application provides a shopping recommendation method, device, equipment and storage medium based on videos, and mainly aims to recommend videos with commodity links to users according to the preference of the users for browsing the videos so as to meet the purchase demands of the users and improve the market conversion rate.
In order to achieve the above object, the present application provides a video-based shopping recommendation method, including:
collecting video activity data in a preset time period of a target user, wherein the video activity data comprise a plurality of data indexes corresponding to related videos of at least one type of product and corresponding index values, and the plurality of data indexes comprise the watching times and watching duration of the related videos of the same type of product, the browsing duration of related comments, the access duration of commodity pages corresponding to the related videos and the searching times of the same type of product;
acquiring the weight of each data index and a score corresponding to the interval to which the index value of the data index belongs;
calculating the label interest indexes of the same first commodity label corresponding to the same kind of products according to the weight and the corresponding score of each data index;
if the label interest index is larger than or equal to a first threshold value, collecting a first target video corresponding to a first commodity label, wherein the first target video carries a product link of a related product corresponding to the first commodity label;
sorting the label interest indexes corresponding to different kinds of products in a descending order, and determining the priority recommendation level of the first target video corresponding to the different kinds of products according to the obtained sorting result;
and sequentially recommending first target videos corresponding to different types of products to the target user according to the priority recommendation level.
In addition, to achieve the above object, the present application also provides a video-based shopping recommendation apparatus, including:
the data collection module is used for collecting video activity data in a preset time period of a target user, the video activity data comprise a plurality of data indexes corresponding to related videos of at least one type of product and corresponding index values, and the plurality of data indexes comprise the watching times and watching duration of the related videos of the similar product, the browsing duration of related comments, the access duration of commodity pages corresponding to the related videos and the searching times of the similar product;
the weight score matching module is used for acquiring the weight of each data index and the score corresponding to the interval to which the index value of the data index belongs;
the calculation module is used for calculating the label interest indexes of the same first commodity label corresponding to the same kind of product according to the weight and the corresponding score of each data index;
the first video searching module is used for searching a first target video corresponding to the first commodity label if the label interest index is larger than or equal to a first threshold value, wherein the first target video carries a product link of a related product corresponding to the first commodity label;
the recommendation level determining module is used for sorting the tag interest indexes corresponding to different types of products in a descending order and determining the priority recommendation level of the first target video corresponding to the different types of products according to the obtained sorting result;
and the first recommending module is used for sequentially recommending first target videos corresponding to different types of products to the target user according to the priority recommending level.
To achieve the above object, the present application also provides a computer device comprising a memory, a processor and computer readable instructions stored on the memory and executable on the processor, the processor executing the computer readable instructions to perform the steps of the video-based shopping recommendation method according to any one of the preceding claims.
To achieve the above object, the present application also provides a computer readable storage medium having computer readable instructions stored thereon, which, when executed by a processor, cause the processor to perform the steps of the video-based shopping recommendation method according to any one of the preceding claims.
According to the shopping recommendation method, the device, the equipment and the storage medium based on the videos, the products interested by the users are predicted according to the behaviors of the users for browsing the videos, the videos corresponding to the interested products are recommended to the users, the potential requirements of the users are effectively mined by optimizing the matching mode of the current shopping requirements, the problem of inaccurate current similarity recommendation is solved, the degree of requirement analysis mining and commodity matching is greatly improved, and the defect of repetitive low-efficiency recommendation caused by fuzzy recommendation limited to similar products is effectively overcome.
Drawings
FIG. 1 is a diagram illustrating an application scenario of a video-based shopping recommendation method according to an embodiment of the present application;
FIG. 2 is a schematic flow chart illustrating a video-based shopping recommendation method according to an embodiment of the present application;
FIG. 3 is a block diagram of a video-based shopping recommendation device according to an embodiment of the present application;
fig. 4 is a block diagram of an internal structure of a computer device according to an embodiment of the present application.
The implementation, functional features and advantages of the objectives of the present application will be further explained with reference to the accompanying drawings.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are some embodiments of the present application, but not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.
The video-based shopping recommendation method can be applied to a shopping recommendation system shown in fig. 1. The shopping recommendation system includes a terminal device 110 and a server side 120 which communicate through a network. The terminal device 110 may be, but is not limited to, various personal computers, notebook computers, smart phones, tablet computers, and portable wearable devices. The server side 120 may be implemented by an independent server or a server cluster composed of a plurality of servers.
The terminal device 110 is installed with a video APP, and the server 120 is an application server corresponding to the video APP. The server 120 may not only provide various videos to the terminal device 110 for the user to watch, but also implement the video-based shopping recommendation method of the present application. Specifically, video activity data in a preset time period of a target user are collected, the video activity data comprise a plurality of data indexes corresponding to related videos of at least one type of product and corresponding index values, and the plurality of data indexes comprise the watching times and watching duration of the related videos of the similar product, the browsing duration of related comments, the access duration of commodity pages corresponding to the related videos and the searching times of the similar product; acquiring the weight of each data index and a score corresponding to the interval to which the index value of the data index belongs; calculating the label interest indexes of the same first commodity label corresponding to the same kind of products according to the weight and the corresponding score of each data index; if the label interest index is larger than or equal to a first threshold value, collecting a first target video corresponding to a first commodity label, wherein the first target video carries a product link of a related product corresponding to the first commodity label; sorting the label interest indexes corresponding to different kinds of products in a descending order, and determining the priority recommendation level of the first target video corresponding to the different kinds of products according to the obtained sorting result; and sequentially recommending first target videos corresponding to different types of products to the target user according to the priority recommendation level.
