阅读说明：本技术 一种基于互联网的流量主播推荐系统 (Flow anchor recommendation system based on Internet ) 是由 游发祥 林健辉 泮圣洁 刘梦凯 黄子祎 于 2020-05-28 设计创作，主要内容包括：本发明公开了一种基于互联网的流量主播推荐系统,包括登录模块、验证模块、数据库、监控模块、采集模块、分析处理模块、推荐排序模块、服务器和发送模块；所述登录模块用于用户登录账号信息,并自动获取登录账号信息,将其一同传输至验证模块,所述数据库内存储有记录账号数据、记录密码数据以及对应的记录账号数据所对应的用户手机号码数据,本发明通过推荐排序模块的设置,对分析模块分析处理的浏览频率均值、休闲时间、真实值、直播频率均值、直播时长均值和工作时间进行推荐排序,从而确定出主播的优选推荐位,增加数据分析的准确性,增加数据的可靠性,增加主播推荐的正确性,节省选取时间,提高工作效率。(The invention discloses an internet-based traffic anchor recommendation system, which comprises a login module, a verification module, a database, a monitoring module, an acquisition module, an analysis processing module, a recommendation sequencing module, a server and a sending module, wherein the login module is used for storing a plurality of types of traffic anchors; the system comprises a login module, a verification module, a database and a recommendation sequencing module, wherein the login module is used for logging in account information of a user, automatically acquiring the login account information and transmitting the login account information and the login account information to the verification module, and the database stores recorded account data, recorded password data and user mobile phone number data corresponding to the recorded account data.)

1. A flow anchor recommendation system based on the Internet is characterized by comprising a login module, a verification module, a database, a monitoring module, an acquisition module, an analysis processing module, a recommendation sequencing module, a server and a sending module;

the login module is used for logging in account information of a user, automatically acquiring the login account information and transmitting the login account information and the login account information to the verification module, and the database stores record account data, record password data and user mobile phone number data corresponding to the corresponding record account data;

the verification module acquires the recorded account data, the recorded password data and the user mobile phone number data from the database, performs security verification on the recorded account data, the recorded password data and the user mobile phone number data together with login account information to obtain a verification passing signal or a verification failure signal, and automatically logs in an account or forbids logging in the account;

the monitoring module is used for monitoring the state of a user account in real time, automatically acquiring account information in the login module and the verification module, generating a corresponding login signal and a corresponding non-login signal according to the login account data and the non-login account data, and sending the login signal and the non-login signal to the acquisition module;

the acquisition module receives the login signal and the non-login signal and identifies and judges the signals, and specifically comprises the following steps:

s1: when the unregistered signal is identified, judging that the user does not perform account login, and not performing browsing data acquisition on account data;

s2: when the login signal is identified, judging that the user logs in through account data, extracting the account data corresponding to the login signal, collecting browsing data of the account data, and automatically acquiring the browsing data;

the acquisition module is also used for acquiring the anchor information and transmitting the anchor information and the browsing data to the analysis processing module; the analysis processing module is used for carrying out analysis processing operation on the anchor information and the browsing data together to obtain a browsing frequency average value, leisure time, a real value, a live broadcasting frequency average value, a live broadcasting time average value and working time, and transmitting the browsing frequency average value, the leisure time, the real value, the live broadcasting frequency average value, the live broadcasting time average value and the working time to the recommendation sorting module together;

the recommendation sorting module is used for performing recommendation sorting operation on the browsing frequency average value, the leisure time, the real value, the live broadcast frequency average value, the live broadcast time average value and the working time to obtain the optimal recommendation anchor sorting and the limited times data and transmitting the optimal recommendation anchor sorting and the limited times data to the sending module through the server;

and the sending module receives the data of the optimal recommended anchor ordering and the limited times and sends the data to the user side.

