Advertisement display method, system and equipment based on page information flow and storage medium

By establishing data connection with the user's display page, obtaining and analyzing page information flow data, combining keyword and price analysis, the problems of insufficient correlation between advertising content and user browsing content are solved, and the correlation of advertising display is improved and rationality improves user experience.

CN118941338AInactive Publication Date: 2024-11-12深圳君颜科技有限公司
View PDF 3 Cites 0 Cited by

Patent Information

Application Number
CN202410961074.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-17
Publication Date
2024-11-12
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

Existing advertising display recommendation technology is difficult to improve the correlation between advertising content and user browsing content at the same time, resulting in failed advertising push and affecting user experience.

Method used

By establishing data connection with the user's display page, obtaining page information flow data, performing keyword analysis and price analysis, analyzing the display of advertisements to be displayed based on keyword matching degree and price range, and advertising display is performed based on the type of display page.

Benefits of technology

It improves the relevance and rationality of advertising content with user browsing content, and enhances the effectiveness and user experience of advertising display recommendations.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118941338A_ABST
    Figure CN118941338A_ABST
Patent Text Reader

Abstract

The invention discloses an advertisement display method, system and device based on page information flow and a storage medium, and relates to the technical field of advertisement display recommendation, and the method comprises the following steps: establishing data connection with a display page of a user, and obtaining page information flow data; performing keyword analysis on the page information flow data to obtain a keyword matching degree; performing price analysis on the page information flow data to obtain a recommended price range; based on the keyword matching degree and the recommended price range, analyzing whether the to-be-displayed advertisement can be displayed; displaying the to-be-displayed advertisement based on the analysis result and the display type of the display page; the method and the device are used for solving the problems that the advertisement pushing fails and the user experience is influenced due to the fact that the correlation between the advertisement content and the user browsing content is insufficient and the price does not conform to the expectation of the user in the existing advertisement display recommendation technology.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of advertisement display recommendation, and specifically to an advertisement display method, system, device and storage medium based on page information flow. Background Art

[0002] Ad display recommendation technology refers to a technology that selects the most suitable advertisements from the ad library and displays them to users by analyzing users' interests, behaviors and preferences, as well as the attributes and characteristics of the advertising content. This technology aims to provide a more personalized and relevant advertising experience, thereby increasing the click-through rate and conversion rate of advertisements.

[0003] Existing advertising display recommendation technologies usually recommend advertisements based on the information flow of pages browsed by users. However, there are usually two problems when recommending advertisements based on the information flow of pages. One is whether the content of the advertisement is the same as or highly correlated with the content browsed by users, and the other is whether the price of the recommended advertisement meets the psychological expectations of users. Both of them play an important role in the success rate of advertisement recommendation, and it is difficult for existing advertising display recommendation technologies to take both into account at the same time. For example, in a patent application with application publication number CN109978584A, a method and device for pushing information flow advertisements are disclosed. The scheme analyzes advertisements to be pushed by analyzing historical data, and advertisement push is intended to recommend content that users are currently interested in, rather than historical content. Therefore, analysis based on historical data will lead to the problem that the pushed advertisements are poorly correlated with the content browsed by users. Existing advertising display recommendation technologies also have the problem that the advertisement content is not sufficiently correlated with the content browsed by users and the price does not meet the expectations of users, resulting in the failure of advertisement push and affecting the user experience. Summary of the invention

[0004] The present invention aims to solve one of the technical problems in the prior art to at least a certain extent, by establishing a data connection with the user's display page, obtaining page information flow data, performing keyword analysis on the page information flow data, obtaining keyword matching, and performing price analysis on the page information flow data to obtain a recommended price range, and then analyzing whether the to-be-displayed advertisement can be displayed based on the keyword matching and the recommended price range, and finally displaying the to-be-displayed advertisement based on the analysis result and the display type of the display page, so as to solve the problems in the existing advertisement display recommendation technology that the advertisement content is not sufficiently related to the user's browsing content and the price does not meet the user's expectations, resulting in the failure of advertisement push and affecting the user's usage experience.

[0005] To achieve the above objectives, in a first aspect, the present application provides an advertisement display method based on a page information flow, comprising the following steps:

[0006] Establish a data connection with the user's display page to obtain page information flow data;

[0007] Perform keyword analysis on page information flow data to obtain keyword matching degree;

[0008] Perform price analysis on page information flow data to obtain a recommended price range;

[0009] Analyze whether the ad to be displayed can be displayed based on the keyword matching degree and recommended price range;

[0010] The display advertisement is displayed based on the analysis result and the display type of the display page.

[0011] Furthermore, establishing a data connection with the user's display page and obtaining page information flow data includes the following sub-steps:

[0012] Establish a data connection with the user's display page;

[0013] Obtain each piece of information in the user's display page and name it display information, and mark the label of the display information as a display label;

[0014] Obtain the displayed information that the user clicks to enter detailed browsing, mark it as interest information, and mark the display tag of the interest information as an interest tag;

[0015] Get the price corresponding to the display information, marked as display price; get the price of the interest information, marked as interest price;

[0016] The page information flow data includes display tags, interest tags, display prices and interest prices.

[0017] Furthermore, performing keyword analysis on the page information flow data to obtain the keyword matching degree includes the following sub-steps:

[0018] The display tags and interest tags are segmented by the intelligent segmentation model to obtain display segmentations and interest segmentations. Each display information corresponds to a set of display segmentations, and each interest information corresponds to a set of interest segmentations.

[0019] The display participle includes a number of display words, and the interest participle includes a number of interest words;

[0020] Find the intersection of all display participles, named display intersection. If the display intersection exists, output a display intersection existence signal. If the display intersection does not exist, output a display intersection non-existence signal.

