Short video platform customer obtaining method based on multi-source data fusion

Abstract

The invention relates to the technical field of short video platform data pushing, and discloses a short video platform customer obtaining method based on multi-source data fusion. The method comprises the following steps: acquiring data information of a current user on each short video platform, analyzing and calculating the data information of the current user on each short video platform, collecting merchant data, calculating the merchant data to obtain a merchant distribution list, and distributing the users according to a list in which the users are interested and the merchant distribution list. According to the method, liking, shielding and comment data of users in different types of short videos are collected from different short video platforms, semantic analysis is performed on the user comment data, different weight values are set for the user comment data, so that the diversity of the analyzed data is ensured, meanwhile, the interested degree of the users in the different types of short videos is calculated, and the user experience is improved. The degree of interest at the position can well reflect the receiving degree of the commodity by the user and can ensure the complete playing rate of video pushing, so that accurate pushing to the client is ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of short video platform data push technology, and specifically to a short video platform customer acquisition method based on multi-source data fusion. Background Art

[0002] Short videos refer to video content that is played on various new media platforms, suitable for viewing while on the move or in a short leisure state, and is pushed frequently. The content integrates themes such as skill sharing, humor, fashion trends, social hot spots, street interviews, public welfare education, advertising creativity and commercial customization. Due to the short content, it can be a separate film or a series of columns. With the popularization of mobile terminals and the acceleration of the Internet, short, flat and fast high-traffic content has gradually gained favor from various platforms.

[0003] Currently, there are many ways for short video platforms to acquire customers, but they all have various defects. These defects affect the benefits and costs of operators. Although these short video platforms have a wide range of influence, the costs during operation are indeed very high. Potential customers cannot collect their information. On the other hand, the quality of customers acquired by adopting the method of casting a wide net and laying out advertisements is relatively low. The reality is that the customer acquisition method that only targets existing customers with demand and customers who are interested in consuming leads to high marketing costs. Summary of the invention

[0004] 1. Technical issues to be resolved In view of the shortcomings of the prior art, the present invention provides a short video platform customer acquisition method based on multi-source data fusion, which has the advantages of accurately pushing customers and solves the above-mentioned technical problems.

[0005] (II) Technical solution To achieve the above object, the present invention provides the following technical solution: a short video platform customer acquisition method based on multi-source data fusion, comprising the following steps: S1. Obtain the data information of the current user on each short video platform, including the user's likes, blocks and comments on each different type of short video; S2. Analyze and calculate the data information of the current user on each short video platform, specifically including calculating the user's interest in different types of short videos by analyzing the user's likes, blocks and comments on each different type of short video. ; S3, determining a user's interested list based on the user's interest in different types of short videos, that is, arranging them from high to low, and updating the list regularly; S4. Collect merchant data and calculate the merchant data to obtain a merchant allocation list; S5. Allocate users according to the user's interest list and the merchant allocation list.

[0006] As a preferred technical solution of the present invention, the expression for obtaining the data information of the current user's likes on short videos on various short video platforms in step S1 is as follows:

[0007]

[0008] in, represents the user's likes dataset, Respectively represent the likes dataset in the first short video platform, , No. Likes dataset from short video platforms, , No. A dataset of likes from short video platforms, Respectively represent The first type of likes data in the likes dataset of short video platforms, , No. The likes dataset in the short video platform Types of likes data, , No. The likes dataset in the short video platform Type of likes data.

[0009] As a preferred technical solution of the present invention, the expression for obtaining the data information of the current user's short video blocking on each short video platform in step S1 is as follows:

[0010]

[0011] in, represents the masked dataset of users, represents the blocked dataset in the first short video platform, , No. Masked datasets in short video platforms, , No. A masked dataset from a short video platform. Respectively represent The first type of masked data in the masked dataset of short video platforms, , No. The first in the masked dataset of short video platforms Type shielding data, , No. The first in the masked dataset of short video platforms Type masking data.

[0012] As a preferred technical solution of the present invention, the steps of obtaining the data information of the current user's short video comments on various short video platforms in step S1 are as follows: S1.1. For users in Among the short video platforms Extract comments of different types of data; S1.2. For the extracted users in Among the short video platforms Semantic analysis is performed on the comments of each type of data, and they are divided into positive comments, negative comments and neutral comments; S1.3. Input the semantic analysis result into the comment calculation function to output the user’s Among the short video platforms The influence value of comments on different types of data.

