A precise advertising push method based on smart water dispenser

Abstract

The invention relates to the technical field of drinking water equipment, and discloses a precise advertising push method based on an intelligent water dispenser, the method comprising the following steps: step 1, displaying a promotional video through an interactive advertising machine, and during the playback, counting the gaze duration of pedestrians, and a data acquisition module collecting an eye movement data set; step 2, whenever the playback of the promotional video ends, the interactive advertising machine guides pedestrians to click on the screen for interaction, and the data acquisition module collects an interaction data set; step 3, the data acquisition module collects a material data set; step 4, a precise push module analyzes and identifies potential users with a long gaze duration, and the intelligent interactive user portrait has high accuracy; step 5, a precise push module analyzes and generates intelligent water dispenser videos that potential users are interested in; step 6, a precise push module generates a push index; step 7, a precise push module controls the interactive advertising machine to switch the corresponding playback content, and the personalized push marketing effect is better.

Description

Technical Field

[0001] The present invention relates to the technical field of drinking water equipment, and in particular to a method for accurately pushing advertisements based on a smart drinking water machine. Background Art

[0002] Smart water dispensers are household appliances that combine modern water purification technology with intelligent control systems to provide users with a safe, healthy and convenient drinking experience. Smart water dispensers are usually equipped with a multi-stage filtration system, including PP cotton, activated carbon, reverse osmosis membrane or ultrafiltration membrane, which can effectively remove impurities, bacteria, viruses, heavy metals, residual chlorine and other harmful substances in the water. Users can get clear and good-tasting drinking water with simple operations. The built-in temperature control function of smart water dispensers allows users to easily choose cold water, hot water or water at a specific temperature according to personal preferences or beverage needs. This function is especially suitable for families who need to brew coffee, tea or quickly make baby formula. Smart water dispensers are equipped with a variety of sensors to monitor water quality indicators such as TDS and pH in real time to ensure that the water quality always meets health standards. When water quality problems are detected or the filter needs to be replaced, smart water dispensers can also automatically notify users through mobile phone apps to ensure safe drinking water. Smart water dispensers support Wi-Fi or Bluetooth connections and can be seamlessly connected to smart home systems. Users can remotely control the water dispenser through a smartphone app to achieve intelligent operations such as timer switching, temperature adjustment, and water volume monitoring. Some high-end models even integrate voice control functions and are compatible with smart assistants such as Amazon Alexa and Google Assistant. As consumers pursue a healthier quality of life, the market prospects of smart water dispensers will continue to be promising.

[0003] At present, the traditional push advertising method based on smart water dispensers is a one-way information transmission and lacks interaction with consumers. As a high-tech product, the ease of use, connectivity, and intelligent control of smart water dispensers require users to have an immersive experience to fully understand and appreciate them. Facing different consumer groups, the accuracy of user portraits needs to be improved. Summary of the invention

[0004] 1. Technical issues to be resolved

[0005] In view of the shortcomings of the prior art, the present invention provides a method for accurately pushing advertisements based on smart water dispensers, which has the advantages of high accuracy of intelligent interactive user portraits and better personalized push marketing effects. It solves the problem that traditional methods for pushing advertisements based on smart water dispensers lack interactive means and low accuracy of user portraits.

[0006] (II) Technical solution

[0007] To achieve the above-mentioned purpose, the present invention provides the following technical solutions: a method for accurately pushing advertisements based on a smart water dispenser, comprising a central control console, an interactive advertising machine and an eye tracking device, wherein the central control console is connected to the interactive advertising machine via a network, and a screen and an eye tracking device are fixedly installed on the front of the interactive advertising machine. The method for accurately pushing advertisements based on a smart water dispenser comprises the following steps:

[0008] Step 1: Display the promotional video P of the smart water dispenser through an interactive advertising machine, and during the playback of the promotional video P, use an eye tracking device to count the gaze duration of pedestrians. The data acquisition module connects to the eye tracking device through the network and forms an eye movement data set with the eye movement feature data;

[0009] Step 2: Whenever the promotional video P ends, the interactive advertising machine switches to the traffic diversion interface to guide pedestrians to click on the screen for interaction. The data collection module connects to the interactive advertising machine through the network and forms an interactive data set with the interactive data;

