CN112381578A - Internet advertisement intelligent recommendation management system based on behavior characteristic recognition - Google Patents
Internet advertisement intelligent recommendation management system based on behavior characteristic recognition Download PDFInfo
- Publication number
- CN112381578A CN112381578A CN202011279208.3A CN202011279208A CN112381578A CN 112381578 A CN112381578 A CN 112381578A CN 202011279208 A CN202011279208 A CN 202011279208A CN 112381578 A CN112381578 A CN 112381578A
- Authority
- CN
- China
- Prior art keywords
- browsing
- advertisement
- interest
- advertisements
- time period
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0241—Advertisements
- G06Q30/0251—Targeted advertisements
- G06Q30/0255—Targeted advertisements based on user history
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- Finance (AREA)
- Strategic Management (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Accounting & Taxation (AREA)
- Development Economics (AREA)
- General Engineering & Computer Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Game Theory and Decision Science (AREA)
- Data Mining & Analysis (AREA)
- Economics (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses an internet advertisement intelligent recommendation management system based on behavior feature recognition, which comprises a user information collection module, a user information preprocessing module, an advertisement classification module, a user interest analysis module, a database, an analysis server and an intelligent recommendation terminal. According to the advertisement recommendation system, the interest advertisement categories of the browsing places of the users in the current browsing time period are obtained through the user information extraction module, the user information preprocessing module, the advertisement classification module and the user interest analysis module and the analysis server, and effective advertisements are screened from the primary, secondary and tertiary interest advertisements stored in the database and recommended to the users, so that the problems that in the prior art, the behavior characteristics of the users cannot be changed, the advertisement recommendation accuracy is low, and the advertisement recommendation efficiency is low are solved.
Description
Technical Field
The invention belongs to the technical field of advertisements, and relates to an internet advertisement intelligent recommendation management system based on behavior feature recognition.
Background
Advertisement, as the name implies, is an advertisement that informs the general public of the society of something. Nowadays, with the continuous development of society and economy, people's social contact has gradually developed to the internet and has occupied most of the daily life of people, thus leading to the gradual decline of old social contact ways and entertainment ways. However, in the current technology, advertisement propaganda in the internet is often mass propaganda, the propaganda contents and the propaganda modes of different users are consistent, and the behavior characteristics of the advertisement cannot be properly changed for the users, the recommendation method does not consider deep mining of the real emotional orientation and the interest degree of the users to the advertisement, the recommended advertisement is difficult to avoid harassment to the users, the accuracy of advertisement recommendation is not high, and the problem of low advertisement propaganda efficiency is caused.
Disclosure of Invention
Aiming at the problems, the invention provides an internet advertisement intelligent recommendation management system based on behavior feature recognition, which counts each advertisement under each interest advertisement category corresponding to each level of interest advertisement of a user through a user information extraction module, a user information preprocessing module, an advertisement classification module and a user interest analysis module and combines an analysis server, calculates and screens out the total number of advertisements recommended to the user every day, and solves the problems that the prior art can not change according to the behavior feature of the user, the recommendation accuracy of the advertisements is low, and the propaganda efficiency is low.
The purpose of the invention can be realized by the following technical scheme:
an internet advertisement intelligent recommendation management system based on behavior feature recognition comprises a user information collection module, a user information preprocessing module, an advertisement classification module, a user interest analysis module, a database, an analysis server and an intelligent recommendation terminal;
the user information preprocessing module is connected with the user information collecting module and the advertisement classifying module, the database is connected with the advertisement classifying module, the intelligent recommending terminal and the analyzing server, and the user interest analyzing module is connected with the advertisement classifying module and the analyzing server;
the user information collection module is used for collecting all historical browse advertisements of the user according to a login account of the user on the advertisement platform, screening the historical browse advertisements of the user according to a preset time period to obtain all the historical browse advertisements in the preset time period, and sending all the historical browse advertisements in the preset time period to the user information preprocessing module;
the user information preprocessing module receives all historical browse advertisements in a preset time period sent by the user information collecting module, divides all the received historical browse advertisements in the preset time period into various historical browse advertisements corresponding to the browse time periods according to the browse time of the various historical browse advertisements, numbers the browse time periods according to a preset sequence, sequentially marks the browse time periods as 1,2, a.
