CN112990986A - Advertisement service information publishing system based on block chain - Google Patents

Advertisement service information publishing system based on block chain Download PDF

Info

Publication number
CN112990986A
CN112990986A CN202110471029.8A CN202110471029A CN112990986A CN 112990986 A CN112990986 A CN 112990986A CN 202110471029 A CN202110471029 A CN 202110471029A CN 112990986 A CN112990986 A CN 112990986A
Authority
CN
China
Prior art keywords
keywords
advertisement
module
click rate
keyword
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
Application number
CN202110471029.8A
Other languages
Chinese (zh)
Other versions
CN112990986B (en
Inventor
资芳
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Guangzhou Yitui Network Technology Co ltd
Original Assignee
Guangzhou Yitui Network Technology Co ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Guangzhou Yitui Network Technology Co ltd filed Critical Guangzhou Yitui Network Technology Co ltd
Priority to CN202110471029.8A priority Critical patent/CN112990986B/en
Publication of CN112990986A publication Critical patent/CN112990986A/en
Application granted granted Critical
Publication of CN112990986B publication Critical patent/CN112990986B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • 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
    • G06Q30/0242—Determining effectiveness of advertisements
    • G06Q30/0245—Surveys
    • 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
    • G06Q30/0256—User search

Abstract

The invention discloses an advertisement service information publishing system based on a block chain, which comprises a data monitoring module, a static presentation management module, a dynamic presentation management module and a platform processing module, the data monitoring module is used for monitoring the effect presented by the advertisement in real time and monitoring the click quantity of the user in real time, thereby adding or deleting the keywords in time and achieving the aim of improving the click rate of the user, the static presentation management module is used for screening the advertisement keywords, thereby improving the exposure rate of the advertisement and the click rate of the user, the dynamic presentation management module is used for carrying out online popularization on the product when the click rate of the user exceeds the preset number of people, therefore, the popularity of the product can be expanded, and the platform processing module is used for analyzing the click rate of the advertisement through each block link point, so that coordination and control can be performed among the modules.