[0021] If a display intersection exists signal, the number of display words in the display intersection is counted and marked as the number of display words, and the number of display words is compared with the first number threshold. If the number of display words is less than the first number threshold, an insufficient display number signal is output; if the number of display words is greater than or equal to the first number threshold, a sufficient display number signal is output;

[0022] If the output signal indicates that the number of displays is insufficient or that the display intersection does not exist, the number of times each display word appears is counted and marked as the number of words, and the display words in the display intersection and several display words with the largest number of displays are combined into a first word group, where the number of display words in the first word group is a first number threshold; if the output signal indicates that the number of displays is sufficient, the display intersection is marked as the first word group;

[0023] Find the intersection of all interest segmentations, named interest intersection. If the interest intersection exists, output an interest intersection existence signal. If the interest intersection does not exist, output an interest intersection non-existence signal.

[0024] If the output interest intersection existence signal, then count the number of interest words in the interest intersection, mark it as the number of interest words, compare the number of interest words with the first quantity threshold, if the number of interest words is less than the first quantity threshold, then output the interest quantity insufficient signal; if the number of interest words is greater than or equal to the first quantity threshold, then output the interest quantity sufficient signal;

[0025] If the output signal of insufficient number of interests or no signal of interest intersection exists, the number of occurrences of each interest word is counted and marked as the number of interests, and the interest words in the interest intersection and several interest words with the largest number of interests are merged into a second phrase, and the number of interest words in the second phrase is the first number threshold; if the output signal of sufficient number of interests is used, the interest intersection is marked as the second phrase;

[0026] The keyword matching degree of the advertisement to be displayed is analyzed based on the first phrase and the second phrase.

[0027] Furthermore, analyzing the keyword matching degree of the advertisement to be displayed based on the first phrase and the second phrase includes the following sub-steps:

[0028] Obtain a label of the advertisement to be displayed, named as an advertisement label, and segment the advertisement label using an intelligent segmentation model to obtain advertisement segmentation, wherein the advertisement segmentation includes a plurality of advertisement words;

[0029] For any advertisement to be displayed, set the keyword matching degree, find out whether the advertisement word exists in the first phrase, if so, increase the keyword matching degree by one, otherwise the keyword matching degree remains unchanged; find out whether the advertisement word exists in the second phrase, if so, increase the keyword matching degree by two, otherwise the keyword matching degree remains unchanged, analyze all advertisement words in the advertisement word segmentation, and obtain the final keyword matching degree;

[0030] Calculate the keyword matching degree of each ad to be displayed.

[0031] Furthermore, performing price analysis on the page information flow data to obtain a recommended price range includes the following sub-steps:

[0032] Count the number of times different display prices appear, and mark it as price frequency;

[0033] A rectangular coordinate system is established with the display price as the X-axis and the price frequency as the Y-axis, named the display frequency coordinate system, the display price and price frequency are entered into the display frequency coordinate system, and the coordinate points in the display frequency coordinate system are named frequency coordinate points;

[0034] Connect the frequency coordinate points through a smooth curve to obtain a frequency change curve;

[0035] Count the number of times different interest prices appear, mark them as interest frequencies, mark the maximum value of the interest price as the price threshold, and mark the interest price with the largest interest frequency as the expected price;

[0036] Calculate (SY-SQ) / SQ and mark the result as the price increase, where SY is the price threshold and SQ is the expected price. Calculate SY×SF+SY and mark the result as the price peak, where SF is the price increase.

[0037] The part of the frequency change curve that is greater than the price peak is removed to obtain a new curve, named the price expectation curve, and the peak of the price expectation curve is obtained, named the curve peak. The largest curve peak is marked as the maximum peak, and the curve peak that is smaller than the maximum peak and closest to the maximum peak is marked as the price valley.

[0038] Use [price valley, price peak] as the recommended price range.

[0039] Furthermore, based on the keyword matching degree and the recommended price range, analyzing whether the advertisement to be displayed can be displayed includes the following sub-steps:

[0040] Get the price range of the advertisement to be displayed, marked as advertisement price;

[0041] Get the intersection of the advertising price and the recommended price range, mark it as price intersection, calculate the difference between the maximum and minimum values ​​in the price intersection, mark it as intersection difference; calculate the difference between the maximum and minimum values ​​of the advertising price, mark it as advertising price difference;

[0042] Calculate the intersection difference and divide it by the advertising price difference, and mark the result as price matching degree;

[0043] Sort the price matching degree from large to small, using the symbol Pn It means that n is a positive integer and n is the serial number of P, and the same price matching degree can be listed in parallel;

[0044] Get the keyword matching degree of the ad to be displayed, sort the keyword matching degrees from large to small, and use the symbol Q m Indicates that m is a positive integer and m is the serial number of Q, and keywords with the same matching degree can be listed in parallel;

[0045] Starting from n=m=1, find P n With Q m Whether the same ad to be displayed exists in the search results, if so, output an ad display signal; if not, output a continue search signal;

[0046] If a continue search signal is output, n is increased by 1 and the search is repeated. If a continue search signal is output again, m is increased by 1 and the search is repeated. This cycle is repeated until an advertisement display signal is output.

[0047] Furthermore, displaying the advertisement to be displayed based on the analysis result and the display type of the display page includes the following sub-steps:

[0048] Displaying the to-be-displayed advertisement of the output advertisement display signal based on the display type of the display page;

[0049] The display types include video and picture types;

[0050] When an ad needs to be displayed again, the ads that have been displayed will no longer be included in the analysis.