[0010] Step 3: The data acquisition module is connected to the central control console through the network, and the video materials related to the smart water dispenser are combined into a material data set;

[0011] Step 4: The precise push module generates the primary user Cjyh based on the eye movement data set analysis, and obtains potential users with longer gaze duration;

[0012] Step 5: The precise push module analyzes and generates a personalized push data set Gxts based on the interaction data set and the material data set, and obtains the smart water dispenser video that the potential user is interested in;

[0013] Step 6: The precise push module is set with a fixed observation period G, and analyzes and generates a push index Tszs to observe the advertising push effect within the period G;

[0014] Step 7: The precise push module controls the interactive advertising machine to switch the corresponding playback content according to the personalized push data group Gxts and the push index Tszs.

[0015] Preferably, in step 1, the expression of the eye movement data set is {Y1 s 、Y2 s 、Y3 s 、...、Yn s}, Y1 s To Yn s The pedestrian eye movement feature data correspond to each time the interactive advertising machine plays the promotional video P, s represents the pedestrian's gaze duration, and 1 to n represent that there are n groups of pedestrian eye movement feature data.

[0016] Preferably, in step 2, the traffic diversion interface is provided with multiple dialog boxes, and each dialog box corresponds to a type of consumer group label, and the expression of the interactive data set is {J1 b 、J2 b 、J3 b ,...,Jh b}, J1 b To Jh b It corresponds to the interactive data each time the pedestrian clicks on the interactive advertising machine screen, b represents the consumer group label corresponding to the dialog box, and 1 to h represent that there are h groups of pedestrian interaction data.

[0017] Preferably, in step 3, the expression of the material data set is {S1 l 、S2 l 、S3 l ,...,Sc l},S1 l To Sc l They correspond in sequence to the smart water dispenser video materials stored and backed up in the central control console, l represents the content type of the video material, and 1 to c represent that there are c groups of smart water dispenser video materials stored and backed up.

[0018] Preferably, in step 4, the calculation formula of the primary user Cjyh is as follows:

[0019] Cjyh=∑ n (Yi s -BZ)

[0020] In the formula, Cjyh represents primary users, Yi s represents the eye movement feature data of the i-th group of pedestrians in the eye movement dataset, BZ represents the fixed parameter in the potential user calculation formula to measure whether the pedestrian's gaze duration meets the standard, that is, the standard gaze duration, ∑ n (Yi s -BZ) means calculating the difference between n groups of pedestrian eye movement feature data and fixed parameters at the same time. If the difference is a positive number, it means that the gaze duration of this group of pedestrians has reached the standard, that is, they are primary users.

[0021] Preferably, in step 5, the calculation formula of the personalized push data group Gxts is as follows:

[0022] YH=∑ h (Ciyh+Jk b )

[0023] s=∑ c (S1 l +S2 l +S3 l +...+Sc l )

[0024]

[0025] In the formula, Gxts represents the personalized push data group, ∑ h (Cjyh+Jk b ) means filtering the corresponding interaction data in the interaction dataset according to the primary user Cjyh, obtaining the consumer group label corresponding to the primary user Cjyh, and marking it as YH, ∑ c (S1 l +S2 l +S3 l +...+Sc l ) represents all the smart water dispenser video materials in the comprehensive material dataset and is marked as S. It means matching the video materials of corresponding content types in the smart water dispenser video material S according to the primary user consumer group label YH, and combining them into a personalized push data group.

[0026] Preferably, in step 6, the push index Tszs is calculated as follows:

[0027]

[0028] In the formula, Tszs represents the push index, Y represents the standby time when the interactive advertising machine switches to the drainage interface, and P+Y represents the sum of the standby time of the smart water dispenser promotion video and the drainage interface, that is, the total push time of a single playback round. The observation period G is divided by the total push duration of a single play round, and the number of rounds LB is obtained. G (Y1 s +Y2 s +Y3 s +...+Yn s ) represents the total time that pedestrians look at the interactive advertising machine screen within the statistical observation period G, It represents the ratio of the total time that pedestrians look at the interactive advertising machine screen to the observation period G. The push index of a single playback round is obtained by dividing the ratio of the total time that pedestrians look at the interactive advertising machine screen to the observation period by the number of rotations.