Fig. 2 is a flowchart illustrating a video-based shopping recommendation method according to an embodiment of the present application. Referring to fig. 2, the method is illustrated as applied to the server side in fig. 1. The video-based shopping recommendation method includes the following steps S100-S600.
S100: the method comprises the steps of collecting video activity data in a preset time period of a target user, wherein the video activity data comprise a plurality of data indexes corresponding to related videos of at least one type of product and corresponding index values, and the data indexes comprise the watching times and watching duration of the related videos of the same type of product, the browsing duration of related comments, the access duration of commodity pages corresponding to the related videos and the searching times of the same type of product.
Specifically, the types of videos watched by the target user in the preset time period are various, so that all watched videos are classified and counted according to the product relevance, and a plurality of data indexes and index values of the relevant videos of each product type can be obtained.
The video activity data represent the attention or participation of the video, and the video activity data source can be obtained by embedding points in the video background. The number of views of the video is the accumulation of the number of effective views of the video, for example, if the user finishes watching the video 80% of the time, the user can mark as 1 effective view. The same video can be effectively watched at most once; the same video can be effectively watched for multiple times, and once watching time reaches a certain time length during the period of re-entering and re-exiting every time, the watching time is counted for one time.
The viewing duration refers to a duration of time that the user stays on the viewed video. The duration of the review comment-related comment refers to the stay duration of the user in the comment interface of the video. The access duration of the commodity page refers to the access duration of the commodity page corresponding to the product link implanted in the related video, and a user can browse the commodity page after clicking a certain shopping link in the video, wherein the commodity page is the commodity page corresponding to the video. The searching times of the similar products refer to the purpose that the user searches the similar products by searching videos corresponding to the similar products, and the accumulated searching times are recorded as the total searching times in the preset time period. When searching videos corresponding to similar products, a user can search through keywords, and the search keywords can be commodity names or commodity models and the like, but are not limited to the commodity names or the commodity models.
The preset time period such as the last month or 1 week or half month, etc. is not limited thereto.
Each video is tagged with a corresponding video tag that reflects or is related to the content of the video to some extent. If a video is a product category related video, its video tag may include a product name, a product model number or a product category or a product usage, etc. And judging which videos correspond to the same kind of products according to the video tags. Similar label sets, such as labels of cat food, are pre-stored in the label library, and the similar labels can be covered into cat food, kitten food, cat food recommendation, imported cat food and the like. For example, the label of the dress can cover a long dress, a medium-long dress, a plain dress and the like. And if the video labels of the videos belong to similar labels in the label library, judging that the videos are related videos of the same type of products.
In addition, according to the pictures extracted from the video, which products appear in the video can be analyzed, and which videos are related to the same products can also be judged.
S200: and acquiring the weight of each data index and the score corresponding to the interval to which the index value of the data index belongs.
Specifically, a weight table and a score table are preset before step S200, and the weight of each data index can be obtained by looking up the weight table. The scores corresponding to the different value intervals of each data index are stored in the score table, so that the value interval of the target index to which the data index belongs can be determined according to the index value of the data index, and the score corresponding to the value interval of the target index is obtained.
S300: and calculating the label interest indexes of the same first commodity label corresponding to the similar products according to the weight and the corresponding score of each data index.
Specifically, the first merchandise tag is a broad class of similar tags. For example, the first commercial label for adult cat food, kitten cat food, cat food recommendations, import cat food, and similar labels is cat food.
The similar labels are divided in the label library, the commodity labels corresponding to the similar labels are also divided, and the same video label may correspond to a plurality of different commodity labels. The tag interest index is the sum of the products of the weights of all data indices and the corresponding scores.
S400: and if the label interest index is larger than or equal to a first threshold value, collecting a first target video corresponding to the first commodity label, wherein the first target video carries a product link of a related product corresponding to the first commodity label.
Specifically, if the tag interest index is greater than or equal to the first threshold, it indicates that the related product corresponding to the first commodity tag is a product in which the user is interested, and therefore, a first target video corresponding to the product in interest may be collected and recommended to the target user. The product categories in which the target user is interested may be multiple, and thus, the first target video gathered may include related videos for multiple different product categories.
The recommended target video can be a short video or a live video. The recommended first target video carrying product link is convenient for the user to directly enter the commodity detail page from the video implanted product link, so that the purchase is convenient, and the purchase rate is promoted.
S500: and sorting the label interest indexes corresponding to the different kinds of products in a descending order, and determining the priority recommendation level of the first target video corresponding to the different kinds of products according to the obtained sorting result.
Specifically, the tag interest index judges the interest degree of the target user in the corresponding product through multiple dimensions, and therefore, the interest degree of the target user in the related product is higher as the tag interest index is higher. According to the method and the device, when the target user is interested in multiple different products at the same time, the priority recommendation level of the first target video corresponding to the different products is determined through the tag interest index. The higher the tag interest index, the higher the priority recommendation level.