[0051] In a second aspect, the present application provides an advertising display system based on a page information flow, comprising a data acquisition module, a keyword analysis module, a price analysis module, a display analysis module and an advertising display module; the data acquisition module, the keyword analysis module, the price analysis module and the advertising display module are respectively connected to the display analysis module data;

[0052] The data acquisition module is used to establish a data connection with the user's display page and obtain page information flow data;

[0053] The keyword analysis module is used to perform keyword analysis on the page information flow data to obtain the keyword matching degree;

[0054] The price analysis module is used to perform price analysis on the page information flow data to obtain a recommended price range;

[0055] The display analysis module is used to analyze whether the advertisement to be displayed can be displayed based on the keyword matching degree and the recommended price range;

[0056] The advertisement display module is used to display the advertisement to be displayed based on the analysis result and the display type of the display page.

[0057] In a third aspect, the present application provides an electronic device, comprising a processor and a memory, wherein the memory stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, the steps in the above method are performed.

[0058] In a fourth aspect, the present application provides a storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps in the above method are performed.

[0059] Beneficial effects of the present invention: The present invention establishes a data connection with the user's display page to obtain display tags and interest tags, analyzes the display tags and interest tags to obtain the first phrase and the second phrase, and then analyzes the first phrase and the second phrase to obtain the keyword matching degree of the advertisement to be displayed. The advantage is that the first phrase reflects the information composition in the page information flow, and the second phrase reflects the information composition in the content that the user is interested in. By analyzing them, the matching degree of the advertisement to be displayed at the page information flow level and the user's interest level, that is, the keyword matching degree, can be obtained, thereby improving the relevance of the advertisement content to the user's browsing content;

[0060] The present invention establishes a data connection with the user's display page to obtain the display price and the interest price, and obtains the recommended price range by analyzing the display price and the interest price. The advantage is that since the price of the commodity browsed by the user is usually the expected price in his mind, the price increase calculated by the interest price can provide a certain increase range for the maximum price. At the same time, the price distribution of such commodities is analyzed by the display price to find the most suitable price valley, so as to screen suitable advertisements to be displayed for the user, so that the price of the commodity in the advertisement is more in line with the user's expectations, thereby improving the effectiveness and rationality of advertisement display recommendation;

[0061] The present invention analyzes the advertisements to be displayed by combining the keyword matching degree and the price matching degree, and displays the advertisements to be displayed with higher rankings in both the keyword matching degree and the price matching degree as recommended advertisements. The advantage is that the keyword matching degree reflects the relevance of the advertisement to be displayed and the content browsed by the user, while the price matching degree reflects whether the price of the advertisement to be displayed is close to the psychological expectation of the user. By combining the two, both can be taken into account, thereby improving the relevance of the advertisement and the rationality of the price. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] Figure 1 is a functional block diagram of the system of the present invention;

[0063] Figure 2 A schematic diagram showing information of the present invention;

[0064] Figure 3 is the display frequency coordinate system of the present invention;

[0065] Figure 4 is a schematic diagram of a price expectation curve of the present invention;

[0066] Figure 5 is a flow chart of the steps of the method of the present invention;

[0067] Figure 6 The figure is a schematic diagram of the structure of an electronic device of the present invention. DETAILED DESCRIPTION

[0068] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0069] Example 1, please refer to Figure 1 As shown, in the first aspect, the present application provides an advertisement display system based on page information flow, including a data acquisition module, a keyword analysis module, a price analysis module, a display analysis module and an advertisement display module; the data acquisition module, the keyword analysis module, the price analysis module and the advertisement display module are respectively connected to the display analysis module data;

[0070] The data acquisition module is used to establish a data connection with the user's display page and obtain page information flow data;

[0071] The data acquisition module is configured with a data acquisition strategy, which includes:

[0072] Establish a data connection with the user's display page;

[0073] See also Figure 2 As shown, each piece of information in the user's display page is obtained and named as display information, and the label of the display information is marked as a display label;

[0074] Obtain the displayed information that the user clicks to enter detailed browsing, mark it as interest information, and mark the display tag of the interest information as an interest tag;

[0075] Get the price corresponding to the display information, marked as display price; get the price of the interest information, marked as interest price;

[0076] Page information flow data includes display tags, interest tags, display prices, and interest prices;

[0077] In actual applications, taking any shopping platform as an example, the definitions of display information, display labels, and display prices are as follows: Figure 2 As shown, if the user clicks on this display information to enter the details page, this display information is marked as interest information, and the corresponding display label and display price are also changed to interest label and interest price. In this embodiment, Figure 2 The display label of the display information is "Newly arrived I512600KF plus Colorful 3060TI ULTRA gaming machine", and the display price is 3888.

[0078] The keyword analysis module is used to perform keyword analysis on the page information flow data to obtain the keyword matching degree; the keyword analysis module includes a phrase analysis unit and a matching degree analysis unit;

[0079] The phrase analysis unit is configured with a phrase analysis strategy, which includes:

[0080] The display tags and interest tags are segmented by the intelligent segmentation model to obtain display segmentations and interest segmentations. Each display information corresponds to a set of display segmentations, and each interest information corresponds to a set of interest segmentations.

[0081] The display participle includes several display words, and the interest participle includes several interest words;

[0082] In actual applications, the intelligent word segmentation model adopts the existing jieba word segmentation technology, and the display words in the display word segmentation obtained by word segmentation include "new arrival", "I5", "12600KF", "plus", "3060TI", "ULTRA", "game" and "whole machine"; the word segmentation of the interest tag is the same as the word segmentation of the display tag, so no specific demonstration is performed in this embodiment;

[0083] Find the intersection of all display participles, named display intersection. If the display intersection exists, output a display intersection existence signal. If the display intersection does not exist, output a display intersection non-existence signal.