The advertisement classification module receives each historical browse advertisement in each browse time period sent by the user information preprocessing module, analyzes and processes each historical browse advertisement in each browse time period to obtain a keyword of each historical browse advertisement in each browse time period, compares the keyword of each historical browse advertisement in each browse time period with advertisement category keywords corresponding to various advertisement categories stored in the database one by one to obtain advertisement categories corresponding to each historical browse advertisement in each browse time period through screening, and forms a browse time period advertisement category set S by using the advertisement categories corresponding to each historical browse advertisement in each browse time period obtained through screeningi(si1,si2,...,sip,...,siq)，sip is expressed as the p advertisement category in the ith browsing time period, and according to various advertisement categories obtained through statistics, all historical browsing advertisements under the same advertisement category are obtained, so that all historical browsing advertisements under various advertisement categories are obtained, and all historical browsing advertisements under various advertisement categories in all browsing time periods and an advertisement category set of browsing time periods are sent to a user interest analysis module;
the user interest analysis module receives historical browse advertisements and browse time period advertisement category sets under various advertisement categories in various browse time periods sent by the advertisement classification module, analyzes the interest advertisement categories of users according to the historical browse advertisements and browse time period advertisement category sets under various advertisement categories in various browse time periods, can establish the interest sets of the users according to processed information, and is convenient for better recommending advertisements which the users like, and the specific analysis process comprises the following steps:
s1: extracting various advertisement categories of each browsing time period in the browsing time period advertisement category set, counting historical browsing advertisements under each advertisement category in each browsing time period to obtain the total number of the historical browsing advertisements under each advertisement category in each browsing time period, and marking each advertisement category in each browsing time period as an interest advertisement category;
s2: extracting browsing time lengths corresponding to the historical browsing advertisements under various interest categories in each browsing time period, accumulating the browsing time lengths corresponding to the historical browsing advertisements under the various interest advertisement categories in each browsing time period, and counting the interest advertisement categories in each browsing time period to obtain total browsing time lengths corresponding to the various interest advertisement categories in each browsing time period;
s3: extracting browsing places corresponding to historical browsing advertisements under various interest categories in each browsing time period, comparing the browsing places corresponding to the historical browsing advertisements under the various interest categories in each browsing time period, analyzing whether the browsing places are the same browsing place, if the browsing places are the same browsing place, the browsing place is the best browsing place corresponding to the interest advertisement category in the browsing time period, and if the browsing places are not the same browsing place, the browsing place with the largest occurrence number is used as the best browsing place corresponding to the advertisement category;
s4: constructing an interest advertisement category browsing parameter set R of the browsing time period by counting the total number of historical browsing advertisements corresponding to various interest advertisement categories in each browsing time period by S1, the total browsing time length corresponding to various interest advertisement categories in each browsing time period obtained by S2 and the optimal browsing location corresponding to various interest advertisement categories in each browsing time period obtained by S3ij(rij1,rij2,...,rijp,...,rijq)，rijp represents the jth browse of the pth interest advertisement category in the ith browse periodParameters j is shown as browsing parameters, j is k1, k2, k3, k1, k2 and k3 are respectively shown as the total number of historical browsing advertisements, the total browsing time length and the optimal browsing location corresponding to the same interest advertisement category;
s5: counting browsing interest values corresponding to various interest advertisement categories in each browsing time period according to the browsing parameter set of interest advertisement categories in the browsing time period constructed by S4, and sending the various interest advertisement categories and the corresponding browsing interest values in each browsing time period to an analysis server;
the database is used for storing each advertisement under the interest advertisement category corresponding to each level of interest advertisement and storing interest coefficients corresponding to the first level, second level and third level of interest advertisements;
the analysis server receives each interest advertisement category and corresponding browse interest value in each browse time period sent by the user interest analysis module, extracts browse interest threshold values corresponding to the primary, secondary and tertiary interest advertisements stored in the database, if the browse interest value corresponding to the interest advertisement category is smaller than the browse interest threshold value corresponding to the primary interest advertisement, the interest advertisement category is a primary interest advertisement, if the browsing interest value corresponding to the interest advertisement category is larger than the browsing interest threshold corresponding to the primary interest advertisement and smaller than the browsing interest threshold corresponding to the secondary interest advertisement, the interest advertisement category is a secondary interest advertisement, if the browsing interest value corresponding to the interest advertisement category is larger than the browsing interest threshold value corresponding to the secondary interest advertisement, the interest advertisement category is a three-level interest advertisement, and the counted first-level, second-level and third-level interest advertisements in each browsing time period are sent to the intelligent recommendation terminal;
the intelligent recommendation terminal receives the first-level, second-level and third-level interest advertisements sent by the analysis server in each browsing time period, extracts each advertisement under the interest advertisement category corresponding to the first-level, second-level and third-level interest advertisements and the interest coefficient corresponding to each level of interest advertisements from the database, screens effective advertisements from each advertisement under the interest advertisement category corresponding to the first-level, second-level and third-level interest advertisements, calculates the recommended advertisement number corresponding to the first-level, second-level and third-level interest advertisements from the screened effective advertisements according to the received interest coefficient corresponding to the first-level, second-level and third-level interest advertisements and the preset total recommended advertisement number per day, and further pushes the effective advertisements corresponding to the first-level, second-level and third-level interest advertisements to the user.
Further, the specific method for screening each historical browsing advertisement in the preset browsing time period in the user information acquisition module comprises the following steps:
h1: obtaining a screening cut-off time point according to a preset browsing time period and a screening starting time point;
h2: according to all the received historical browse advertisements in the preset time period, extracting browsing completion time points corresponding to all the historical browse advertisements of the user;
h3: matching the extracted browsing completion time point corresponding to each historical browsing advertisement of the user with the screening start time point and the screening stop time point, judging whether the browsing completion time point corresponding to each historical browsing advertisement of the user is in the screening start time point and the screening stop time point, if so, retaining the historical browsing advertisement of the user, and if not, removing the historical browsing advertisement of the user until obtaining each historical browsing advertisement in each browsing time period.
