Scenarized advertisement triggering method, terminal and equipment linked with equipment of Internet of Things

Abstract

The invention relates to a scene advertisement triggering method, terminal and equipment linked with Internet of Things equipment, and the method comprises the steps: obtaining terminal perception data associated with a user in real time; the terminal sensing data are input into a preset scene recognition model for feature correlation analysis, the scene recognition model comprises a scene mapping rule base, and a three-dimensional digital label of a physical scene is output; matching an advertisement content database according to the three-dimensional digital label, and obtaining an associated advertisement material set; and according to the associated advertisement material set, generating and pushing adaptive scene advertisement content based on user terminal equipment attributes so as to achieve the purposes of deep binding of the advertisement content and a real-time scene and cross-terminal equipment adaptive accurate pushing.

Description

Technical Field

[0001] The present invention relates to the fields of Internet of Things technology and smart advertising technology, and in particular to a scenario-based advertising triggering method, terminal and device linked to Internet of Things devices. Background Art

[0002] Against the backdrop of the rapid development of IoT technology, the popularity of smart homes and wearable devices has made digital perception of physical scenes possible. However, existing advertising push technologies still rely mainly on static analysis of user portraits or status feedback of a single device, lacking in-depth understanding of multi-device collaborative scenarios and real-time response capabilities. Traditional advertising systems often match content based on fixed rules or historical behaviors, making it difficult to capture the immediate changes in user needs in dynamic scenarios, resulting in a disconnect between advertising content and physical scenes. In addition, existing solutions have technical gaps in cross-device data fusion, scene label generation, and advertising format adaptation, making it impossible to achieve closed-loop optimization from scene perception to content triggering. With the increasing demand for user privacy protection, how to achieve accurate advertising triggering under non-invasive data collection has become a technical problem that the industry urgently needs to solve. Summary of the Invention

[0003] The main purpose of the present invention is to provide a scenario-based advertising triggering method, terminal and device linked to IoT devices. By real-time integration of IoT device status data and user behavior characteristics, a dynamic digital mapping of the physical scene is constructed to achieve deep binding of advertising content with real-time scenes and adaptive and precise push across terminal devices.

[0004] To achieve the above objectives, the present invention provides a scenario-based advertising triggering method for IoT device linkage, comprising the following steps:

[0005] Obtain user-associated terminal perception data in real time;

[0006] Inputting the terminal perception data into a preset scene recognition model for feature association analysis, wherein the scene recognition model includes a scene mapping rule library and outputs a three-dimensional digital label of the physical scene;

[0007] Matching an advertisement content database according to the three-dimensional digital tag to obtain a collection of associated advertisement materials;

[0008] According to the associated advertising material set, adaptive scenario-based advertising content is generated and pushed based on the user terminal device attributes.

[0009] Furthermore, the step of obtaining the terminal perception data associated with the user in real time includes:

[0010] Collect user-associated terminal perception data based on the IoT device communication interface, including status collaboration data of smart home device groups and multi-source sensor data of wearable devices;

[0011] Among them, the state collaborative data obtains device operation mode parameters and environmental perception data in real time through the inter-device communication protocol, and the multi-source sensor data includes a set of user motion trajectory characteristics and physiological indicator fluctuation characteristics.

[0012] Furthermore, the step of inputting the terminal perception data into a preset scene recognition model for feature association analysis and outputting a three-dimensional digital label of the physical scene includes:

[0013] Extracting device state combination features from terminal perception data, including spatiotemporal correlation features between smart home device operating mode parameters and environmental perception data, and temporal matching features between wearable device motion trajectory features and physiological indicator fluctuation features;

[0014] The combined features are pattern matched with a pre-built scene mapping rule library to generate a three-dimensional digital label including a timestamp, spatial coordinates, and user behavior pattern encoding.

[0015] Furthermore, the steps of constructing the scene recognition model include:

[0016] Collect operating mode parameters, environmental perception data, and wearable device sensor data of smart home device groups in historical data sets;

[0017] The mapping relationship between device state combination features and physical scene labels is trained through a supervised learning algorithm to generate a scene mapping rule library;

[0018] Based on the spatiotemporal correlation features in the user behavior data, the matching weight distribution of the scene mapping rule library is dynamically optimized.