[0084] If there is a signal in the output display intersection, the number of display words in the display intersection is counted and marked as the number of display words. The number of display words is compared with the first quantity threshold. If the number of display words is less than the first quantity threshold, an insufficient display quantity signal is output; if the number of display words is greater than or equal to the first quantity threshold, a sufficient display quantity signal is output; in actual applications, the first quantity threshold is set to enhance the directivity of keywords. The directivity of a single word is too broad. For example, in the computer field, multiple components are included, and users have different requirements for different components. Usually, the requirements for components such as CPU and graphics card are more stringent. For example, in the field of skin care products, there are many types of skin care products. Taking hand cream as an example, when searching for hand cream, "hand cream" must be a display intersection, and hand cream has different brands and different main ingredients. The user's requirements for brands and ingredients can further enhance the directivity of keywords, thereby selecting suitable advertisements. Through a large amount of analysis, it is found that the value range of the first quantity threshold is 3 to 5. In this embodiment, the first quantity threshold is set to 4;

[0085] If the output display quantity is insufficient or the display intersection does not have a signal, the number of times each display word appears is counted and marked as the word quantity, and the display words in the display intersection and several display words with the largest display quantity are combined into a first phrase, and the number of display words in the first phrase is the first quantity threshold; if the output display quantity is sufficient, the display intersection is marked as the first phrase;

[0086] In actual applications, since the number of display words is too large and inconvenient to display, they are not displayed in this embodiment. The display intersection is obtained by comparison as "whole machine", and there is a display intersection, and a display intersection existence signal is output. The display intersection contains only one display word, that is, the number of display words is 1. The number of display words obtained by comparison is less than the first quantity threshold, and a display quantity insufficient signal is output; since the number of display words is large, only the top 5 display words in terms of word quantity are listed in this embodiment to illustrate the specific processing process; the number of words obtained by statistics is sorted in order from large to small, and the top 5 display words are "whole machine", "game", "I5", "16G" and "I7" respectively, and the display intersection is "whole machine", which is the same as the first-ranked display word, so it can only be calculated once, and the first phrase obtained by integration is "whole machine", "game", "I5" and "16G"; the reason for not directly obtaining the first phrase by sorting is that the number of display words in the display intersection may be greater than the first quantity threshold, and the role of intersection in relevance is more important than the number of word appearances, so it is necessary to first determine whether the display intersection meets the conditions;

[0087] Find the intersection of all interest segmentations, named interest intersection. If the interest intersection exists, output an interest intersection existence signal. If the interest intersection does not exist, output an interest intersection non-existence signal.

[0088] If the output interest intersection existence signal, then count the number of interest words in the interest intersection, mark it as the number of interest words, compare the number of interest words with the first quantity threshold, if the number of interest words is less than the first quantity threshold, then output the interest quantity insufficient signal; if the number of interest words is greater than or equal to the first quantity threshold, then output the interest quantity sufficient signal;

[0089] If the outputted interest quantity is insufficient or there is no signal of interest intersection, the number of occurrences of each interest word is counted and marked as the interest quantity, and the interest words in the interest intersection and several interest words with the largest interest quantity are merged into a second phrase, and the number of interest words in the second phrase is the first quantity threshold; if the outputted interest quantity signal is sufficient, the interest intersection is marked as the second phrase;

[0090] Analyzing the keyword matching degree of the advertisement to be displayed based on the first phrase and the second phrase;

[0091] In actual application, the interest intersection is found to be "13600KF", "4060TI", "32G" and "WIFI", and there is an interest intersection. An interest intersection existence signal is output. The interest intersection contains four interest words, that is, the number of interest words is 4. By comparison, the number of interest words is equal to the first quantity threshold, and a sufficient interest quantity signal is output. "13600KF", "4060TI", "32G" and "WIFI" are combined and marked as the second phrase;

[0092] The matching analysis unit is configured with a matching analysis strategy, which includes:

[0093] Obtain the label of the advertisement to be displayed, name it as the advertisement label, and segment the advertisement label through the intelligent word segmentation model to obtain advertisement segmentation, which includes several advertisement words;

[0094] For any advertisement to be displayed, set the keyword matching degree, find out whether the advertisement word exists in the first phrase, if so, increase the keyword matching degree by one, otherwise the keyword matching degree remains unchanged; find out whether the advertisement word exists in the second phrase, if so, increase the keyword matching degree by two, otherwise the keyword matching degree remains unchanged, analyze all advertisement words in the advertisement word segmentation, and obtain the final keyword matching degree;

[0095] Calculate the keyword matching degree of each ad to be displayed;

[0096] In actual application, the advertisement tag "I5 13600KF+4060TI 32G memory e-sports game machine" is obtained. Through word segmentation, the advertisement words in the advertisement segmentation include "I5", "13600KF", "+", "4060TI", "32G", "memory", "e-sports", "game" and "machine"; it is found that "I5" exists in the first phrase, and the keyword matching degree is increased by one, which is 1 at this time. It is found that "I5" does not exist in the second phrase, that is, the keyword matching degree remains unchanged, and is still 1. Then "13600KF" is analyzed, and it is found that "13600KF" does not exist in the first phrase, but exists in the second phrase. Therefore, the keyword matching degree is increased by 2, and the keyword matching degree is 3 at this time. And so on, until all advertisement words are analyzed, and the final keyword matching degree is 9, and the keyword matching degree of each advertisement to be displayed is calculated. Due to the large amount of data, it is not specifically displayed in this embodiment.

[0097] The price analysis module is used to perform price analysis on page information flow data to obtain a recommended price range;

[0098] The price analysis module is configured with price analysis strategies, which include:

[0099] Count the number of times different display prices appear, and mark it as price frequency;

[0100] See also Figure 3 As shown, a rectangular coordinate system is established with the display price as the X-axis and the price frequency as the Y-axis, named the display frequency coordinate system, the display price and price frequency are entered into the display frequency coordinate system, and the coordinate points in the display frequency coordinate system are named frequency coordinate points;

[0101] Connect the frequency coordinate points through a smooth curve to obtain a frequency change curve;

[0102] In actual applications, due to the large fluctuation of prices, this embodiment does not show specific numerical values, but only roughly summarizes them through the frequency change curve. The frequency coordinate system is obtained by constructing Figure 3 As shown, the curve is the frequency change curve;

[0103] Count the number of times different interest prices appear, mark them as interest frequencies, mark the maximum value of the interest price as the price threshold, and mark the interest price with the largest interest frequency as the expected price;

[0104] Calculate (SY-SQ) / SQ and mark the result as the price increase, where SY is the price threshold and SQ is the expected price. Calculate SY×SF+SY and mark the result as the price peak, where SF is the price increase.