[0029] Preferably, in step seven, the precision push module transmits the personalized push data group Gxts to the central control console via the network, and the central control console controls the interactive advertising machine via the network to play the smart water dispenser video that potential users are interested in in the personalized push data group Gxts.

[0030] Preferably, in step seven, the precise push module is provided with an index threshold. If the push index Tszs is lower than the index threshold, it indicates that the advertising push effect of the current promotion video P of the smart water dispenser is poor. The precise push module updates the promotion video P, and the data collection module performs data collection for a single observation period G. The precise push module generates the push index Tszs again until the push index Tszs is higher than or included in the index threshold.

[0031] Preferably, in step seven, the precise push module updates the promotion video P, and the update process is as follows:

[0032] 50% of the content in the promotional video P is retained, and video materials of any content type in the material data set are randomly extracted to replace the remaining 50% of the content in the promotional video P. The content length of the updated promotional video P is consistent with that of the original promotional video P.

[0033] Compared with the prior art, the present invention provides a method for accurately pushing advertisements based on a smart water dispenser, which has the following beneficial effects:

[0034] 1. The present invention displays the promotional video P of the smart water dispenser through an interactive advertising machine, and during the playback of the promotional video P, the gaze duration of pedestrians is counted through an eye tracking device, and the data acquisition module is connected to the eye tracking device through a network to collect eye movement data sets. Whenever the promotional video P ends, the interactive advertising machine switches to the drainage interface to guide pedestrians to click on the screen for interaction. The data acquisition module is connected to the interactive advertising machine through a network to collect the interactive data set. The data acquisition module is connected to the central control console through a network to obtain video materials related to the smart water dispenser, and composes them into a material data set. The precise push module generates a primary user Cjyh based on the eye movement data set analysis, obtains potential users with a longer gaze duration, effectively identifies intended users, and has a high accuracy of intelligent interactive user portraits.

[0035] 2. The present invention uses a precise push module to analyze and generate a personalized push data group Gxts according to the interactive data set and the material data set, and obtains the smart water dispenser video that potential users are interested in. The precise push module is set with an observation period G of a fixed duration, and analyzes and generates a push index Tszs, and observes the advertising push effect within the period G, so as to facilitate the subsequent update of the promotion video P according to the proportion of the total gaze time in a single round and the observation period G, so as to achieve the best advertising push effect. The precise push module controls the interactive advertising machine to switch the corresponding playback content according to the personalized push data group Gxts and the push index Tszs, and the personalized push marketing effect is better. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 It is a step diagram of the method of the present invention. DETAILED DESCRIPTION

[0037] 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.

[0038] Since the traditional push advertising method based on smart water dispensers is a one-way information transmission and lacks interaction with consumers, smart water dispensers are high-tech products whose ease of use, connectivity, and intelligent control require users to have an immersive experience to fully understand and appreciate them. Facing different consumer groups, the accuracy of user portraits needs to be improved. Therefore, a precise push advertising method based on smart water dispensers is provided. Please refer to Figure 1 , including a central console, an interactive advertising machine and an eye tracking device. The central console is connected to the interactive advertising machine through a network. A screen and an eye tracking device are fixedly installed on the front of the interactive advertising machine to realize data interconnection between the central console and the interactive advertising machine. The precise advertising push method based on the smart water dispenser includes the following steps:

[0039] Step 1: Display the promotional video P of the smart water dispenser through an interactive advertising machine. The interactive advertising machine can not only display and promote information, but also interact with users, provide personalized experience and real-time feedback, thereby greatly enhancing the effect and influence of the advertisement. During the playback of the promotional video P, the eye tracking device is used to count the gaze duration of pedestrians. The data acquisition module is connected to the eye tracking device through the network, and the eye movement feature data is formed into an eye movement data set. The expression of the eye movement data set is {Y1 s 、Y2 s 、Y3 s 、...、Yn s}, Y1 s To Yn s The pedestrian eye movement feature data corresponding to each time the interactive advertising machine plays the promotional video P, s represents the pedestrian's gaze duration, 1 to n represent that there are n groups of pedestrian eye movement feature data, and the eye tracking device can accurately locate the pupil position and analyze the person's gaze point, so as to facilitate the subsequent rapid judgment of whether the pedestrian has an interest in or purchase intention for the smart water dispenser promotional video P;