S600: and sequentially recommending first target videos corresponding to different types of products to the target user according to the priority recommendation level.
Specifically, the first target video with higher priority recommendation level is preferentially recommended to the target user, so that the target user can preferentially see the video of the product in most interest, and the purpose of improving the conversion rate of commodity purchase is achieved.
Certainly, if there are more first target videos, the first target videos may also be divided into multiple groups, each group includes some first target videos corresponding to all kinds of products, and the first target videos in each group are sorted according to the priority recommendation level in the recommendation order. All the first target videos are played in turn according to the sequence of one group, so that the problem that the user is tired when watching the first target videos of only one type of products within a period of time can be solved, and the aim that the target videos corresponding to the products which are more interested by the user are played preferentially can be achieved.
According to the method and the device, the product which the user is interested in is predicted according to the behavior of the user browsing the video, the video corresponding to the product which the user is interested in is recommended to the user, optimization of the matching mode of the current shopping requirement is achieved, the potential requirement of the user is effectively mined, the problem that the current similarity recommendation is inaccurate is solved, the matching degree of the requirement analysis mining and the recommended commodity is greatly improved, and the defect that repeated low-efficiency recommendation is caused by fuzzy recommendation limited to similar products is effectively overcome. The market conversion rate is improved while the purchase demand of the user is met.
In one embodiment, the method for collecting the watching times of the related videos of the same type of products comprises the following steps:
and monitoring the video watching behaviors of the target user through the buried point, and accumulating the watching times of videos with the actual watching time length reaching the first preset proportion in the related videos of the similar products to obtain the watching times of the related videos of the similar products.
Specifically, the video watched by the target user in the preset time period has various subjects, some videos are purely entertainment videos without product recommendation, some videos with product recommendation properties, some videos with product links, and some videos without product links. The types of products recommended by videos with product recommendation can be various, so that videos corresponding to similar products need to be analyzed in a targeted manner according to categories, and whether users are interested in the similar products corresponding to the videos can be found out.
In addition, the user generally skips or scratches uninteresting videos, so that whether the user is interested in the video content can be judged according to the time length for watching a section of video, and further whether commodities appearing in the video are interested can be judged, and therefore the embodiment restricts the effective watching performance of the videos which are used for analyzing and mining the user preferences. For videos with short watching time, the user can be judged to be uninterested in the products or subject matters, and the videos cannot be used as videos for analyzing and mining the requirements of the user. The first preset proportion may be specifically set according to actual conditions, and for example, 80% and 90% and the like are not limited thereto. Only videos having an effective viewing time period of 80% or more of the total time period of the video, for example, are counted for one viewing. Whether the same video can be counted into the watching times for multiple times can be determined according to actual conditions.
The embodiment selects the video which is effectively watched, so that the user requirements can be more accurately mined, the conflict emotion brought to the user by blind recommendation is avoided, and the user experience is improved.
In one embodiment, gathering a first target video corresponding to a first merchandise tag in step S400 includes:
extracting seller information of a seller concerned by a target user;
extracting a video tag of a seller video of a concerned seller according to the seller information;
and screening out a first target video which is matched with the first commodity label and carries the product link from the seller videos.
Specifically, a seller concerned by a target user, namely a concerned related blogger, matches a video tag with high matching degree or high correlation with a first commodity tag based on the first commodity tag corresponding to a commodity interested by the user, finds out a video in which the video tag is matched with the first commodity tag in a video released by the concerned blogger as a target video, and pushes the target video to the target user.
The video tags of the seller videos can be obtained through the product pages corresponding to the product links implanted in the seller videos, and also can be extracted directly from the tags of the seller videos, so that whether the video tags in the seller videos are matched with the first commodity tags or not is judged. Wherein, a mapping relation exists between the video label and the commodity label. The mapping relation of various sub-commodity labels or variants of the commodity labels corresponding to the commodity labels is stored in the label library, and the video labels are the sub-commodity labels or the variants or synonyms of the commodity labels.
The first target video may be a short video or a live video.
In consideration of the fact that the user has higher trust degree on the concerned bloggers or sellers, but the number of videos issued by the bloggers or sellers is large, and the target user cannot easily look over a plurality of videos one by one to find the target product link, therefore, the target video of the sellers or the bloggers trusted by the user is automatically recommended according to the mined user preference, and the conversion rate can be further improved.
In one embodiment, gathering a first target video corresponding to a first merchandise tag in step S400 includes:
screening out candidate videos which are matched with the first commodity label and carry product links from videos which are not watched by a target user;
extracting commodity pricing of commodities corresponding to product links in the candidate videos;
and screening out a first target video carrying product links with different commodity pricing from the candidate videos according to the commodity pricing.
Specifically, the candidate video carries the product link, so that the commodity pricing can be extracted from the product page corresponding to the product link.
The candidate videos can be divided into price intervals according to commodity pricing, then are recombined according to commodity price intervals, each group comprises a plurality of candidate videos, the candidate videos are located in different price intervals, each video recommendation is a candidate video recommended in one combination, and after the candidate videos of the combination are recommended, the candidate videos in the next combination are recommended, and the like.