Furthermore, the method for the user interest analysis module to count the browsing duration according to each historical browsing advertisement specifically comprises the steps of extracting a browsing start time point and an browsing end time point from the browsing record corresponding to each marked historical browsing advertisement, and subtracting the browsing start time point from the browsing end time point to obtain the browsing duration corresponding to each historical browsing advertisement.
Further, the calculation formula of the browsing interest value corresponding to each interest advertisement category is
Is expressed as a browsing interest value, r, corresponding to the p-th interest advertisement categoryk1p is expressed as the total number of browsed pieces, r, corresponding to the p-th interest advertisement categoryk2p is expressed as a p-th interest advertisement category pairThe total time of browsing.
Furthermore, the size sequence of the interest coefficients corresponding to the primary, secondary and tertiary interest advertisements is xi 1 < xi 2 < xi 3.
Further, the calculation formula of the advertisement recommendation number corresponding to the first-level, second-level and third-level interest advertisements is lambdaε＝με*F0,λεIs expressed as the number of recommended advertisements, mu, corresponding to the first-level, second-level and third-level interest advertisement categoriesεIs expressed as interest coefficients corresponding to the first level, second level and third level interest advertisement categories, F0The advertisement recommendation is expressed as a preset total number of recommended advertisements per day, epsilon is expressed as interest advertisement grade, and epsilon is expressed as first grade, second grade and third grade respectively.
Furthermore, the method for selecting effective advertisements from the advertisements under the interest advertisement categories corresponding to the first-level, second-level and third-level interest advertisements stored in the database by the intelligent recommendation terminal is to acquire a browsing place and a browsing time period logged by a current user, match the browsing place with the best browsing place corresponding to each interest advertisement category in each browsing time period in the browsing time period interest advertisement category browsing parameter set of the browsing time period, acquire the interest advertisement category of the browsing place where the current user logs in the browsing time period, further screen out effective advertisements from the advertisements under the interest advertisement categories corresponding to the first-level, second-level and third-level interest advertisements stored in the database, and complete the total number of advertisement recommendations preset every day.
Has the advantages that:
(1) according to the advertisement recommendation method and device, the interest advertisement category of the browsing place where the user is located in the current browsing time period is obtained through the user information extraction module, the user information preprocessing module, the advertisement classification module and the user interest analysis module and the analysis server, and then effective advertisements are screened from the primary, secondary and tertiary interest advertisements stored in the database and recommended to the user, so that the problems that in the prior art, the behavior characteristics of the user cannot be changed, the advertisement recommendation accuracy is low, and the advertisement recommendation efficiency is low are solved.
(2) According to the invention, in the user information collection module, some useless information is removed by processing and integrating the information of the user, so that the calculation of subsequent information is facilitated;
(3) according to the invention, the user interest analysis module can establish the interest set of the user according to the processed information, so that the favorite advertisements of the user can be recommended better.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art that other drawings can be obtained according to the drawings without creative efforts.
FIG. 1 is a 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.
Please refer to fig. 1, which illustrates an internet advertisement intelligent recommendation management system based on behavior feature recognition, comprising a user information collection module, a user information preprocessing module, an advertisement classification module, a user interest analysis module, a database, an analysis server and an intelligent recommendation terminal;
the user information preprocessing module is connected with the user information collecting module and the advertisement classifying module, the database is connected with the advertisement classifying module, the intelligent recommending terminal and the analyzing server, and the user interest analyzing module is connected with the advertisement classifying module and the analyzing server;
the user information collection module is used for collecting all historical browse advertisements of the user according to a login account of the user on the advertisement platform, screening the historical browse advertisements of the user according to a preset time period to obtain all historical browse advertisements in the preset time period, removing some useless information by processing and integrating the information of the user to facilitate the calculation of subsequent information, and sending all the historical browse advertisements in the preset time period to the user information preprocessing module;
the user information preprocessing module receives all historical browse advertisements in a preset time period sent by the user information collecting module, divides all the received historical browse advertisements in the preset time period according to the browse time of the historical browse advertisements, divides the historical browse advertisements corresponding to the browse time periods into various historical browse advertisements corresponding to the browse time periods, numbers the browse time periods according to a preset sequence, and sequentially marks the browse time periods as 1,2, a.