[0013] As a preferred technical solution of the present invention, the expression of the comment calculation function in step S1.3 is as follows:

[0014] in, Indicates that the user Among the short video platforms The influence value of comments on the data type, Indicates that the user is Among the short video platforms The comments of this type of data are positive comments. Indicates that the user Among the short video platforms The comments of this type of data are neutral comments. Indicates that the user is Among the short video platforms The comments of this type of data are negative comments.

[0015] As a preferred technical solution of the present invention, in step S2, the expression for calculating the user's interest in different types of short videos by analyzing the user's likes, blocks and comments on each different type of short video is as follows:

[0016] in, Indicates the results of the likes data analysis. Indicates the analysis results of comment data. Indicates the result of shielding data analysis. Indicates The level of interest in each type.

[0017] As a preferred technical solution of the present invention, the likes data analysis result and comment on the data analysis results The expression is as follows:

[0018]

[0019] in, Indicates that the user is Among the short video platforms The influence value of comments on the data type, Indicates The likes dataset in the short video platform Types of likes data, Indicates the sum of data from all short video platforms. Represents a short video platform of the number of people online when the user made a like action, Represents a short video platform The sum of the number of people online on all short video platforms when the user makes a like action, Indicates the results of the likes data analysis. Indicates the results of comment data analysis.

[0020] As a preferred technical solution of the present invention, the shielding data analysis result The expression is as follows:

[0021] in, Indicates The likes dataset in the short video platform Type masking data, Indicates the results of shielded data analysis.

[0022] As a preferred technical solution of the present invention, the merchant data in step S4 includes the type data of the merchant's goods and the degree of product praise, and the specific expression is as follows:

[0023]

[0024] in, represents the positive reviews dataset, Respectively represent the type of the first merchant's product, , No. The type of merchant's products, , No. The type of merchant's products, Respectively represent The positive review data of the first product in the product type of the merchant, , No. Of the merchant's product types Good reviews of products, , No. Of the merchant's product types The positive reviews of the product.

[0025] As a preferred technical solution of the present invention, the specific steps of step S5 are as follows: S5.1. The positive reviews dataset Product types in short video platforms Types of data were compared and the positive reviews data set was excluded. Medium and short video platforms Types other than types; S5.2, sorting the data set of positive reviews after elimination, and selecting the top three types of the user's list of interest in step S3; S5.3. Recommend the top three commodities according to the interest level ratio of the top three categories in the user's interest list.

[0026] Compared with the prior art, the present invention provides a short video platform customer acquisition method based on multi-source data fusion, which has the following beneficial effects: The present invention collects the data of users' likes, blocks and comments on each different type of short videos from different short video platforms, performs semantic analysis on the user comment data, and divides them into positive comments, negative comments and neutral comments, and sets different weight values ​​for them, thereby ensuring the diversity of the analyzed data. At the same time, the degree of user interest in different types of short videos is calculated, and the degree of interest here can well reflect the user's acceptance of this type of product and can ensure the completion rate of video push, thereby ensuring accurate push to customers. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 It is a schematic diagram of the process of the present invention. DETAILED DESCRIPTION

[0028] 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.

[0029] See also Figure 1 , a short video platform customer acquisition method based on multi-source data fusion, comprising the following steps: S1. Obtain the data information of the current user on each short video platform, including the user's likes, blocks and comments on each different type of short video; The expression for obtaining the data information of the current user's likes on short videos on various short video platforms in step S1 is as follows:

[0030]

[0031] in, represents the user's likes dataset, Respectively represent the likes dataset in the first short video platform, , No. Likes dataset from short video platforms, , No. A dataset of likes from short video platforms, Respectively represent The first type of likes data in the likes dataset of short video platforms, , No. The likes dataset in the short video platform Types of likes data, , No. The likes dataset in the short video platform Types of likes data; The expression for obtaining the data information of the current user's short video blocking on various short video platforms in step S1 is as follows:

[0032]