[0105] See also Figure 4 As shown, the part of the frequency change curve that is greater than the price peak is removed to obtain a new curve, named the price expectation curve, and the peak of the price expectation curve is obtained, named the curve peak. The largest curve peak is marked as the maximum peak, and the curve peak that is smaller than the maximum peak and closest to the maximum peak is marked as the price valley.

[0106] Take [price valley, price peak] as the recommended price range;

[0107] In practical applications, the maximum value of the interest price represents the price limit that the user can currently accept, but it may also increase, so it is necessary to calculate the price increase. The interest price with the largest interest frequency reflects that the user has browsed the expected price the most times, which can reflect the user's expectations of the price of such goods to a certain extent. Through statistics, the price threshold SY is 6000, the expected price SQ is 5500, and the price increase is 0.091. The calculation result is rounded to three decimal places, and the price peak is further calculated to be 6546. The part of the frequency change curve greater than 6545 is removed to obtain the price expectation curve as shown below: Figure 4 As shown, based on Figure 4 The maximum peak is obtained as Figure 4 As shown, the price valley is 5428, that is, the recommended price range is [5428, 6546].

[0108] The display analysis module is used to analyze whether the advertisement to be displayed can be displayed based on the keyword matching degree and the recommended price range;

[0109] The display analysis module is configured with display analysis strategies, which include:

[0110] Get the price range of the advertisement to be displayed, marked as advertisement price;

[0111] Get the intersection of the advertising price and the recommended price range, mark it as price intersection, calculate the difference between the maximum and minimum values ​​in the price intersection, mark it as intersection difference; calculate the difference between the maximum and minimum values ​​of the advertising price, mark it as advertising price difference;

[0112] Calculate the intersection difference and divide it by the advertising price difference, and mark the result as price matching degree;

[0113] In actual applications, the price of a product usually starts at XX yuan, so there is a certain range of prices. The obtained advertising price is [5399, 6499], and the price intersection of the advertising price and the recommended price is [5428, 6499]. The intersection difference is calculated to be 1071, and the advertising price difference is 1100. Further calculation shows that the price matching degree is 0.9736, and the calculation result is rounded to 4 decimal places.

[0114] Sort the price matching degree from large to small, using the symbol P n It means that n is a positive integer and n is the serial number of P, and the same price matching degree can be listed in parallel;

[0115] Get the keyword matching degree of the ad to be displayed, sort the keyword matching degrees from large to small, and use the symbol Q m Indicates that m is a positive integer and m is the serial number of Q, and keywords with the same matching degree can be listed in parallel;

[0116] Starting from n=m=1, find P n With Q m Whether the same ad to be displayed exists in the search results, if so, output an ad display signal; if not, output a continue search signal;

[0117] If a continue search signal is output, n is increased by 1 and the search is repeated. If a continue search signal is output again, m is increased by 1 and the search is repeated. The process is repeated until an advertisement display signal is output.

[0118] In practical applications, P is obtained by sorting 1 Including Ad A and Ad B, P 2 Including advertising C, P 3 Includes advertising D; Q 1 Includes ads C, Q 2 Includes Ad A, Ad D, and Ad Q 3 Include ad B; find P 1 With Q 1 There is no same ad to be displayed in the output, so the output continues to search for the signal, increases n by 1, and searches for P 1 , P 2 With Q 1 Whether there is the same ad to be displayed in P 2 Zhong and Q 1 Advertisement C is included in both, so an advertisement display signal is output and advertisement C needs to be displayed.

[0119] The advertisement display module is used to display the advertisement to be displayed based on the analysis results and the display type of the display page;

[0120] The advertisement display module is configured with advertisement display strategies, which include:

[0121] Displaying the to-be-displayed advertisement of the output advertisement display signal based on the display type of the display page;

[0122] Display types include video and picture types;

[0123] When an ad needs to be displayed again, the ads that have been displayed will no longer be included in the analysis;

[0124] In practical applications, such as Figure 2 As shown in , the display type is image type, so Ad C is displayed in the display page according to the image type, and Ad C no longer participates in the analysis of the next advertisement to be displayed. If the user exits the display page at this time and enters again, Ad C can continue to participate in the analysis.

[0125] Example 2, please refer to Figure 5 As shown, in the second aspect, the present application provides an advertisement display method based on a page information flow, comprising the following steps:

[0126] Step S1, establishing a data connection with the user's display page to obtain page information flow data; Step S1 includes the following sub-steps:

[0127] Step S101, establishing a data connection with the user's display page;

[0128] Step S102, obtaining each piece of information in the user's display page and naming it as display information, and marking the label of the display information as a display label;

[0129] Step S103, obtaining the displayed information that the user clicks to enter detailed browsing, marking it as interest information, and marking the display tag of the interest information as an interest tag;

[0130] Step S104, obtaining the price corresponding to the display information, marking it as the display price; obtaining the price of the interest information, marking it as the interest price;

[0131] Step S105, the page information flow data includes display tags, interest tags, display prices and interest prices;

[0132] Step S2, performing keyword analysis on the page information flow data to obtain keyword matching degree; Step S2 includes the following sub-steps:

[0133] Step S201, segmenting the display tags and interest tags through an intelligent segmentation model to obtain display segmentations and interest segmentations, where each display information corresponds to a group of display segmentations, and each interest information corresponds to a group of interest segmentations;

