Advertisement marketing system based on data source positioning

Abstract

The invention relates to the technical field of advertisement marketing, in particular to an advertisement marketing system based on data source positioning. The system comprises an interest marketing unit, a fatigue attenuation unit, a variety delivery unit and a calculation adjustment unit. The fatigue degree module is used for calculating a fatigue value of the user in contact with the matched interest point commodity advertisement according to the interest attenuation frequency and the advertisement contact frequency, judging whether the user generates fatigue degree on the matched interest point commodity advertisement by utilizing the fatigue value of the user in contact with the matched interest point commodity advertisement, and calculating the fatigue value to consider the factor; according to the method, a large amount of advertisements can be prevented from being continuously put when the interest of the user is low, advertisement putting better fits the current interest state of the user, the acceptability of marketing advertisements is improved, meanwhile, more appropriate opportunities are selected to put advertisements according to the current interest state of the user, the user is promoted to purchase interest point commodities in the marketing advertisements, and the user experience is improved. And the click rate and the conversion rate of the advertisement are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of advertising marketing, and more specifically, to an advertising marketing system based on data source positioning. Background Art

[0002] Advertising marketing based on data source positioning helps advertisers achieve precise marketing by integrating and analyzing multiple data sources. With the continuous progress of technology and the in-depth application, through reasonable implementation and continuous optimization, advertisers can better reach the target audience. When analyzing the behavior data of users' short video swiping, it is found that users are interested in mobile products in the video. Advertising marketers match the user's interest point products with the advertisement library, match the advertisements containing the interest point products and put them on the user side. Over time, users may experience a decline in interest or a transfer of interest in the interest point products. When users' interest in the interest point products declines, advertising marketers continue to conduct excessive marketing and investment when the users' interest in the interest point products is already very low, which increases the interference to users, resulting in a decrease in the users' favorability and trust in advertising marketers and a reduction in the user experience effect. If users have already transferred their interest to other products, advertising marketers increase advertising investment, resulting in a waste of advertising resources, making users feel fatigued with the advertised products and causing users to dislike the advertisements and advertising marketers. Therefore, we provide an advertising marketing system based on data source positioning. Summary of the Invention

[0003] The purpose of the present invention is to provide an advertising marketing system based on data source positioning to solve the problems raised in the above background art.

[0004] To achieve the above purpose, the present invention provides an advertising marketing system based on data source positioning, including an interest marketing unit, a decay and fatigue unit, a diverse placement unit, and a calculation and adjustment unit;

[0005] The interest marketing unit obtains the video data preferred by users for data source positioning, excavates the videos watched by users and the number of times of watching the videos to calculate the interest degree of the videos and analyzes the interest point products that users are interested in the videos, then matches the advertisements containing the interest point products with the advertisement library, formulates a marketing placement strategy according to the matched interest point product advertisements and puts them on the users, and records the number of times users contact the matched interest point product advertisements and the frequency of placing the matched interest point product advertisements;

[0006] The attenuation fatigue unit is used to receive the number of times of interest point product advertisements that are contact-matched in the interest marketing unit and the interest point products, calculate the advertisement contact frequency and the interest attenuation frequency of the user for the interest point products, analyze whether there is an interest transfer situation of the user for the interest point products by using the interest attenuation frequency. When there is no command for the interest transfer situation, calculate the fatigue value of the user's contact with the interest point product advertisements that are contact-matched according to the interest attenuation frequency and the advertisement contact frequency, and use the fatigue value of the user's contact with the interest point product advertisements that are contact-matched to determine whether the user has a fatigue degree for the interest point product advertisements that are matched;

[0007] The diversified placement unit is used to receive the command of the interest transfer situation in the attenuation fatigue unit. The diversified placement unit obtains the interest point product advertisements that are matched from the interest marketing unit and calculates the similarity, calculates the score of the advertisement diversity by using the calculated similarity, and then uses the calculated advertisement diversity score to judge the diversity degree of the similar advertisement set;

[0008] The calculation adjustment unit is used to receive the command that the user has a fatigue degree for the interest point product advertisements that are matched in the attenuation fatigue unit. The calculation adjustment unit obtains the fatigue value of the contact with the interest point product advertisements that are contact-matched from the attenuation fatigue unit and the frequency of placing the interest point product advertisements that are matched in the interest marketing unit, calculates the advertisement placement frequency adjustment factor according to the fatigue value of the contact with the interest point product advertisements that are contact-matched, uses the frequency of placing the interest point product advertisements that are matched and the advertisement placement frequency adjustment factor to calculate the adjusted advertisement placement frequency, and controls the interest point product advertisements that are matched according to the calculated adjusted advertisement placement frequency.