[0019] Furthermore, the step of matching the three-dimensional digital tag with the advertisement content database to obtain a set of associated advertisement materials includes:

[0020] Activate the multi-level index structure of the advertising database based on the spatiotemporal dimension features in the three-dimensional digital labels;

[0021] Filter the emotional attributes of advertising materials based on user behavior pattern coding;

[0022] The final output associated advertising creative set is determined by the preset association strength threshold in the scene-advertising mapping graph.

[0023] Furthermore, the step of generating and pushing adapted scenario-based advertising content based on the associated advertising material set and user terminal device attributes includes:

[0024] Extracting graphic and text materials, audio and video materials, and interactive control templates that match the three-dimensional scene label from the associated advertising material set;

[0025] Dynamically blending the graphic and text materials with the audio and video materials in proportion to generate initial advertising content with appropriate proportions of media elements based on the display interface size and audio output module parameters of the user terminal device;

[0026] Embedding an interactive control in the initial advertisement content based on the user behavior pattern encoding in the three-dimensional digital label, wherein the triggering logic of the interactive control is synchronized with the user's current device operation mode;

[0027] When the mobile state characteristics of the user terminal device are detected, the audio and video materials are converted into voice broadcast control instructions and the rendering of graphic materials is suppressed to generate a voice-priority advertising content package;

[0028] The advertising content package is pushed to the user terminal device to complete the adaptation and push of the scenario-based advertising content.

[0029] Furthermore, after the step of generating and pushing adapted scenario-based advertising content based on user terminal device attributes, the method further includes:

[0030] Monitor user interaction behavior data on scenario-based advertising content, extracting indicators such as ad exposure duration, control triggering frequency, and voice command response rate;

[0031] Calculating a relevance score between the advertisement content and the physical scene based on the indicators;

[0032] Dynamically adjust the association strength threshold in the scene-advertising mapping graph based on the association score.

[0033] The present invention also provides a scenario-based advertising triggering terminal linked to an Internet of Things device, comprising:

[0034] Data acquisition module, used to obtain user-associated terminal perception data in real time;

[0035] A scene recognition module is used to input the terminal perception data into a preset scene recognition model for feature association analysis and output a three-dimensional digital label of the physical scene;

[0036] An advertisement matching module, configured to match an advertisement content database according to the three-dimensional digital tags to obtain a set of associated advertisement materials;

[0037] The content output module is used to generate and push adapted scenario-based advertising content based on the associated advertising material set and user terminal device attributes.

[0038] The present invention also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the steps of the scenario-based advertising triggering method for linking the above-mentioned Internet of Things devices are implemented.

[0039] The present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the scenario-based advertising triggering method for the linkage of the above-mentioned Internet of Things devices.

[0040] The scenario-based advertising triggering method, terminal and device for the linkage of IoT devices provided by the present invention have the following beneficial effects: the present invention constructs a three-dimensional digital mapping system of physical scenes through the collaborative analysis of the status of IoT device groups and the biometric perception of wearable devices, thereby realizing the deep binding of advertising triggering with real-time scenes. The scene recognition model based on the fusion of spatiotemporal-behavioral multi-dimensional features improves the accuracy of scene classification in complex environments, so that the advertising content is highly consistent with the context of the scene in which the user is located. Through dynamic proportional fusion and interactive logic synchronization technology, the adaptive presentation of advertising content between different terminal devices is ensured, and the multimodal interactive experience is optimized. In addition, the introduction of a feedback-driven scenario-advertising association strength optimization mechanism enables the system to continuously adapt to the evolution of user behavior patterns and form a self-evolving advertising push strategy. Compared with traditional solutions, the present invention has achieved a qualitative improvement in advertising relevance, real-time responsiveness and user acceptance, providing reliable technical support for precision marketing in the IoT environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 This is a flow chart of a scenario-based advertising triggering method for IoT device linkage according to an embodiment of the present invention;

[0042] Figure 2 This is a structural block diagram of a scenario-based advertising triggering terminal linked to an IoT device in one embodiment of the present invention;

[0043] Figure 3 It is a schematic block diagram of the structure of a computer device according to an embodiment of the present invention.