In consideration of different budgets and psychological prices of different users, the embodiment recommends various videos carrying product links with different prices to the target user, provides a wide price selectable range, avoids the recommended price from being too low or too high to match consumption psychology, and promotes conversion of video orders.
In one embodiment, the method further comprises:
crawling comments of a target user under a watched video, and screening out target comments carrying the link intention of the sought product;
obtaining product information of a product which is interested by a target user according to the target comment;
acquiring a second commodity label corresponding to a product in which the target user is interested according to the product information;
screening out a second target video with the video tag matched with the second commodity tag from videos which are not watched by the target user;
and recommending the second target video to the target user.
Specifically, interesting products of the target user are obtained from videos corresponding to the target comments, and/or a preset number of high-quality comments before the videos corresponding to the target comments are traversed to obtain the interesting products of the target user.
For example, if the target user reviews "a link to cat food," it may be determined that the product of interest to the target user is cat food. If the target user reviews "link up," then traversing the top 50 high-quality reviews under the video, for example, may mine product information for the product of interest to the target user from these high-quality reviews.
The obtained comment of the product link represents the very clear purchasing intention of the target user, so that the real requirement of the user can be very clearly mined according to the target comment. According to the method and the device, the product information of the interested product is mined according to the comments of the target user in the video, the certainty is high, convenience is brought to potential users with strong purchasing willingness, and the order conversion rate is improved.
In one embodiment, the method further comprises:
if the product information of the product in which the target user is interested cannot be acquired according to the target comment, extracting pictures from the video corresponding to the target comment according to frames so as to screen out at least one different target product picture from the extracted pictures;
screening a third target video from videos which are not watched by the target user, wherein the product picture in the third target video and the target product picture have the same or similar product;
and recommending the third target video to the target user.
Specifically, if a product interested by the target user cannot be estimated according to the comment of the target user and the comment of the video, the video corresponding to the same or similar product is searched in a picture searching manner according to the product picture in the video, and then the video is recommended to the target user.
Generally, a video has a cover, whether a product in a target product picture appears or not can be identified according to the cover of the video, and whether the video is a third target video or not can be further judged.
And candidate product pictures can be extracted from the product pages corresponding to the product links carried by the video, the target product pictures and the candidate product pictures are matched or identified, and whether the video is a third target video can be judged.
In addition, based on the image recognition technology, product information of a product which the target user is interested in can be acquired according to a target product picture extracted from a video watched by the user, and the corresponding target video can be matched according to the product information and recommended to the target user.
According to the method and the device, when the user cannot accurately express the real purchasing demand in a language or cannot mine the real purchasing intention of the user through comment, similar videos carrying product links are recommended to the target user in a picture searching mode, so that the user can quickly acquire the purchasing links or channels of related products, the loss of business or market caused by the fact that the purchasing links cannot be known is reduced, the user purchasing experience is greatly improved, and the purchasing conversion rate is improved.
In one embodiment, the method further comprises:
if the product information of the product in which the target user is interested cannot be acquired according to the target comment, extracting pictures from the video corresponding to the target comment according to frames so as to screen out at least one different candidate product picture from the extracted pictures;
displaying the candidate product pictures for selection by a target user;
determining a target product picture of a product in which the target user is interested from the candidate product pictures according to the selection result of the candidate user;
screening a fourth target video from videos which are not watched by the target user, wherein the product picture in the fourth target video and the target product picture have the same or similar product;
and recommending the fourth target video to the target user.
Specifically, after a specific product category is obtained through analysis, the specific product category is analyzed and compared with a video picture, and similar product recommendation is performed in a picture searching mode. If the specific categories are unknown, multiple categories of commodities acquired in the video can be recommended and recommended to the user in batches; it is also contemplated to click through a pop-up, for example, to let the user directly select: "help me find similar products in video: trousers, jacket and shoes "to improve accuracy. And after the real requirements of the target user are determined, continuously recommending the videos implanted with the related commodity shopping links to the user.
The recommendation methods in the above embodiments may be used alternatively or in a plurality of overlapping manners, and the specific recommendation manner may be determined according to the availability of actual data and the user requirement that can be mined, so as to accurately mine the user requirement through one or more manners to recommend an accurate video to a user.
According to the method and the device, the product which the user is interested in is predicted according to the behavior of the user browsing the video, and the video corresponding to the product which is interested in is recommended to the user, so that the potential requirements of the user are effectively mined by optimizing the matching mode of the current shopping requirements, the problem of inaccurate recommendation of the current similarity is solved, the degree of requirement analysis mining and commodity matching is greatly improved, and the defect of repetitive low-efficiency recommendation caused by fuzzy recommendation limited to similar products is effectively overcome.
The optimization of the current shopping demand matching mode can well dig out the potential demands of the user, the problem of inaccurate similarity recommendation at present is solved, the trust of the user to the concerned blogger and the analysis of the comment content of the user are fully utilized, the technical capabilities of NLP and commodity interception in video are utilized, the degree of demand analysis digging and commodity matching is greatly improved, the fuzzy and repetitive low-efficiency recommendation limited to similar products is not only realized, the purchase demand of the user is met, and the trust and good sensitivity of the user to a platform are improved, and meanwhile, the market conversion rate is improved.