[0033] in, represents the masked dataset of users, represents the blocked dataset in the first short video platform, , No. Masked datasets in short video platforms, , No. A masked dataset from a short video platform. Respectively represent The first type of masked data in the masked dataset of short video platforms, , No. The first in the masked dataset of short video platforms Type shielding data, , No. The first in the masked dataset of short video platforms Type shielding data; The steps for obtaining the data information of the current user's short video comments on various short video platforms in step S1 are as follows: S1.1. For users in Among the short video platforms Extract comments of different types of data; S1.2. For the extracted users in Among the short video platforms Semantic analysis is performed on the comments of each type of data, and they are divided into positive comments, negative comments and neutral comments; S1.3. Input the semantic analysis results into the comment calculation function Output the user in Among the short video platforms The influence value of comments of data of different types; Indicates that the user is Among the short video platforms The influence value of comments on the data type, Indicates that the user is Among the short video platforms The comments of this type of data are positive comments. Indicates that the user is Among the short video platforms The comments of this type of data are neutral comments. Indicates that the user is Among the short video platforms The comments of this type of data are negative comments; S2. Analyze and calculate the data information of the current user on each short video platform, specifically including calculating the user's interest in different types of short videos by analyzing the user's likes, blocks and comments on each different type of short video. ; In step S2, the expression for calculating the user's interest in different types of short videos by analyzing the user's likes, blocks and comments on each different type of short video is as follows:

[0034] in, Indicates the results of the likes data analysis. Indicates the analysis results of comment data. Indicates the result of shielding data analysis. Indicates The level of interest in each type.

[0035] Among them, the analysis results of the likes data and comment on the data analysis results The expression is as follows:

[0036]

[0037]

[0038] in, Indicates that the user is Among the short video platforms The influence value of comments on the data type, Indicates The likes dataset in the short video platform Types of likes data, Indicates the sum of data from all short video platforms. Represents a short video platform of the number of people online when the user made a like action, Represents a short video platform The sum of the number of people online on all short video platforms when the user makes a like action, Indicates the results of the likes data analysis. Indicates the analysis results of comment data. Indicates The likes dataset in the short video platform Type masking data, It indicates the result of the blocked data analysis, and analyzes the user's interest in a certain type through the likes, comments and blocks.

[0039] S3, determining a user's interested list based on the user's interest in different types of short videos, that is, arranging them from high to low, and updating the list regularly; S4: Collect merchant data and calculate the merchant data to obtain a merchant allocation list. The merchant data in step S4 includes the type data of the merchant's products and the product praise , ,in, represents the positive reviews dataset, Respectively represent the type of the first merchant's product, , No. The type of merchant's products, , No. The type of merchant's products, Respectively represent The positive review data of the first product in the product type of the merchant, , No. Of the merchant's product types Good reviews of products, , No. Of the merchant's product types The positive reviews of the product.

[0040] S5. Assign users according to their interest lists and merchant allocation lists. First, the positive review dataset is Product types in short video platforms Types of data were compared and the positive reviews data set was excluded. Medium and short video platforms For the types other than the types, the removed favorable comment data set is sorted, and the top three types in the user's list of interest in step S3 are selected, and the top three products are recommended according to the interest level ratio of the top three types in the user's list of interest. For example, the interest level of the top three products is , , , then select the top three products with the highest praise in the merchant product type data, a total of 9 products, for advertising push, and the push ratio is : : , and the push ratio between the three products of the same type is : : .

[0041] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A short video platform customer acquisition method based on multi-source data fusion, characterized by: The following steps are involved: S1. Obtain the data information of the current user on each short video platform, including the user's likes, blocks and comments on each different type of short video; S2. Analyze and calculate the data information of the current user on each short video platform, specifically including calculating the user's interest in different types of short videos by analyzing the user's likes, blocks and comments on each different type of short video. ; S3, determining a user's interested list based on the user's interest in different types of short videos, that is, arranging them from high to low, and updating the list regularly; S4. Collect merchant data, calculate the merchant data, and obtain a merchant allocation list; S5. Allocate users according to the user's interest list and the merchant allocation list.