Description

Advertisement service information publishing system based on block chain
Technical Field
The invention relates to the technical field of advertisements, in particular to an advertisement service information publishing system based on a block chain.
Background
When a new product is put into an existing user, the characteristics of the product need to be recommended to a large number of users through advertisements, but when the new product is popularized, the new product needs to be positioned and popularized, corresponding keywords are set according to the adapted product, and data identification is carried out on click rates of different keywords, so that the advertisements can be exposed according to the different keywords, when the click rate of the corresponding keyword is detected to be lower than the preset times, the keyword needs to be adjusted, but when the keyword is adjusted, the keyword with the same meaning as the keyword needing to be deleted still appears, or after part of the keywords are set, the keywords need to be recombined and separated, so that the exposure rate of the advertisements is improved, the first step of product popularization is achieved, meanwhile, the advertisements also have fatigue period, when the product is known by the user, the keywords of the advertisements need to be adjusted in time, the method and the device have the advantages that the user cannot dislike the popularization of the product, so that the keywords of the advertisement are very important to the product when the product is popularized.
Disclosure of Invention
The present invention is directed to an advertisement service information distribution system based on a block chain, so as to solve the problems set forth in the background art.
In order to solve the technical problems, the invention provides the following technical scheme: an advertisement service information publishing system based on a block chain comprises a data monitoring module, a static presentation management module, a dynamic presentation management module and a platform processing module, wherein the data monitoring module is used for monitoring the effect presented by an advertisement in real time and monitoring the click rate of a user in real time so as to add or delete key words in time and achieve the aim of improving the click rate of the user, the static presentation management module is used for screening advertisement key words so as to improve the exposure rate of the advertisement and the click rate of the user, the dynamic presentation management module is used for promoting a product on line when the click rate of the user exceeds a preset number of people so as to expand the popularity of the product, the platform processing module is used for analyzing the click rate of the advertisement through each block chain link so as to coordinate and control the modules, the data monitoring module comprises a user search record storage unit, an advertisement click rate data monitoring unit and a feedback unit, wherein the user search record storage unit is used for recording and storing keywords of the same product when a user searches the same product so as to adjust the keywords of the product per se, the advertisement click rate data monitoring unit is used for monitoring the number of times of browsing the advertisement by clicking currently, so that the playing effect of the advertisement of the current product can be analyzed, when the click rate of the current advertisement is detected to exceed the preset number of times, the current advertisement is sent to the dynamic presentation management module through the platform processing module to be popularized on line, when the number of times of browsing the advertisement is detected to be lower than the preset number of times, the current advertisement is sent to the static presentation management module through the platform processing module to adjust the keywords, and the feedback unit is used for sending the data of the advertisement to the platform, so as to receive a next step instruction, the static presentation management module comprises a keyword aggregation module, a semantic keyword screening module, a keyword dimension reduction screening module and a fatigue period replacement module, wherein the keyword aggregation module is used for monitoring the browsing amount of each keyword of the current product and adjusting the keyword, the semantic keyword screening module is used for extracting the keyword with the same semantic meaning from the set when the long-term click rate of the keyword stored in the keyword aggregation module is detected to be lower, the keyword dimension reduction screening module is used for performing dimension reduction processing on the keyword with the lower click rate in the period, storing the keyword with the lower click rate in the set, recombining, splitting or moving the keyword in the set to present the sequence of the keyword, and the fatigue period replacement module is used for when the click rate of the keyword after adjustment is detected to be lower than a preset value, the system comprises a platform processing module, a data monitoring module, a static presentation management module and a dynamic presentation management module, wherein the platform processing module is connected with the data monitoring module, the static presentation management module and the dynamic presentation management module, the dynamic presentation management module comprises a customer group analysis module and a private customized advertisement analysis module, the customer group analysis module is used for analyzing according to the customer group clicking the advertisement at present and judging that the advertisement is a young man, a young woman or an old person group, the private customized advertisement analysis module is used for setting negative keywords aiming at the search habits of different users so as to promote the click rate of the advertisement, and the platform processing module is connected with the data monitoring module, the static presentation management module and the dynamic presentation management.
Further, set the current set of keywords as Q = { Q = }1,q2,q3...qnD = { D } for the current set of click rates for different keywords1,d2,d3...dnH, when the click rate d of the keyword is detectedkWhen the semantic content is smaller than the preset value, screening the keywords with the same semantics as the keywords, and determining the keywords with the same semantics as the keywords by the following steps:
z01, segmenting the current keywords, deleting the low-frequency keywords after segmentation, and setting different vectors according to different keywords;
z02, taking the current other key word vectors as the center, and judging the cosine similarity until traversing the key words with similar semantemes;
z03, hiding the same semantic keywords in the set;
wherein: the cosine similarity is expressed as
Figure 143672DEST_PATH_IMAGE001
When is coming into contact with
Figure 529522DEST_PATH_IMAGE002
The closer the value of (1) is, the higher the similarity value of the keywords in the representative set with other keywords is; when in use
Figure 667243DEST_PATH_IMAGE002