It should be understood that, the sequence numbers of the steps in the foregoing embodiments do not imply an execution sequence, and the execution sequence of each process should be determined by its function and inherent logic, and should not constitute any limitation to the implementation process of the embodiments of the present application.
FIG. 4 is a block diagram of a video-based shopping recommendation device according to an embodiment of the present application. Referring to fig. 4, the apparatus includes:
the data collection module 100 is configured to collect video activity data within a preset time period of a target user, where the video activity data includes multiple data indexes and corresponding index values corresponding to related videos of at least one type of product, and the multiple data indexes include viewing times and viewing durations of the related videos of the similar product, browsing durations of related comments, access durations of a commodity page corresponding to the related video, and search times of the similar product;
the weight score matching module 200 is configured to obtain the weight of each data index and a score corresponding to an interval to which an index value of the data index belongs;
the calculating module 300 is configured to calculate tag interest indexes of the same first commodity tag corresponding to the same kind of product according to the weight and the corresponding score of each data index;
the first video searching module 400 is configured to collect a first target video corresponding to a first commodity label if the label interest index is greater than or equal to a first threshold, where the first target video carries a product link of a related product corresponding to the first commodity label;
the recommendation level determining module 500 is configured to perform descending order sorting on the tag interest indexes corresponding to different types of products, and determine a priority recommendation level of a first target video corresponding to different types of products according to an obtained sorting result;
the first recommending module 600 is configured to recommend first target videos corresponding to different types of products to a target user in sequence according to the priority recommendation level.
In one embodiment, the data collection module 100 is specifically configured to:
and monitoring the video watching behaviors of the target user through the buried point, and accumulating the watching times of videos with the actual watching time length reaching the first preset proportion in the related videos of the similar products to obtain the watching times of the related videos of the similar products.
In one embodiment, the first video search module 400 specifically includes:
the first extraction module is used for extracting seller information of a seller concerned by a target user;
the second extraction module is used for extracting the video tags of the seller videos of the concerned sellers according to the seller information;
the first screening module is used for screening out a first target video which is matched with the first commodity label and carries the product link from the seller videos.
In one embodiment, the first video search module 400 specifically includes:
the second screening module is used for screening candidate videos which are matched with the first commodity label and carry product links from videos which are not watched by the target user;
the third extraction module is used for extracting commodity pricing of commodities corresponding to product links in the candidate videos;
and the third screening module is used for screening out the first target video carrying the product links with different commodity pricing from the candidate videos according to the commodity pricing.
In one embodiment, the apparatus further comprises:
the data crawling module is used for crawling comments of a target user under a watched video and screening out target comments carrying the link intention of the sought product;
the text analysis module is used for acquiring product information of a product in which the target user is interested according to the target comment;
the first searching module is used for acquiring a second commodity label corresponding to a product in which the target user is interested according to the product information;
the fourth screening module is used for screening out a second target video with the video tag matched with the second commodity tag from videos which are not watched by the target user;
and the second recommending module is used for recommending the second target video to the target user.
In one embodiment, the apparatus further comprises:
the image extraction module is used for extracting images according to frames from a video corresponding to the target comment to screen at least one different target product image from the extracted images if the product information of the product in which the target user is interested cannot be acquired according to the target comment;
the fifth screening module is used for screening a third target video from videos which are not watched by the target user, and a product picture in the third target video and the target product picture have the same or similar product;
and the third recommending module is used for recommending the third target video to the target user.
In one embodiment, the apparatus further comprises:
the image extraction module is used for extracting images according to frames from a video corresponding to the target comment to screen out at least one different candidate product image from the extracted images if the product information of the product in which the target user is interested cannot be acquired according to the target comment;
the picture display module is used for displaying candidate product pictures for selection of a target user;
the determining module is used for determining a target product picture of a product in which the target user is interested from the candidate product pictures according to the selection result of the candidate user;
the sixth screening module is used for screening a fourth target video from videos which are not watched by the target user, and a product picture in the fourth target video and the target product picture have the same or similar product;
and the fourth recommending module is used for recommending the fourth target video to the target user.
Wherein the meaning of "first" and "second" in the above modules/units is only to distinguish different modules/units, and is not used to define which module/unit has higher priority or other defining meaning. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or modules is not necessarily limited to those steps or modules explicitly listed, but may include other steps or modules not explicitly listed or inherent to such process, method, article, or apparatus, and such that a division of modules presented in this application is merely a logical division and may be implemented in a practical application in a further manner.
For specific limitations of the video-based shopping recommendation device, reference may be made to the above limitations of the video-based shopping recommendation method, which are not described herein again. The modules in the video-based shopping recommendation device can be wholly or partially implemented by software, hardware and a combination thereof. The modules can be embedded in a hardware form or independent from a processor in the computer device, and can also be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.