2. The short video platform customer acquisition method based on multi-source data fusion according to claim 1, characterized in that: The expression for obtaining the data information of the current user's likes on short videos on various short video platforms in step S1 is as follows: in, represents the user's likes dataset, Respectively represent the likes dataset in the first short video platform, , No. Likes dataset from short video platforms, , No. A dataset of likes from short video platforms, Respectively represent The first type of likes data in the likes dataset of short video platforms, , No. The likes dataset in the short video platform Types of likes data, , No. The likes dataset in the short video platform Type of likes data.

3. The short video platform customer acquisition method based on multi-source data fusion according to claim 1, characterized in that: The expression for obtaining the data information of the current user's short video blocking on each short video platform in step S1 is as follows: in, represents the masked dataset of users, represents the blocked dataset in the first short video platform, , No. A masked dataset from a short video platform, , No. A masked dataset from a short video platform. Respectively represent The first type of masked data in the masked dataset of short video platforms, , No. The first in the masked dataset of short video platforms Type shielding data, , No. The first in the masked dataset of short video platforms Type masking data.

4. The short video platform customer acquisition method based on multi-source data fusion according to claim 1, characterized in that: The steps of obtaining the data information of the current user's short video comments on various short video platforms in step S1 are as follows: S1.

1. For users in Among the short video platforms Extract comments of different types of data; S1.

2. For the extracted users in Among the short video platforms Semantic analysis of comments of each type of data is performed and they are divided into positive comments, negative comments and neutral comments. S1.

3. Input the semantic analysis result into the comment calculation function to output the user’s Among the short video platforms The influence value of comments on different types of data.

5. The short video platform customer acquisition method based on multi-source data fusion according to claim 4 is characterized by: The expression of the comment calculation function in step S1.3 is as follows: in, Indicates that the user Among the short video platforms The influence value of comments on the data type, Indicates that the user Among the short video platforms The comments of this type of data are positive comments. Indicates that the user Among the short video platforms The comments of this type of data are neutral comments. Indicates that the user Among the short video platforms The comments of this type of data are negative comments.

6. The short video platform customer acquisition method based on multi-source data fusion according to claim 1, characterized in that: In step S2, the expression for calculating the user's interest in different types of short videos by analyzing the user's likes, blocks and comments on each different type of short video is as follows: in, Indicates the results of the likes data analysis. Indicates the analysis results of comment data. Indicates the result of shielding data analysis. Indicates The level of interest in each type.

7. The short video platform customer acquisition method based on multi-source data fusion according to claim 6, characterized in that: The analysis results of the likes data and comment on the data analysis results The expression is as follows: in, Indicates that the user Among the short video platforms The influence value of comments on the data type, Indicates The likes dataset in the short video platform Types of likes data, Indicates the sum of data from all short video platforms. Represents a short video platform of the number of people online when the user made a like action, Represents a short video platform The sum of the number of people online on all short video platforms when the user makes a like action, Indicates the results of the likes data analysis. Indicates the results of comment data analysis.

8. The short video platform customer acquisition method based on multi-source data fusion according to claim 6, characterized in that: The shielded data analysis results The expression is as follows: in, Indicates The likes dataset in the short video platform Type masking data, Indicates the results of shielded data analysis.

9. The short video platform customer acquisition method based on multi-source data fusion according to claim 6, characterized in that: The merchant data in step S4 includes the type data of the merchant's goods and the degree of product praise, and the specific expression is as follows: in, represents the positive reviews dataset, Respectively represent the type of the first merchant's product, , No. The type of merchant's products, , No. The type of merchant's products. Respectively represent The positive review data of the first product in the product type of the merchant, , No. Of the merchant's product types Good reviews of products, , No. Of the merchant's product types The positive reviews of the product.

10. A short video platform customer acquisition method based on multi-source data fusion according to claim 9, characterized in that: The specific steps of step S5 are as follows: S5.

1. The positive reviews dataset Product types in short video platforms Types of data were compared and the positive reviews data set was excluded. Medium and short video platforms Types other than types; S5.2, sorting the data set of positive reviews after elimination, and selecting the top three types of the user's list of interest in step S3; S5.

3. Recommend the top three commodities according to the interest level ratio of the top three categories in the user's interest list.

Images