Is much larger than or much smaller than 1, the lower the similarity between the keywords in the representative set and other keywords, x refers to the keyword vector, y refers to other keyword vectors,
Figure 129448DEST_PATH_IMAGE002
refers to the angle between the keyword vectors,
Figure 336307DEST_PATH_IMAGE003
is referred to as the module of the keyword vector,
Figure 542161DEST_PATH_IMAGE004
refers to the modulus of other keyword vectors.
Furthermore, the dimensionality reduction processing is carried out on the keywords with lower click rate, and the dimensionality reduction processing steps are as follows:
z001, setting the dimensionality of the current keyword as a vector matrix Z with n rows and m columns;
z002, calculating the mean value of each line of the current vector matrix and obtaining a new vector matrix Z';
z003, calculating an eigenvalue and an eigenvector value according to a covariance matrix formula;
z004: and arranging a vector matrix Z ' according to the characteristic vector values from large to small, splitting, combining and adjusting the sequence of the keywords to be subjected to dimension reduction, wherein the dimension of the vector matrix is n ' rows and m ' columns.
Further, in the step Z004, the original non-recombined keyword set is set to M = { M1, M2, M3.. mn }, and the split and combined keyword set is M '= { ma + mb, ma1+ mb1.. man + mbn }, when it is detected that the user searches for the same product, the keyword set H = { c1+ d1, c2+ d2, c3+ d3... cn + dn } of the product searched by the user and the matching degree value sim of the recombined keyword set M' are analyzed, when the matching degree value sim is greater than a preset threshold, the platform processing module recommends an advertisement with a higher matching degree to the user, sorts the advertisement according to the matching degree, and when the matching value sim is less than the preset threshold, the advertisement content is replaced.
The conditions required to be met by the negative keywords are set as follows:
a1, when the matching degree value of the keyword set H and M' of the product searched by the user is larger than a preset threshold value, and the click rate of the user is larger than the click rate of the keyword set before the keywords are not split and combined and is smaller than a second preset threshold value;
a2, adding negative different keywords aiming at different crowds when the click rate of the split and combined keywords is less than half of the click rate of the applicable crowds;
the second preset click rate refers to a sum of click rates before and after splitting of the keyword.
When the matched value sim after the recombination is detected to be smaller than the preset threshold value, the first-time advertisement promotion time before the recombination of the keywords is T1According to the method, the interest degree value generated by detecting the click rate of the user on the advertisement by the platform processing module is W1, and the interest degree value generated by detecting the click rate of the user on the advertisement within the preset time is W2When the keyword after the recombination is detected, the interest degree generated by the click rate of the user to the advertisement is W3Wherein W is3>W1>W2When SW is detected3-SW2<SW2-SW1And then, when the fatigue degree of the user to the advertisement is larger than a preset threshold value, the interest degree of the user is reduced after the keywords are recombined, so that the content of the advertisement needs to be replaced, and the time is divided into three segments, wherein SW1、SW2、SW3The interest degrees of the user for the advertisement in the former period, the interest degrees of the user for the advertisement in the middle period and the interest degrees of the user for the advertisement in the later period are respectively.
When detecting that the recombined matching value sim is larger than a preset threshold value, the platform processing module automatically matches suffix words suitable for different users behind the keywords according to the same type of product search records specified by the users, wherein the suffix words are not equal to negative keywords;
after adding, verifying the advertisement click rate brought by the added suffix words in real time, judging whether the added suffix words are matched with the object, investigating the click rate brought by the suffix words of other products of the same type, wherein the suffix words are based on the characteristics contained in the product history watched by the user or based on the user, if a skirt is recommended, adding proper larger or smaller suffix words to the suffix words according to the number of codes of the skirt purchased by the user in the history record, and improving the click rate of the user.
The system comprises the following steps:
d1, using a data monitoring module to record the keywords of the historical data when the user searches for the same product, simultaneously monitoring the click rate caused by adding or deleting different keywords, and monitoring the advertisement effect presented by the advertisement in real time, thereby achieving the purpose of improving the click rate of the user;
d2: using a static presentation management module, retrieving keywords with similar meanings, deleting or adding the retrieved similar keywords, reducing the dimensions of different keywords, re-splitting the keywords after the dimensions are reduced, combining and adjusting the sequence, judging the exposure of the current advertisement according to the result, and judging whether the click rate of a user is improved or not;
d3, using dynamic process module, when the click rate of the advertisement user exceeds the preset click rate, carrying out the directional analysis according to different groups, and setting different negative keywords according to different users, thereby further promoting the click rate of the advertisement.
Compared with the prior art, the invention has the following beneficial effects:
1. by arranging the data monitoring module, keywords in historical data when a user searches for the same product can be recorded, click rate caused by adding or deleting different keywords can be monitored, and advertising effect presented by advertisements can be monitored in real time, so that the aim of improving the click rate of the user is fulfilled.
2. Using a static presentation management module, retrieving keywords with similar meanings, deleting or adding the retrieved similar keywords, reducing the dimensions of different keywords, re-splitting the keywords after the dimensions are reduced, combining and adjusting the sequence, judging the exposure of the current advertisement according to the exposure, and judging whether the click rate of a user is improved or not, so that the user can know the content in the product and stabilize the basic amount of clients;
3. by using the dynamic processing module, when the click rate of the issued advertisement user exceeds the preset click rate, the directional analysis is carried out according to different groups, and different negative keywords are set according to different users, so that the click of the advertisement can be further promoted.