[0134] Step S202, the display segmentation includes a number of display words, and the interest segmentation includes a number of interest words;

[0135] Step S203, finding the intersection of all display participles, naming it the display intersection, if the display intersection exists, outputting a display intersection existence signal, if the display intersection does not exist, outputting a display intersection non-existence signal;

[0136] Step S204: if a display intersection existence signal is output, the number of display words in the display intersection is counted and marked as the number of display words, and the number of display words is compared with a first number threshold. If the number of display words is less than the first number threshold, an insufficient display number signal is output; if the number of display words is greater than or equal to the first number threshold, a sufficient display number signal is output;

[0137] Step S205: if the output signal is that the display quantity is insufficient or the display intersection does not exist, the number of times each display word appears is counted and marked as the word quantity, and the display words in the display intersection and the display words with the largest display quantity are combined into a first phrase, and the number of display words in the first phrase is a first quantity threshold; if the output signal is that the display quantity is sufficient, the display intersection is marked as the first phrase;

[0138] Step S206, finding the intersection of all interest segmentations, naming it as interest intersection, if the interest intersection exists, outputting an interest intersection existence signal, if the interest intersection does not exist, outputting an interest intersection non-existence signal;

[0139] Step S207, if the interest intersection existence signal is output, the number of interest words in the interest intersection is counted, marked as the number of interest words, and the number of interest words is compared with the first number threshold. If the number of interest words is less than the first number threshold, an insufficient interest number signal is output; if the number of interest words is greater than or equal to the first number threshold, a sufficient interest number signal is output;

[0140] Step S208: If the output signal is that the number of interests is insufficient or the interest intersection does not exist, the number of occurrences of each interest word is counted and marked as the number of interests, and the interest words in the interest intersection and several interest words with the largest number of interests are merged into a second phrase, and the number of interest words in the second phrase is the first number threshold; if the output signal is that the number of interests is sufficient, the interest intersection is marked as the second phrase;

[0141] Step S209, analyzing the keyword matching degree of the advertisement to be displayed based on the first phrase and the second phrase;

[0142] Step S209 includes the following sub-steps:

[0143] Step S2091, obtaining a label of the advertisement to be displayed, naming it an advertisement label, and segmenting the advertisement label through an intelligent word segmentation model to obtain advertisement segmentation, which includes a number of advertisement words;

[0144] Step S2092: for any advertisement to be displayed, set the keyword matching degree, find out whether the advertisement word exists in the first phrase, if so, increase the keyword matching degree by one, otherwise the keyword matching degree remains unchanged; find out whether the advertisement word exists in the second phrase, if so, increase the keyword matching degree by two, otherwise the keyword matching degree remains unchanged, analyze all advertisement words in the advertisement word segmentation, and obtain the final keyword matching degree;

[0145] Step S2093, calculating the keyword matching degree of each advertisement to be displayed;

[0146] Step S3, performing price analysis on the page information flow data to obtain a recommended price range; Step S3 includes the following sub-steps:

[0147] Step S301, counting the number of times different display prices appear, and marking it as price frequency;

[0148] Step S302, establishing a rectangular coordinate system with the display price as the X-axis and the price frequency as the Y-axis, named the display frequency coordinate system, entering the display price and the price frequency into the display frequency coordinate system, and naming the coordinate points in the display frequency coordinate system as frequency coordinate points;

[0149] Step S303, connecting the frequency coordinate points through a smooth curve to obtain a frequency change curve;

[0150] Step S304, counting the number of occurrences of different interest prices, marking them as interest frequencies, marking the maximum value of the interest prices as the price threshold, and marking the interest price with the maximum interest frequency as the expected price;

[0151] Step S305, calculate (SY-SQ) / SQ, and mark the calculation result as the price increase, where SY is the price threshold, SQ is the expected price, calculate SY×SF+SY, and mark the calculation result as the price peak, where SF is the price increase;

[0152] Step S306, remove the part of the frequency change curve that is greater than the price peak value, obtain a new curve, name it the price expectation curve, obtain the peak of the price expectation curve, name it the curve peak, mark the largest curve peak as the maximum peak, and mark the curve peak that is smaller than the maximum peak and closest to the maximum peak as the price valley value;

[0153] Step S307, taking [valley price, peak price] as the recommended price range;

[0154] Step S4, based on the keyword matching degree and the recommended price range, analyzing whether the advertisement to be displayed can be displayed; Step S4 includes the following sub-steps:

[0155] Step S401, obtaining a price range of the advertisement to be displayed, and marking it as advertisement price;

[0156] Step S402, obtaining the intersection of the advertisement price and the recommended price range, marking it as the price intersection, calculating the difference between the maximum and minimum values ​​in the price intersection, marking it as the intersection difference; calculating the difference between the maximum and minimum values ​​of the advertisement price, marking it as the advertisement price difference;

[0157] Step S403, calculating the intersection difference divided by the advertisement price difference, and marking the calculation result as the price matching degree;

[0158] Step S404: sort the price matching degrees in descending order, using the symbol P n It means that n is a positive integer and n is the serial number of P, and the same price matching degree can be listed in parallel;

[0159] Step S405: Obtain the keyword matching degree of the advertisement to be displayed, sort the keyword matching degrees in descending order, and sort the keyword matching degrees by symbol Q. m Indicates that m is a positive integer and m is the serial number of Q, and keywords with the same matching degree can be listed in parallel;

[0160] Step S406, starting from n=m=1, search for P n With Q m Whether the same ad to be displayed exists in the search results, if so, output an ad display signal; if not, output a continue search signal;

[0161] Step S407: if a continue search signal is output, n is increased by 1 and the search is repeated; if a continue search signal is output again, m is increased by 1 and the search is repeated, and the process is repeated until an advertisement display signal is output;

[0162] Step S5, displaying the advertisement to be displayed based on the analysis result and the display type of the display page; Step S5 includes the following sub-steps:

[0163] Step S501, displaying the to-be-displayed advertisement outputting the advertisement display signal based on the display type of the display page;

[0164] Step S502, display types include video and picture types;

[0165] Step S503: When the advertisement needs to be displayed again, the advertisement to be displayed that has been displayed will no longer be analyzed.

[0166] Example 3, please refer to Figure 6 As shown, Figure 6The structural diagram of an electronic device is illustrated, and the electronic device may include: a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus. The memory stores computer-readable instructions, and the processor may call the instructions in the memory. When the computer-readable instructions are executed by the processor, the steps in the method for displaying advertisements based on page information flow are executed to implement the following functions: establish a data connection with the user's display page to obtain page information flow data; perform keyword analysis on the page information flow data to obtain keyword matching; perform price analysis on the page information flow data to obtain a recommended price range; analyze whether the advertisement to be displayed can be displayed based on the keyword matching and the recommended price range; and display the advertisement to be displayed based on the analysis results and the display type of the display page.

[0167] In addition, the logic instructions in the above-mentioned memory can be implemented in the form of software functional units and can be stored in a computer-readable storage medium when sold or used as an independent product. Based on this understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art, and the computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a disk or an optical disk.

[0168] Embodiment 4, the present application also provides a computer program product, the computer program product includes a computer program stored on a computer-readable storage medium, the computer program includes program instructions, when the program instructions are executed by the computer, the computer can execute the advertising display method based on the page information flow provided by the above methods, the method includes: establishing a data connection with the user's display page to obtain page information flow data; performing keyword analysis on the page information flow data to obtain keyword matching degree; performing price analysis on the page information flow data to obtain a recommended price range; based on the keyword matching degree and the recommended price range, analyzing whether the advertisement to be displayed can be displayed; based on the analysis result and the display type of the display page, displaying the advertisement to be displayed.

[0169] Embodiment 5, the present application also provides a computer-readable storage medium, the present application provides a storage medium, on which a computer program is stored, and when the computer program is executed by the processor, the steps in the above-mentioned method for displaying advertisements based on page information flow are executed. Through the above-mentioned technical scheme, when the computer program is executed by the processor, the method in any optional implementation of the above-mentioned embodiment is executed to realize the following functions: establish a data connection with the user's display page to obtain page information flow data; perform keyword analysis on the page information flow data to obtain keyword matching degree; perform price analysis on the page information flow data to obtain a recommended price range; analyze whether the advertisement to be displayed can be displayed based on the keyword matching degree and the recommended price range; display the advertisement to be displayed based on the analysis result and the display type of the display page.

[0170] Through the description of the above implementation methods, the embodiments of the present invention can be provided as methods, systems or computer program products. Based on such an understanding, the above technical solutions can be essentially or partly contributed to the prior art in the form of software products, which can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, an optical disk, etc., and include several instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0171] In the embodiments provided in the present application, it should be understood that the disclosed system or method can be implemented in other ways. The embodiments described above are only schematic. For example, the division of modules or units is only a logical function division. There may be other division methods in actual implementation. For example, multiple modules or units can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some communication interfaces, and the indirect coupling or communication connection of systems, modules and units can be electrical, mechanical or other forms.

[0172] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. An advertisement display method based on page information flow, characterized in that: The steps include: Establish a data connection with the user's display page to obtain page information flow data; Perform keyword analysis on page information flow data to obtain keyword matching degree; Perform price analysis on page information flow data to obtain a recommended price range; Analyze whether the ad to be displayed can be displayed based on the keyword matching degree and recommended price range; The display advertisement is displayed based on the analysis result and the display type of the display page.

2. The method for displaying advertisements based on page information flow according to claim 1, characterized in that: Establishing a data connection with the user's display page and obtaining page information flow data includes the following sub-steps: Establish a data connection with the user's display page; Obtain each piece of information in the user's display page and name it display information, and mark the label of the display information as a display label; Obtain the displayed information that the user clicks to enter detailed browsing, mark it as interest information, and mark the display tag of the interest information as an interest tag; Get the price corresponding to the display information, marked as display price; get the price of the interest information, marked as interest price; The page information flow data includes display tags, interest tags, display prices and interest prices.

3. The method for displaying advertisements based on page information flow according to claim 2, characterized in that: Performing keyword analysis on page information flow data to obtain keyword matching includes the following sub-steps: The display tags and interest tags are segmented by the intelligent segmentation model to obtain display segmentations and interest segmentations. Each display information corresponds to a set of display segmentations, and each interest information corresponds to a set of interest segmentations. The display participle includes a number of display words, and the interest participle includes a number of interest words; Find the intersection of all display participles, named display intersection. If the display intersection exists, output a display intersection existence signal. If the display intersection does not exist, output a display intersection non-existence signal. If a display intersection exists signal, the number of display words in the display intersection is counted and marked as the number of display words, and the number of display words is compared with the first number threshold. If the number of display words is less than the first number threshold, an insufficient display number signal is output; If the number of displayed words is greater than or equal to the first quantity threshold, a sufficient display quantity signal is output; If the output signal indicates that the number of displays is insufficient or that the display intersection does not exist, the number of times each display word appears is counted and marked as the number of words, and the display words in the display intersection and several display words with the largest number of displays are combined into a first word group, where the number of display words in the first word group is a first number threshold; if the output signal indicates that the number of displays is sufficient, the display intersection is marked as the first word group; Find the intersection of all interest segmentations, named interest intersection. If the interest intersection exists, output an interest intersection existence signal. If the interest intersection does not exist, output an interest intersection non-existence signal. If the output interest intersection existence signal, then count the number of interest words in the interest intersection, mark it as the number of interest words, compare the number of interest words with the first quantity threshold, if the number of interest words is less than the first quantity threshold, then output the interest quantity insufficient signal; if the number of interest words is greater than or equal to the first quantity threshold, then output the interest quantity sufficient signal; If the output signal of insufficient number of interests or no signal of interest intersection exists, the number of occurrences of each interest word is counted and marked as the number of interests, and the interest words in the interest intersection and several interest words with the largest number of interests are merged into a second phrase, and the number of interest words in the second phrase is the first number threshold; if the output signal of sufficient number of interests is used, the interest intersection is marked as the second phrase; The keyword matching degree of the advertisement to be displayed is analyzed based on the first phrase and the second phrase.

4. The method for displaying advertisements based on page information flow according to claim 3, characterized in that: Analyzing the keyword matching degree of the advertisement to be displayed based on the first phrase and the second phrase includes the following sub-steps: Obtain a label of the advertisement to be displayed, named as an advertisement label, and segment the advertisement label using an intelligent segmentation model to obtain advertisement segmentation, wherein the advertisement segmentation includes a plurality of advertisement words; For any advertisement to be displayed, set the keyword matching degree, and find out whether the advertisement word exists in the first phrase. If so, increase the keyword matching degree by one, otherwise, the keyword matching degree remains unchanged; Find whether the advertising word exists in the second phrase. If so, increase the keyword matching degree by two. Otherwise, the keyword matching degree remains unchanged. Analyze all advertising words in the advertising word segmentation to obtain the final keyword matching degree. Calculate the keyword matching degree of each ad to be displayed.

5. The method for displaying advertisements based on page information flow according to claim 4, characterized in that: Performing price analysis on page information flow data to obtain a recommended price range includes the following sub-steps: Count the number of times different display prices appear, and mark it as price frequency; A rectangular coordinate system is established with the display price as the X-axis and the price frequency as the Y-axis, named the display frequency coordinate system, the display price and price frequency are entered into the display frequency coordinate system, and the coordinate points in the display frequency coordinate system are named frequency coordinate points; Connect the frequency coordinate points through a smooth curve to obtain a frequency change curve; Count the number of times different interest prices appear, mark them as interest frequencies, mark the maximum value of the interest price as the price threshold, and mark the interest price with the largest interest frequency as the expected price; Calculate (SY-SQ) / SQ and mark the result as the price increase, where SY is the price threshold and SQ is the expected price. Calculate SY×SF+SY and mark the result as the price peak, where SF is the price increase. The part of the frequency change curve that is greater than the price peak is removed to obtain a new curve, named the price expectation curve, and the peak of the price expectation curve is obtained, named the curve peak. The largest curve peak is marked as the maximum peak, and the curve peak that is smaller than the maximum peak and closest to the maximum peak is marked as the price valley. Use [price valley, price peak] as the recommended price range.

6. The method for displaying advertisements based on page information flow according to claim 5, characterized in that: Based on the keyword matching degree and the recommended price range, analyzing whether the ad to be displayed can be displayed includes the following sub-steps: Get the price range of the advertisement to be displayed, marked as advertisement price; Get the intersection of the advertised price and the recommended price range, mark it as price intersection, calculate the difference between the maximum and minimum values ​​in the price intersection, and mark it as intersection difference; Calculate the difference between the maximum and minimum advertising prices, and mark it as the advertising price difference; Calculate the intersection difference and divide it by the advertising price difference, and mark the result as price matching degree; Sort the price matching degree from large to small, using the symbol P n It means that n is a positive integer and n is the serial number of P, and the same price matching degree can be listed in parallel; Get the keyword matching degree of the ad to be displayed, sort the keyword matching degrees from large to small, and use the symbol Q m Indicates that m is a positive integer and m is the serial number of Q, and keywords with the same matching degree can be listed in parallel; Starting from n=m=1, find P n With Q m Whether the same ad to be displayed exists in the search results, if so, output an ad display signal; if not, output a continue search signal; If a continue search signal is output, n is increased by 1 and the search is repeated. If a continue search signal is output again, m is increased by 1 and the search is repeated. This cycle is repeated until an advertisement display signal is output.

7. The method for displaying advertisements based on page information flow according to claim 6, characterized in that: Displaying the display ad based on the analysis results and the display type of the display page includes the following sub-steps: Displaying the to-be-displayed advertisement of the output advertisement display signal based on the display type of the display page; The display types include video and picture types; When an ad needs to be displayed again, the ads that have been displayed will no longer be included in the analysis.

8. An advertisement display system based on page information flow, used to implement the advertisement display method based on page information flow according to any one of claims 1 to 7, characterized in that: It includes a data acquisition module, a keyword analysis module, a price analysis module, a display analysis module and an advertisement display module; the data acquisition module, the keyword analysis module, the price analysis module and the advertisement display module are respectively connected with the display analysis module data; The data acquisition module is used to establish a data connection with the user's display page and obtain page information flow data; The keyword analysis module is used to perform keyword analysis on the page information flow data to obtain the keyword matching degree; The price analysis module is used to perform price analysis on the page information flow data to obtain a recommended price range; The display analysis module is used to analyze whether the advertisement to be displayed can be displayed based on the keyword matching degree and the recommended price range; The advertisement display module is used to display the advertisement to be displayed based on the analysis result and the display type of the display page.

9. An electronic device, characterized in that: The method comprises a processor and a memory, wherein the memory stores computer-readable instructions. When the computer-readable instructions are executed by the processor, the steps in the method according to any one of claims 1 to 7 are executed.

10. A storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps in the method according to any one of claims 1 to 7 are executed.

Citation Information

Patent Citations

  • Data processing method and apparatus

    CN105718184A

  • Advertisement marketing recommendation method based on data analysis

    CN116911926A

  • Tantalum capacitor and manufacturing methhod the same

    KR1020220132819A