2. The internet-based traffic anchor recommendation system according to claim 1, wherein the specific verification process of the security verification is as follows:

the method comprises the following steps: acquiring login account information, marking data input by a user in an account input field as account data, marking the account data as ZHi, i is 1,2,3.. No. n1, marking data input by the user in the account password input field as password data, marking the password data as MMi, i is 1,2,3.. No. n1, and sequentially marking recorded account data, recorded password data and user mobile phone number data as: JZl, JMl and YSl, where l is 1,2,3.. n2, and JZl, JMl and YSl correspond one to one;

step two: acquiring account data, matching the account data with recorded account data, judging that the account exists and generating a storage signal when the account data is matched from the recorded account data, and judging that the account does not exist and generating a non-signal when the account data cannot be matched from the recorded account data;

step three: receiving the first storage signal and the second non-signal in the second step, identifying the first storage signal and the second non-signal, automatically jumping to an account registration interface when the second non-signal is identified, and performing password verification on the first storage signal when the first storage signal is identified, specifically:

k1: extracting password data corresponding to the account data, and comparing the password data with record password data corresponding to the record account data;

k2: judging the comparison result according to the data comparison in the K1, specifically: when the matching result of the password data and the recorded password data is consistent, judging that the password of the user account is correct, and when the matching result of the password data and the recorded password data is inconsistent, judging that the password of the user account is wrong, and automatically acquiring the user mobile phone number data corresponding to the user account data;

k3: and sending verification information like the user mobile phone number data, and verifying the account again according to the verification information.

3. The internet-based traffic anchor recommendation system according to claim 1, wherein the specific operation procedure of the analysis processing operation is as follows:

h1: acquiring anchor information, marking live broadcast names of anchors in the anchor as anchor name data, marking the number of audiences displayed by the anchors in the anchor as display popularity data, marking the actual audience number of the anchors in the anchor as actual popularity data, marking the number of live broadcasts per month of the anchors in the anchor as live broadcast time length data, marking the time point of live broadcasts per time of the anchors in the anchor as live broadcast time data, marking the classification of the anchors in the anchor as anchor classification data, acquiring browsing data, marking the browsing classification data in the anchor as classification data, marking the browsing times of users in the anchor as browsing times data, and marking the browsing time point of users in the anchor as browsing time data;

h2: extracting anchor category data, anchor name data, display popularity data, actual popularity data, live broadcast frequency data, live broadcast duration data, live broadcast time data, category data, browsing frequency data and browsing time data in the H1, and sequentially marking the data as: WLv, WMv, WXv, WRv, WCv, WSv, WJv, LZc, LSc, and LJc, v being 1,2,3.. No. n3, c being 1,2,3.. No. n 4;

h3: extracting the type data browsed by the user each time, marking the times of the type data, sequencing the type data according to a descending order, calculating the difference value of the sequenced browsing times of the type data, and calculating the difference value of the browsing times of the type data with the largest numerical value, wherein the method specifically comprises the following steps:

performing difference calculation on the data of the first sorted numerical value and the data of the second sorted numerical value, performing difference calculation on the data of the second sorted numerical value and the data of the third sorted numerical value, and performing difference calculation on the data of the third sorted numerical value and the data of the fourth sorted numerical value;

firstly, selecting sort data in a first sequence, secondly, setting a preset value M, comparing the preset value M with e1, when M is less than e1, judging that the difference value between the sort data and the sort data in the first sequence is small, selecting data in a second sequence at the same time, otherwise, judging that the difference value between the sort data and the sort data in the first sequence is large, and selecting the data in the third sequence is not simultaneously selected, on the premise that M is less than e1, comparing M with e2, on the premise that M is less than e2, and on the premise that M is less than e1 and M is less than e2, selecting data in the fourth sequence when M is less than e 3;

extracting the final selection result and calibrating the final selection result into a selection category;

h4: acquiring browsing frequency data, and bringing the browsing frequency data into a calculation formula: vc_{A Liu}LSc/Q, where Vc_{A Liu}Expressed as browsing frequency, Q as days per month, and Vc_{A Liu}Carry over into the calculation:wherein PVc_{A Liu}The average value of the browsing frequency, namely the average value of the browsing frequency, acquires browsing time data, counts the occurrence frequency of each browsing time, selects the time point with the largest occurrence frequency, and marks the time point as the leisure time;