[0044] The purpose, features and advantages of the present invention will be further described with reference to the accompanying drawings and in conjunction with the embodiments. DETAILED DESCRIPTION

[0045] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0046] Reference Figure 1, which is a flow chart of a scenario-based advertising triggering method for IoT device linkage proposed by the present invention, comprising the following steps:

[0047] S1, real-time acquisition of user-associated terminal perception data;

[0048] S2, inputting the terminal perception data into a preset scene recognition model for feature association analysis, wherein the scene recognition model includes a scene mapping rule library and outputs a three-dimensional digital label of the physical scene;

[0049] S3, matching the advertising content database according to the three-dimensional digital label to obtain a set of related advertising materials;

[0050] S4: Generate and push scenario-specific advertising content based on the associated advertising material set and user terminal device attributes.

[0051] In one embodiment, for step S1,

[0052] The steps for obtaining user-associated terminal perception data in real time include:

[0053] Collect user-associated terminal perception data based on the IoT device communication interface, including status collaboration data of smart home device groups and multi-source sensor data of wearable devices;

[0054] Among them, the state collaborative data obtains device operation mode parameters and environmental perception data in real time through the inter-device communication protocol, and the multi-source sensor data includes a set of user motion trajectory characteristics and physiological indicator fluctuation characteristics.

[0055] In the specific implementation, a multi-source data collection channel is constructed through the IoT device communication interface (such as MQTT / CoAP protocol), in which the status collaborative data of the smart home device group is collected in real time using the inter-device communication protocol (such as ZigBee 3.0), and the device operation mode parameters are defined as a set M d ={m1,m2,...,m n}, where m i The operating parameters of the i-th type of equipment (such as air conditioning cooling power m1∈[0,3500]W, lighting color temperature m2∈[2700,6500]W) and environmental perception data are modeled as a triplet E e =(T,H,L), where T is temperature, H is humidity, and L is light intensity. Wearable devices transmit multi-source sensor data via Bluetooth 5.2, including: accelerometer trajectory where a i is the three-axis acceleration component, GPS position sequence G p ={(x1, y1),(x2, y2),...} represents the longitude and latitude coordinates of the user's movement trajectory, heart rate fluctuation Hr =|Δh / Δt|, where Δh is the heart rate change and Δt is the sampling interval.

[0056] The performance comparison of multi-source data acquisition is as follows:

[0057]

[0058] Experimental data shows that after using Kalman filtering to suppress sensor noise, the environmental data error rate is ≤1.2%, and the signal-to-noise ratio of wearable device data is improved to 28dB, significantly better than traditional threshold filtering solutions. The cross-protocol data fusion delay is 120±15ms, and the data fusion delay (wearable device) is 80±8ms, meeting the timeliness requirements of real-time scene triggering. This embodiment obtains terminal perception data through the real-time collection mechanism of the IoT communication protocol stack, achieving collaborative acquisition of multi-source heterogeneous data, and solving the problem of scene perception lag caused by device data silos in traditional advertising push.

[0059] In one embodiment, for step S2,

[0060] The step of inputting the terminal perception data into a preset scene recognition model for feature association analysis and outputting a three-dimensional digital label of the physical scene includes:

[0061] Extracting device state combination features from terminal perception data, including spatiotemporal correlation features between smart home device operating mode parameters and environmental perception data, and temporal matching features between wearable device motion trajectory features and physiological indicator fluctuation features;

[0062] The combined features are pattern matched with a pre-built scene mapping rule library to generate a three-dimensional digital label including a timestamp, spatial coordinates, and user behavior pattern encoding.