H1: obtaining a screening cut-off time point according to a preset browsing time period and a screening starting time point;
h2: according to all the received historical browse advertisements in the preset time period, extracting browsing completion time points corresponding to all the historical browse advertisements of the user;
h3: matching the extracted browsing completion time point corresponding to each historical browsing advertisement of the user with the screening start time point and the screening stop time point, judging whether the browsing completion time point corresponding to each historical browsing advertisement of the user is within the screening start time point and the screening stop time point, if so, retaining the historical browsing advertisement of the user, and if not, removing the historical browsing advertisement of the user until obtaining each historical browsing advertisement in each browsing time period;
the advertisement classification module receives each historical browse advertisement in each browse time period sent by the user information preprocessing module, analyzes and processes each historical browse advertisement in each browse time period, obtains the key words of each historical browse advertisement in each browse time period, and keys of each historical browse advertisement in each browse time periodComparing the words with advertisement category keywords corresponding to various advertisement categories stored in the database one by one, screening to obtain advertisement categories corresponding to various historical browse advertisements in each browsing time period, and forming a browsing time period advertisement category set S by using the advertisement categories corresponding to various historical browse advertisements in each browsing time period obtained by screeningi(si1,si2,...,sip,...,siq)，sip is expressed as the p advertisement category in the ith browsing time period, and according to various advertisement categories obtained through statistics, all historical browsing advertisements under the same advertisement category are obtained, so that all historical browsing advertisements under various advertisement categories are obtained, and all historical browsing advertisements under various advertisement categories in all browsing time periods and an advertisement category set of browsing time periods are sent to a user interest analysis module;
the user interest analysis module receives each historical browse advertisement and browse time period advertisement category set under each advertisement category in each browse time period sent by the advertisement classification module, and analyzes the interest advertisement category of the user according to each historical browse advertisement and browse time period advertisement category set under each advertisement category in each browse time period, wherein the specific analysis process comprises the following steps:
s1: extracting various advertisement categories of each browsing time period in the browsing time period advertisement category set, counting historical browsing advertisements under each advertisement category in each browsing time period to obtain the total number of the historical browsing advertisements under each advertisement category in each browsing time period, and marking each advertisement category in each browsing time period as an interest advertisement category;
s2: extracting browsing duration corresponding to each historical browsing advertisement under each interest category in each browsing time period, accumulating the browsing duration corresponding to each historical browsing advertisement under each interest advertisement category in each browsing time period, counting each interest advertisement category in each browsing time period to obtain total browsing duration corresponding to each interest advertisement category in each browsing time period, extracting a browsing start time point and an browsing end time point from a browsing record corresponding to each marked historical browsing advertisement according to a method for counting browsing duration of each historical browsing advertisement by using a user interest analysis module, and subtracting the browsing start time point from the browsing end time point to obtain the browsing duration corresponding to each historical browsing advertisement;
s3: extracting browsing places corresponding to historical browsing advertisements under various interest categories in each browsing time period, comparing the browsing places corresponding to the historical browsing advertisements under the various interest categories in each browsing time period, analyzing whether the browsing places are the same browsing place, if the browsing places are the same browsing place, the browsing place is the best browsing place corresponding to the interest advertisement category in the browsing time period, and if the browsing places are not the same browsing place, the browsing place with the largest occurrence number is used as the best browsing place corresponding to the advertisement category;
s4: constructing an interest advertisement category browsing parameter set R of the browsing time period by counting the total number of historical browsing advertisements corresponding to various interest advertisement categories in each browsing time period by S1, the total browsing time length corresponding to various interest advertisement categories in each browsing time period obtained by S2 and the optimal browsing location corresponding to various interest advertisement categories in each browsing time period obtained by S3ij(rij1,rij2,...,rijp,...,rijq)，rijp is expressed as the jth browsing parameter of the pth interest advertisement category in the ith browsing period, j is expressed as the browsing parameter, j is k1, k2, k3, k1, k2 and k3 are respectively expressed as the total number of historical browsing advertisements, the total browsing duration and the optimal browsing location corresponding to the same interest advertisement category;
s5: according to the browsing parameter set of interest advertisement categories in the browsing time period constructed by S4, the browsing interest values corresponding to the interest advertisement categories in each browsing time period are counted, and the calculation formula of the browsing interest value corresponding to each interest advertisement category is
Is expressed as a browsing interest value, r, corresponding to the p-th interest advertisement categoryk1p is expressed as the p-th interest advertisement categoryCorresponding total number of views, rk2p is expressed as the total browsing duration corresponding to the p-th interest advertisement category, and the various interest advertisement categories and the corresponding browsing interest values in the browsing time periods are sent to the analysis server;
