A method for accurate traffic generation based on big data analysis of short videos on psychological counseling

Abstract

The invention discloses a method for accurate drainage of psychological counseling short videos through big data analysis, and specifically relates to the technical field of big data analysis. The invention classifies psychological short videos according to audience groups, classifies existing users and collects data according to their actual conditions, extracts usage information of other users of various electronic channels through big data, analyzes user portraits, and extracts primary target users with high similarity based on user characteristics, constructs a correlation matrix of user characteristic indicators and a similarity matrix of current users, extracts advanced target users from primary target users by adopting a Pearson correlation coefficient analysis algorithm, and formulates a differentiated drainage strategy in combination with specific customer conditions, and pushes relevant psychological counseling short videos in a targeted manner according to the classifications and preferences of different target users, so that accurate drainage processing can be performed separately according to the usage conditions of different users based on big data.

Description

Technical Field

[0001] The present invention relates to the technical field of big data analysis, and more specifically, to a method for accurately conducting big data analysis on short videos of psychological counseling. Background Art

[0002] Human beings are complex organisms, and there are various reasons for mental health problems, some are genetic and some are caused later in life. Especially with the continuous increase in the double pressure of people's daily life and work, if they do not find suitable venting targets in time to relieve their troubles, they will suffer from depression, anxiety, depression and other problems. If these "troubles" are not dealt with in time, it is very likely that stubborn patients will easily suffer from mental illness and other problems.

[0003] Psychological counseling refers to the process of using psychological methods to provide psychological assistance to those who have problems with psychological adaptation and seek to solve them. Those who need to solve problems and come to seek help are called visitors or consultants, and the consulting experts who provide help are called consultants. Visitors describe, ask questions and discuss their own psychological discomfort or psychological barriers with consultants through communication media such as language and text. With the consultants' support and help, they find out the causes of psychological problems through joint discussions, analyze the crux of the problems, and then seek conditions and countermeasures to get out of the predicament and solve the problems, so as to restore psychological balance, improve the ability to adapt to the environment, and promote physical and mental health.

[0004] A psychological counselor is someone who can communicate with people, guide the secrets "hidden" in their hearts, and give effective suggestions. However, some people treat psychological counseling with a "colored plate" myopic glasses, which makes some people afraid to pay attention to their own mental health problems. Therefore, on the rapidly developing online platform, the participation of psychological counseling short videos can help many people understand their own psychological problems online, but with the rapid development of short videos, it is difficult for psychological counseling short videos to stand out quickly, so some publicity and drainage methods are needed to promote them. The existing online sales publicity and drainage methods have a unified publicity method, and a large number of advertisements with a single form and content are placed on everyone's short video platform, while psychological counseling short videos are difficult to accurately push to users in more short videos according to different users. Therefore, it is of great significance to study a big data analysis and precise drainage method for psychological counseling short videos to solve the above problems. Summary of the invention

[0005] In order to overcome the above-mentioned defects of the prior art, the present invention provides a big data analysis and precise traffic diversion method for psychological counseling short videos. The technical problem to be solved by the present invention is: the existing online sales and promotion traffic diversion methods have a unified promotional method, and a large number of advertisements with a single form and content are placed on everyone's short video platform, while it is difficult to accurately push psychological counseling short videos to users among more short videos according to different users.

[0006] To achieve the above purpose, the present invention provides the following technical solution: a method for accurate drainage of psychological counseling short video by big data analysis, comprising the following methods:

[0007] S1. Classify the psychological counseling short videos according to the video content, and divide them into categories according to the audience groups, such as the pressure of studying and working for teenagers and children, the pressure of employees, divorced families, husband-wife relationships, family relationships, social pressure and other categories. Optimize the search engines, improve the natural ranking of keywords in major search engines, obtain natural traffic, and classify them according to the actual situation of existing users, such as the pressure of studying and working for teenagers and children, the pressure of employees, divorced families, husband-wife relationships, family relationships, social pressure and other categories, and collect the usage data of existing users respectively.

[0008] S2. At the same time, through big data extraction and effective integration of user usage information of various electronic channels, according to the social information, transaction records and browsing information of different users on this platform, collect the social data, browsing data and transaction record information of different users on social media, and extract the features after sorting and integrating them. Then, based on the obtained data, segment the user group, explore in more depth the inherent characteristics of users' use of various electronic channels, personal status, shared content, and liked interactive content, and make predictions and inferences on users' consumption tendencies and browsing preferences.