H5：acquiring and selecting the category data, matching the category data with the anchor category data, selecting one or more corresponding anchor category data, extracting anchor name data, display popularity data, actual popularity data, live broadcast frequency data, live broadcast duration data and live broadcast time data under the anchor category data, and bringing the display popularity data and the actual popularity data in the anchor name data into a calculation formula together: vv_TrueWRv/WXv, in which Vv_TrueAnd expressing as a real proportion, namely a real value, and bringing the live broadcast times data into a calculation formula: vv_{Straight bar}WCv/Q, wherein Vv_{Straight bar}Expressed as live frequency and substituted into the calculation:

4. The internet-based traffic anchor recommendation system according to claim 1, wherein the specific operation procedure of the recommendation ranking operation is as follows:

g1: acquiring leisure time and working time, matching the leisure time and the working time, selecting corresponding anchor name data, and extracting a real value, a live broadcast frequency mean value and a live broadcast time mean value corresponding to the anchor name data;

g2: sorting from large to small according to the real value, the live frequency mean value and the live time mean value, selecting the first three anchor name data in the real value, the live frequency mean value and the live time mean value sorting, marking the live frequency mean value sorting of the three as a first recommendation position, marking the live time mean value sorting as a second recommendation position, marking the real value sorting as a third recommendation position, namely, when the three are sorted identically, recommending according to the order of the live frequency mean value, the live time mean value and the real value sorting, and listing the optimal recommended anchor sorting according to the anchor name data corresponding to the recommendation order;

g3: and extracting the average value of the browsing frequency, recommending the user in a limited number of times according to the average value of the browsing frequency, and calibrating the average value as limited number of times data, namely, the number of times of recommending the anchor to the user in a period of time does not exceed a preset value.

Technical Field

The invention relates to the technical field of live broadcast recommendation, in particular to a flow anchor recommendation system based on the Internet.

Background

With the development of society, the living standard of people is gradually increased, and then the increasing pressure of the living pressure is followed, and after relaxation, people usually choose to watch some live programs to relax themselves.

The existing anchor recommendation system can frequently recommend an anchor to a user at irregular time and cannot reasonably push the anchor in reasonable time, so that the user feels dislike, and the anchor recommendation can not be carried out according to the account login condition of the user, the habit of the user and the attention category of the user.

Disclosure of Invention

The invention aims to provide an internet-based traffic anchor recommendation system, which verifies account login safety of a user through the arrangement of a login module and a verification module, generates a corresponding signal, acquires browsing data through a collection module according to the memorability identification judgment of the collection module, solves the problem that data acquisition cannot be carried out according to whether the account is logged in the prior art, increases the account safety, ensures the accuracy of data acquisition, saves the time consumed by data acquisition errors, improves the working efficiency, carries out recommendation sequencing on a browsing frequency mean value, a leisure time, a true value, a live broadcast frequency mean value, a live broadcast time mean value and working time analyzed and processed by an analysis module through the arrangement of a recommendation sequencing module, determines an optimal recommendation position of an anchor, and solves the problem that the recommendation priority cannot be rapidly sequenced in the prior art, the accuracy of data analysis is improved, the reliability of data is improved, the accuracy of anchor recommendation is improved, the selection time is saved, and the working efficiency is improved.

The purpose of the invention can be realized by the following technical scheme: a flow anchor recommendation system based on the Internet comprises a login module, a verification module, a database, a monitoring module, a collection module, an analysis processing module, a recommendation sequencing module, a server and a sending module;

the login module is used for logging in account information of a user, automatically acquiring the login account information and transmitting the login account information and the login account information to the verification module, and the database stores record account data, record password data and user mobile phone number data corresponding to the corresponding record account data;

the verification module acquires the recorded account data, the recorded password data and the user mobile phone number data from the database, performs security verification on the recorded account data, the recorded password data and the user mobile phone number data together with login account information to obtain a verification passing signal or a verification failure signal, and automatically logs in an account or forbids logging in the account;

the monitoring module is used for monitoring the state of a user account in real time, automatically acquiring account information in the login module and the verification module, generating a corresponding login signal and a corresponding non-login signal according to the login account data and the non-login account data, and sending the login signal and the non-login signal to the acquisition module;