the database is used for storing various advertisements under interest advertisement categories corresponding to interest advertisements of all levels, and storing interest coefficients corresponding to interest advertisements of the first level, the second level and the third level, wherein the size sequence of the interest coefficients corresponding to the interest advertisements of the first level, the second level and the third level is xi 1 < xi 2 < xi 3;
the analysis server receives each interest advertisement category and corresponding browse interest value in each browse time period sent by the user interest analysis module, extracts browse interest threshold values corresponding to the primary, secondary and tertiary interest advertisements stored in the database, if the browse interest value corresponding to the interest advertisement category is smaller than the browse interest threshold value corresponding to the primary interest advertisement, the interest advertisement category is a primary interest advertisement, if the browsing interest value corresponding to the interest advertisement category is larger than the browsing interest threshold corresponding to the primary interest advertisement and smaller than the browsing interest threshold corresponding to the secondary interest advertisement, the interest advertisement category is a secondary interest advertisement, if the browsing interest value corresponding to the interest advertisement category is larger than the browsing interest threshold value corresponding to the secondary interest advertisement, the interest advertisement category is a three-level interest advertisement, and the counted first-level, second-level and third-level interest advertisements in each browsing time period are sent to the intelligent recommendation terminal;
the intelligent recommendation terminal receives the first-level, second-level and third-level interest advertisements sent by the analysis server in each browsing time period, extracts the advertisements under the interest advertisement categories corresponding to the first-level, second-level and third-level interest advertisements and the interest coefficients corresponding to the interest advertisements in each level from the database, and screens effective advertisements from the advertisements under the interest advertisement categories corresponding to the first-level, second-level and third-level interest advertisementsMatching the best browsing place corresponding to the category to obtain the interest advertisement category of the browsing place in the current user login browsing time period, further screening effective advertisements from all advertisements under the interest advertisement categories corresponding to the first-level, second-level and third-level interest advertisements stored in the database, finishing the total advertisement recommendation number preset every day, and calculating the advertisement recommendation number corresponding to the first-level, second-level and third-level interest advertisements from the screened effective advertisements according to the preset total advertisement recommendation number according to the interest coefficients corresponding to the received first-level, second-level and third-level interest advertisements and the interest coefficients corresponding to the first-level, second-level and third-level interest advertisements, wherein the calculation formula of the advertisement recommendation number corresponding to the first-level, second-level and third-level interest advertisements is lambdaε＝με*F0,λεIs expressed as the number of recommended advertisements, mu, corresponding to the first-level, second-level and third-level interest advertisement categoriesεIs expressed as interest coefficients corresponding to the first level, second level and third level interest advertisement categories, F0The total recommended number of the advertisements is represented as preset daily advertisement recommendation, epsilon represents interest advertisement level, epsilon is I, II, III, I, II and III are respectively represented as primary, secondary and tertiary, and then the intelligent recommendation terminal pushes effective advertisements corresponding to the primary, secondary and tertiary interest advertisements to the user.
According to the advertisement recommendation system, the interest advertisement categories of the browsing places of the users in the current browsing time period are obtained through the user information extraction module, the user information preprocessing module, the advertisement classification module and the user interest analysis module and the analysis server, and effective advertisements are screened from the primary, secondary and tertiary interest advertisements stored in the database and recommended to the users, so that the problems that in the prior art, the behavior characteristics of the users cannot be changed, the advertisement recommendation accuracy is low, and the advertisement recommendation efficiency is low are solved.
The foregoing is merely exemplary and illustrative of the principles of the present invention and various modifications, additions and substitutions of the specific embodiments described herein may be made by those skilled in the art without departing from the principles of the present invention or exceeding the scope of the claims set forth herein.
Claims (7)
1. The utility model provides an internet advertisement intelligence recommendation management system based on behavioral characteristic discernment which characterized in that: the system comprises a user information collection module, a user information preprocessing module, an advertisement classification module, a user interest analysis module, a database, an analysis server and an intelligent recommendation terminal;
the user information preprocessing module is connected with the user information collecting module and the advertisement classifying module, the database is connected with the advertisement classifying module, the intelligent recommending terminal and the analyzing server, and the user interest analyzing module is connected with the advertisement classifying module and the analyzing server;
the user information collection module is used for collecting all historical browse advertisements of the user according to a login account of the user on the advertisement platform, screening the historical browse advertisements of the user according to a preset time period to obtain all the historical browse advertisements in the preset time period, and sending all the historical browse advertisements in the preset time period to the user information preprocessing module;
the user information preprocessing module receives all historical browse advertisements in a preset time period sent by the user information collecting module, divides all the received historical browse advertisements in the preset time period into various historical browse advertisements corresponding to the browse time periods according to the browse time of the various historical browse advertisements, numbers the browse time periods according to a preset sequence, sequentially marks the browse time periods as 1,2, a.
The advertisement classification module receives each historical browse advertisement in each browse time period sent by the user information preprocessing module, analyzes and processes each historical browse advertisement in each browse time period to obtain the keyword of each historical browse advertisement in each browse time period, compares the keyword of each historical browse advertisement in each browse time period with the advertisement category keyword corresponding to each advertisement category stored in the database one by one, and screens the advertisement category keywords to obtain each historical browse advertisement in each browse time periodBrowsing advertisement corresponding advertisement category, and forming browsing time period advertisement category set S by using advertisement categories corresponding to each historical browsing advertisement in each browsing time period obtained by screeningi(si1,si2,...,sip,...,siq)，sip is expressed as the p advertisement category in the ith browsing time period, and according to various advertisement categories obtained through statistics, all historical browsing advertisements under the same advertisement category are obtained, so that all historical browsing advertisements under various advertisement categories are obtained, and all historical browsing advertisements under various advertisement categories in all browsing time periods and an advertisement category set of browsing time periods are sent to a user interest analysis module;
the user interest analysis module receives each historical browse advertisement and browse time period advertisement category set under each advertisement category in each browse time period sent by the advertisement classification module, and analyzes the interest advertisement category of the user according to each historical browse advertisement and browse time period advertisement category set under each advertisement category in each browse time period, and the specific analysis process comprises the following steps:
s1: extracting various advertisement categories of each browsing time period in the browsing time period advertisement category set, counting historical browsing advertisements under each advertisement category in each browsing time period to obtain the total number of the historical browsing advertisements under each advertisement category in each browsing time period, and marking each advertisement category in each browsing time period as an interest advertisement category;
s2: extracting browsing time lengths corresponding to the historical browsing advertisements under various interest categories in each browsing time period, accumulating the browsing time lengths corresponding to the historical browsing advertisements under the various interest advertisement categories in each browsing time period, and counting the interest advertisement categories in each browsing time period to obtain total browsing time lengths corresponding to the various interest advertisement categories in each browsing time period;
s3: extracting browsing places corresponding to historical browsing advertisements under various interest categories in each browsing time period, comparing the browsing places corresponding to the historical browsing advertisements under the various interest categories in each browsing time period, analyzing whether the browsing places are the same browsing place, if the browsing places are the same browsing place, the browsing place is the best browsing place corresponding to the interest advertisement category in the browsing time period, and if the browsing places are not the same browsing place, the browsing place with the largest occurrence number is used as the best browsing place corresponding to the advertisement category;
s4: constructing an interest advertisement category browsing parameter set R of the browsing time period by counting the total number of historical browsing advertisements corresponding to various interest advertisement categories in each browsing time period by S1, the total browsing time length corresponding to various interest advertisement categories in each browsing time period obtained by S2 and the optimal browsing location corresponding to various interest advertisement categories in each browsing time period obtained by S3ij(rij1,rij2,...,rijp,...,rijq)，rijp is expressed as the jth browsing parameter of the pth interest advertisement category in the ith browsing period, j is expressed as the browsing parameter, j is k1, k2, k3, k1, k2 and k3 are respectively expressed as the total number of historical browsing advertisements, the total browsing duration and the optimal browsing location corresponding to the same interest advertisement category;
s5: counting browsing interest values corresponding to various interest advertisement categories in each browsing time period according to the browsing parameter set of interest advertisement categories in the browsing time period constructed by S4, and sending the various interest advertisement categories and the corresponding browsing interest values in each browsing time period to an analysis server;
the database is used for storing each advertisement under the interest advertisement category corresponding to each level of interest advertisement and storing interest coefficients corresponding to the first level, second level and third level of interest advertisements;
the analysis server receives each interest advertisement category and corresponding browse interest value in each browse time period sent by the user interest analysis module, extracts browse interest threshold values corresponding to the primary, secondary and tertiary interest advertisements stored in the database, if the browse interest value corresponding to the interest advertisement category is smaller than the browse interest threshold value corresponding to the primary interest advertisement, the interest advertisement category is a primary interest advertisement, if the browsing interest value corresponding to the interest advertisement category is larger than the browsing interest threshold corresponding to the primary interest advertisement and smaller than the browsing interest threshold corresponding to the secondary interest advertisement, the interest advertisement category is a secondary interest advertisement, if the browsing interest value corresponding to the interest advertisement category is larger than the browsing interest threshold value corresponding to the secondary interest advertisement, the interest advertisement category is a three-level interest advertisement, and the counted first-level, second-level and third-level interest advertisements in each browsing time period are sent to the intelligent recommendation terminal;
the intelligent recommendation terminal receives the first-level, second-level and third-level interest advertisements sent by the analysis server in each browsing time period, extracts each advertisement under the interest advertisement category corresponding to the first-level, second-level and third-level interest advertisements and the interest coefficient corresponding to each level of interest advertisements from the database, screens effective advertisements from each advertisement under the interest advertisement category corresponding to the first-level, second-level and third-level interest advertisements, calculates recommended advertisement numbers corresponding to the first-level, second-level and third-level interest advertisements from the screened effective advertisements according to the received interest coefficients corresponding to the first-level, second-level and third-level interest advertisements and the preset total recommended advertisement number per day, and then pushes the effective advertisements corresponding to the first-level, second-level and third-level interest advertisements to the user.
2. The internet advertisement intelligent recommendation management system based on behavior feature recognition as claimed in claim 1, wherein: the specific method for screening each historical browsing advertisement in the preset browsing time period in the user information acquisition module comprises the following steps:
h1: obtaining a screening cut-off time point according to a preset browsing time period and a screening starting time point;
h2: according to all the received historical browse advertisements in the preset time period, extracting browsing completion time points corresponding to all the historical browse advertisements of the user;
h3: matching the extracted browsing completion time point corresponding to each historical browsing advertisement of the user with the screening start time point and the screening stop time point, judging whether the browsing completion time point corresponding to each historical browsing advertisement of the user is in the screening start time point and the screening stop time point, if so, retaining the historical browsing advertisement of the user, and if not, removing the historical browsing advertisement of the user until obtaining each historical browsing advertisement in each browsing time period.