[0009] S3. Through algorithm analysis based on user characteristics, we can obtain the primary target user group that can be attracted and has similar user characteristic indicators and user habits to the designated platform or website, that is, the primary customer group with attraction potential. At the same time, based on the five dimensions of user network access time, terminal type, traffic activity, traffic usage rate and traffic usage characteristics, we segment users based on big data association analysis and divide the target users into the categories of teenagers and children’s study and work pressure, employees’ pressure, divorced families, husband-wife relationships, family relationships, social pressure and others.

[0010] S4. Based on the channel drainage indicator system, a correlation matrix of user characteristic indicators and a similarity matrix of current users are constructed based on the user's age range, personality orientation, life satisfaction, current stress, alcohol and drug intake, and gender information.

[0011] S5. Use the user-based collaborative filtering-Pearson correlation coefficient analysis algorithm to obtain the target user group that is similar to the current user's characteristic indicators, that is, the target user group with drainage potential. Then, based on customer characteristics and combined with specific customer situations, formulate differentiated drainage strategies, and push relevant psychological counseling short videos according to the categories and preferences of different target users.

[0012] As a further solution of the present invention: the browsing data includes the time the user stays on different short videos and the frequency of interaction.

[0013] As a further solution of the present invention: the network access time dimension subdivides users into smartphone users and existing users who have been networked for more than one month, and the terminal type dimension subdivides users into smartphone users, computer users, non-smartphone users and children's watch users.

[0014] As a further solution of the present invention: the traffic activity dimension subdivides users into high-active users, silent and low-active users, among which silent and low-active users of smart phones are potential users; the traffic usage rate dimension subdivides users into high-usage rate users and low-usage rate users, among which high-active usage rate users are user groups with Internet habits and potential traffic needs; the traffic usage characteristics dimension subdivides users into idle users, temporary users and inhibited users.

[0015] As a further solution of the present invention: in the process of sorting and integrating, the missing values ​​and outliers in the data are first processed, and then the redundant data is removed, and the data is classified and discretized to determine the division of each interval, and the entropy method is used to determine the weight and score of each indicator.

[0016] As a further solution of the present invention: the specific implementation steps of the entropy method are as follows:

[0017] Assuming that the data has n rows of records and m variables, the data can be represented by an n*m matrix A with n rows and m columns, that is, n rows of records and m feature columns;

[0018] A=[x1,...xm]

[0019] The normalization of data is the minimum and maximum value standardization;

[0020] xij represents the i-th row and j-th column element of matrix A:

[0021] Calculate the proportion of the i-th record under the j-th indicator:

[0022] Calculate the entropy value of the jth indicator:

[0023] Calculate the coefficient of variation of the j-th indicator:

[0024] gj=1-ej

[0025] Calculate the weight of the j-th indicator.

[0026] As a further solution of the present invention: in S5, after the psychological counseling short video is pushed, 50-150 users with potential for attracting traffic are set, and the psychological counseling short videos are classified and pushed according to the user analysis results, and data testing is performed.

[0027] As a further solution of the present invention: after the data test, the collected data of the operations of the users who are not interested are analyzed according to the test results, the factors causing the users not to visit are analyzed, and the user guidance and attractiveness are improved according to the factors, and the strategy is adjusted for the users who are not interested among the target users.

[0028] The beneficial effects of the present invention are as follows: the present invention classifies psychological short videos according to audience groups, and classifies and collects data on existing users according to their actual conditions. At the same time, it extracts usage information of other users of various electronic channels through big data, analyzes user portraits, and compares user habits of other channels with existing user habits based on user characteristics, extracts primary target users with high similarity, and performs segmentation. Then, based on the channel drainage index system, it constructs a correlation matrix of user characteristic indicators and a similarity matrix of current users, and uses the Pearson correlation coefficient analysis algorithm to extract advanced target users from primary target users. In combination with specific customer conditions, it formulates differentiated drainage strategies, and pushes relevant psychological counseling short videos according to the classifications and preferences of different target users, so that accurate drainage processing can be performed separately according to the usage of different users based on big data. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 It is a schematic diagram of the process of the present invention. DETAILED DESCRIPTION

[0030] The technical solutions in the embodiments of the present invention are described clearly and completely below. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. 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.

[0031] A method for accurate drainage of psychological counseling short videos through big data analysis, including the following methods:

[0032] S1. Classify the psychological counseling short videos according to the video content, and divide them into categories according to the audience groups, such as the pressure of studying and working for teenagers and children, the pressure of employees, divorced families, husband-wife relationships, family relationships, social pressure and other categories. Optimize the search engines, improve the natural ranking of keywords in major search engines, obtain natural traffic, and classify them according to the actual situation of existing users, such as the pressure of studying and working for teenagers and children, the pressure of employees, divorced families, husband-wife relationships, family relationships, social pressure and other categories, and collect the usage data of existing users respectively.