Fig. 4 is a block diagram of an internal structure of a computer device according to an embodiment of the present application. The computer device may specifically be the server side in fig. 1. As shown in fig. 4, the computer apparatus includes a processor, a memory, a network interface, an input device, and a display screen connected through a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory includes a storage medium and an internal memory. The storage medium may be a nonvolatile storage medium or a volatile storage medium. The storage medium stores an operating system and may also store computer readable instructions that, when executed by the processor, may cause the processor to implement a video-based shopping recommendation method. The internal memory provides an environment for the operating system and execution of computer readable instructions in the storage medium. The internal memory may also have computer readable instructions stored thereon that, when executed by the processor, cause the processor to perform a method for video-based shopping recommendation. The network interface of the computer device is used for communicating with an external server through a network connection. The display screen of the computer equipment can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer equipment can be a touch layer covered on the display screen, a key, a track ball or a touch pad arranged on the shell of the computer equipment, an external keyboard, a touch pad or a mouse and the like.
In one embodiment, a computer device is provided, comprising a memory, a processor, and computer readable instructions (e.g., a computer program) stored on the memory and executable on the processor, the processor implementing the steps of the video-based shopping recommendation method in the above embodiments when executing the computer readable instructions, such as the steps S100 to S600 shown in fig. 2 and other extensions of the method and related steps. Alternatively, the processor, when executing the computer readable instructions, implements the functions of the modules/units of the video-based shopping recommendation device in the above embodiments, such as the functions of the modules 100 to 600 shown in fig. 3. To avoid repetition, further description is omitted here.
The Processor may be a Central Processing Unit (CPU), other general purpose Processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field-Programmable gate array (FPGA) or other Programmable logic device, discrete gate or transistor logic device, discrete hardware component, etc. The general purpose processor may be a microprocessor or the processor may be any conventional processor or the like, the processor being the control center of the computer device and the various interfaces and lines connecting the various parts of the overall computer device.
The memory may be used to store computer readable instructions and/or modules, and the processor may implement various functions of the computer apparatus by executing or executing the computer readable instructions and/or modules stored in the memory and invoking data stored in the memory. The memory may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required by at least one function (such as a sound playing function, an image playing function, etc.), and the like; the storage data area may store data (such as audio data, video data, etc.) created according to the use of the cellular phone, etc.
The memory may be integrated in the processor or may be provided separately from the processor.
Those skilled in the art will appreciate that the architecture shown in fig. 4 is merely a block diagram of some of the structures associated with the disclosed aspects and is not intended to limit the computing devices to which the disclosed aspects apply, as particular computing devices may include more or less components than those shown, or may combine certain components, or have a different arrangement of components.
In one embodiment, a computer readable storage medium is provided having computer readable instructions stored thereon which, when executed by a processor, implement the steps of the video-based shopping recommendation method of the above embodiments, such as the steps S100-S600 shown in fig. 2 and extensions of other extensions and related steps of the method. Alternatively, the computer readable instructions, when executed by the processor, implement the functions of the modules/units of the video-based shopping recommendation device in the above embodiments, such as the functions of the modules 100-600 shown in fig. 3. To avoid repetition, further description is omitted here.
It will be understood by those of ordinary skill in the art that all or part of the processes of the methods of the embodiments described above may be implemented by instructing associated hardware to implement computer readable instructions, which may be stored in a computer readable storage medium, and when executed, may include processes of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in the embodiments provided herein may include non-volatile and/or volatile memory, among others. Non-volatile memory can include read-only memory (ROM), Programmable ROM (PROM), Electrically Programmable ROM (EPROM), Electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), Dynamic RAM (DRAM), Synchronous DRAM (SDRAM), double-rate SDRAM (DDRSDRAM), Enhanced SDRAM (ESDRAM), synchronous Link (Synchlink) DRAM (SLDRAM), Rambus Direct RAM (RDRAM), direct bus dynamic RAM (DRDRAM), and bus dynamic RAM (RDRAM).
It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, apparatus, article, or method that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, apparatus, article, or method. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, apparatus, article, or method that includes the element.
The above-mentioned serial numbers of the embodiments of the present application are merely for description and do not represent the merits of the embodiments. Through the above description of the embodiments, those skilled in the art will clearly understand that the method of the above embodiments can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware, but in many cases, the former is a better implementation manner. Based on such understanding, the technical solution of the present application may be substantially or partially embodied in the form of a software product, which is stored in a storage medium (e.g., ROM/RAM, magnetic disk, optical disk) as described above and includes instructions for enabling a terminal device (e.g., a mobile phone, a computer, a server, or a network device) to execute the method according to the embodiments of the present application.
The above description is only a preferred embodiment of the present application, and not intended to limit the scope of the present application, and all modifications of equivalent structures and equivalent processes, which are made by the contents of the specification and the drawings of the present application, or which are directly or indirectly applied to other related technical fields, are included in the scope of the present application.
Claims (10)
1. A video-based shopping recommendation method, the method comprising:
collecting video activity data in a preset time period of a target user, wherein the video activity data comprise a plurality of data indexes corresponding to related videos of at least one kind of products and corresponding index values, and the data indexes comprise the watching times and watching duration of the related videos of the same kind of products, the browsing duration of related comments, the access duration of commodity pages corresponding to the related videos and the searching times of the same kind of products;
acquiring the weight of each data index and a score corresponding to the interval to which the index value of the data index belongs;
calculating the label interest index of the same first commodity label corresponding to the similar product according to the weight and the corresponding score of each data index;
if the label interest index is larger than or equal to a first threshold value, collecting a first target video corresponding to the first commodity label, wherein the first target video carries a product link of a related product corresponding to the first commodity label;
sorting the label interest indexes corresponding to different kinds of products in a descending order, and determining the priority recommendation level of the first target video corresponding to the different kinds of products according to the obtained sorting result;
and sequentially recommending first target videos corresponding to different types of products to the target user according to the priority recommendation level.