[0063] In the specific implementation, the device state combination features are extracted from the terminal perception data collected in step S1. For the smart home device group, the spatiotemporal correlation features are defined as:

[0064]

[0065] Where m i ∈M d Indicates equipment operating parameters (such as air conditioning power, lighting color temperature), E e =(T,H,L) is the environmental perception data (temperature, humidity, light), a i ∈[0,1] is the device weight coefficient optimized by supervised learning; for wearable devices, the timing matching characteristics are defined as:

[0066]

[0067] Where, is the acceleration trajectory modulus (a i are the three-axis acceleration components), is the heart rate fluctuation rate (Δh is the heart rate change, Δt is the sampling interval), and the integration interval [t0, t1] is determined by the start and end time of the user's behavior pattern. s with C t Input the scene mapping rule base for pattern matching and calculate the comprehensive matching degree M = λ·C s +(1+λ)·C t (λ∈[0,1]), when M≥θ (experimental calibration threshold θ=0.72), label generation is triggered and the three-dimensional digital label L is output. 3D =(t,P,B), where t is the timestamp, P = (X,Y) is the Gauss-Krüger projection space coordinate, and B is the user behavior pattern feature value compressed using Huffman coding. This implemented scene recognition model addresses the ad triggering bias caused by single-dimensional labels in traditional scene recognition by fusing the spatiotemporal correlation features of multi-source data with the temporal sequence features of behavior.

[0068] In one embodiment, the step of constructing a scene recognition model includes:

[0069] Collect operating mode parameters, environmental perception data, and wearable device sensor data of smart home device groups in historical data sets;

[0070] The mapping relationship between device state combination features and physical scene labels is trained through a supervised learning algorithm to generate a scene mapping rule library;

[0071] Based on the spatiotemporal correlation features in the user behavior data, the matching weight distribution of the scene mapping rule library is dynamically optimized.

[0072] Specifically, build a historical training dataset covering multiple device states

[0073]

[0074] Where, The operating parameters of the kth group of smart home devices (such as air conditioning cooling power Humidifier humidity set value For environmental perception data (temperature T (k) ∈[-10,40]℃, humidity H (k) ∈[0,100]%RH, light intensity L (k) ∈[0,2000]lx), For wearable device sensor data (acceleration trajectory modulus, GPS position sequence, heart rate fluctuation rate), is the manually labeled physical scene category label (such as 1 for cooking scene, 2 for fitness scene, and C for the total number of scenes). train Perform supervised learning, and the training goal is to minimize the scene recognition loss function:

[0075]

[0076] Where, is the spatiotemporal correlation feature, is the temporal matching feature, P(c|C s ,C t ) is the posterior probability of scene c. The initial scene mapping rule base is generated by feature importance evaluation The rule item r j Formalized as a logical expression (such as "IF C s >0.6AND C t <1.2THENL scene =2"), initial weight w j Output by random forest. Further based on user behavior feedback data ( For three-dimensional digital labeling, I (m) = {exposure duration, click-through rate, conversion rate} are advertising interaction indicators) Dynamically optimize rule weights and update the formula:

[0077]

[0078] Where, the learning rate η = 0.01, the partial derivative The scene recognition model construction method implemented in this paper achieves accurate modeling and adaptive updating of physical scenes through hierarchical data training and dynamic weight optimization mechanism.

[0079] In one embodiment, for step S3,

[0080] The step of matching the advertisement content database according to the three-dimensional digital label to obtain a set of associated advertisement materials includes:

[0081] Activate the multi-level index structure of the advertising database based on the spatiotemporal dimension features in the three-dimensional digital labels;

[0082] Filter the emotional attributes of advertising materials based on user behavior pattern coding;

[0083] The final output associated advertising creative set is determined by the preset association strength threshold in the scene-advertising mapping graph.

[0084] In the specific implementation, based on the three-dimensional digital label L 3D=(t,P,B) The spatiotemporal dimension features (timestamp t and spatial coordinate P) in activate the multi-level index structure of the advertising database:

[0085] Index(t,P)=TimeTree(t)×GeoHash(P)

[0086] Where TimeTree(t) is the time dimension B+ tree index (time granularity Δ(t) = 1 hour, GeoHash(P) encodes the coordinates (X, Y) into a 7-bit Geohash string (accuracy ±1.2 meters). Based on the user behavior pattern code B (Huffman coding value), the user behavior category (such as fitness B = 0010, movie watching B = 1101) is analyzed, and the pre-trained emotion matching model EmoScore (A c ,B)=σ(W·[Embed(A c ); Embed(B)]), where A c is the emotional label of the advertising material (such as positive Emo = 0.8, neutral Emo = 0.5), Embed(·) is the word vector embedding function, W is the weight matrix, and σ is the Sigmoid activation function. The emotional tendency attribute of the advertising material is filtered by EmoScore (A c ,B)≥β(β=0.65calibrated by AB test). Define the advertising material A in the scene-advertising mapping diagram c With scene label L scene The strength of association:

[0087]

[0088] Where, Represents scenes in historical data The set of clicked ads, K is the number of scene samples. The final output is the associated advertising material set A out ={A c |S(A c ,L scene )≥θ}, with a retention threshold of θ=0.4 (determined by ROC curve analysis, AUC=0.87). This implementation achieves precise binding of advertising content and physical scenes through spatiotemporal-behavioral collaborative indexing and dynamic association strength evaluation.

[0089] In one embodiment, for step S4,

[0090] The step of generating and pushing adapted scenario-based advertising content based on the associated advertising material set and user terminal device attributes includes:

[0091] Extracting graphic and text materials, audio and video materials, and interactive control templates that match the three-dimensional scene label from the associated advertising material set;

[0092] Dynamically blending the graphic and text materials with the audio and video materials in proportion to generate initial advertising content with appropriate proportions of media elements based on the display interface size and audio output module parameters of the user terminal device;

[0093] Embedding an interactive control in the initial advertisement content based on the user behavior pattern encoding in the three-dimensional digital label, wherein the triggering logic of the interactive control is synchronized with the user's current device operation mode;

[0094] When the mobile state characteristics of the user terminal device are detected, the audio and video materials are converted into voice broadcast control instructions and the rendering of graphic materials is suppressed to generate a voice-priority advertising content package;

[0095] The advertising content package is pushed to the user terminal device to complete the adaptation and push of the scenario-based advertising content.

[0096] In a specific implementation, from the associated advertising creative set A out Extraction and 3D digitization of labels L 3D Matching material elements, resolution set of graphic materials R img ={(w i ,h i}, satisfying the device screen size constraint w i ≤W device ,h i ≤H device , the audio sampling rate f of the audio and video material audio ∈[44.1,192]kHz, video frame rate f audio =30fps, touch area coordinate set C of interactive control template touch ={(x j ,y j ,r j )},r j is the response radius. The initial advertising content is generated through the dynamic proportional fusion algorithm:

[0097] Q layout =arg maxQ(α·Readability(Q)+(1-α)·AudioVisual Score(Q))

[0098] Where Q is the candidate layout solution, Readability calculates the readability score based on font size and contrast, AudioVisualScore evaluates the audio and video synchronization quality, and α = 0.7 (calibrated by eye tracking experiments). Interactive controls are embedded based on the user behavior pattern code B, and the control trigger logic is synchronized with the device operation mode. Trigger delay constraint T response =β·T avg , where Tavg The average interval time between historical user operations (e.g., the click interval T for smart watches) avg =0.8s), β=0.9 is the tolerance coefficient; gesture mapping rule, when B is decoded as a fitness scene, the acceleration trajectory feature A t >2.5m / s 2 Trigger the control response to avoid accidental touch. In the mobile state detection (wearable device acceleration variance Var (A t )>1.2m 2 / s 4 ), perform voice-first conversion:

[0099]

[0100] Where TTS is the text-to-speech engine (generation rate 1.2× real-time, This indicates audio stream mixing, while disabling graphic rendering to reduce GPU load. Finally, the advertising content package is pushed through the edge node, and the transmission protocol is dynamically selected based on the network status (latency < 100ms in a 5G environment). This embodiment achieves deep adaptation of advertising formats to user devices and behavior status through multimodal material fusion and interactive logic synchronization technology.

[0101] In one embodiment, after the step of generating and pushing adapted scenario-based advertising content based on user terminal device attributes, the method further includes:

[0102] Monitor user interaction behavior data on scenario-based advertising content, extracting indicators such as ad exposure duration, control triggering frequency, and voice command response rate;

[0103] Calculating a relevance score between the advertisement content and the physical scene based on the indicators;

[0104] Dynamically adjust the association strength threshold in the scene-advertising mapping graph based on the association score.

[0105] In the specific implementation, after the advertising content is pushed, the user interaction behavior data stream D is monitored in real time. feedback ={(τ exp ,f click ,ρ voice )}, where τ exp ∈[0,T max ] is the advertising exposure time (T max =30s, f click is the control trigger frequency, ρ voice The response rate to voice commands. The matching degree between advertisement and scene is quantified by the relevance scoring model:

[0106]

[0107] Where, w1=0.5, w2=0.3, w3=0.2, τ norm =τ exp / T max When S rel When θ<0.6, the association strength threshold θ in the scene-advertisement mapping graph is dynamically lowered. new =θ old -γ·(0.6-S rel ), and the learning rate γ = 0.05, to ensure that low-relevance ads are gradually eliminated. This embodiment implements adaptive iteration of advertising strategies through a relevance scoring model.

[0108] Reference Figure 2 , is a structural block diagram of a scenario-based advertising triggering terminal linked to an IoT device in one embodiment of the present invention, including:

[0109] Data acquisition module, used to obtain user-associated terminal perception data in real time;

[0110] A scene recognition module is used to input the terminal perception data into a preset scene recognition model for feature association analysis and output a three-dimensional digital label of the physical scene;

[0111] An advertisement matching module, configured to match an advertisement content database according to the three-dimensional digital tags to obtain a set of associated advertisement materials;

[0112] The content output module is used to generate and push adapted scenario-based advertising content based on the associated advertising material set and user terminal device attributes.

[0113] For the specific implementation of each module in the above device example, please refer to the above method embodiment, which will not be repeated here.

[0114] Reference Figure 3 In an embodiment of the present invention, a computer device is also provided. The computer device may be a server, and its internal structure may be as follows: Figure 3 As shown. The computer device includes a processor, memory, display screen, input device, network interface and database connected via a system bus. The processor of the computer design is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and computer program in the non-volatile storage medium. The database of the computer device is used to store the corresponding data in this embodiment. The network interface of the computer device is used to communicate with an external terminal via a network connection. When the computer program is executed by the processor, the above method is implemented.

[0115] Those skilled in the art will understand that Figure 3 The structure shown in the figure is merely a block diagram of a portion of the structure related to the solution of the present invention and does not constitute a limitation on the computer device to which the solution of the present invention is applied.

[0116] An embodiment of the present invention further provides a computer-readable storage medium having a computer program stored thereon, which implements the above-described method when executed by a processor. It is understood that the computer-readable storage medium in this embodiment can be a volatile readable storage medium or a non-volatile readable storage medium.

[0117] To summarize, the present invention obtains user-associated terminal perception data in real time; inputs the terminal perception data into a preset scene recognition model for feature association analysis, wherein the scene recognition model includes a scene mapping rule library and outputs a three-dimensional digital label of the physical scene; matches the advertising content database according to the three-dimensional digital label to obtain a set of associated advertising materials; based on the associated advertising material set, generates and pushes adapted scenario-based advertising content based on the user terminal device attributes, so as to achieve deep binding of advertising content with real-time scenes and the purpose of adaptive and precise push across terminal devices.

[0118] Those skilled in the art will appreciate that all or part of the processes in the above-described method embodiments can be implemented by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the above-described method embodiments. Among them, any reference to memory, storage, database, or other media provided by the present invention and used in the embodiments can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double-speed SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct RAM bus dynamic RAM (DRDRAM), and RAMbus dynamic RAM.

[0119] It should be noted that, in this document, the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, apparatus, article, or method comprising a series of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, apparatus, article, or method. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the process, apparatus, article, or method comprising the element.

[0120] The above description is only a preferred embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made by using the contents of the present invention description and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A scenario-based advertising triggering method for IoT device linkage, characterized in that: The following steps are involved: Obtain user-associated terminal perception data in real time; Inputting the terminal perception data into a preset scene recognition model for feature association analysis, wherein the scene recognition model includes a scene mapping rule library and outputs a three-dimensional digital label of the physical scene; Matching an advertisement content database according to the three-dimensional digital tag to obtain a collection of associated advertisement materials; According to the associated advertising material set, adaptive scenario-based advertising content is generated and pushed based on the user terminal device attributes.

2. The scenario-based advertising triggering method for IoT device linkage according to claim 1 is characterized in that: The step of acquiring user-associated terminal perception data in real time includes: Collect user-associated terminal perception data based on the IoT device communication interface, including status collaboration data of smart home device groups and multi-source sensor data of wearable devices; Among them, the state collaborative data obtains device operation mode parameters and environmental perception data in real time through the inter-device communication protocol, and the multi-source sensor data includes a set of user motion trajectory characteristics and physiological indicator fluctuation characteristics.

3. The scenario-based advertising triggering method for IoT device linkage according to claim 1 is characterized in that: The step of inputting the terminal perception data into a preset scene recognition model for feature association analysis and outputting a three-dimensional digital label of the physical scene includes: Extracting device state combination features from terminal perception data, including spatiotemporal correlation features between smart home device operating mode parameters and environmental perception data, and temporal matching features between wearable device motion trajectory features and physiological indicator fluctuation features; The combined features are pattern matched with a pre-built scene mapping rule library to generate a three-dimensional digital label including a timestamp, spatial coordinates, and user behavior pattern encoding.

4. The scenario-based advertising triggering method for IoT device linkage according to claim 3 is characterized in that: The steps of constructing the scene recognition model include: Collect operating mode parameters, environmental perception data, and wearable device sensor data of smart home device groups in historical data sets; The mapping relationship between device state combination features and physical scene labels is trained through a supervised learning algorithm to generate a scene mapping rule library; Based on the spatiotemporal correlation features in the user behavior data, the matching weight distribution of the scene mapping rule library is dynamically optimized.

5. The scenario-based advertising triggering method for IoT device linkage according to claim 1, characterized in that: The step of matching the advertising content database according to the three-dimensional digital label to obtain a set of associated advertising materials includes: Activate the multi-level index structure of the advertising database based on the spatiotemporal dimension features in the three-dimensional digital labels; Filter the emotional attributes of advertising materials based on user behavior pattern coding; The final output associated advertising creative set is determined by the preset association strength threshold in the scene-advertising mapping graph.

6. The scenario-based advertising triggering method for IoT device linkage according to claim 1, characterized in that: The step of generating and pushing adapted scenario-based advertising content based on the associated advertising material set and user terminal device attributes includes: Extracting graphic and text materials, audio and video materials, and interactive control templates that match the three-dimensional scene label from the associated advertising material set; Dynamically blending the graphic and text materials with the audio and video materials in proportion to generate initial advertising content with appropriate proportions of media elements based on the display interface size and audio output module parameters of the user terminal device; Embedding an interactive control in the initial advertisement content based on the user behavior pattern encoding in the three-dimensional digital label, wherein the triggering logic of the interactive control is synchronized with the user's current device operation mode; When the mobile state characteristics of the user terminal device are detected, the audio and video materials are converted into voice broadcast control instructions and the rendering of graphic materials is suppressed to generate a voice-priority advertising content package; The advertising content package is pushed to the user terminal device to complete the adaptation and push of the scenario-based advertising content.

7. The scenario-based advertising triggering method for IoT device linkage according to claim 1, characterized in that: After the step of generating and pushing adapted scenario-based advertising content based on user terminal device attributes, the method further includes: Monitor user interaction behavior data on scenario-based advertising content, extracting indicators such as ad exposure duration, control triggering frequency, and voice command response rate; Calculating a relevance score between the advertisement content and the physical scene based on the indicators; Dynamically adjust the association strength threshold in the scene-advertising mapping graph based on the association score.

8. A scenario-based advertising triggering terminal linked to IoT devices, characterized in that: include: Data acquisition module, used to obtain user-associated terminal perception data in real time; A scene recognition module is used to input the terminal perception data into a preset scene recognition model for feature association analysis and output a three-dimensional digital label of the physical scene; An advertisement matching module, configured to match an advertisement content database according to the three-dimensional digital tags to obtain a set of associated advertisement materials; The content output module is used to generate and push adapted scenario-based advertising content based on the associated advertising material set and user terminal device attributes.

9. A computer device comprising a memory and a processor, wherein a computer program is stored in the memory, wherein: When the processor executes the computer program, the steps of the scenario-based advertising triggering method for linking IoT devices as described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the scenario-based advertising triggering method for linking IoT devices according to any one of claims 1 to 7 are implemented.

Images