[0033] S2. At the same time, through big data extraction and effective integration of user usage information of various electronic channels, according to the social information, transaction records and browsing information of different users on this platform, collect the social data, browsing data and transaction record information of different users on social media, and extract the features after sorting and integrating them. Then, based on the obtained data, segment the user group, explore in more depth the inherent characteristics of users' use of various electronic channels, personal status, shared content, and liked interactive content, and make predictions and inferences on users' consumption tendencies and browsing preferences.

[0034] S3. Through algorithm analysis based on user characteristics, we can obtain the primary target user group that can be attracted and has similar user characteristic indicators and user habits to the designated platform or website, that is, the primary customer group with attraction potential. At the same time, based on the five dimensions of user network access time, terminal type, traffic activity, traffic usage rate and traffic usage characteristics, we segment users based on big data association analysis and divide the target users into the categories of teenagers and children’s study and work pressure, employees’ pressure, divorced families, husband-wife relationships, family relationships, social pressure and others.

[0035] S4. Based on the channel drainage indicator system, a correlation matrix of user characteristic indicators and a similarity matrix of current users are constructed based on the user's age range, personality orientation, life satisfaction, current stress, alcohol and drug intake, and gender information.

[0036] S5. Use the user-based collaborative filtering-Pearson correlation coefficient analysis algorithm to obtain the target user group that is similar to the current user's characteristic indicators, that is, the target user group with drainage potential. Then, based on customer characteristics and combined with specific customer situations, formulate differentiated drainage strategies, and push relevant psychological counseling short videos according to the categories and preferences of different target users.

[0037] Browsing data includes the time users spend on different short videos and the frequency of their interactions.

[0038] The network access time dimension divides users into smartphone users and existing users who have been connected to the network for more than one month. The terminal type dimension divides users into smartphone users, computer users, non-smartphone users and children's watch users.

[0039] The traffic activity dimension divides users into high-active users, silent and low-active users, among which silent and low-active smartphone users are potential users. The traffic usage dimension divides users into high-usage users and low-usage users, among which high-active users are user groups with Internet habits and potential traffic demand. The traffic usage characteristics dimension divides users into off-peak users, temporary users and inhibited users.

[0040] In the process of sorting and integration, the missing values ​​and outliers in the data are first processed, and then the redundant data is removed, and the data is classified and discretized to determine the division of each interval. The entropy method is used to determine the weight and score of each indicator.

[0041] The specific implementation steps of the entropy method are as follows:

[0042] Assuming that the data has n rows of records and m variables, the data can be represented by an n*m matrix A with n rows and m columns, that is, n rows of records and m feature columns;

[0043] A=[x1,...xm]

[0044] The normalization of data is the minimum and maximum value standardization;

[0045] xij represents the i-th row and j-th column element of matrix A:

[0046] Calculate the proportion of the i-th record under the j-th indicator:

[0047] Calculate the entropy value of the jth indicator:

[0048] Calculate the coefficient of variation of the j-th indicator:

[0049] gj=1-ej

[0050] Calculate the weight of the j-th indicator.

[0051] In S5, after the psychological counseling short video is pushed, 50-150 users with potential for traffic generation are set up, and the psychological counseling short videos are classified and pushed according to the user analysis results to conduct data testing.

[0052] After the data test, the collected data of the uninterested users’ operations are analyzed according to the test results to find out the factors that lead to the users’ non-visit. Based on the factors, the user guidance and attractiveness are improved, and the strategy is adjusted for the uninterested users among the target users.

[0053] The present invention classifies psychological short videos according to audience groups, and classifies and collects data on existing users according to their actual conditions. At the same time, it extracts usage information of other users of various electronic channels through big data, analyzes user portraits, and compares user habits of other channels with existing user habits based on user characteristics, extracts primary target users with high similarity, and performs segmentation. Then, based on the channel drainage index system, it constructs a correlation matrix of user characteristic indicators and a similarity matrix of current users, and uses the Pearson correlation coefficient analysis algorithm to extract advanced target users from primary target users. In combination with specific customer conditions, it formulates differentiated drainage strategies, and pushes relevant psychological counseling short videos according to the classifications and preferences of different target users, so that accurate drainage processing can be performed separately according to the usage conditions of different users based on big data.