[0009] As a further improvement of this technical solution, the interest marketing unit includes an interest degree module and a marketing placement module;

[0010] The interest degree module obtains the video data liked by the user through web crawlers, uses big data technology to locate the data source of the obtained user-liked video data, locates the source of the video data, and then uses data mining technology to mine the videos watched by the user, the watched videos liked, and the video comment content data at the source of the video data, records the number of times of watching videos, the number of liked videos, and the number of video comments and calculates the interest degree of the videos, analyzes whether the user is interested in the videos by using the calculated video interest degree. When the calculated video interest degree is greater than the set video interest degree, it is analyzed that the user is interested in the videos;

[0011] The marketing placement module is used to receive the command that the user is interested in the video analyzed by the interest module. The marketing placement module obtains the viewed videos, liked viewed videos, and video comment content data from the interest module and analyzes the interest point products that the user is interested in the video. Then, it matches the interest point products with the advertisement library to find the advertisements containing the interest point products, formulates a marketing placement strategy based on the matched interest point product advertisements, and then places the matched interest point product advertisements to the user side according to the formulated marketing placement strategy, and records the number of times the user contacts the matched interest point product advertisements, the time period for placing the matched interest point product advertisements, the time when the user contacts the matched interest point product advertisements, the time interval between adjacent contacts of the matched interest point product advertisements, and the frequency of placing the matched interest point product advertisements.

[0012] As a further improvement of this technical solution, the attenuation fatigue unit includes a contact attenuation module and a fatigue degree module;

[0013] The contact attenuation module is used to receive the number of times the user contacts the matched interest point product advertisements, the time when the user contacts the matched interest point product advertisements, and the interest point products in the marketing placement module. It calculates the advertisement contact frequency according to the number of times the user contacts the matched interest point product advertisements, and then obtains the time interval from the last contact of the matched interest point product advertisements to the current contact of the matched interest point product advertisements according to the time when the user contacts the matched interest point product advertisements. It calculates the interest attenuation frequency of the user for the interest point products according to the time interval by using the exponential attenuation function, and then analyzes whether there is an interest transfer situation of the user for the interest point products by using the interest attenuation frequency. When the interest attenuation frequency is greater than the set interest attenuation frequency, it is analyzed that there is an interest transfer situation of the user for the interest point products. When the interest attenuation frequency is less than the set interest attenuation frequency, it is analyzed that there is no interest transfer situation of the user for the interest point products;

[0014] The fatigue degree module is used to receive the command that there is no interest transfer situation in the contact attenuation module. The fatigue degree module obtains the interest attenuation frequency, advertisement contact frequency, and the matched interest point product advertisements in the marketing placement module from the contact attenuation module, calculates the fatigue value of the user's contact with the matched interest point product advertisements according to the interest attenuation frequency and advertisement contact frequency, and judges whether the user has a fatigue degree for the matched interest point product advertisements by using the fatigue value of the user's contact with the matched interest point product advertisements. When the fatigue value of the user's contact with the matched interest point product advertisements is greater than the set fatigue value, it is determined that the user has a fatigue degree for the matched interest point product advertisements.

[0015] As a further improvement of this technical solution, the implementation principle of calculating the fatigue value of the user's contact with the matched interest point product advertisements in the fatigue degree module:

[0016] Collect the interest attenuation frequency and the advertisement contact frequency and weight coefficients 、 Calculate the fatigue value of the user's exposure to the point-of-interest product advertisement , and the specific algorithm formula is:

[0017] ;

[0018] Wherein, and The value ranges of are between , and .

[0019] As a further improvement of the present technical solution, the diversified placement unit is used to receive the command with interest transfer in the contact attenuation module. The diversified placement unit obtains the matching point-of-interest product advertisement from the marketing placement module, enables the point-of-interest product advertisement matched by the user to be rematched with the advertisement library, rematches the advertisement similar to the point-of-interest product advertisement matched by the user, inputs the matched point-of-interest product advertisement and the rematched advertisement into the decision tree model. The decision tree model learns the input data and outputs the corresponding feature vector, calculates the similarity between the matched point-of-interest product advertisement and the rematched advertisement according to the feature vector, and uses the calculated similarity to judge the similarity level between the matched point-of-interest product advertisement and the rematched advertisement. When the calculated similarity is greater than the set standard similarity, it is determined that the similarity between the matched point-of-interest product advertisement and the rematched advertisement is high, combines the matched point-of-interest product advertisement and the rematched advertisement into a similar advertisement set, calculates the score of the advertisement diversity according to the calculated similarity, the rematched advertisement, the matched point-of-interest product advertisement, and the similar advertisement set, and uses the calculated advertisement diversity score to judge the diversity degree of the similar advertisement set.