Drawings
The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain the principles of the invention and not to limit the invention. In the drawings:
FIG. 1 is a block chain-based advertisement service information distribution system module composition diagram of the present invention;
fig. 2 is a schematic diagram illustrating steps of an advertisement service information distribution system based on a block chain according to 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. 1-2, the present invention provides a technical solution: an advertisement service information publishing system based on a block chain comprises a data monitoring module, a static presentation management module, a dynamic presentation management module and a platform processing module, wherein the data monitoring module is used for monitoring the effect presented by an advertisement in real time and monitoring the click rate of a user in real time so as to add or delete key words in time and achieve the aim of improving the click rate of the user, the static presentation management module is used for screening advertisement key words so as to improve the exposure rate of the advertisement and the click rate of the user, the dynamic presentation management module is used for promoting a product on line when the click rate of the user exceeds a preset number of people so as to expand the popularity of the product, the platform processing module is used for analyzing the click rate of the advertisement through each block chain link so as to coordinate and control the modules, the data monitoring module comprises a user search record storage unit, an advertisement click rate data monitoring unit and a feedback unit, wherein the user search record storage unit is used for recording and storing keywords of the same product when a user searches the same product so as to adjust the keywords of the product per se, the advertisement click rate data monitoring unit is used for monitoring the number of times of browsing the advertisement by clicking currently, so that the playing effect of the advertisement of the current product can be analyzed, when the click rate of the current advertisement is detected to exceed the preset number of times, the current advertisement is sent to the dynamic presentation management module through the platform processing module to be popularized on line, when the number of times of browsing the advertisement is detected to be lower than the preset number of times, the current advertisement is sent to the static presentation management module through the platform processing module to adjust the keywords, and the feedback unit is used for sending the data of the advertisement to the platform, so as to receive a next step instruction, the static presentation management module comprises a keyword aggregation module, a semantic keyword screening module, a keyword dimension reduction screening module and a fatigue period replacement module, wherein the keyword aggregation module is used for monitoring the browsing amount of each keyword of the current product and adjusting the keyword, the semantic keyword screening module is used for extracting the keyword with the same semantic meaning from the set when the long-term click rate of the keyword stored in the keyword aggregation module is detected to be lower, the keyword dimension reduction screening module is used for performing dimension reduction processing on the keyword with the lower click rate in the period, storing the keyword with the lower click rate in the set, recombining, splitting or moving the keyword in the set to present the sequence of the keyword, and the fatigue period replacement module is used for when the click rate of the keyword after adjustment is detected to be lower than a preset value, the system comprises a platform processing module, a data monitoring module, a static presentation management module and a dynamic presentation management module, wherein the platform processing module is connected with the data monitoring module, the static presentation management module and the dynamic presentation management module, the dynamic presentation management module comprises a customer group analysis module and a private customized advertisement analysis module, the customer group analysis module is used for analyzing according to the customer group clicking the advertisement at present and judging that the advertisement is a young man, a young woman or an old person group, the private customized advertisement analysis module is used for setting negative keywords aiming at the search habits of different users so as to promote the click rate of the advertisement, and the platform processing module is connected with the data monitoring module, the static presentation management module and the dynamic presentation management.
Further, set the current set of keywords as Q = { Q = }1,q2,q3...qnD = { D } for the current set of click rates for different keywords1,d2,d3...dnH, when the click rate d of the keyword is detectedkWhen the semantic content is smaller than the preset value, screening the keywords with the same semantics as the keywords, and determining the keywords with the same semantics as the keywords by the following steps:
z01, segmenting the current keywords, deleting the low-frequency keywords after segmentation, and setting different vectors according to different keywords;
z02, taking the current other key word vectors as the center, and judging the cosine similarity until traversing the key words with similar semantemes;
z03, hiding the same semantic keywords in the set;
wherein: the cosine similarity is expressed as
Figure 268808DEST_PATH_IMAGE001
When is coming into contact with
Figure 167494DEST_PATH_IMAGE002
The closer the value of (1) is, the higher the similarity value of the keywords in the representative set with other keywords is; when in use
Figure 127229DEST_PATH_IMAGE002
Is much larger than or much smaller than 1, the lower the similarity between the keywords in the representative set and other keywords, x refers to the keyword vector, y refers to other keyword vectors,
Figure 605615DEST_PATH_IMAGE002
refers to the angle between the keyword vectors,
Figure 717927DEST_PATH_IMAGE003
is referred to as the module of the keyword vector,
Figure 771203DEST_PATH_IMAGE004
refers to the modulus of other keyword vectors;
by analyzing cosine similarity values, whether keywords of similar sentences exist in each set is judged, and when the measured cosine similarity value is close to 1, the keywords with similar semantics can be judged to exist in the set.
Furthermore, the dimensionality reduction processing is carried out on the keywords with lower click rate, and the dimensionality reduction processing steps are as follows:
z001, setting the dimensionality of the current keyword as a vector matrix Z with n rows and m columns;
z002, calculating the mean value of each line of the current vector matrix and obtaining a new vector matrix Z';
z003, calculating an eigenvalue and an eigenvector value according to a covariance matrix formula;