[0054] Finally, a few points should be explained: Although the present invention has been described in detail above with reference to the above description and specific embodiments, on the basis of the present invention, the above embodiments are only used to illustrate the technical solution of the present invention, rather than to limit it; although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the above embodiments, or replace some or all of the technical features therein with equivalents; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for accurate drainage of psychological counseling short videos through big data analysis, characterized in that: The following methods are included: S1. Classify the psychological counseling short videos according to the video content, and divide them into categories according to the audience groups: adolescents and children's study and work pressure, employees' pressure, divorced families, husband-wife relationships, family relationships, social pressure, and other categories. Optimize the search engine, improve the natural ranking of keywords in major search engines, obtain natural traffic, and classify them according to the actual situation of existing users, and divide them into categories according to the actual situation of existing users: adolescents and children's study and work pressure, employees' pressure, divorced families, husband-wife relationships, family relationships, social pressure, and other categories, and collect usage data of existing users respectively; S2. At the same time, through big data, we extract and effectively integrate the user usage information of various electronic channels. According to the social information, transaction records and browsing information of different users on this platform, we collect the social data, browsing data and transaction record information of different users on social media, and extract the features after sorting and integrating them. Then, we segment the user groups according to the obtained data, and explore in depth the inherent characteristic rules, personal status, shared content and liked interactive content of users using various electronic channels, and make predictions and inferences on users' consumption tendencies and browsing preferences. S3. Based on user characteristics, algorithm analysis is used to obtain the primary target user group that can be diverted and has similar user characteristic indicators and user habits to the designated platform or website, that is, the primary customer group with diversion potential. At the same time, based on the five dimensions of user network access time, terminal type, traffic activity, traffic usage rate and traffic usage characteristics, users are segmented based on big data association analysis, and the target users are divided into the categories of adolescents and children's study and work pressure, employees' pressure, divorced families, husband-wife relationships, family relationships, social pressure and others. S4. Based on the channel drainage index system, the correlation matrix of user characteristic indicators and the similarity matrix of current users are constructed according to the user's age range, personality orientation, life satisfaction, current stress, alcohol and drug intake, and gender information; S5. Adopt the user-based collaborative filtering-Pearson correlation coefficient analysis algorithm to obtain the target user groups with similar characteristic indicators to the current users, that is, the target user groups with drainage potential. Then, based on customer characteristics and specific customer situations, formulate differentiated drainage strategies, and push relevant psychological counseling short videos according to the categories and preferences of different target users. The network access time dimension subdivides users into smartphone users and existing users who have been networked for more than one month, and the terminal type dimension subdivides users into smartphone users, computer users, non-smartphone users, and children's watch users; The traffic activity dimension subdivides users into highly active users, silent and low-active users, among which silent and low-active users of smart phones are potential users. The traffic usage dimension subdivides users into high-usage users and low-usage users, among which high-active users are users with Internet habits and potential traffic demand. The traffic usage characteristics dimension subdivides users into idle users, temporary users and inhibited users. In the process of collating and integrating, missing values ​​and outliers in the data are first processed, and then redundant data is removed, and the data is classified and discretized to determine the division of each interval, and the entropy method is used to determine the weight and score of each indicator; The specific implementation steps of the entropy method are as follows: Assuming that the data has n rows of records and m variables, the data can be represented by an n*m matrix A with n rows and m columns, that is, n rows of records and m feature columns; A=[x1,...xm] The normalization of data is the minimum and maximum value standardization; xij represents the i-th row and j-th column element of matrix A: Calculate the proportion of the i-th record under the j-th indicator: Calculate the entropy value of the jth indicator: Calculate the coefficient of variation of the j-th indicator: gj=1-ej Calculate the weight of the j-th indicator.

2. According to claim 1, a method for accurate drainage of psychological counseling short video by big data analysis is characterized by: The browsing data includes the time users spend on different short videos and the frequency of their interactions.

3. According to claim 1, a method for accurate drainage of psychological counseling short video by big data analysis is characterized by: In S5, after the psychological counseling short video is pushed, 50-150 users with potential for attracting traffic are set, and the psychological counseling short video is classified and pushed according to the user analysis results to conduct data testing.

4. According to claim 3, a method for accurate drainage of psychological counseling short video big data analysis is characterized by: After the data test, the collected data of the operations of the uninterested users are analyzed according to the test results to find out the factors causing the users not to visit, and the user guidance and attractiveness are improved according to the factors, and the strategy is adjusted for the uninterested users among the target users.

Images