[0020] As a further improvement of the present technical solution, the specific judgment situations in the diversified placement unit include:

[0021] Situation ①: When the calculated advertisement diversity score is greater than the set standard advertisement diversity score, it is determined that the difference between the advertisement elements in the similar advertisement set is greater and the diversity is better, and then the advertisements in the similar advertisement set are placed to the user side;

[0022] Situation ②: When the calculated advertisement diversity score is less than the set standard advertisement diversity score, it is determined that the difference between the advertisement elements in the similar advertisement set is smaller.

[0023] As a further improvement of the present technical solution, the algorithm formula for calculating the score of the advertisement diversity in the diversified placement unit is:

[0024] .

[0025] As a further improvement of the present technical solution, the calculation and adjustment unit is used to receive the command that the user has a fatigue degree for the matched interest point product advertisement in the fatigue degree module. The calculation and adjustment unit obtains the interest decay frequency from the contact attenuation module, the advertisement contact frequency, and the fatigue value of contacting the matched interest point product advertisement from the fatigue degree module, and obtains the frequency of placing the matched interest point product advertisement from the marketing placement module. The advertisement placement frequency adjustment factor is calculated according to the interest decay frequency, the advertisement contact frequency, and the fatigue value of contacting the matched interest point product advertisement. The calculation of adjusting the advertisement placement frequency is carried out by using the frequency of placing the matched interest point product advertisement and the advertisement placement frequency adjustment factor, and the matched interest point product advertisement is controlled and adjusted according to the calculated adjusted advertisement placement frequency.

[0026] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0027] 1. In the advertising and marketing system based on data source positioning, the contact attenuation module calculates the advertisement contact frequency according to the number of times of contacting the matched interest point product advertisement, and then obtains the time interval from the last time of contacting the matched interest point product advertisement to the current time of contacting the matched interest point product advertisement according to the time of the number of times of contacting the matched interest point product advertisement. The interest decay frequency of the user for the interest point product is calculated by using the time interval, and then whether there is an interest transfer situation of the user for the interest point product is analyzed by using the interest decay frequency. By calculating the interest decay frequency, it is possible to avoid continuing excessive marketing when the user's interest in the interest point product in the marketing advertisement is already very low, reduce the disturbance to the user, enhance the user's favorability and trust in the enterprise. At the same time, for the interest point products with a relatively fast interest decay or the user has already had an interest transfer, the advertisement investment is reduced to avoid waste of resources.

[0028] 2. In the advertising and marketing system based on data source positioning, the fatigue degree module is used to receive the command that there is no interest transfer situation in the contact attenuation module. The fatigue degree module obtains the interest decay frequency, the advertisement contact frequency, and the matched interest point product advertisement in the marketing placement module from the contact attenuation module, calculates the fatigue value of the user contacting the matched interest point product advertisement according to the interest decay frequency and the advertisement contact frequency, and uses the fatigue value of the user contacting the matched interest point product advertisement to judge whether the user has a fatigue degree for the matched interest point product advertisement. By considering this factor through calculating the fatigue value, it is possible to avoid continuing to place a large number of such advertisements when the user's interest is low, make the advertisement placement more in line with the user's current interest state, improve the acceptance of the marketing advertisement. At the same time, according to the user's current interest state, a more appropriate timing for placing advertisements is selected to promote the user's behavior of purchasing the interest point product in the marketing advertisement, and the click-through rate and conversion rate of the advertisement are improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 It is a block diagram of the present invention;

[0030] Figure 2 This is the block diagram of the module unit of the present invention.

[0031] The meanings of the various labels in the figure are as follows:

[0032] 1. Interest marketing unit; 11. Interest degree module; 12. Marketing investment module;

[0033] 2. Attenuation fatigue unit; 21. Contact attenuation module; 22. Fatigue degree module;

[0034] 3. Diverse investment unit; 4. Calculation adjustment unit. Specific implementation manner

[0035] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in the present invention without creative efforts shall fall within the protection scope of the present invention.

[0036] Embodiment 1

[0037] The present invention provides an advertising marketing system based on data source positioning. Please refer to Figure 1 , including an interest marketing unit 1, an attenuation fatigue unit 2, a diverse investment unit 3, and a calculation adjustment unit 4;

[0038] The interest marketing unit 1 obtains video data of user preferences for data source location, mines the videos watched by the user and the number of times the videos are watched to calculate the interest degree of the videos and analyzes the interest points of the goods in which the user is interested in the videos, then matches with the advertisement library to find advertisements containing the interest point goods, formulates a marketing placement strategy according to the matched interest point goods advertisements and places them to the user, records the number of times the user contacts the matched interest point goods advertisements and the frequency of placing the matched interest point goods advertisements; the attenuation fatigue unit 2 is used to receive the number of times the user contacts the matched interest point goods advertisements and the interest point goods in the interest marketing unit 1, calculates the advertisement contact frequency and the interest attenuation frequency of the user for the interest point goods, uses the interest attenuation frequency to analyze whether there is an interest transfer situation for the user for the interest point goods, when there is no command for the interest transfer situation, calculates the fatigue value of the user contacting the matched interest point goods advertisements according to the interest attenuation frequency and the advertisement contact frequency, and uses the fatigue value of the user contacting the matched interest point goods advertisements to judge whether the user has a fatigue degree for the matched interest point goods advertisements; the diverse placement unit 3 is used to receive the command of the interest transfer situation in the attenuation fatigue unit 2, the diverse placement unit 3 obtains the matched interest point goods advertisements from the interest marketing unit 1 and calculates the similarity, calculates the score of advertisement diversity by using the calculated similarity, and then uses the calculated advertisement diversity score to judge the diversity degree of the similar advertisement set; the calculation adjustment unit 4 is used to receive the command that the user has a fatigue degree for the matched interest point goods advertisements in the attenuation fatigue unit 2, the calculation adjustment unit 4 obtains the fatigue value of contacting the matched interest point goods advertisements from the attenuation fatigue unit 2 and the frequency of placing the matched interest point goods advertisements in the interest marketing unit 1, calculates the advertisement placement frequency adjustment factor according to the fatigue value of contacting the matched interest point goods advertisements, uses the frequency of placing the matched interest point goods advertisements and the advertisement placement frequency adjustment factor to calculate the adjustment of the advertisement placement frequency, and controls the matched interest point goods advertisements according to the calculated adjusted advertisement placement frequency.

[0039] The following is the refinement of the above units. Please refer to Figure 2 ;

[0040] The interest marketing unit 1 includes an interest degree module 11 and a marketing placement module 12;

[0041] The interest degree module 11 obtains video data of user preferences through web crawlers, uses big data technology to locate the data source of the obtained user preference video data, locates the origin of the video data source (such as a website, APP), and then uses data mining technology to mine the videos watched by the user, the liked watched videos, and the video comment content data at the origin of the video data source, and records the number of times the videos are watched , the number of liked videos and the number of video comments , according to the number of times the videos are watched , The number of liked videos , The number of video comments Calculate the interest level of the video, and use the calculated video interest level and the set video interest level to analyze whether the user is interested in the video. When the calculated video interest level is greater than the set video interest level, it is analyzed that the user is interested in the video. When the calculated video interest level is less than the set video interest level, it is analyzed that the user is not interested in the video;

[0042] The implementation principle of calculating the video interest level:

[0043] Step ①, Collect the number of times the video is watched , The number of liked videos and the number of video comments data and input them into the machine learning model. The machine model learns the input data and outputs the weight of the number of views , The like weight and the comment weight ;

[0044] Step ②, According to the number of times the video is watched , The number of liked videos , The number of video comments , The weight of the number of views , The like weight and the comment weight calculate the user's interest level in the video, and obtain the calculated video interest level , The specific algorithm formula:

[0045] ;

[0046] Among them, this formula is used to calculate the interest level of the video. By calculating the video interest level, it is possible to accurately identify the user group with strong interest in specific types of videos;

[0047] The marketing placement module 12 is used to receive the command that the user is interested in the video analyzed in the interest level module 11. The marketing placement module 12 obtains the data of the watched video, the liked watched video, and the comment content of the video from the interest level module 11. According to the watched video, the liked watched video, and the comment content of the video, it analyzes the interest point products (including clothes, shoes, food) that the user is interested in the video. Then, it matches the interest point products with the advertisement library, matches the advertisements containing the interest point products, formulates a marketing placement strategy according to the matched interest point product advertisements, and then places the matched interest point product advertisements to the user side and records the number of times the user contacts the matched interest point product advertisements 、Time period for delivering matched point-of-interest product advertisements 、Number of times of exposure to matched point-of-interest product advertisements and time 、Time interval between adjacent times of exposure to matched point-of-interest product advertisements (Interval between the time when the user first exposes to a matched point-of-interest product advertisement and the time when the user first exposes to a matched point-of-interest product advertisement) and frequency of delivering matched point-of-interest product advertisements ;

[0048] The attenuation fatigue unit 2 includes a contact attenuation module 21 and a fatigue degree module 22;

[0049] The contact attenuation module 21 is used to receive the number of times of exposure to matched point-of-interest product advertisements in the marketing delivery module 12 、Number of times of exposure to matched point-of-interest product advertisements and time 、Time period for delivering matched point-of-interest product advertisements 、Time interval between adjacent times of exposure to matched point-of-interest product advertisements and point-of-interest products, and calculate the advertisement exposure frequency according to the number of times of exposure to matched point-of-interest product advertisements 、Time period for delivering matched point-of-interest product advertisements and time interval between adjacent times of exposure to matched point-of-interest product advertisements Calculate the advertisement exposure frequency , where refers to the time interval between the i-th adjacent times of exposure to matched point-of-interest product advertisements. Obtain the time interval from the last exposure to a matched point-of-interest product advertisement to the current exposure to a matched point-of-interest product advertisement according to the time of exposure to matched point-of-interest product advertisements (If the user last saw the advertisement 3 days ago, then (t = 3) days). Use the exponential decay function to calculate the interest decay frequency of the user for the point-of-interest product according to the time interval , where refers to the decay coefficient refers to a mathematical constant, known as the natural constant, also called Euler's number. It is a very important constant in the fields of mathematics and science. Then use the interest decay frequency Analyze whether there is an interest transfer of the user for the POI (Point of Interest) product with the set interest decay frequency. When the interest decay frequency is greater than the set interest decay frequency, it is analyzed that there is an interest transfer of the user for the POI product. When the interest decay frequency is less than the set interest decay frequency, it is analyzed that there is no interest transfer of the user for the POI product. By calculating the interest decay frequency, it is possible to avoid excessive marketing when the user's interest in the product is already very low, reduce the disturbance to the user, which can enhance the user's favorability and trust in the enterprise, strengthen the long-term relationship between the user and the enterprise, and improve the user experience effect. For example, after the interest of a fitness club member in a certain fitness course decays, the club no longer frequently pushes promotional information about relevant courses, but provides other services that more conform to the member's current interests. At the same time, for products with a relatively fast interest decay or for which the user has already had an interest transfer, the advertising investment is reduced to avoid waste of resources;

[0050] Example of calculating the advertising exposure frequency:

[0051] When the time period for placing ads for the matched POI product is one day (24 hours), the number of times the user is exposed to the matched POI product ads is 3 times, the time interval between the user's first and second exposures to the matched POI product ads is 3 hours, and the time interval between the second and third exposures to the matched POI product ads is 5 hours, then the advertising exposure frequency is:

[0052] ;

[0053] The fatigue module 22 is used to receive the command of no interest transfer situation in the contact decay module 21, and the fatigue module 22 obtains the interest decay frequency from the contact decay module 21 and the advertising exposure frequency and the matched POI product ads in the marketing placement module 12 , and input the interest decay frequency and the advertising exposure frequency into the decision tree model. The decision tree model learns the input data and outputs the corresponding weight coefficients , . According to the interest decay frequency , the advertising exposure frequency and the weight coefficients , Calculate the fatigue value of the user's exposure to the interest point product advertisement that matches, and use the fatigue value of the user's exposure to the interest point product advertisement that matches and the set fatigue value for judgment to determine whether the user has a fatigue degree for the interest point product advertisement that matches. When the fatigue value of the user's exposure to the interest point product advertisement that matches is greater than the set fatigue value, it is determined that the user has a fatigue degree for the interest point product advertisement that matches. When the fatigue value of the user's exposure to the interest point product advertisement that matches is less than the set fatigue value, it is determined that the user does not have a fatigue degree for the interest point product advertisement that matches. By calculating the fatigue value and considering this factor, it is possible to accurately understand the user's acceptance of the interest point product advertisement that matches at different time points. For example, within a period of time after a user purchases a certain electronic product, the user's interest in advertisements for this type of product will rapidly decline. By calculating the fatigue value in combination with the interest decay frequency, it is possible to avoid continuing to a large number of such advertisements when the user's interest is low, making the advertisement placement more in line with the user's current interest state, and selecting a more appropriate timing to place advertisements according to the user's current interest state. When the fatigue value is low, the user's acceptance of the advertisement is high. At this time, placing the product advertisement that matches the interest point can improve the click-through rate and conversion rate of the advertisement. For example, when the user has just developed an interest in a certain sports brand and the frequency of exposure to advertisements of this brand is low, placing relevant advertisements is more likely to attract the user's attention and promote the purchase behavior;

[0054] The implementation principle of calculating the fatigue value of the user's exposure to the interest point product advertisement that matches in the fatigue degree module 22:

[0055] Collect the interest decay frequency and the advertisement exposure frequency and the weight coefficient 、 Calculate the fatigue value of the user's exposure to the interest point product advertisement that matches , and the specific algorithm formula:

[0056] ;

[0057] Among them, and The value ranges of are between , and . This formula is used to calculate the fatigue value of the user's exposure to the interest point product advertisement that matches. By calculating the fatigue value and considering this factor, advertising marketers can accurately understand the user's acceptance of the interest point product advertisement that matches at different time points. By calculating the fatigue value in combination with the interest decay frequency, it is possible to avoid continuing to a large number of such advertisements when the user's interest is low, making the advertisement placement more in line with the user's current interest state;

[0058] The diverse placement unit 3 is used to receive the command with an interest transfer situation in the exposure decay module 21, and the diverse placement unit 3 obtains the interest point product advertisement that matches from the marketing placement module 12 , the interest point product advertisement matched with the user Perform another match with the advertisement library, and re-match the advertisements similar to the interest point product advertisement matched with the user to obtain the re-matched advertisements , the matched interest point product advertisement and the re-matched advertisement Input into the decision tree model, and the decision tree model learns the input data and outputs the corresponding feature vectors , , according to the feature vectors and Calculate the similarity between the matched interest point product advertisement and the re-matched advertisement to obtain the calculated similarity , where refers to the dot product of the feature vectors and the feature vector , refers to the Euclidean norm of the feature vector (i.e., the length of the feature vector ), refers to the Euclidean norm of the feature vector (i.e., the length of the feature vector ), ), use the calculated similarity to judge with the set standard similarity, and judge the similarity level between the matched interest point product advertisement and the re-matched advertisement When the calculated similarity is greater than the set standard similarity, it is determined that the similarity between the matched interest point product advertisement and the re-matched advertisement is high, and the matched interest point product advertisement and the re-matched advertisement are combined into a similar advertisement set (the similar advertisement set refers to advertisements with similar elements but different interest point products in the advertisements), record the number of advertisements in the similar advertisement set , , according to the calculated similarity , the re-matched advertisement , the number of advertisements in the similar advertisement set , the matched interest point product advertisement , the similar advertisement set Calculate the score of advertisement diversity, and use the calculated advertisement diversity score Judging the diversity degree of a similar advertisement set based on a set standard advertisement diversity score. When advertisements have rich content and forms, it is easier to stimulate the purchasing desire of the audience. For example, when promoting a smart phone, in addition to showing the appearance and basic functions of the phone, different advertisements can also be used to display its different selling points such as camera effects, gaming performance, and battery life, meeting the concerns of different audiences and thus enhancing the advertisement effect;

[0059] The specific judgment situations in the diverse placement unit 3 include:

[0060] Situation ①: When the calculated advertisement diversity score is greater than the set standard advertisement diversity score, it is determined that the difference between advertisement elements in the similar advertisement set is greater and the diversity is better. Then, the advertisements in the similar advertisement set are placed to the user side;

[0061] Situation ②: When the calculated advertisement diversity score is less than the set standard advertisement diversity score, it is determined that the difference between advertisement elements in the similar advertisement set is smaller;

[0062] The score algorithm formula for calculating the advertisement diversity in the diverse placement unit 3:

[0063] ;

[0064] Among them, this formula is used to calculate the score of advertisement diversity. Diversified advertisements can attract the attention of more audiences and improve the click-through rate and conversion rate of advertisements.

[0065] The calculation adjustment unit 4 is used to receive the command that the user has a fatigue degree for the interest point commodity advertisement matched in the fatigue degree module 22. The calculation adjustment unit 4 obtains the interest decay frequency from the contact attenuation module 21 , the advertisement contact frequency and the fatigue value of contacting the interest point commodity advertisement matched in the fatigue degree module 22 , obtains the frequency of placing the interest point commodity advertisement matched from the marketing placement module 12 , and calculates the advertisement placement frequency adjustment factor according to the interest decay frequency , the advertisement contact frequency , the fatigue value of contacting the interest point commodity advertisement matched , calculates the advertisement placement frequency adjustment factor , and uses the frequency of placing the interest point commodity advertisement matched and the advertisement placement frequency adjustment factor to calculate and adjust the advertisement placement frequency, and obtains the calculated adjusted advertisement placement frequency , control and adjust the matched point-of-interest product advertisements according to the calculated advertisement delivery frequency adjustment, reasonably control the advertisement delivery frequency, avoid over-marketing, and enable the audience to feel the professionalism of the brand and the attention to the user experience. When the advertisement appears at an appropriate frequency and provides valuable information to the audience, it helps to establish a good brand image and enhance the trust and favor of the audience towards the brand;

[0066] Algorithm formula for calculating the advertisement delivery frequency adjustment factor:

[0067] ;

[0068] Among them, the company uses it to calculate the advertisement delivery frequency adjustment factor. By adjusting the delivery frequency, it reduces unnecessary advertisement interference and enables the audience to obtain a more comfortable and pleasant experience when browsing the media or using the application. When the audience feels that the advertisement is customized for them and will not overly disturb them, they will be more willing to interact with the brand, thereby promoting the long-term relationship development between the brand and the user.

[0069] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art of this industry should understand that the present invention is not limited by the above embodiments. The above embodiments and the descriptions in the specification are only preferred examples of the present invention and are not used to limit the present invention. Without departing from the spirit and scope of the present invention, the present invention will have various changes and improvements, and these changes and improvements all fall within the scope of the present invention claimed. The scope of protection claimed by the present invention is defined by the appended claims and their equivalents.

Claims

1. An advertising and marketing system based on data source location, characterized in that: It includes an interest marketing unit (1), a decay fatigue unit (2), a diversified delivery unit (3), and a calculation adjustment unit (4); The interest marketing unit (1) obtains video data preferred by users for data source positioning, mines the videos watched by users and the number of times of watching videos to calculate the interest degree of the videos and analyzes the interest point products that users are interested in the videos, then matches with the advertisement library to find advertisements containing the interest point products, formulates a marketing delivery strategy according to the matched interest point product advertisements and delivers them to users, records the number of times users contact the matched interest point product advertisements and the frequency of delivering the matched interest point product advertisements; The decay fatigue unit (2) is used to receive the number of times of contacting the matched interest point product advertisements and the interest point products in the interest marketing unit (1), calculate the advertisement contact frequency and the interest decay frequency of users for the interest point products, use the interest decay frequency to analyze whether there is an interest transfer situation of users for the interest point products, when there is no command for an interest transfer situation, calculate the fatigue value of users contacting the matched interest point product advertisements according to the interest decay frequency and the advertisement contact frequency, and use the fatigue value of users contacting the matched interest point product advertisements to judge whether users have a fatigue degree for the matched interest point product advertisements; The diversified delivery unit (3) is used to receive the command of an interest transfer situation in the decay fatigue unit (2), the diversified delivery unit (3) obtains the matched interest point product advertisements from the interest marketing unit (1) and calculates the similarity, uses the calculated similarity to calculate the score of advertisement diversity, and then uses the calculated advertisement diversity score to judge the diversity degree of the similar advertisement set; The calculation adjustment unit (4) is used to receive the command that users have a fatigue degree for the matched interest point product advertisements in the decay fatigue unit (2), the calculation adjustment unit (4) obtains the fatigue value of contacting the matched interest point product advertisements from the decay fatigue unit (2) and the frequency of delivering the matched interest point product advertisements in the interest marketing unit (1), calculates the advertisement delivery frequency adjustment factor according to the fatigue value of contacting the matched interest point product advertisements, uses the frequency of delivering the matched interest point product advertisements and the advertisement delivery frequency adjustment factor to calculate the adjustment of the advertisement delivery frequency, and controls the matched interest point product advertisements according to the calculated adjusted advertisement delivery frequency.

2. The advertising and marketing system based on data source location according to claim 1, wherein: The interest marketing unit (1) includes an interest degree module (11) and a marketing delivery module (12); The interest degree module (11) obtains video data preferred by users through web crawlers, uses big data technology to perform data source positioning on the obtained video data preferred by users, locates the source of the video data, then uses data mining technology to mine the videos watched by users, the watched videos with likes, and the video comment content data at the source of the video data, records the number of times of watching videos, the number of liked videos, and the number of video comments and calculates the interest degree of the videos, uses the calculated video interest degree to analyze whether users are interested in the videos, and when the calculated video interest degree is greater than the set video interest degree, analyzes that users are interested in the videos; The marketing placement module (12) is used to receive the command that the user is interested in the video analyzed by the interest module (11). The marketing placement module (12) obtains the viewed videos, liked viewed videos, and video comment content data from the interest module (11) and analyzes the interest point products that the user is interested in the video. Then, it matches the interest point products with the advertisement library to find the advertisements containing the interest point products, formulates a marketing placement strategy based on the matched interest point product advertisements, and then places the matched interest point product advertisements to the user side according to the formulated marketing placement strategy. It records the number of times the user contacts the matched interest point product advertisements, the time period for placing the matched interest point product advertisements, the time when contacting the matched interest point product advertisements, the time interval between adjacent contacts of the matched interest point product advertisements, and the frequency of placing the matched interest point product advertisements.

3. The advertising and marketing system based on data source location according to claim 2, wherein: The attenuation fatigue unit (2) includes a contact attenuation module (21) and a fatigue degree module (22); The contact attenuation module (21) is used to receive the number of times of contacting the matched interest point product advertisements, the time of contacting the matched interest point product advertisements, and the interest point products in the marketing placement module (12). It calculates the advertisement contact frequency according to the number of times of contacting the matched interest point product advertisements, and then obtains the time interval from the last contact of the matched interest point product advertisements to the current contact of the matched interest point product advertisements according to the time of contacting the matched interest point product advertisements. It calculates the interest attenuation frequency of the user for the interest point products according to the time interval by using the exponential attenuation function, and then analyzes whether there is an interest transfer situation of the user for the interest point products by using the interest attenuation frequency. When the interest attenuation frequency is greater than the set interest attenuation frequency, it is analyzed that there is an interest transfer situation of the user for the interest point products. When the interest attenuation frequency is less than the set interest attenuation frequency, it is analyzed that there is no interest transfer situation of the user for the interest point products; The fatigue degree module (22) is used to receive the command that there is no interest transfer situation in the contact attenuation module (21). The fatigue degree module (22) obtains the interest attenuation frequency, the advertisement contact frequency, and the matched interest point product advertisements in the marketing placement module (12) from the contact attenuation module (21). It calculates the fatigue value of the user's contact with the matched interest point product advertisements according to the interest attenuation frequency and the advertisement contact frequency, and uses the fatigue value of the user's contact with the matched interest point product advertisements to judge whether the user has a fatigue degree for the matched interest point product advertisements. When the fatigue value of the user's contact with the matched interest point product advertisements is greater than the set fatigue value, it is determined that the user has a fatigue degree for the matched interest point product advertisements.

4. The advertising and marketing system based on data source location according to claim 3, wherein: The implementation principle of calculating the fatigue value of the user's contact with the matched interest point product advertisements in the fatigue degree module (22): Collect the interest decay frequency , the advertisement exposure frequency and the weight coefficient , Calculate the fatigue value of the user's exposure to the interest point products' advertisements that match , and the specific algorithm formula is: ； Among them, and The value ranges are between and, and .

5. The advertising and marketing system based on data source location according to claim 3, wherein: The diverse delivery unit (3) is used to receive the command indicating the presence of interest transfer in the contact attenuation module (21). The diverse delivery unit (3) obtains the interest point product advertisements that match in the marketing delivery module (12), re-matches the user-matched interest point product advertisements with the advertisement library, and re-matches the advertisements similar to the user-matched interest point product advertisements. The matched interest point product advertisements and the re-matched advertisements are input into the decision tree model. The decision tree model learns the input data and outputs the corresponding feature vectors. The similarity between the matched interest point product advertisements and the re-matched advertisements is calculated based on the feature vectors. The similarity between the matched interest point product advertisements and the re-matched advertisements is judged according to the calculated similarity. When the calculated similarity is greater than the set standard similarity, it is determined that the similarity between the matched interest point product advertisements and the re-matched advertisements is high. The matched interest point product advertisements and the re-matched advertisements are combined into a similar advertisement set. The advertisement diversity score is calculated based on the calculated similarity, the re-matched advertisements, the matched interest point product advertisements, and the similar advertisement set. The diversity degree of the similar advertisement set is judged using the calculated advertisement diversity score.

6. The advertising and marketing system based on data source location according to claim 5, wherein: The specific judgment situations in the diverse delivery unit (3) include: Situation ①: When the calculated advertisement diversity score is greater than the set standard advertisement diversity score, it is determined that the greater the difference between the advertisement elements in the similar advertisement set, the better the diversity. Then, the advertisements in the similar advertisement set are delivered to the user terminal. Situation ②: When the calculated advertisement diversity score is less than the set standard advertisement diversity score, it is determined that the difference between the advertisement elements in the similar advertisement set is smaller.

7. The advertising and marketing system based on data source positioning according to claim 5, wherein: The algorithm formula for calculating the advertisement diversity score in the diverse delivery unit (3): 。 8. The advertising and marketing system based on data source location according to claim 5, wherein: The calculation adjustment unit (4) is used to receive the command indicating that the user has developed fatigue with the matched interest point product advertisements in the fatigue module (22). The calculation adjustment unit (4) obtains the interest attenuation frequency from the contact attenuation module (21), the advertisement contact frequency, and the fatigue value of contacting the matched interest point product advertisements from the fatigue module (22). It obtains the frequency of delivering the matched interest point product advertisements from the marketing delivery module (12). The advertisement delivery frequency adjustment factor is calculated based on the interest attenuation frequency, the advertisement contact frequency, and the fatigue value of contacting the matched interest point product advertisements. The calculation of adjusting the advertisement delivery frequency is performed using the frequency of delivering the matched interest point product advertisements and the advertisement delivery frequency adjustment factor. The matched interest point product advertisements are controlled and adjusted according to the calculated adjusted advertisement delivery frequency.

Images