2. The method according to claim 1, wherein the collection method of the number of times of watching related videos of the same kind of products comprises:
and monitoring the video watching behaviors of the target user through the buried point, and accumulating the watching times of videos with the actual watching time length reaching a first preset proportion in the related videos of the similar products to obtain the watching times of the related videos of the similar products.
3. The method of claim 1, wherein said gathering a first target video corresponding to the first merchandise tag comprises:
extracting seller information of a seller concerned by the target user;
extracting a video tag of a seller video of the concerned seller according to the seller information;
and screening out a first target video which is matched with the first commodity label and carries a product link from the seller videos.
4. The method of claim 1, wherein said gathering a first target video corresponding to the first merchandise tag comprises:
screening out candidate videos which are matched with the first commodity label and carry product links from videos which are not watched by the target user;
extracting commodity pricing of commodities corresponding to the product links in the candidate videos;
and screening out a first target video carrying product links with different commodity pricing from the candidate videos according to the commodity pricing.
5. The method of claim 1, further comprising:
crawling the comments of the target user under the watched video, and screening out the target comments carrying the link intention of the sought product;
obtaining product information of a product which is interested by the target user according to the target comment;
acquiring a second commodity label corresponding to a product in which the target user is interested according to the product information;
screening out a second target video with a video tag matched with the second commodity tag from videos which are not watched by the target user;
and recommending the second target video to the target user.
6. The method of claim 5, further comprising:
if the product information of the product which the target user is interested in cannot be acquired according to the target comment, extracting pictures from the video corresponding to the target comment according to frames so as to screen at least one different target product picture from the extracted pictures;
screening a third target video from videos which are not watched by the target user, wherein product pictures in the third target video have the same or similar products with the target product pictures;
recommending the third target video to the target user.
7. The method of claim 5, further comprising:
if the product information of the product in which the target user is interested cannot be acquired according to the target comment, extracting pictures from the video corresponding to the target comment according to frames so as to screen out at least one different candidate product picture from the extracted pictures;
displaying the candidate product picture for the target user to select;
determining a target product picture of a product in which the target user is interested from the candidate product pictures according to the selection result of the candidate user;
screening a fourth target video from videos which are not watched by the target user, wherein a product picture in the fourth target video and the target product picture have the same or similar product;
recommending the fourth target video to the target user.
8. A video-based shopping recommendation apparatus, the apparatus comprising:
the system comprises a data collection module, a search module and a search module, wherein the data collection module is used for collecting video activity data in a preset time period of a target user, the video activity data comprises a plurality of data indexes and corresponding index values corresponding to related videos of at least one type of product, and the data indexes comprise the watching times and watching duration of the related videos of the same type of product, the browsing duration of related comments, the access duration of commodity pages corresponding to the related videos and the search times of the same type of product;
the weight score matching module is used for acquiring the weight of each data index and the score corresponding to the interval to which the index value of the data index belongs;
the calculation module is used for calculating the label interest indexes of the same first commodity label corresponding to the similar product according to the weight and the corresponding score of each data index;
the first video searching module is used for collecting a first target video corresponding to the first commodity label if the label interest index is larger than or equal to a first threshold value, wherein the first target video carries a product link of a related product corresponding to the first commodity label;
the recommendation level determining module is used for sorting the tag interest indexes corresponding to different types of products in a descending order and determining the priority recommendation level of the first target video corresponding to the different types of products according to the obtained sorting result;
and the first recommending module is used for sequentially recommending first target videos corresponding to different types of products to the target user according to the priority recommending level.
9. A computer device comprising a memory, a processor, and computer readable instructions stored on the memory and executable on the processor, wherein the processor when executing the computer readable instructions performs the steps of the video-based shopping recommendation method of any one of claims 1-7.
10. A computer readable storage medium having computer readable instructions stored thereon, which when executed by a processor, cause the processor to perform the steps of the video-based shopping recommendation method of any one of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210040493.6A CN114219586A (en) | 2022-01-14 | 2022-01-14 | Shopping recommendation method, device, equipment and storage medium based on video |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210040493.6A CN114219586A (en) | 2022-01-14 | 2022-01-14 | Shopping recommendation method, device, equipment and storage medium based on video |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114219586A true CN114219586A (en) | 2022-03-22 |
Family
ID=80708217
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210040493.6A Pending CN114219586A (en) | 2022-01-14 | 2022-01-14 | Shopping recommendation method, device, equipment and storage medium based on video |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114219586A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114866693A (en) * | 2022-04-15 | 2022-08-05 | 苏州清睿智能科技股份有限公司 | Information interaction method and device based on intelligent terminal |
| CN116828270A (en) * | 2023-07-05 | 2023-09-29 | 联通沃音乐文化有限公司 | Intelligent segmentation labeling method and system for video data stream |
| CN117495513A (en) * | 2024-01-03 | 2024-02-02 | 深圳市智百威科技发展有限公司 | Shopping recommendation method and system based on big data and artificial intelligence |
| CN117575662A (en) * | 2024-01-17 | 2024-02-20 | 深圳市微购科技有限公司 | Commercial intelligent business decision support system and method based on video analysis |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120246000A1 (en) * | 2009-12-15 | 2012-09-27 | Yarvis Mark D | Techniques to capture context and location information and utilize heuristics to turn location tracked over time and context information into semantic location information |
| CN105512331A (en) * | 2015-12-28 | 2016-04-20 | 海信集团有限公司 | Video recommending method and device |
| CN111767429A (en) * | 2020-06-29 | 2020-10-13 | 北京奇艺世纪科技有限公司 | Video recommendation method and device and electronic equipment |
| CN113420181A (en) * | 2021-06-23 | 2021-09-21 | 未鲲(上海)科技服务有限公司 | Video recommendation method and device, computer equipment and storage medium |
-
2022
- 2022-01-14 CN CN202210040493.6A patent/CN114219586A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120246000A1 (en) * | 2009-12-15 | 2012-09-27 | Yarvis Mark D | Techniques to capture context and location information and utilize heuristics to turn location tracked over time and context information into semantic location information |
| CN105512331A (en) * | 2015-12-28 | 2016-04-20 | 海信集团有限公司 | Video recommending method and device |
| CN111767429A (en) * | 2020-06-29 | 2020-10-13 | 北京奇艺世纪科技有限公司 | Video recommendation method and device and electronic equipment |
| CN113420181A (en) * | 2021-06-23 | 2021-09-21 | 未鲲(上海)科技服务有限公司 | Video recommendation method and device, computer equipment and storage medium |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114866693A (en) * | 2022-04-15 | 2022-08-05 | 苏州清睿智能科技股份有限公司 | Information interaction method and device based on intelligent terminal |
| CN114866693B (en) * | 2022-04-15 | 2024-01-05 | 苏州清睿智能科技股份有限公司 | Information interaction method and device based on intelligent terminal |
| CN116828270A (en) * | 2023-07-05 | 2023-09-29 | 联通沃音乐文化有限公司 | Intelligent segmentation labeling method and system for video data stream |
| CN116828270B (en) * | 2023-07-05 | 2024-04-09 | 联通沃音乐文化有限公司 | Intelligent segmentation labeling method for video data stream |
| CN117495513A (en) * | 2024-01-03 | 2024-02-02 | 深圳市智百威科技发展有限公司 | Shopping recommendation method and system based on big data and artificial intelligence |
| CN117495513B (en) * | 2024-01-03 | 2024-03-29 | 深圳市智百威科技发展有限公司 | Shopping recommendation method and system based on big data and artificial intelligence |
| CN117575662A (en) * | 2024-01-17 | 2024-02-20 | 深圳市微购科技有限公司 | Commercial intelligent business decision support system and method based on video analysis |
| CN117575662B (en) * | 2024-01-17 | 2024-06-07 | 深圳市微购科技有限公司 | Commercial intelligent business decision support system and method based on video analysis |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105989004B (en) | Information delivery preprocessing method and device | |
| CN109325179B (en) | Content promotion method and device | |
| CN114219586A (en) | Shopping recommendation method, device, equipment and storage medium based on video | |
| US8423539B2 (en) | Advertising medium determination device method therefor | |
| US9002895B2 (en) | Systems and methods for providing modular configurable creative units for delivery via intext advertising | |
| US20160364762A1 (en) | Method and system for creating an audience list based on user behavior data | |
| US8626602B2 (en) | Consumer shopping and purchase support system and marketplace | |
| CN109582876B (en) | Tourist industry user portrait construction method and device and computer equipment | |
| US20110320429A1 (en) | Systems and methods for augmenting a keyword of a web page with video content | |
| US20100030648A1 (en) | Social media driven advertisement targeting | |
| CN109753601B (en) | Method and device for determining click rate of recommended information and electronic equipment | |
| US20110015996A1 (en) | Systems and Methods For Providing Keyword Related Search Results in Augmented Content for Text on a Web Page | |
| CN111325609A (en) | Commodity recommendation list determining method and device, electronic equipment and storage medium | |
| CN113744016B (en) | Object recommendation method and device, equipment and storage medium | |
| CN111061979B (en) | User tag pushing method and device, electronic equipment and medium | |
| CN116561415A (en) | Travel recommendation system and method based on big data | |
| US20150120634A1 (en) | Information processing device, information processing method, and program | |
| Piccardi et al. | On the Value of Wikipedia as a Gateway to the Web | |
| CN113032668A (en) | Product recommendation method, device and equipment based on user portrait and storage medium | |
| US20140214541A1 (en) | Method and system for user-controlled rendering of mobile advertisements | |
| US20150142782A1 (en) | Method for associating metadata with images | |
| CN112015970A (en) | Product recommendation method, related equipment and computer storage medium | |
| CN113297471B (en) | Data object tag generation method, data object searching device and electronic equipment | |
| CN115168700A (en) | Information flow recommendation method, system and medium based on pre-training algorithm | |
| CN112307298B (en) | Method and device for generating personal brand label |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |