Client mining method and system based on short video

Abstract

The invention discloses a customer mining method and system based on a short video, and relates to the technical field of customer mining, and the method comprises the following steps: step 1, short video data acquisition, step 2, user watching behavior analysis, step 3, user homepage analysis, step 4, user address analysis and step 5, platform recommendation processing. According to the method, when the short video user watches other videos related to products of clients, the attention degree of the short video user to the videos is analyzed, so that potential clients are accurately grasped, loss of the potential clients is prevented, video stream pushing is carried out, and the short video users in which regions enough pay attention to the products of the clients are judged, so that the user experience is improved. And the offline actual conditions of the regions are comprehensively analyzed, and the competitiveness of the purchasing power and the commodity discount strength of other related merchants is used as a reference for the development planning of the customer, so that the operation cost of the customer is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of customer mining, and in particular to a customer mining method and system based on short videos. Background Art

[0002] Traditional customer discovery methods generally use offline surveys, and customers often find it difficult to conduct accurate customer discovery analysis. In addition, for customers who want to increase the number of offline merchants selling goods, it is difficult to make development plans based on huge amounts of data. Short videos are loved by the public due to their simplicity, and increasing exposure through short videos has become the choice of more and more merchants. Therefore, it is necessary for customers to conduct customer discovery and development planning through short video data.

[0003] Prior art, such as the invention patent application with announcement number: CN116976958A, discloses a potential customer mining method based on short video data, which method includes: retrieving from a database the account information corresponding to each short video that has been browsed in the previous time period; obtaining the customer personal information corresponding to each account information; generating browsing record information; matching the content type whose browsing time corresponding to the customer ranks before a preset ranking, and identifying the customer as a primary potential customer; obtaining the customer's basic personal information; identifying the content type corresponding to the customer as an ultimate potential customer, and generating corresponding ultimate potential data.

[0004] Prior art, such as the invention patent application with announcement number: CN111737517A, discloses a method and system for instant recommendation based on short videos, which includes: receiving a matching judgment request from a preset client in a terminal device; calculating the matching degree between the user and the short video, and judging whether the matching degree is greater than a preset threshold; obtaining at least one recommended short video related to the content of the short video; and displaying the same playback interface of the short video by the preset client. Based on the method provided by this invention, the potential preferences of users are reflected in a timely manner to improve the user experience while alleviating the information cocoon.

[0005] It can be seen from the above scheme that the current customer mining method lacks certain attention to the analysis of short video users' preferences for customers based on their operating behaviors when watching other videos related to merchants. Customers will advertise their products in the form of short videos. If some short video users are interested in the customer's products, they will often watch other videos related to the customer's products. The more videos they watch, the more likely these short video users are to become potential customers. If these users are not accurately grasped, it is easy to cause potential customer loss, thereby reducing customer sales. At the same time, there is a lack of attention to the development planning based on short video users' preferences for products and the actual offline conditions in the region. When customers make development plans for offline merchant product placement, in addition to considering the degree of preference of users in the region for the products, they also need to consider the purchasing power of users in the region. If a large number of products are not placed in a large number of offline merchants based on actual conditions, it is easy to have a situation of oversupply, thereby increasing the customer's operating costs. Summary of the invention

[0006] The purpose of the present invention is to provide a customer mining method and system based on short videos, which solves the problems existing in the background technology.

[0007] In order to solve the above technical problems, the present invention adopts the following technical solutions: The first aspect of the present invention provides a customer mining method based on short videos, including: Step 1. Short video data acquisition: Obtain the IP address consent sharing status of each viewing user of each promotional video of the customer, so as to screen each allowed user of each promotional video of the customer, obtain the like status, collection status, number of views and viewing time of each user of each promotional video of the customer, and obtain each viewing video of each user of each promotional video of the customer within the target time period.

[0008] Step 2. User viewing behavior analysis: Analyze the video attention coefficient of each user allowed by each promotional video of the client.

[0009] Step 3. User homepage analysis: Analyze the image correlation coefficients of each promotional video of the client that allows users to watch each video within the target time period, and evaluate the short video attention coefficients of each promotional video of the client that allows users to watch related products.

[0010] Step 4. User address analysis: obtain the regions to which each user of each promotional video of the customer belongs, analyze each user in each region of each promotional video of the customer, evaluate the adaptability coefficient of each region of each promotional video of the customer, analyze each video streaming region of the customer, determine whether the customer has plans to add offline merchants, and if so, obtain the per capita GDP value of each video streaming region of the customer, the ratio of the online average price to the offline average price of each other related customer, analyze each recommended development region of the customer, and evaluate the estimated number of offline merchants to be added in each recommended development region of the customer;

[0011] Step 5. Platform recommendation processing: The short video platform sends the customer’s video streaming areas, recommended development areas and the corresponding estimated number of offline merchants to the customer’s email.

[0012] Preferably, the specific analysis method of analyzing the video attention coefficient of each user allowed for each promotional video of the client is as follows: obtaining the like attention adjustment parameter value corresponding to each like status and the collection attention adjustment parameter value corresponding to each collection status from the local database, and mapping the like attention adjustment parameter value a of each user allowed for each promotional video of the client according to the like status and collection status of each user allowed for each promotional video of the client. xn 、Collection attention parameter value b xn , where x represents the number of each promotional video, x=1,2,...,y, y is a positive integer greater than 2, and n represents the number of each allowed user, n=1,2,...,m, m is a positive integer greater than 2.

[0013] Get the playing time of each promotional video of the customer from the local database c x , total number of views x , based on the number of views allowed by each user of each promotional video of the client xn and viewing time xn , analyze the video attention coefficient of each promotional video of the client that can be used by users

[0014] Preferably, the image correlation coefficients of each promotional video of a client that allows the user to watch each video within a target time period are analyzed by a specific analysis method: based on each promotional video of a client that allows the user to watch each video within a target time period, each image of each promotional video of a client that allows the user to watch each video within the target time period is extracted.

[0015] Obtain each image feature of the customer's related products from the local database, determine whether each image of each video that the user is allowed to watch in the target time period of each promotional video of the customer appears each image feature of the related product, if a certain image of a certain video that the user is allowed to watch in the target time period of a certain promotional video of the customer appears a certain image feature of the related product, then mark the image as a target image, thereby screening each target image of each video that the user is allowed to watch in the target time period of each promotional video of the customer, and count the number of images of each video that the user is allowed to watch in the target time period of each promotional video of the customer xni and the number of target images k xni , where i represents the number of each watched video, i=1,2,...,j, and j is a positive integer greater than 2.

[0016] Analyze the image correlation coefficient of each promotional video of the client, which allows users to watch each video within the target time period

[0017] Preferably, the evaluation method for evaluating the short video attention coefficient of each promotional video of the customer that allows the user to pay attention to the relevant product is as follows: obtaining the like status and collection status of each promotional video of the customer that allows the user to watch each video within the target time period from the local database, and mapping the like attention adjustment parameter value l of each promotional video of the customer that allows the user to watch each video within the target time period according to the like attention adjustment parameter value corresponding to each like status and the collection attention adjustment parameter value corresponding to each collection status. xni 、Collection attention parameter value r xni .

[0018] Obtain from the local database the labels of the customer's promotional videos that allow the user to watch the videos within the target time period and the customer's target labels. If a label of a promotional video of the customer that allows the user to watch the videos within the target time period is consistent with a target label, mark the label as a relevant label, thereby filtering the relevant labels of the customer's promotional videos that allow the user to watch the videos within the target time period, and counting the number of labels of the customer's promotional videos that allow the user to watch the videos within the target time period. xni and the number of related labels t xni .

[0019] The image correlation coefficient δ of each promotional video of the client that allows the user to watch each video within the target time period xni , calculate the short video attention coefficient of each promotional video of the customer that allows users to pay attention to the relevant products

[0020] Preferably, the adaptability coefficients of each region of each promotional video of the customer are evaluated by a specific evaluation method: based on the coefficients ε of each promotional video of the customer that allow users to pay attention to the short videos of the relevant products xn , and based on the video attention coefficient β of each promotional video of the customer xn , calculate the personal adaptation coefficient φ of each user for each promotional video of the customer xn =ε xn +β xn .

[0021] Based on each allowed user in each region of each promotional video of the client, the personal adaptation coefficient of each allowed user in each region of each promotional video of the client is mapped, and the personal adaptation coefficients of all allowed users in each region of each promotional video of the client are counted and used as the adaptation coefficients of each region of each promotional video of the client.

[0022] Preferably, the specific analysis method for analyzing the regions where each video of the client is pushed is as follows: based on the adaptation coefficient u of each region of each promotional video of the client xp , where p represents the number of each region, p = 1, 2, ..., q, q is a positive integer greater than 2, and the average adaptation coefficient of each region of the customer is calculated Where y represents the number of promotional videos.

[0023] The adaptation coefficient average value threshold is obtained from the local database, and the adaptation coefficient average value of each region of the customer is compared with the adaptation coefficient average value threshold. If the adaptation coefficient average value of a region of the customer is greater than the adaptation coefficient average value threshold, the region is marked as a video streaming region, thereby screening the customer's video streaming regions.

[0024] Preferably, the specific analysis method for analyzing the recommended development areas of the customer is as follows: according to the average adaptation coefficient of each area of ​​the customer, extract the average adaptation coefficient T of each video streaming area of ​​the customer N , and based on this, analyze the reference coefficient D of offline merchants recommended by customers in each video streaming area N , where N represents the number of each video streaming area, N=1,2,...,M, and M is a positive integer greater than 2.

[0025] Get the average per capita GDP value A from the local database, the ratio value B of the customer's online average price to the offline average price, and the per capita GDP value P of each video streaming area of ​​the customer. N , the ratio of the online average price to the offline average price of other relevant customers Q NR , where R represents the number of other related customers, R = 1, 2, ..., S, S is a positive integer greater than 2, and the recommended development coefficient of each video streaming area of ​​the customer is calculated Where S is the number of other relevant customers.

[0026] The recommended development coefficient threshold is obtained from the local database, and the recommended development coefficient of each video streaming area of ​​the customer is compared with the recommended development coefficient threshold. If the recommended development coefficient of a video streaming area of ​​the customer is greater than the recommended development coefficient threshold, the video streaming area is marked as a recommended development area, thereby screening the recommended development areas of the customer.

[0027] Preferably, the analysis of the reference coefficients of offline merchants recommended by the customer in each video streaming area is specifically performed by obtaining the ideal number of offline merchants corresponding to each adaptation coefficient interval from the local database, and mapping the ideal number of offline merchants E in each video streaming area of ​​the customer. N .

[0028] Get the number of offline merchants F in each video streaming area from the local database N , average sales per unit time G N and average passenger flow H N , average sales volume of offline merchants in all regions N and average passenger flow K N , analyze the reference coefficients of offline merchants recommended by customers in various video streaming areas

[0029]

[0030] Preferably, the estimated increase in the number of offline merchants in each recommended development region of the client is evaluated by a specific evaluation method: based on the recommended development coefficients of each video streaming region of the client, the recommended development coefficients of each recommended development region of the client are extracted.

[0031] The estimated number of offline merchants to be added in each recommended development coefficient interval is obtained from the local database, and the estimated number of offline merchants to be added in each recommended development area of ​​the customer is mapped.

[0032] The second aspect of the present invention provides a customer mining system that executes the short video-based customer mining method, including: a short video data acquisition module, which is used to obtain the IP address consent sharing status of each viewing user of each promotional video of the customer, so as to screen each allowed user of each promotional video of the customer, obtain the like status, collection status, number of views and viewing time of each user of each promotional video of the customer, and obtain each viewing video of each user of each promotional video of the customer within a target time period.

[0033] The user viewing behavior analysis module is used to analyze the video attention coefficient of each user allowed by each promotional video of the client.

[0034] The user homepage analysis module is used to analyze the image correlation coefficients of each promotional video of the client that allows users to watch each video within a target time period, and to evaluate the short video attention coefficients of each promotional video of the client that allows users to watch related products.

[0035] The user address analysis module is used to obtain the regions to which each user of each promotional video of the customer belongs, analyze each user of each promotional video of the customer in each region, evaluate the adaptability coefficient of each promotional video of the customer in each region, analyze each video streaming region of the customer, determine whether the customer has a plan to add offline merchants, and if so, obtain the per capita GDP value of each video streaming region of the customer, the ratio of the online average price to the offline average price of each other related customer, analyze each recommended development region of the customer, and evaluate the estimated number of offline merchants to be added in each recommended development region of the customer;

[0036] The platform recommendation processing module is used for the short video platform to send the customer's video streaming areas, recommended development areas and their corresponding estimated number of offline merchants to the customer's mailbox.

[0037] The beneficial effects of the present invention are as follows: (1) Step 1 of the present invention is to obtain short video data, by obtaining various data of each promotional video of the customer, so as to facilitate subsequent analysis.

[0038] (2) Step 2. User viewing behavior analysis and Step 3. User homepage analysis of the present invention, while paying attention to the video attention coefficient of short video users for each of the client's promotional videos, also analyzes other videos watched by these short video users within the target time period, and analyzes the degree of attention these short video users pay to other videos related to the client's products, so as to accurately grasp potential customers, prevent potential customer loss, and thus increase customer sales.

[0039] (3) Step 4 of the present invention: User address analysis. By analyzing the IP address of the short video user, the region where the short video user is located is determined, thereby determining which regions of short video users are sufficiently concerned about the client's products, and comprehensively analyzing the actual offline conditions in these regions. The purchasing power and the competitiveness of the discounts of other related merchants' products are used as a reference for the client's development plan. If the client already has stores in these regions, the actual conditions of these stores will be analyzed, thereby reducing the occurrence of oversupply due to too many offline stores and reducing the client's operating costs.

[0040] (4) Step 5 of the present invention: Platform recommendation processing, by sending the analyzed data to the client, thereby facilitating the client to make development plans. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0042] Figure 1 It is a schematic diagram of the method flow of the present invention.

[0043] Figure 2 It is a schematic diagram of the system module of the present invention. DETAILED DESCRIPTION

[0044] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0045] Reference Figure 1 As shown, the first aspect of the present invention provides a customer mining method based on short videos, including: Step 1. Short video data acquisition: obtaining the IP address consent sharing status of each viewing user of each promotional video of the customer, so as to screen each allowed user of each promotional video of the customer, obtain the like status, collection status, number of views and viewing time of each user of each promotional video of the customer, and obtain each viewing video of each user of each promotional video of the customer within the target time period.

[0046] In a specific embodiment, the method for obtaining the IP address sharing status of each user who views each promotional video of the client is as follows: obtaining the IP address sharing status of each user who views each promotional video of the client from the short video platform terminal.

[0047] In a specific embodiment, the specific screening method for screening the allowed users of each promotional video of the customer is as follows: if the IP address of a viewing user of a certain promotional video of the customer agrees to share, the viewing user is marked as an allowed user, thereby screening the allowed users of each promotional video of the customer.

[0048] In a specific embodiment, the like status, collection status, number of views and viewing time of each promotional video of the customer that each user is allowed to watch, and each viewing video of each promotional video of the customer that each user is allowed to watch within a target time period are obtained by a specific acquisition method: the like status, collection status, number of views and viewing time of each promotional video of the customer that each user is allowed to watch are obtained from a local database, and each viewing video of each promotional video of the customer that each user is allowed to watch within a target time period is obtained from a short video platform terminal.

[0049] It should be noted that the local database is used to store the like status, collection status, number of views and viewing time of each promotional video of the customer, the like attention adjustment parameter value corresponding to each like status, the collection attention adjustment parameter value corresponding to each collection status, the playback time of each promotional video of the customer, the total number of views, the image features of the customer's related products, the like status and collection status of each promotional video of the customer that allows the user to watch each video within the target time period, the tags of each promotional video of the customer that allows the user to watch each video within the target time period, the target tags of the customer, the IP address range included in each region, the adaptation coefficient average value threshold, the appropriate average per capita GDP value of the customer's products, the ratio of the customer's average online price to the average offline price, the recommended development coefficient threshold, the ideal number of offline merchants corresponding to each adaptation coefficient interval, the number of offline merchants of the customer in each video streaming area, the average sales and average customer flow per unit time, the average sales and average customer flow of offline merchants of the customer in all regions, and the estimated number of offline merchants increased in each recommended development coefficient interval.

[0050] It should be noted that the IP address sharing consent status includes: agree to share, disagree to share.

[0051] It should be noted that the like status includes: liked and not liked.

[0052] It should be noted that the collection status includes: collected and not collected.

[0053] Step 1 of the present invention is to obtain short video data, by obtaining various data of each promotional video of the customer, so as to facilitate subsequent analysis.

[0054] Step 2. User viewing behavior analysis: Analyze the video attention coefficient of each user allowed by each promotional video of the client.

[0055] In a specific embodiment of the present invention, the video attention coefficient of each user allowed for each promotional video of the client is analyzed, and the specific analysis method is: obtaining the like attention adjustment parameter value corresponding to each like status and the collection attention adjustment parameter value corresponding to each collection status from the local database, and mapping the like attention adjustment parameter value a of each user allowed for each promotional video of the client according to the like status and collection status of each user allowed for each promotional video of the client. xn 、Collection attention parameter value b xn , where x represents the number of each promotional video, x=1,2,...,y, y is a positive integer greater than 2, and n represents the number of each allowed user, n=1,2,...,m, m is a positive integer greater than 2.

[0056] Get the playing time of each promotional video of the customer from the local database c x , total number of views x , based on the number of views allowed by each user of each promotional video of the client xn and viewing time xn , analyze the video attention coefficient of each promotional video of the client that can be used by users

[0057] Step 3. User homepage analysis: Analyze the image correlation coefficients of each promotional video of the client that allows users to watch each video within the target time period, and evaluate the short video attention coefficients of each promotional video of the client that allows users to watch related products.

[0058] In a specific embodiment of the present invention, the image correlation coefficients of each promotional video of a customer that allows the user to watch each video within a target time period are analyzed, and the specific analysis method is: based on each promotional video of a customer that allows the user to watch each video within the target time period, each image of each promotional video of a customer that allows the user to watch each video within the target time period is extracted.

[0059] Obtain each image feature of the customer's related products from the local database, determine whether each image of each video that the user is allowed to watch in the target time period of each promotional video of the customer appears each image feature of the related product, if a certain image of a certain video that the user is allowed to watch in the target time period of a certain promotional video of the customer appears a certain image feature of the related product, then mark the image as a target image, thereby screening each target image of each video that the user is allowed to watch in the target time period of each promotional video of the customer, and count the number of images of each video that the user is allowed to watch in the target time period of each promotional video of the customer xni and the number of target images k xni , where i represents the number of each watched video, i=1,2,...,j, and j is a positive integer greater than 2.

[0060] Analyze the image correlation coefficient of each promotional video of the client, which allows users to watch each video within the target time period

[0061] In a specific embodiment, the method for determining whether image features of related commodities appear in each image of each promotional video of the client that allows the user to watch the video within a target time period is as follows: by using an existing method for determining whether image features appear in an image, it is possible to determine whether image features of related commodities appear in each image of each promotional video of the client that allows the user to watch the video within a target time period.

[0062] It should be noted that if the customer's product is processed meat food, the image features include: image features of the customer's processed meat outer packaging, image features of other customers' processed meat outer packaging, image features of meat-related foods, and other image features.

[0063] In a specific embodiment of the present invention, the evaluation method for evaluating the short video attention coefficient of each promotional video of the customer that allows the user to pay attention to the relevant product is as follows: the like status and collection status of each promotional video of the customer that allows the user to watch each video within the target time period are obtained from the local database, and the like attention adjustment parameter value corresponding to each like status and the collection attention adjustment parameter value corresponding to each collection status are mapped to obtain the like attention adjustment parameter value l of each promotional video of the customer that allows the user to watch each video within the target time period. xni 、Collection attention parameter value r xni .

[0064] Obtain from the local database the labels of the customer's promotional videos that allow the user to watch the videos within the target time period and the customer's target labels. If a label of a promotional video of the customer that allows the user to watch the videos within the target time period is consistent with a target label, mark the label as a relevant label, thereby filtering the relevant labels of the customer's promotional videos that allow the user to watch the videos within the target time period, and counting the number of labels of the customer's promotional videos that allow the user to watch the videos within the target time period. xni and the number of related labels t xni .

[0065] The image correlation coefficient δ of each promotional video of the client that allows the user to watch each video within the target time period xni , calculate the short video attention coefficient of each promotional video of the customer that allows users to pay attention to the relevant products

[0066] Step 2. User viewing behavior analysis and step 3. User homepage analysis of the present invention, while paying attention to the video attention coefficient of short video users for each promotional video of the client, also analyzes other videos watched by these short video users within the target time period, and analyzes the degree of attention of these short video users to other videos related to the client's products, so as to accurately grasp potential customers, prevent potential customer loss, and thus increase customer sales.

[0067] Step 4. User address analysis: obtain the regions where each allowed user of each promotional video of the customer belongs, analyze the users in each region of each promotional video of the customer, evaluate the adaptability coefficient of each promotional video of the customer in each region, analyze the streaming regions of each video of the customer, determine whether the customer has plans to add offline merchants, if so, obtain the per capita GDP value of each video streaming region of the customer, the ratio of the online average price to the offline average price of other related customers, analyze the recommended development regions of the customer, and evaluate the estimated increase in the number of offline merchants in each recommended development region of the customer.

[0068] In a specific embodiment, the per capita GDP value of each video streaming area of ​​the customer and the ratio of the online average price to the offline average price of each other related customer are obtained by the following specific method: the per capita GDP value of each video streaming area of ​​the customer can be obtained through the financial management platform of each video streaming area, and the ratio of the online average price to the offline average price of each other related customer in each video streaming area of ​​the customer can be obtained through the short video platform terminal.

[0069] In a specific embodiment, the region to which each allowed user of each promotional video of the customer belongs is obtained by: obtaining the IP address of each allowed user of each promotional video of the customer from the short video platform terminal, obtaining the IP address range included in each region from the local database, and mapping to obtain the region to which each allowed user of each promotional video of the customer belongs.

[0070] In a specific embodiment, the specific analysis method for analyzing each allowed user in each region of each promotional video of the client is as follows: mapping each allowed user in each region of each promotional video of the client according to the region to which each allowed user belongs.

[0071] In a specific embodiment of the present invention, the adaptability coefficients of each region of each promotional video of the customer are evaluated, and the specific evaluation method is: based on the coefficient ε of each promotional video of the customer that allows users to pay attention to the short video of the relevant product xn , and based on the video attention coefficient β of each promotional video of the customer xn , calculate the personal adaptation coefficient φ of each user for each promotional video of the customer xn =εxn +β xn .

[0072] Based on each allowed user in each region of each promotional video of the client, the personal adaptation coefficient of each allowed user in each region of each promotional video of the client is mapped, and the personal adaptation coefficients of all allowed users in each region of each promotional video of the client are counted and used as the adaptation coefficients of each region of each promotional video of the client.

[0073] In a specific embodiment of the present invention, the specific analysis method for analyzing the regions where each video is pushed by the client is as follows: based on the adaptation coefficient u of each region of each promotional video of the client xp , where p represents the number of each region, p = 1, 2, ..., q, q is a positive integer greater than 2, and the average adaptation coefficient of each region of the customer is calculated Where y represents the number of promotional videos.

[0074] The adaptation coefficient average value threshold is obtained from the local database, and the adaptation coefficient average value of each region of the customer is compared with the adaptation coefficient average value threshold. If the adaptation coefficient average value of a region of the customer is greater than the adaptation coefficient average value threshold, the region is marked as a video streaming region, thereby screening the customer's video streaming regions.

[0075] In a specific embodiment of the present invention, the specific analysis method for analyzing the recommended development areas of the customer is as follows: based on the average adaptability coefficients of the customer's various areas, extract the average adaptability coefficients T of the customer's various video streaming areas. N , and based on this, analyze the reference coefficient D of offline merchants recommended by customers in each video streaming area N , where N represents the number of each video streaming area, N=1,2,...,M, and M is a positive integer greater than 2.

[0076] Get the average per capita GDP value A from the local database, the ratio value B of the customer's online average price to the offline average price, and the per capita GDP value P of each video streaming area of ​​the customer. N , the ratio of the online average price to the offline average price of other relevant customers Q NR , where R represents the number of other related customers, R = 1, 2, ..., S, S is a positive integer greater than 2, and the recommended development coefficient of each video streaming area of ​​the customer is calculated Where S is the number of other relevant customers.

[0077] The recommended development coefficient threshold is obtained from the local database, and the recommended development coefficient of each video streaming area of ​​the customer is compared with the recommended development coefficient threshold. If the recommended development coefficient of a video streaming area of ​​the customer is greater than the recommended development coefficient threshold, the video streaming area is marked as a recommended development area, thereby screening the recommended development areas of the customer.

[0078] In a specific embodiment of the present invention, the analysis of the reference coefficients of offline merchants recommended by the customer in each video streaming area is specifically performed by obtaining the ideal number of offline merchants corresponding to each adaptation coefficient interval from the local database, and mapping the ideal number of offline merchants E in each video streaming area of ​​the customer. N .

[0079] Get the number of offline merchants F in each video streaming area from the local database N , average sales per unit time G N and average passenger flow H N , average sales volume of offline merchants in all regions N and average passenger flow K N , analyze the reference coefficients of offline merchants recommended by customers in various video streaming areas

[0080]

[0081] In a specific embodiment of the present invention, the estimated increase in the number of offline merchants in each recommended development area of ​​the client is evaluated by a specific evaluation method: based on the recommended development coefficient of each video streaming area of ​​the client, the recommended development coefficient of each recommended development area of ​​the client is extracted.

[0082] The estimated number of offline merchants to be added in each recommended development coefficient interval is obtained from the local database, and the estimated number of offline merchants to be added in each recommended development area of ​​the customer is mapped.

[0083] Step 4 of the present invention is user address analysis. By analyzing the IP address of the short video user, the region where the short video user is located is determined, thereby determining which regions of short video users are sufficiently concerned about the client's products, and comprehensively analyzing the actual offline conditions in these regions. The purchasing power and the competitiveness of the discount strength of goods of other related merchants are used as a reference for the client's development plan. If the client itself has stores in these regions, the actual conditions of these stores will be analyzed, thereby reducing the occurrence of oversupply due to too many offline stores and reducing the client's operating costs.

[0084] Step 5. Platform recommendation processing: The short video platform sends the customer’s video streaming areas, recommended development areas and the corresponding estimated number of offline merchants to the customer’s email.

[0085] Step 5 of the present invention is platform recommendation processing, which facilitates the development planning of the customers by sending the analyzed data to the customers.

[0086] Reference Figure 2 As shown, the second aspect of the present invention provides a customer mining system for executing the short video-based customer mining method, including: a short video data acquisition module, a user viewing behavior analysis module, a user homepage analysis module, a user address analysis module, a platform recommendation processing module and a local database.

[0087] It should be noted that the short video data acquisition module is connected to the user viewing behavior analysis module, the user viewing behavior analysis module is connected to the user homepage analysis module, the user homepage analysis module is connected to the user address analysis module, the user address analysis module is connected to the platform recommendation processing module, and the local database is connected to the short video data acquisition module, the user viewing behavior analysis module, the user homepage analysis module, and the user address analysis module.

[0088] The short video data acquisition module is used to obtain the IP address sharing consent status of each viewing user of the customer's each promotional video, so as to screen each allowed user of the customer's each promotional video, obtain the like status, collection status, number of views and viewing time of each user of the customer's each promotional video, and obtain each viewing video of each user of the customer's each promotional video within a target time period.

[0089] The user viewing behavior analysis module is used to analyze the video attention coefficient of each user allowed by each promotional video of the client.

[0090] The user homepage analysis module is used to analyze the image correlation coefficients of each promotional video of the client that allows users to watch each video within a target time period, and to evaluate the short video attention coefficients of each promotional video of the client that allows users to pay attention to related products.

[0091] The user address analysis module is used to obtain the regions to which each allowed user of each promotional video of the customer belongs, analyze each user in each region of each promotional video of the customer, evaluate the adaptability coefficient of each promotional video of the customer in each region, analyze each video streaming region of the customer, determine whether the customer has plans to add offline merchants, and if so, obtain the per capita GDP value of each video streaming region of the customer, the ratio of the online average price to the offline average price of each other related customer, analyze the recommended development regions of the customer, and evaluate the estimated increase in the number of offline merchants in each recommended development region of the customer.

[0092] The platform recommendation processing module is used for the short video platform to send the client's video streaming areas, recommended development areas and their corresponding estimated number of offline merchants to the client's mailbox.

[0093] The above contents are merely examples and explanations of the concept of the present invention. Those skilled in the art may make various modifications or additions to the specific embodiments described or replace them in a similar manner. As long as they do not deviate from the concept of the invention or exceed the scope defined by the present invention, they shall all fall within the protection scope of the present invention.

Claims

1. A customer mining method based on short videos, characterized in that: include: Step 1. Acquisition of short video data: Acquire the IP address consent sharing status of each viewing user of each promotional video of the client, thereby screening each user allowed to view each promotional video of the client, acquire the like status, collection status, number of views and viewing time of each user allowed to view each promotional video of the client, and acquire each viewing video of each promotional video of the client that each user allowed to view within the target time period; Step 2. User viewing behavior analysis: Analyze the video attention coefficient of each user allowed by each promotional video of the client; Step 3. User homepage analysis: Analyze the image correlation coefficients of each promotional video of the client that allows users to watch each video within the target time period, and evaluate the short video attention coefficients of each promotional video of the client that allows users to pay attention to related products; Step 4. User address analysis: obtain the regions to which each user of each promotional video of the customer belongs, analyze each user in each region of each promotional video of the customer, evaluate the adaptability coefficient of each region of each promotional video of the customer, analyze each video streaming region of the customer, determine whether the customer has plans to add offline merchants, and if so, obtain the per capita GDP value of each video streaming region of the customer, the ratio of the online average price to the offline average price of each other related customer, analyze each recommended development region of the customer, and evaluate the estimated number of offline merchants to be added in each recommended development region of the customer; Step 5. Platform recommendation processing: The short video platform sends the customer’s video streaming areas, recommended development areas and the corresponding estimated number of offline merchants to the customer’s email.

2. The customer mining method based on short videos according to claim 1 is characterized in that: The specific analysis method of analyzing the video attention coefficient of each user of each promotional video of the client is as follows: Obtain the like and attention adjustment parameter values ​​corresponding to each like status and the collection and attention adjustment parameter values ​​corresponding to each collection status from the local database, and map the like and attention adjustment parameter values ​​a for each promotional video of the client to the like and collection status of each user allowed by the client. xn 、Collection attention parameter value b xn , where x represents the number of each promotional video, x=1,2,...,y, y is a positive integer greater than 2, n represents the number of each allowed user, n=1,2,...,m, m is a positive integer greater than 2; Get the playing time of each promotional video of the customer from the local database c x , total number of views x , based on the number of views allowed by each user of each promotional video of the client xn and viewing time xn , analyze the video attention coefficient of each promotional video of the client that can be used by users 3. The customer mining method based on short videos according to claim 1 is characterized in that: The specific analysis method of analyzing the image correlation coefficients of each promotional video of the client that allows the user to watch each video within the target time period is as follows: Extracting images of videos that are allowed to be viewed by users within a target time period from each promotional video of the client; Obtain each image feature of the customer's related products from the local database, determine whether each image of each video that the user is allowed to watch in the target time period of each promotional video of the customer appears each image feature of the related product, if a certain image of a certain video that the user is allowed to watch in the target time period of a certain promotional video of the customer appears a certain image feature of the related product, then mark the image as a target image, thereby screening each target image of each video that the user is allowed to watch in the target time period of each promotional video of the customer, and count the number of images of each video that the user is allowed to watch in the target time period of each promotional video of the customer xni and the number of target images k xni , where i represents the number of each video watched, i=1,2,...,j, j is a positive integer greater than 2; Analyze the image correlation coefficient of each promotional video of the client, which allows users to watch each video within the target time period 4. The customer mining method based on short videos according to claim 1 is characterized in that: The specific evaluation method of evaluating the short video attention coefficient of each promotional video of the client that allows users to pay attention to the relevant products is as follows: The like status and collection status of each promotional video of the client that allows the user to watch each video within the target time period are obtained from the local database, and the like and attention adjustment parameter value l of each promotional video of the client that allows the user to watch each video within the target time period is mapped according to the like and attention adjustment parameter value corresponding to each like status and the collection and attention adjustment parameter value corresponding to each collection status. xni 、Collection attention parameter value r xni ; Obtain from the local database the labels of the customer's promotional videos that allow the user to watch the videos within the target time period and the customer's target labels. If a label of a promotional video of the customer that allows the user to watch the videos within the target time period is consistent with a target label, mark the label as a relevant label, thereby filtering the relevant labels of the customer's promotional videos that allow the user to watch the videos within the target time period, and counting the number of labels of the customer's promotional videos that allow the user to watch the videos within the target time period. xni and the number of related labels t xni ; The image correlation coefficient δ of each promotional video of the client that allows the user to watch each video within the target time period xni , calculate the short video attention coefficient of each promotional video of the customer that allows users to pay attention to the relevant products 5. The customer mining method based on short videos according to claim 3 is characterized in that: The specific evaluation method of evaluating the adaptability coefficient of each region of each promotional video of the client is as follows: Based on the short video attention coefficient ε of each promotional video of the customer, which allows users to pay attention to the relevant products xn , and based on the video attention coefficient β of each promotional video of the customer xn , calculate the personal adaptation coefficient φ of each user for each promotional video of the customer xn =ε xn +β xn ; Based on each allowed user in each region of each promotional video of the client, the personal adaptation coefficient of each allowed user in each region of each promotional video of the client is mapped, and the personal adaptation coefficients of all allowed users in each region of each promotional video of the client are counted and used as the adaptation coefficients of each region of each promotional video of the client.

6. The customer mining method based on short videos according to claim 2 is characterized in that: The specific analysis method for analyzing the video streaming areas of the customers is as follows: Based on the adaptability coefficient u of each region of each customer's promotional video xp , where p represents the number of each region, p = 1, 2, ..., q, q is a positive integer greater than 2, and the average adaptation coefficient of each region of the customer is calculated Where y represents the number of promotional videos; The adaptation coefficient average value threshold is obtained from the local database, and the adaptation coefficient average value of each region of the customer is compared with the adaptation coefficient average value threshold. If the adaptation coefficient average value of a region of the customer is greater than the adaptation coefficient average value threshold, the region is marked as a video streaming region, thereby screening the customer's video streaming regions.

7. The customer mining method based on short videos according to claim 1 is characterized in that: The specific analysis method for analyzing the recommended development areas of the customers is as follows: According to the average adaptation coefficient of each region of the customer, extract the average adaptation coefficient T of each video streaming region of the customer N , and based on this, analyze the reference coefficient D of offline merchants recommended by customers in each video streaming area N , where N represents the number of each video streaming area, N = 1, 2, ..., M, and M is a positive integer greater than 2; Get the average per capita GDP value A from the local database, the ratio value B of the customer's online average price to the offline average price, and the per capita GDP value P of each video streaming area of ​​the customer. N , the ratio of the online average price to the offline average price of other relevant customers Q NR , where R represents the number of other related customers, R = 1, 2, ..., S, S is a positive integer greater than 2, and the recommended development coefficient of each video streaming area of ​​the customer is calculated Where S is the number of other relevant customers; The recommended development coefficient threshold is obtained from the local database, and the recommended development coefficient of each video streaming area of ​​the customer is compared with the recommended development coefficient threshold. If the recommended development coefficient of a video streaming area of ​​the customer is greater than the recommended development coefficient threshold, the video streaming area is marked as a recommended development area, thereby screening the recommended development areas of the customer.

8. The customer mining method based on short videos according to claim 7 is characterized in that: The specific analysis method of analyzing the reference coefficients of offline merchants recommended by customers in various video streaming areas is as follows: The ideal number of offline merchants corresponding to each adaptation coefficient interval is obtained from the local database, and the ideal number of offline merchants E in each video streaming area of ​​the client is mapped. N ; Get the number of offline merchants F in each video streaming area from the local database N , average sales per unit time G N and average passenger flow H N , average sales volume of offline merchants in all regions N and average passenger flow K N , analyze the reference coefficients of offline merchants recommended by customers in various video streaming areas 9. The customer mining method based on short videos according to claim 7 is characterized in that: The specific evaluation method for the estimated increase in the number of offline merchants in each recommended development area of ​​the evaluation client is as follows: Extract the recommended development coefficients of each recommended development region of the customer based on the recommended development coefficients of each video streaming region of the customer; The estimated number of offline merchants to be added in each recommended development coefficient interval is obtained from the local database, and the estimated number of offline merchants to be added in each recommended development area of ​​the customer is mapped.

10. A customer mining system for executing the short video-based customer mining method according to any one of claims 1 to 9, comprising: The short video data acquisition module is used to obtain the IP address consent sharing status of each viewing user of each promotional video of the customer, so as to screen each allowed user of each promotional video of the customer, obtain the like status, collection status, number of views and viewing time of each user of each promotional video of the customer, and obtain each viewing video of each promotional video of the customer that each user is allowed to watch within a target time period; A user viewing behavior analysis module is used to analyze the video attention coefficient of each user allowed by each promotional video of the client; The user homepage analysis module is used to analyze the image correlation coefficients of each promotional video of the client that allows the user to watch each video within the target time period, and evaluate the short video attention coefficient of each promotional video of the client that allows the user to pay attention to the relevant products; The user address analysis module is used to obtain the regions to which each user of each promotional video of the customer belongs, analyze each user of each promotional video of the customer in each region, evaluate the adaptability coefficient of each promotional video of the customer in each region, analyze each video streaming region of the customer, determine whether the customer has a plan to add offline merchants, and if so, obtain the per capita GDP value of each video streaming region of the customer, the ratio of the online average price to the offline average price of each other related customer, analyze each recommended development region of the customer, and evaluate the estimated number of offline merchants to be added in each recommended development region of the customer; The platform recommendation processing module is used for the short video platform to send the customer's video streaming areas, recommended development areas and their corresponding estimated number of offline merchants to the customer's mailbox.

Images