z004: and arranging a vector matrix Z ' according to the characteristic vector values from large to small, splitting, combining and adjusting the sequence of the keywords to be subjected to dimension reduction, wherein the dimension of the vector matrix is n ' rows and m ' columns.
The keywords with low click rate are judged, the keywords are analyzed and processed when no keywords with similar sentences appear, original vectors are set, the keywords after splitting and the keywords before splitting can be compared, the keywords are subjected to dimensionality reduction by using a PCA method, data after dimensionality reduction can be mutually independent, when the sequence is adjusted, a null keyword needs to be preset, when the keywords needing to be changed are changed, the keywords needing to be changed are stored in the null values so as to be changed in time, when the keywords are split by using the PCA, the null values of the keywords need to be independently set for the contents needing to be split, the vectors projected into the corresponding matrix are one-dimensional vector matrices, the vectors can be combined by adding and subtracting the vectors, and the covariance matrix formula is a symmetric matrix, the existing characteristic vectors can be judged through the set formula.
Further, in the step Z004, setting an original un-recombined keyword set to M = { M1, M2, M3.. mn }, where the split and combined keyword set is M '= { ma + mb, ma1+ mb1.. man + mbn }, when it is detected that the user searches for the same product, analyzing a keyword set H = { c1+ d1, c2+ d2, c3+ d3... cn + dn } of the product searched by the user and a matching degree value sim of the recombined keyword set M', when the matching degree value sim is greater than a preset threshold, recommending an advertisement with a higher matching degree to the user by the platform processing module, sorting according to the matching degree from large to small, and when the matching value sim is less than the preset threshold, replacing the advertisement content;
the conditions required to be met by the negative keywords are set as follows:
a1, when the matching degree value of the keyword set H and M' of the product searched by the user is larger than a preset threshold value, and the click rate of the user is larger than the click rate of the keyword set before the keywords are not split and combined and is smaller than a second preset threshold value;
a2, adding negative different keywords aiming at different crowds when the click rate of the split and combined keywords is less than half of the click rate of the applicable crowds;
the second preset click rate is the sum of the click rates before and after the splitting of the keyword;
the negative keyword is that the advertiser does not see related words for the designated user, and the advertiser needs to set the negative keyword on the assumption that the keyword of the current cup is light and convenient but the advertiser does not want to see the convenient advertising word for the user.
When the matched value sim after the recombination is detected to be smaller than the preset threshold value, the first-time advertisement promotion time before the recombination of the keywords is T1According to the method, the interest degree value generated by detecting the click rate of the user on the advertisement by the platform processing module is W1, and the interest degree value generated by detecting the click rate of the user on the advertisement within the preset time is W2When the keyword after the recombination is detected, the interest degree generated by the click rate of the user to the advertisement is W3Wherein W is3>W1>W2When SW is detected3-SW2<SW2-SW1And then, when the fatigue degree of the user to the advertisement is larger than a preset threshold value, the interest degree of the user is reduced after the keywords are recombined, so that the content of the advertisement needs to be replaced, and the time is divided into three segments, wherein SW1、SW2、SW3The interest degrees of the user for the advertisement in the former period, the interest degrees of the user for the advertisement in the middle period and the interest degrees of the user for the advertisement in the later period are respectively;
and when detecting that the recombined matching value sim is larger than a preset threshold value, the platform processing module automatically matches suffix words suitable for different users behind the keywords according to the search records of the same type of products specified by the users, wherein the suffix words are not equal to the negative keywords.
The system comprises the following steps:
d1, using a data monitoring module to record the keywords of the historical data when the user searches for the same product, simultaneously monitoring the click rate caused by adding or deleting different keywords, and monitoring the advertisement effect presented by the advertisement in real time, thereby achieving the purpose of improving the click rate of the user;
d2: using a static presentation management module, retrieving keywords with similar meanings, deleting or adding the retrieved similar keywords, reducing the dimensions of different keywords, re-splitting the keywords after the dimensions are reduced, combining and adjusting the sequence, judging the exposure of the current advertisement according to the result, and judging whether the click rate of a user is improved or not;
d3, using dynamic process module, when the click rate of the advertisement user exceeds the preset click rate, carrying out the directional analysis according to different groups, and setting different negative keywords according to different users, thereby further promoting the click rate of the advertisement.
Example 1: setting a notebook computer, wherein keywords need to be set for the notebook computer, the set keywords are { lightweight, light in weight, large in memory, capable of playing large games and high in processing speed of a processor }, the click rate in the current keyword set is set to be {10,12,50, 90 and 20}, and the minimum keywords are determined to be { lightweight } according to the detected click rate;
the keywords in the set are judged through cosine similarity, the phrase similarity of light and light is detected to be high, the keywords are hidden, the keyword with the lowest click rate in the rest keywords is processed, the processor speed is high, the keywords are split into the processor, speed, processing and fast, the keywords are set according to the same type of products, the keyword combination with high click rate of the keywords related to the processor, such as the processor fast, can be correspondingly set, the speed and the processing are hidden, the user is not watched, the current processor and fast vectors are combined, one-dimensional vectors are promoted, and the click rate can be increased.
It is noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus.
Finally, it should be noted that: although the present invention has been described in detail with reference to the foregoing embodiments, it will be apparent to those skilled in the art that changes may be made in the embodiments and/or equivalents thereof without departing from the spirit and scope of the invention. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims (8)