3. The internet advertisement intelligent recommendation management system based on behavior feature recognition as claimed in claim 1, wherein: the method for counting the browsing time length according to each historical browsing advertisement by the user interest analysis module specifically comprises the steps of extracting a browsing start time point and an browsing end time point from a browsing record corresponding to each marked historical browsing advertisement, and subtracting the browsing start time point from the browsing end time point to obtain the browsing time length corresponding to each historical browsing advertisement.
4. The internet advertisement intelligent recommendation management system based on behavior feature recognition as claimed in claim 1, wherein: the calculation formula of the browsing interest value corresponding to each interest advertisement category is
Is expressed as a browsing interest value, r, corresponding to the e-th interest advertisement categoryk1e is expressed as the total number of browsed items, r, corresponding to the e-th interest advertisement categoryk2And e represents the total browsing time length corresponding to the e-th interest advertisement category.
5. The internet advertisement intelligent recommendation management system based on behavior feature recognition as claimed in claim 1, wherein: the size sequence of the interest coefficients corresponding to the first-level, second-level and third-level interest advertisements is xi 1 < xi 2 < xi 3.
6. The internet advertisement intelligent recommendation management system based on behavior feature recognition as claimed in claim 1, wherein: the calculation formula of the recommended number of the advertisements corresponding to the first-level, second-level and third-level interest advertisements is lambdaε＝με*F0,λεExpressed as first, second and third gradeNumber of recommended advertisements, mu, corresponding to interesting advertisement categoriesεIs expressed as interest coefficients corresponding to the first level, second level and third level interest advertisement categories, F0The advertisement recommendation is expressed as a preset total number of recommended advertisements per day, epsilon is expressed as interest advertisement grade, and epsilon is expressed as first grade, second grade and third grade respectively.
7. The internet advertisement intelligent recommendation management system based on behavior feature recognition as claimed in claim 1, wherein: the method for selecting the effective advertisements from the advertisements under the interest advertisement categories corresponding to the first-level, second-level and third-level interest advertisements stored in the database by the intelligent recommendation terminal is to acquire the browsing place and the browsing time period of the current user login, match the optimal browsing place corresponding to each interest advertisement category in each browsing time period in the interest advertisement category browsing parameter set of the browsing time period, acquire the interest advertisement category of the browsing place of the current user login in the browsing time period, further screen out the effective advertisements from the advertisements under the interest advertisement categories corresponding to the first-level, second-level and third-level interest advertisements stored in the database, and complete the total advertisement recommendation number preset every day.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011279208.3A CN112381578B (en) | 2020-11-16 | 2020-11-16 | Internet advertisement intelligent recommendation management system based on behavior characteristic recognition |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011279208.3A CN112381578B (en) | 2020-11-16 | 2020-11-16 | Internet advertisement intelligent recommendation management system based on behavior characteristic recognition |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112381578A true CN112381578A (en) | 2021-02-19 |
| CN112381578B CN112381578B (en) | 2021-05-14 |
Family
ID=74585289
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011279208.3A Active CN112381578B (en) | 2020-11-16 | 2020-11-16 | Internet advertisement intelligent recommendation management system based on behavior characteristic recognition |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112381578B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113158048A (en) * | 2021-04-22 | 2021-07-23 | 武汉申子仟电子商务有限公司 | Mobile internet browsing content intelligent recommendation method, system, equipment and storage medium based on feature recognition |
| CN113327140A (en) * | 2021-08-02 | 2021-08-31 | 深圳小蝉文化传媒股份有限公司 | Video advertisement putting effect intelligent analysis management system based on big data analysis |
| CN113781113A (en) * | 2021-09-09 | 2021-12-10 | 杭州爆米花鹰眼科技有限责任公司 | Chained information pushing system and method |
| CN114187036A (en) * | 2021-11-30 | 2022-03-15 | 深圳市喂车科技有限公司 | Internet advertisement intelligent recommendation management system based on behavior characteristic recognition |
| CN116109353A (en) * | 2022-12-12 | 2023-05-12 | 广州圆原元网络科技有限公司 | Mobile phone application store advertisement delivery management platform |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105528711A (en) * | 2014-09-29 | 2016-04-27 | 银橙(上海)信息技术有限公司 | Method and system for intelligent bidding for Internet advertising |
| CN110135897A (en) * | 2019-04-24 | 2019-08-16 | 上海淇毓信息科技有限公司 | A kind of building of Internet advertising, loading method and background management system |
| CN111008869A (en) * | 2019-12-05 | 2020-04-14 | 秒针信息技术有限公司 | Advertisement recommendation method and device, electronic equipment and storage medium |