the acquisition module receives the login signal and the non-login signal and identifies and judges the signals, and specifically comprises the following steps:

s1: when the unregistered signal is identified, judging that the user does not perform account login, and not performing browsing data acquisition on account data;

s2: when the login signal is identified, judging that the user logs in through account data, extracting the account data corresponding to the login signal, collecting browsing data of the account data, and automatically acquiring the browsing data;

the acquisition module is also used for acquiring the anchor information and transmitting the anchor information and the browsing data to the analysis processing module; the analysis processing module is used for carrying out analysis processing operation on the anchor information and the browsing data together to obtain a browsing frequency average value, leisure time, a real value, a live broadcasting frequency average value, a live broadcasting time average value and working time, and transmitting the browsing frequency average value, the leisure time, the real value, the live broadcasting frequency average value, the live broadcasting time average value and the working time to the recommendation sorting module together;

the recommendation sorting module is used for performing recommendation sorting operation on the browsing frequency average value, the leisure time, the real value, the live broadcast frequency average value, the live broadcast time average value and the working time to obtain the optimal recommendation anchor sorting and the limited times data and transmitting the optimal recommendation anchor sorting and the limited times data to the sending module through the server;

and the sending module receives the data of the optimal recommended anchor ordering and the limited times and sends the data to the user side.

As a further improvement of the invention: the specific verification process of the security verification is as follows:

the method comprises the following steps: acquiring login account information, marking data input by a user in an account input field as account data, marking the account data as ZHi, i is 1,2,3.. No. n1, marking data input by the user in the account password input field as password data, marking the password data as MMi, i is 1,2,3.. No. n1, and sequentially marking recorded account data, recorded password data and user mobile phone number data as: JZl, JMl and YSl, where l is 1,2,3.. n2, and JZl, JMl and YSl correspond one to one;

step two: acquiring account data, matching the account data with recorded account data, judging that the account exists and generating a storage signal when the account data is matched from the recorded account data, and judging that the account does not exist and generating a non-signal when the account data cannot be matched from the recorded account data;

step three: receiving the first storage signal and the second non-signal in the second step, identifying the first storage signal and the second non-signal, automatically jumping to an account registration interface when the second non-signal is identified, and performing password verification on the first storage signal when the first storage signal is identified, specifically:

k1: extracting password data corresponding to the account data, and comparing the password data with record password data corresponding to the record account data;

k2: judging the comparison result according to the data comparison in the K1, specifically: when the matching result of the password data and the recorded password data is consistent, judging that the password of the user account is correct, and when the matching result of the password data and the recorded password data is inconsistent, judging that the password of the user account is wrong, and automatically acquiring the user mobile phone number data corresponding to the user account data;

k3: and sending verification information like the user mobile phone number data, and verifying the account again according to the verification information.

As a further improvement of the invention: the specific operation process of the analysis processing operation comprises the following steps:

h1: acquiring anchor information, marking live broadcast names of anchors in the anchor as anchor name data, marking the number of audiences displayed by the anchors in the anchor as display popularity data, marking the actual audience number of the anchors in the anchor as actual popularity data, marking the number of live broadcasts per month of the anchors in the anchor as live broadcast time length data, marking the time point of live broadcasts per time of the anchors in the anchor as live broadcast time data, marking the classification of the anchors in the anchor as anchor classification data, acquiring browsing data, marking the browsing classification data in the anchor as classification data, marking the browsing times of users in the anchor as browsing times data, and marking the browsing time point of users in the anchor as browsing time data;

h2: extracting anchor category data, anchor name data, display popularity data, actual popularity data, live broadcast frequency data, live broadcast duration data, live broadcast time data, category data, browsing frequency data and browsing time data in the H1, and sequentially marking the data as: WLv, WMv, WXv, WRv, WCv, WSv, WJv, LZc, LSc, and LJc, v being 1,2,3.. No. n3, c being 1,2,3.. No. n 4;

h3: extracting the type data browsed by the user each time, marking the times of the type data, sequencing the type data according to a descending order, calculating the difference value of the sequenced browsing times of the type data, and calculating the difference value of the browsing times of the type data with the largest numerical value, wherein the method specifically comprises the following steps:

performing difference calculation on the data of the first sorted numerical value and the data of the second sorted numerical value, performing difference calculation on the data of the second sorted numerical value and the data of the third sorted numerical value, and performing difference calculation on the data of the third sorted numerical value and the data of the fourth sorted numerical value;

firstly, selecting sort data in a first sequence, secondly, setting a preset value M, comparing the preset value M with e1, when M is less than e1, judging that the difference value between the sort data and the sort data in the first sequence is small, selecting data in a second sequence at the same time, otherwise, judging that the difference value between the sort data and the sort data in the first sequence is large, and selecting the data in the third sequence is not simultaneously selected, on the premise that M is less than e1, comparing M with e2, on the premise that M is less than e2, and on the premise that M is less than e1 and M is less than e2, selecting data in the fourth sequence when M is less than e 3;

extracting the final selection result and calibrating the final selection result into a selection category;

h4: acquiring browsing frequency data, and bringing the browsing frequency data into a calculation formula: vc_{A Liu}LSc/Q, where Vc_{A Liu}Expressed as browsing frequency, Q as days per month, and Vc_{A Liu}Carry over into the calculation:wherein PVc_{A Liu}The average value of the browsing frequency, namely the average value of the browsing frequency, acquires browsing time data, counts the occurrence frequency of each browsing time, selects the time point with the largest occurrence frequency, and marks the time point as the leisure time;

h5: acquiring and selecting the category data, matching the category data with the anchor category data, selecting one or more corresponding anchor category data, extracting anchor name data, display popularity data, actual popularity data, live broadcast frequency data, live broadcast duration data and live broadcast time data under the anchor category data, and bringing the display popularity data and the actual popularity data in the anchor name data into a calculation formula together: vv_TrueWRv/WXv, in which Vv_TrueAnd expressing as a real proportion, namely a real value, and bringing the live broadcast times data into a calculation formula: vv_{Straight bar}WCv/Q, wherein Vv_{Straight bar}Expressed as live frequency and substituted into the calculation:wherein, PVv_{Straight bar}The average value of the live broadcast frequency, namely the live broadcast frequency average value, is expressed, and the live broadcast time length data is brought into a calculation formula:

wherein, P_WSvThe average value of the live broadcast time length data, namely the live broadcast time length average value, is used for acquiring the live broadcast time data, counting the occurrence times of the live broadcast time data, selecting the time point with the maximum occurrence times, and calibrating the time point as the working time.

As a further improvement of the invention: the specific operation process of the recommended sorting operation is as follows:

g1: acquiring leisure time and working time, matching the leisure time and the working time, selecting corresponding anchor name data, and extracting a real value, a live broadcast frequency mean value and a live broadcast time mean value corresponding to the anchor name data;

g2: sorting from large to small according to the real value, the live frequency mean value and the live time mean value, selecting the first three anchor name data in the real value, the live frequency mean value and the live time mean value sorting, marking the live frequency mean value sorting of the three as a first recommendation position, marking the live time mean value sorting as a second recommendation position, marking the real value sorting as a third recommendation position, namely, when the three are sorted identically, recommending according to the order of the live frequency mean value, the live time mean value and the real value sorting, and listing the optimal recommended anchor sorting according to the anchor name data corresponding to the recommendation order;

g3: and extracting the average value of the browsing frequency, recommending the user in a limited number of times according to the average value of the browsing frequency, and calibrating the average value as limited number of times data, namely, the number of times of recommending the anchor to the user in a period of time does not exceed a preset value.

The invention has the beneficial effects that:

(1) the method comprises the steps that a login module user logs in account information, automatically acquires the login account information and transmits the login account information and the login account information to a verification module, the verification module acquires recorded account data, recorded password data and user mobile phone number data from a database, the recorded account data, the recorded password data and the user mobile phone number data are subjected to security verification together with the login account information to obtain a verification passing signal or a verification failure signal, and the login account is automatically logged in or is prohibited; the monitoring module monitors the state of a user account in real time, automatically acquires account information in the login module and the verification module, wherein the account information comprises login account data, non-login account data and account data corresponding to the login account data, and generates a corresponding login signal and a non-login signal according to the login account data and the non-login account data; the acquisition module receives the login signal and the non-login signal, identifies and judges the login signal and the non-login signal, acquires browsing data according to a judgment result and automatically acquires the browsing data; the acquisition module is used for acquiring anchor information, verifying account login safety of a user through the arrangement of the login module and the verification module and generating a corresponding signal, and the acquisition module is used for identifying and judging according to the memorability of the user, so that browsing data is acquired, the account safety is improved, the accuracy of data acquisition is guaranteed, the time consumed by data acquisition errors is saved, and the working efficiency is improved;

(2) the analysis processing module is used for carrying out analysis processing operation on the anchor information and the browsing data together to obtain a browsing frequency average value, leisure time, a real value, a live broadcasting frequency average value, a live broadcasting time average value and working time; the recommendation sorting module carries out recommendation sorting operation on the browsing frequency mean value, the leisure time, the real value, the live broadcast frequency mean value, the live broadcast time mean value and the working time to obtain the optimal recommendation anchor sorting and the limited times data, and the optimal recommendation anchor sorting and the limited times data are transmitted to the sending module through the server; the sending module receives the optimal recommended anchor sorting and limited times data and sends the optimal recommended anchor sorting and limited times data to the user side; through the arrangement of the recommendation sorting module, the browsing frequency average value, the leisure time, the real value, the live broadcast frequency average value, the live broadcast time average value and the working time which are analyzed and processed by the analysis module are recommended and sorted, so that the preferred recommendation position of the anchor is determined, the accuracy of data analysis is increased, the reliability of data is increased, the accuracy of anchor recommendation is increased, the selection time is saved, and the working efficiency is improved.

Drawings

The invention will be further described with reference to the accompanying drawings.

FIG. 1 is a system block diagram of the present invention.

Detailed Description

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

Referring to fig. 1, the present invention is a traffic anchor recommendation system based on the internet, including a login module, a verification module, a database, a monitoring module, an acquisition module, an analysis processing module, a recommendation sorting module, a server and a sending module;

the login module is used for logging in account information of a user, automatically acquiring the login account information and transmitting the login account information and the login account information to the verification module, and the database stores the record account data, the record password data and the user mobile phone number data corresponding to the corresponding record account data;

the verification module acquires the recorded account data, the recorded password data and the user mobile phone number data from the database, and carries out security verification on the recorded account data, the recorded password data and the user mobile phone number data together with login account information, and the specific verification process of the security verification is as follows:

the method comprises the following steps: acquiring login account information, marking data input by a user in an account input field as account data, marking the account data as ZHi, i is 1,2,3.. No. n1, marking data input by the user in the account password input field as password data, marking the password data as MMi, i is 1,2,3.. No. n1, and sequentially marking recorded account data, recorded password data and user mobile phone number data as: JZl, JMl and YSl, where l is 1,2,3.. n2, and JZl, JMl and YSl correspond one to one;

step two: acquiring account data, matching the account data with recorded account data, judging that the account exists and generating a storage signal when the account data is matched from the recorded account data, and judging that the account does not exist and generating a non-signal when the account data cannot be matched from the recorded account data;

step three: receiving the first storage signal and the second non-signal in the second step, identifying the first storage signal and the second non-signal, automatically jumping to an account registration interface when the second non-signal is identified, and performing password verification on the first storage signal when the first storage signal is identified, specifically:

k1: extracting password data corresponding to the account data, and comparing the password data with record password data corresponding to the record account data;

k2: judging the comparison result according to the data comparison in the K1, specifically: when the matching result of the password data and the recorded password data is consistent, judging that the password of the user account is correct, and when the matching result of the password data and the recorded password data is inconsistent, judging that the password of the user account is wrong, and automatically acquiring the user mobile phone number data corresponding to the user account data;

k3: sending verification information like the user mobile phone number data, and carrying out re-verification on the account number according to the verification information to generate a verification passing signal or a verification failure signal;