1. An advertisement service information publishing system based on a block chain is characterized in that: the system comprises a data monitoring module, a static presentation management module, a dynamic presentation management module and a platform processing module, wherein the data monitoring module is used for monitoring the effect of the advertisement in real time and monitoring the click rate of a user in real time, the static presentation management module is used for screening advertisement keywords, the dynamic presentation management module is used for promoting products on line when the click rate of the user exceeds the preset number of people, the platform processing module is used for analyzing the click rate of the advertisement through each block chain link point, the data monitoring module comprises a user search record storage unit, an advertisement click rate data monitoring unit and a feedback unit, the user search record storage unit is used for recording and storing the keywords of the same product when the user searches the same product, and the advertisement click rate data monitoring unit is used for monitoring the number of times of browsing the advertisement by clicking currently, when the click rate of the current advertisement is detected to exceed the preset times, the current advertisement is sent to a dynamic presentation management module through a platform processing module for online promotion, when the number of times of browsing the advertisement is detected to be lower than the preset times, the current advertisement is sent to a static presentation management module through the platform processing module to adjust the keywords, the feedback unit is used for sending the advertisement data to the platform processing module in real time, the static presentation management module comprises a keyword assembly module, a semantic keyword screening module, a keyword dimension reduction screening module and a fatigue period replacement module, the keyword assembly module is used for monitoring the browsing amount of each keyword of the current product and adjusting the keywords, the semantic keyword screening module is used for extracting the keywords with the same semantics in the assembly when the long-term click rate of the keywords stored in the keyword assembly module is detected to be lower, the keyword dimension reduction screening module is used for carrying out dimension reduction processing on the keywords with lower click rate in the period, storing the keywords with lower click rate in a set, recombining, splitting or moving the keywords in the set to display the sequence of the keywords, the fatigue period replacement module is used for resetting the stored keywords when detecting that the click rate of the adjusted keywords is lower than a preset value, the dynamic presentation management module comprises a customer group analysis module and a private customized advertisement analysis module, the customer group analysis module is used for analyzing according to the customer group which clicks the advertisement at present and judging that the keywords are young men, young women or old people, the private customized advertisement analysis module is used for setting negative keywords according to the search habits of different users, and the platform processing module, the data monitoring module, the fatigue period replacement module and the platform processing module are used for analyzing the keywords and the platform processing module, the data, The static presentation management module is connected with the dynamic presentation management module.
2. The method of claim 1The advertisement service information publishing system based on the block chain is characterized in that: set the set of current keywords as Q = { Q = { (Q) }1,q2,q3...qnD = { D } for the current set of click rates for different keywords1,d2,d3...dnH, when the click rate d of the keyword is detectedkWhen the semantic content is smaller than the preset value, screening the keywords with the same semantics as the keywords, and determining the keywords with the same semantics as the keywords by the following steps:
z01, segmenting the current keywords, deleting the low-frequency keywords after segmentation, and setting different vectors according to different keywords;
z02, taking the current other key word vectors as the center, and judging the cosine similarity until traversing the key words with similar semantemes;
z03, hiding the same semantic keywords in the set;
wherein: the cosine similarity is expressed as
Figure 853045DEST_PATH_IMAGE001
When is coming into contact with
Figure 135122DEST_PATH_IMAGE002
The closer the value of (1) is, the higher the similarity value of the keywords in the representative set with other keywords is; when in use
Figure 351208DEST_PATH_IMAGE002
Is much larger than or much smaller than 1, the lower the similarity between the keywords in the representative set and other keywords, x refers to the keyword vector, y refers to other keyword vectors,
Figure 326117DEST_PATH_IMAGE003
refers to the angle between the keyword vectors,
Figure 745597DEST_PATH_IMAGE004
is referred to as the module of the keyword vector,
Figure 80633DEST_PATH_IMAGE005
refers to the modulus of other keyword vectors.
3. The system according to claim 2, wherein the system comprises: and performing dimensionality reduction processing on the keywords with lower click rate, wherein the dimensionality reduction processing comprises the following steps:
z001, setting the dimensionality of the current keyword as a vector matrix Z with n rows and m columns;
z002, calculating the mean value of each line of the current vector matrix and obtaining a new vector matrix Z';
z003, calculating an eigenvalue and an eigenvector value according to a covariance matrix formula;
z004: and arranging a vector matrix Z ' according to the characteristic vector values from large to small, splitting, combining and adjusting the sequence of the keywords to be subjected to dimension reduction, wherein the dimension of the vector matrix is n ' rows and m ' columns.
4. The system according to claim 3, wherein the system comprises: in step Z004, the original un-recombined keyword set is set as M = { M1, M2, M3.. mn }, and the split and combined keyword set is set as M' = { M =a+mb,ma1+mb1...man+mbnAnd when the condition that the user searches for the same product is detected, analyzing a keyword set H = { c) of the product searched by the user1+d1,c2+d2,c3+d3...cn+dnAnd when the matching degree value sim is larger than a preset threshold value, the platform processing module recommends advertisements with higher matching degree to the user, sorts the advertisements according to the matching degree from large to small, and replaces advertisement promotion contents when the matching value sim is smaller than the preset threshold value.
5. The system according to claim 4, wherein the system comprises: the conditions required to be met by the negative keywords are set as follows:
a1, when the matching degree value of the keyword set H and M' of the product searched by the user is larger than a preset threshold value, and the click rate of the user is larger than the click rate of the keyword set before the keywords are not split and combined and is smaller than a second preset threshold value;
a2, adding negative different keywords aiming at different crowds when the click rate of the split and combined keywords is less than half of the click rate of the applicable crowds;
the second preset click rate refers to a sum of click rates before and after splitting of the keyword.
6. The system according to claim 4, wherein the system comprises: when the matched value sim after the recombination is detected to be smaller than the preset threshold value, the first-time advertisement promotion time before the recombination of the keywords is T1According to the method, the interest degree value generated by detecting the click rate of the user on the advertisement by the platform processing module is W1, and the interest degree value generated by detecting the click rate of the user on the advertisement within the preset time is W2When the keyword after the recombination is detected, the interest degree generated by the click rate of the user to the advertisement is W3Wherein W is3>W1>W2When SW is detected3-SW2<SW2-SW1And then, when the fatigue degree of the user to the advertisement is larger than a preset threshold value, the interest degree of the user is reduced after the keywords are recombined, so that the content of the advertisement needs to be replaced, and the time is divided into three segments, wherein SW1、SW2、SW3The interest degrees of the user for the advertisement in the former period, the interest degrees of the user for the advertisement in the middle period and the interest degrees of the user for the advertisement in the later period are respectively.
7. The system according to claim 4, wherein the system comprises: and when detecting that the recombined matching value sim is larger than a preset threshold value, the platform processing module automatically matches suffix words suitable for different users behind the keywords according to the search records of the same type of products specified by the users, wherein the suffix words are not equal to the negative keywords.
8. The system according to claim 1, wherein the system comprises: the system comprises the following steps:
d1, using a data monitoring module to record the keywords of the historical data when the user searches for the same product, simultaneously monitoring the click rate caused by adding or deleting different keywords, and monitoring the advertisement effect presented by the advertisement in real time, thereby achieving the purpose of improving the click rate of the user;
d2: using a static presentation management module, retrieving keywords with similar meanings, deleting or adding the retrieved similar keywords, reducing the dimensions of different keywords, re-splitting the keywords after the dimensions are reduced, combining and adjusting the sequence, judging the exposure of the current advertisement according to the result, and judging whether the click rate of a user is improved or not;
d3, using dynamic process module, when the click rate of the advertisement user exceeds the preset click rate, carrying out the directional analysis according to different groups, and setting different negative keywords according to different users, thereby further promoting the click rate of the advertisement.
CN202110471029.8A 2021-04-29 2021-04-29 Advertisement service information publishing system based on block chain Active CN112990986B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110471029.8A CN112990986B (en) 2021-04-29 2021-04-29 Advertisement service information publishing system based on block chain

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110471029.8A CN112990986B (en) 2021-04-29 2021-04-29 Advertisement service information publishing system based on block chain

Publications (2)

Publication Number Publication Date
CN112990986A true CN112990986A (en) 2021-06-18
CN112990986B CN112990986B (en) 2021-07-16

Family

ID=76336564

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110471029.8A Active CN112990986B (en) 2021-04-29 2021-04-29 Advertisement service information publishing system based on block chain

Country Status (1)

Country Link
CN (1) CN112990986B (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113570413A (en) * 2021-07-28 2021-10-29 杭州王道控股有限公司 Method and device for generating advertisement keywords, storage medium and electronic equipment
CN115423536A (en) * 2022-10-31 2022-12-02 北京学橙归来科技有限公司 Control method for publishing advertisement information in user management system

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105912686A (en) * 2016-04-18 2016-08-31 上海珍岛信息技术有限公司 Search engine marketing bid method and system based on machine learning
CN109886748A (en) * 2019-02-20 2019-06-14 广州宏数科技有限公司 A kind of advertisement intelligent put-on method and system
CN111882370A (en) * 2020-09-27 2020-11-03 武汉卓尔数字传媒科技有限公司 Advertisement recommendation method and device and electronic equipment
KR102206518B1 (en) * 2020-05-29 2021-01-22 (주) 알트소프트 Banner advertisement service system that determines the priority for each banner advertisement in the advertisement area corresponding to the search word according to the number of grid points included in the reference region

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105912686A (en) * 2016-04-18 2016-08-31 上海珍岛信息技术有限公司 Search engine marketing bid method and system based on machine learning
CN109886748A (en) * 2019-02-20 2019-06-14 广州宏数科技有限公司 A kind of advertisement intelligent put-on method and system
KR102206518B1 (en) * 2020-05-29 2021-01-22 (주) 알트소프트 Banner advertisement service system that determines the priority for each banner advertisement in the advertisement area corresponding to the search word according to the number of grid points included in the reference region
CN111882370A (en) * 2020-09-27 2020-11-03 武汉卓尔数字传媒科技有限公司 Advertisement recommendation method and device and electronic equipment

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
郑屹: "基于模型融合的广告推荐算法的研究与实现", 《中国优秀硕士学位论文全文数据库》 *

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113570413A (en) * 2021-07-28 2021-10-29 杭州王道控股有限公司 Method and device for generating advertisement keywords, storage medium and electronic equipment
CN113570413B (en) * 2021-07-28 2023-12-05 杭州王道控股有限公司 Advertisement keyword generation method and device, storage medium and electronic equipment
CN115423536A (en) * 2022-10-31 2022-12-02 北京学橙归来科技有限公司 Control method for publishing advertisement information in user management system
CN115423536B (en) * 2022-10-31 2023-01-17 北京学橙归来科技有限公司 Control method for publishing advertisement information in user management system

Also Published As

Publication number Publication date
CN112990986B (en) 2021-07-16

Similar Documents

Publication Publication Date Title
EP3779841B1 (en) Method, apparatus and system for sending information, and computer-readable storage medium
US7885952B2 (en) Cloaking detection utilizing popularity and market value
CN104866474B (en) Individuation data searching method and device
US9706008B2 (en) Method and system for efficient matching of user profiles with audience segments
CN101641697B (en) Related search queries for a webpage and their applications
KR100650404B1 (en) On-line Advertising System And Method
US7912868B2 (en) Advertisement placement method and system using semantic analysis
CN103544216B (en) The information recommendation method and system of a kind of combination picture material and keyword
US7831474B2 (en) System and method for associating an unvalued search term with a valued search term
CN109189904A (en) Individuation search method and system
KR101132942B1 (en) Methods and systems for determining a meaning of a document to match the document to conte
CN107180093B (en) Information searching method and device and timeliness query word identification method and device
US20100057536A1 (en) System And Method For Providing Community-Based Advertising Term Disambiguation
US20070143278A1 (en) Context-based key phrase discovery and similarity measurement utilizing search engine query logs
US20110066615A1 (en) Personalization engine for building a user profile
US20090024554A1 (en) Method For Matching Electronic Advertisements To Surrounding Context Based On Their Advertisement Content
CN112990986B (en) Advertisement service information publishing system based on block chain
CN104679771A (en) Individual data searching method and device
CN109885773A (en) A kind of article personalized recommendation method, system, medium and equipment
KR20070053282A (en) Method and apparatus for responding to end-user request for information
CN111444304A (en) Search ranking method and device
Hillard et al. The sum of its parts: reducing sparsity in click estimation with query segments
CN113837842A (en) Commodity recommendation method and equipment based on user behavior data
CN110766486A (en) Method and device for determining item category
US7949576B2 (en) Method of providing product database

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
GR01 Patent grant
GR01 Patent grant