[0040] Step 2. Whenever the promotional video P ends, the interactive advertising machine switches to the traffic diversion interface to guide pedestrians to click on the screen for interaction. The traffic diversion interface is set with multiple dialog boxes. The text content in the dialog box can be set as question sentences, for example, does it take too long to heat the water? Is it too troublesome to adjust the temperature manually? Is there always scale on the home kettle? Each dialog box corresponds to a consumer group label. For example, the dialog box that takes too long to heat the water corresponds to the consumer group label that needs to obtain drinking water conveniently and quickly, and the dialog box that is too troublesome to adjust the temperature manually corresponds to the consumer group label that needs to drink water at a specific temperature. The data acquisition module connects to the interactive advertising machine through the network and combines the interactive data into an interactive data set. The expression of the interactive data set is {J1 b 、J2 b 、J3 b ,...,Jh b}, J1 b To Jh b The interactive data corresponding to each time a pedestrian clicks on the interactive advertising machine screen is shown in sequence. b represents the consumer group label corresponding to the dialog box. The consumer group label includes but is not limited to those with strong health awareness, smart life, family users, and special drinking water needs. 1 to h represent that there are h groups of pedestrian interactive data.

[0041] Step 3: The data acquisition module connects to the central control console through the network and forms a material data set with the video materials related to the smart water dispenser. The expression of the material data set is {S1 l 、S2 l 、S3 l ,...,Sc l},S1 l To Sc l They correspond to the smart water dispenser video materials stored and backed up in the central control console. The smart water dispenser video materials are sliced ​​contents that are different from the promotional video P. The sliced ​​contents are more targeted and promote a certain function or a certain part of the material of the parts of the smart water dispenser. l represents the content type of the video material, that is, the type label corresponding to the sliced ​​content. The type label corresponds to the above consumer group label. For example, the sliced ​​content of the material type of the parts of the smart water dispenser should be pushed to the consumer group with strong health awareness. 1 to c represent that there are c groups of smart water dispenser video materials stored and backed up;

[0042] Step 4: The precise push module generates the primary user Cjyh based on the eye movement data set analysis. The calculation formula is as follows:

[0043] Cjyh=∑ n (Yi s -BZ)

[0044] In the formula, Cjyh represents primary users, Yi srepresents the eye movement feature data of the i-th group of pedestrians in the eye movement dataset, BZ represents the fixed parameter in the potential user calculation formula to measure whether the pedestrian's gaze duration meets the standard, that is, the standard gaze duration, ∑ n (Yi s -BZ) indicates that the difference between n groups of pedestrian eye movement feature data and fixed parameters is calculated simultaneously. If the difference is a positive number, it means that the gaze duration of this group of pedestrians has reached the standard, that is, they are primary users. Potential users with longer gaze duration are obtained, and potential users are effectively identified. The intelligent interactive user portrait is highly accurate.

[0045] Step 5: The precise push module analyzes and generates a personalized push data set Gxts based on the interaction data set and the material data set. The calculation formula is as follows:

[0046] YH=∑ h (ciyh+Jk b )

[0047] S=∑ c (S1 l +S2 l +S3 l +...+Sc l )

[0048]

[0049] In the formula, Gxts represents the personalized push data group, ∑ h (Cjyh+Jk b ) means filtering the corresponding interaction data in the interaction dataset according to the primary user Cjyh, obtaining the consumer group label corresponding to the primary user Cjyh, and marking it as YH, ∑ c (S1 l +S2 l +S3 l +...+Sc l ) represents all the smart water dispenser video materials in the comprehensive material dataset and is marked as S. It means matching the video materials of corresponding content types in the smart water dispenser video material S according to the primary user consumer group label YH, and combining them into a personalized push data group to obtain the smart water dispenser videos that potential users are interested in;

[0050] Step 6: The precise push module sets a fixed-length observation period G, and analyzes and generates a push index Tszs to observe the advertising push effect within the period G. The calculation formula is as follows:

[0051]

[0052] In the formula, Tszs represents the push index, Y represents the standby time when the interactive advertising machine switches to the drainage interface, and P+Y represents the sum of the standby time of the smart water dispenser promotion video and the drainage interface, that is, the total push time of a single playback round. The observation period G is divided by the total push duration of a single play round, and the number of rounds LB is obtained. G (Y1 s +Y2 s +Y3 s +...+Yn s ) represents the total time that pedestrians look at the interactive advertising machine screen within the statistical observation period G, It represents the ratio of the total time that pedestrians look at the interactive advertising machine screen to the observation period G. The digital representation shows the proportion of the total gaze time to the observation period G. The push index of a single play round is obtained by dividing the total time that pedestrians look at the interactive advertising machine screen by the observation period and the number of rounds. This facilitates the subsequent update of the promotion video P according to the ratio of the total gaze time in a single round to the observation period G, so as to achieve the best advertising push effect.

[0053] Step 7. The precise push module controls the interactive advertising machine to switch the corresponding playback content according to the personalized push data group Gxts and the push index Tszs. The precise push module transmits the personalized push data group Gxts to the central control console through the network. The central control console controls the interactive advertising machine through the network to play the smart water dispenser video that potential users are interested in in the personalized push data group Gxts. The personalized push marketing effect is better. The precise push module is set with an index threshold. If the push index Tszs is lower than the index threshold, it means that the advertising push effect of the current promotion video P of the smart water dispenser is poor. The precise push module updates the promotion video P. The update process is as follows:

[0054] 50% of the content in the promotion video P is retained, and video materials of any content type in the material data set are randomly extracted to replace the remaining 50% of the content in the promotion video P. The content length of the updated promotion video P is consistent with that of the original promotion video P. The data collection module then performs data collection for a single observation period G, and the precise push module generates the push index Tszs again until the push index Tszs is higher than or included in the index threshold, and the personalized push marketing effect is better.

[0055] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for accurately pushing advertisements based on a smart water dispenser, comprising a central control console, an interactive advertising machine and an eye tracking device, wherein the central control console is connected to the interactive advertising machine via a network, and a screen and an eye tracking device are fixedly installed on the front of the interactive advertising machine, and the method is characterized in that: The precise advertising push method based on the smart water dispenser includes the following steps: Step 1: Display the promotional video P of the smart water dispenser through an interactive advertising machine, and during the playback of the promotional video P, use an eye tracking device to count the gaze duration of pedestrians. The data acquisition module connects to the eye tracking device through the network and forms an eye movement data set with the eye movement feature data; Step 2: Whenever the promotional video P ends, the interactive advertising machine switches to the traffic diversion interface to guide pedestrians to click on the screen for interaction. The data collection module connects to the interactive advertising machine through the network and forms an interactive data set with the interactive data; Step 3: The data acquisition module is connected to the central control console through the network, and the video materials related to the smart water dispenser are combined into a material data set; Step 4: The precise push module generates the primary user Cjyh based on the eye movement data set analysis, and obtains potential users with longer gaze duration; Step 5: The precise push module analyzes and generates a personalized push data set Gxts based on the interaction data set and the material data set, and obtains the smart water dispenser video that the potential user is interested in; Step 6: The precise push module is set with a fixed observation period G, and analyzes and generates a push index Tszs to observe the advertising push effect within the period G; Step 7: The precise push module controls the interactive advertising machine to switch the corresponding playback content according to the personalized push data group Gxts and the push index Tszs.

2. According to claim 1, a precise advertising push method based on a smart water dispenser is characterized by: In step 1, the expression of the eye movement data set is {Y1 s 、Y2 s 、Y3 s 、...、Yn s }, Y1 S To Yn s The pedestrian eye movement feature data correspond to each time the interactive advertising machine plays the promotional video P, s represents the pedestrian's gaze duration, and 1 to n represent that there are n groups of pedestrian eye movement feature data.

3. According to claim 1, a precise advertising method based on a smart water dispenser is characterized in that: In step 2, the traffic diversion interface is provided with multiple dialog boxes, and each dialog box corresponds to a consumer group label. The expression of the interactive data set is {J1 b 、J2 b 、J3 b ,...,Jh b }, J1 b To Jh b It corresponds to the interactive data each time the pedestrian clicks on the interactive advertising machine screen, b represents the consumer group label corresponding to the dialog box, and 1 to h represent that there are h groups of pedestrian interaction data.

4. According to claim 3, a precise advertising push method based on a smart water dispenser is characterized in that: In step 3, the expression of the material data set is {S1 l 、S2 l 、S3 l ,...,Sc l },S1 l To Sc l They correspond in sequence to the smart water dispenser video materials stored and backed up in the central control console, l represents the content type of the video material, and 1 to c represent that there are c groups of smart water dispenser video materials stored and backed up.

5. According to claim 2, a precise advertising push method based on a smart water dispenser is characterized in that: In step 4, the calculation formula of the primary user Cjyh is as follows: Cjyh=∑ n (Yi s -BZ) In the formula, Cjyh represents primary users, Yi s represents the eye movement feature data of the i-th group of pedestrians in the eye movement dataset, BZ represents the fixed parameter in the potential user calculation formula to measure whether the pedestrian's gaze duration meets the standard, that is, the standard gaze duration, ∑ n (Yi s -BZ) means calculating the difference between n groups of pedestrian eye movement feature data and fixed parameters at the same time. If the difference is a positive number, it means that the gaze duration of this group of pedestrians has reached the standard, that is, they are primary users.

6. According to claim 4, a precise advertising push method based on a smart water dispenser is characterized in that: In step 5, the calculation formula of the personalized push data group Gxts is as follows: YH=∑ h (Cjyh+Jk b , S=∑ c (S1 l +S2 l +S3 l +...+Sc l ) In the formula, Gxts represents the personalized push data group, ∑ h (Cjyh+Jk b ) means filtering the corresponding interaction data in the interaction dataset according to the primary user Cjyh, obtaining the consumer group label corresponding to the primary user Cjyh, and marking it as YH, ∑ c (S1 l +S2 l +S3 l +...+Sc l ) represents all the smart water dispenser video materials in the comprehensive material dataset and is marked as S. It means matching the video materials of corresponding content types in the smart water dispenser video material S according to the primary user consumer group label YH, and combining them into a personalized push data group.

7. According to claim 2, a precise advertising push method based on a smart water dispenser is characterized in that: In step 6, the push index Tszs is calculated as follows: In the formula, Tszs represents the push index, Y represents the standby time when the interactive advertising machine switches to the drainage interface, and P+Y represents the sum of the standby time of the smart water dispenser promotion video and the drainage interface, that is, the total push time of a single playback round. The observation period G is divided by the total push duration of a single play round, and the number of rounds LB is obtained. G (Y1 s +Y2 s +Y3 s +...+Yn s ) represents the total time that pedestrians look at the interactive advertising machine screen within the statistical observation period G, It represents the ratio of the total time that pedestrians look at the interactive advertising machine screen to the observation period G. The push index of a single playback round is obtained by dividing the ratio of the total time that pedestrians look at the interactive advertising machine screen to the observation period by the number of rotations.

8. According to claim 1, a precise advertising push method based on a smart water dispenser is characterized by: In the step seven, the precision push module transmits the personalized push data group Gxts to the central control console through the network, and the central control console controls the interactive advertising machine through the network to play the smart water dispenser video that potential users are interested in in the personalized push data group Gxts.

9. According to claim 1, a precise advertising push method based on a smart water dispenser is characterized by: In the step seven, an index threshold is set for the precise push module. If the push index Tszs is lower than the index threshold, it indicates that the advertising push effect of the current promotion video P of the smart water dispenser is poor. The precise push module updates the promotion video P, and the data collection module performs data collection for a single observation period G. The precise push module generates the push index Tszs again until the push index Tszs is higher than or included in the index threshold.

10. The method for accurately pushing advertisements based on a smart water dispenser according to claim 1, characterized in that: In step 7, the precise push module updates the promotion video P, and the update process is as follows: 50% of the content in the promotional video P is retained, and video materials of any content type in the material data set are randomly extracted to replace the remaining 50% of the content in the promotional video P. The content length of the updated promotional video P is consistent with that of the original promotional video P.

Images