the monitoring module is used for monitoring the state of a user account in real time, automatically acquiring account information in the login module and the verification module, generating a corresponding login signal and a corresponding non-login signal according to the login account data and the non-login account data, and sending the login signal and the non-login signal to the acquisition module;

the acquisition module receives the login signal and the unregistered signal, and identifies and judges the login signal and the unregistered signal, specifically:

s1: when the unregistered signal is identified, judging that the user does not perform account login, and not performing browsing data acquisition on account data;

s2: when the login signal is identified, judging that the user logs in through account data, extracting the account data corresponding to the login signal, collecting browsing data of the account data, and automatically acquiring the browsing data;

the acquisition module is also used for acquiring the anchor information and transmitting the anchor information and the browsing data to the analysis processing module;

the analysis processing module is used for carrying out analysis processing operation on the anchor information and the browsing data together, and the specific operation process of the analysis processing operation is as follows:

h1: acquiring anchor information, marking live broadcast names of anchors in the anchor as anchor name data, marking the number of audiences displayed by the anchors in the anchor as display popularity data, marking the actual audience number of the anchors in the anchor as actual popularity data, marking the number of live broadcasts per month of the anchors in the anchor as live broadcast time length data, marking the time point of live broadcasts per time of the anchors in the anchor as live broadcast time data, marking the classification of the anchors in the anchor as anchor category data, acquiring browsing data, marking the browsing category data in the anchor as category data, marking the browsing times of users in the anchor as browsing time data, and marking the browsing time point of users in the anchor as browsing time data;

h2: extracting anchor category data, anchor name data, display popularity data, actual popularity data, live broadcast frequency data, live broadcast duration data, live broadcast time data, category data, browsing frequency data and browsing time data in the H1, and sequentially marking the data as: WLv, WMv, WXv, WRv, WCv, WSv, WJv, LZc, LSc, and LJc, v being 1,2,3.. No. n3, c being 1,2,3.. No. n 4;

h3: extracting the type data browsed by the user each time, marking the times of the type data, sequencing the type data according to a descending order, calculating the difference value of the sequenced browsing times of the type data, and calculating the difference value of the browsing times of the type data with the largest numerical value, wherein the method specifically comprises the following steps:

performing difference calculation on the data of the first sorted numerical value and the data of the second sorted numerical value, performing difference calculation on the data of the second sorted numerical value and the data of the third sorted numerical value, and performing difference calculation on the data of the third sorted numerical value and the data of the fourth sorted numerical value;

firstly, selecting sort data in a first sequence, secondly, setting a preset value M, comparing the preset value M with e1, when M is less than e1, judging that the difference value between the sort data and the sort data in the first sequence is small, selecting data in a second sequence at the same time, otherwise, judging that the difference value between the sort data and the sort data in the first sequence is large, and selecting the data in the third sequence is not simultaneously selected, on the premise that M is less than e1, comparing M with e2, on the premise that M is less than e2, and on the premise that M is less than e1 and M is less than e2, selecting data in the fourth sequence when M is less than e 3;

extracting the final selection result and calibrating the final selection result into a selection category;

h4: acquiring browsing frequency data, and bringing the browsing frequency data into a calculation formula: vc_{A Liu}LSc/Q, where Vc_{A Liu}Expressed as browsing frequency, Q as days per month, and Vc_{A Liu}Carry over into the calculation:

wherein PVc_{A Liu}The average value of the browsing frequency, namely the average value of the browsing frequency, acquires browsing time data, counts the occurrence frequency of each browsing time, selects the time point with the largest occurrence frequency, and marks the time point as the leisure time;

h5: acquiring and selecting the category data, matching the category data with the anchor category data, selecting one or more corresponding anchor category data, extracting anchor name data, display popularity data, actual popularity data, live broadcast frequency data, live broadcast duration data and live broadcast time data under the anchor category data, and bringing the display popularity data and the actual popularity data in the anchor name data into a calculation formula together: vv_TrueWRv/WXv, in which Vv_TrueAnd expressing as a real proportion, namely a real value, and bringing the live broadcast times data into a calculation formula: vv_{Straight bar}WCv/Q, wherein Vv_{Straight bar}Expressed as live frequency and substituted into the calculation:wherein, PVv_{Straight bar}The average value of the live broadcast frequency, namely the live broadcast frequency average value, is expressed, and the live broadcast time length data is brought into a calculation formula:wherein, P_WSvThe method comprises the steps of obtaining live broadcast time data, counting the occurrence times of the live broadcast time data, selecting a time point with the maximum occurrence times, and calibrating the time point as working time, wherein the time point is represented as an average value of the live broadcast time data, namely a live broadcast time average value;

h6: transmitting the browsing frequency mean value, the leisure time, the real value, the live broadcast frequency mean value, the live broadcast duration mean value and the working time to a recommendation sorting module;

the recommendation sorting module is used for performing recommendation sorting operation on the browsing frequency mean value, the leisure time, the real value, the live broadcast frequency mean value, the live broadcast time mean value and the working time, and the specific operation process of the recommendation sorting operation is as follows:

g1: acquiring leisure time and working time, matching the leisure time and the working time, selecting corresponding anchor name data, and extracting a real value, a live broadcast frequency mean value and a live broadcast time mean value corresponding to the anchor name data;

g2: sorting from large to small according to the real value, the live frequency mean value and the live time mean value, selecting the first three anchor name data in the real value, the live frequency mean value and the live time mean value sorting, marking the live frequency mean value sorting of the three as a first recommendation position, marking the live time mean value sorting as a second recommendation position, marking the real value sorting as a third recommendation position, namely, when the three are sorted identically, recommending according to the order of the live frequency mean value, the live time mean value and the real value sorting, and listing the optimal recommended anchor sorting according to the anchor name data corresponding to the recommendation order;

g3: extracting a browsing frequency mean value, carrying out limited-time recommendation on the user according to the browsing frequency mean value, and calibrating the browsing frequency mean value into limited-time data, wherein the number of times of recommending the anchor to the user in a period of time does not exceed a preset value;

g4: transmitting the data of the optimal recommended anchor ordering and the limited times to a sending module through a server;

and the sending module receives the data of the optimal recommended anchor ordering and the limited times and sends the data to the user side.

When the system works, a login module user logs in account information, automatically acquires the login account information and transmits the login account information and the login account information to a verification module, the verification module acquires recorded account data, recorded password data and user mobile phone number data from a database and carries out security verification together with the login account information to obtain a verification passing signal or a verification failure signal, and the user logs in the account or forbids to log in the account automatically; the monitoring module monitors the state of a user account in real time, automatically acquires account information in the login module and the verification module, wherein the account information comprises login account data, non-login account data and account data corresponding to the login account data, and generates a corresponding login signal and a non-login signal according to the login account data and the non-login account data; the acquisition module receives the login signal and the unregistered signal, and identifies and judges the login signal and the unregistered signal, specifically: when the unregistered signal is identified, judging that the user does not perform account login, and not performing browsing data acquisition on account data; when the login signal is identified, judging that the user logs in through account data, extracting the account data corresponding to the login signal, collecting browsing data of the account data, and automatically acquiring the browsing data; the acquisition module is also used for acquiring the anchor information and transmitting the anchor information and the browsing data to the analysis processing module; the analysis processing module performs analysis processing operation on the anchor information and the browsing data together to obtain a browsing frequency average value, leisure time, a real value, a live broadcasting frequency average value, a live broadcasting time average value and working time, and transmits the browsing frequency average value, the leisure time, the real value, the live broadcasting frequency average value, the live broadcasting time average value and the working time to the recommendation sorting module together; the recommendation sorting module carries out recommendation sorting operation on the browsing frequency mean value, the leisure time, the real value, the live broadcast frequency mean value, the live broadcast time mean value and the working time to obtain the optimal recommendation anchor sorting and the limited times data, and the optimal recommendation anchor sorting and the limited times data are transmitted to the sending module through the server; and the sending module receives the data of the optimal recommended anchor ordering and the limited times and sends the data to the user side.

The foregoing is merely exemplary and illustrative of the present invention and various modifications, additions and substitutions may be made by those skilled in the art to the specific embodiments described without departing from the scope of the invention as defined in the following claims.