| CN111582951A (en) * | 2020-05-28 | 2020-08-25 | 广州蓝深科技有限公司 | Advertisement putting system and method for cloud electronic commerce |
| US10803486B1 (en) * | 2014-04-24 | 2020-10-13 | Amazon Technologies, Inc. | Item recommendations based on media content |
-
2020
- 2020-11-16 CN CN202011279208.3A patent/CN112381578B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10803486B1 (en) * | 2014-04-24 | 2020-10-13 | Amazon Technologies, Inc. | Item recommendations based on media content |
| CN105528711A (en) * | 2014-09-29 | 2016-04-27 | 银橙(上海)信息技术有限公司 | Method and system for intelligent bidding for Internet advertising |
| CN110135897A (en) * | 2019-04-24 | 2019-08-16 | 上海淇毓信息科技有限公司 | A kind of building of Internet advertising, loading method and background management system |
| CN111008869A (en) * | 2019-12-05 | 2020-04-14 | 秒针信息技术有限公司 | Advertisement recommendation method and device, electronic equipment and storage medium |
| CN111582951A (en) * | 2020-05-28 | 2020-08-25 | 广州蓝深科技有限公司 | Advertisement putting system and method for cloud electronic commerce |
Non-Patent Citations (1)
| Title |
|---|
| 付小飞: "《基于用户画像的移动广告推荐技术的研究与应用》", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113158048A (en) * | 2021-04-22 | 2021-07-23 | 武汉申子仟电子商务有限公司 | Mobile internet browsing content intelligent recommendation method, system, equipment and storage medium based on feature recognition |
| CN113327140A (en) * | 2021-08-02 | 2021-08-31 | 深圳小蝉文化传媒股份有限公司 | Video advertisement putting effect intelligent analysis management system based on big data analysis |
| CN113781113A (en) * | 2021-09-09 | 2021-12-10 | 杭州爆米花鹰眼科技有限责任公司 | Chained information pushing system and method |
| CN113781113B (en) * | 2021-09-09 | 2022-06-21 | 杭州爆米花鹰眼科技有限责任公司 | Chained information pushing system and method |
| CN114187036A (en) * | 2021-11-30 | 2022-03-15 | 深圳市喂车科技有限公司 | Internet advertisement intelligent recommendation management system based on behavior characteristic recognition |
| CN114187036B (en) * | 2021-11-30 | 2022-10-11 | 深圳市喂车科技有限公司 | Internet advertisement intelligent recommendation management system based on behavior characteristic recognition |
| CN116109353A (en) * | 2022-12-12 | 2023-05-12 | 广州圆原元网络科技有限公司 | Mobile phone application store advertisement delivery management platform |
| CN116109353B (en) * | 2022-12-12 | 2023-08-08 | 广州圆原元网络科技有限公司 | Mobile phone application store advertisement delivery management platform |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112381578B (en) | 2021-05-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112381578B (en) | Internet advertisement intelligent recommendation management system based on behavior characteristic recognition | |
| CN112417302B (en) | Big data-based information content intelligent analysis recommendation processing system | |
| CN106156127B (en) | Method and device for selecting data content to push to terminal | |
| CN103744928B (en) | A kind of network video classification method based on history access record | |
| CN101334773B (en) | Method for filtrating search engine searching result | |
| CN107220295A (en) | A kind of people's contradiction reconciles case retrieval and mediation strategy recommends method | |
| CN103425650B (en) | Recommend searching method and system | |
| CN108615177B (en) | Electronic terminal personalized recommendation method based on weighting extraction interestingness | |
| CN113204709A (en) | Short video search matching recommendation method and system based on multidimensional data depth comparison analysis and computer storage medium | |
| CN107451148A (en) | Video classification method and device and electronic equipment | |
| CN110287292B (en) | Judgment criminal measuring deviation degree prediction method and device | |
| CN101944111B (en) | Method and device for searching news video | |
| CN114915468B (en) | Intelligent analysis and detection method for network crime based on knowledge graph | |
| CN113327140A (en) | Video advertisement putting effect intelligent analysis management system based on big data analysis | |
| CN112149422B (en) | Dynamic enterprise news monitoring method based on natural language | |
| CN103714120B (en) | A kind of system that user interest topic is extracted in the access record from user url | |
| CN108737524B (en) | Enterprise information pushing method | |
| CN104182539A (en) | Abnormal information batch processing method and system | |
| CN113158048A (en) | Mobile internet browsing content intelligent recommendation method, system, equipment and storage medium based on feature recognition | |
| CN113282641A (en) | Webpage search data information intelligent classification management method and system based on user behavior deep analysis and computer storage medium | |
| CN116861063B (en) | Method for exploring commercial value degree of social media hot search | |
| CN110413770A (en) | Group's message is referred to the method and device of group topic | |
| CN113177164B (en) | Multi-platform collaborative new media content monitoring and management system based on big data | |
| CN113781113B (en) | Chained information pushing system and method | |
| CN113159363B (en) | Event trend prediction method based on historical news reports |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| TA01 | Transfer of patent application right | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20210426 Address after: 510308 Room 203, 996 Xingang East Road, Haizhu District, Guangzhou City, Guangdong Province Applicant after: Guangdong Guangzhou Automobile Digital Marketing Co.,Ltd. Address before: 210043 197 Heyue Road, Baguazhou street, Qixia District, Nanjing City, Jiangsu Province Applicant before: Nanjing Bangfeng Intelligent Technology Co.,Ltd. |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |