Advertisement putting effect evaluation and analysis system and method based on data analysis

Abstract

The invention belongs to the field of advertisement analysis, and provides an advertisement putting effect evaluation and analysis system and method based on data analysis, and the method comprises the steps: constructing a cognitive state vector of a user based on multi-source behavior data before and after the contact between the user and an advertisement; modeling is conducted on user cognitive state vector changes caused by advertisement putting, cognitive intervention vectors of the advertisements are generated, and the cognitive intervention vectors represent the variable quantity of the cognitive states of the users after the users make contact with the advertisements; based on the language feedback, the visual behavior and the emotion track of the user, evaluating the consistency between the user behavior response and the cognitive intervention vector direction, and outputting a multi-modal behavior consistency score; tracking a behavior path of the user from advertisement click to final conversion, forming a conversion chain feature, and identifying a path length, a path node type and a path response time delay; and fusing the cognitive intervention vector, the multi-modal behavior consistency score and the conversion chain feature, and calculating a cognitive effect comprehensive score of advertisement putting for putting effect evaluation.

Description

Technical Field

[0001] The present invention belongs to the field of advertising analysis, and in particular relates to a system and method for evaluating and analyzing the effect of advertising delivery based on data analysis. Background Art

[0002] With the continuous evolution of digital advertising technology, advertisers' evaluation of advertising effectiveness has shifted from traditional, coarse-grained metrics like impression volume and click-through rate to more sophisticated requirements such as user intent identification, behavior prediction, and precise attribution. Current mainstream ad evaluation methods mostly rely on explicit metrics such as user clicks, page dwell time, and conversions. While these can reflect a certain degree of user response, they struggle to capture the underlying impact of advertising on user cognition.

[0003] Cognitive psychology research shows that users' response to advertising isn't a momentary impulse; rather, it involves a complex psychological process involving information perception, interest stimulation, cognitive processing, intention formation, and ultimately behavioral execution. Judging advertising effectiveness solely based on click behavior or conversion results often fails to reveal the true impact of advertising on users' cognitive systems. This is especially true for promotions of high-value products or services, where users may experience cognitive bias after initial exposure to an ad but not immediately complete the purchase. This potential cognitive intervention is often overlooked in traditional evaluation systems.

[0004] In addition, current advertising evaluation systems generally lack the systematic integration of user multimodal behavioral data (such as voice, expression, interaction mode, etc.), and are unable to achieve comprehensive modeling of the entire process from psychology to behavior to conversion, which in turn limits the optimization space of advertising delivery strategies. Summary of the Invention

[0005] In order to solve the problems in the prior art, the present invention provides an advertising effect evaluation and analysis system based on data analysis, comprising:

[0006] A user cognitive state acquisition module is used to construct a user cognitive state vector based on multi-source behavioral data before and after the user is exposed to the advertisement;

[0007] An advertising intervention modeling module is used to model the change in the user's cognitive state vector caused by advertising delivery and generate an advertising cognitive intervention vector, wherein the cognitive intervention vector represents the change in the user's cognitive state after being exposed to the advertisement;

[0008] a multimodal consistency analysis module for evaluating the consistency between the user's behavioral response and the direction of the cognitive intervention vector based on the user's language feedback, visual behavior, and emotional trajectory, and outputting a multimodal behavior consistency score;

[0009] The conversion chain tracking module is used to track the user's behavioral path from ad click to final conversion, form conversion chain features, and identify path length, path node type, and path response delay;

[0010] The effect score generation module is used to integrate the cognitive intervention vector, multimodal behavior consistency score and conversion chain characteristics to calculate the comprehensive score of the cognitive effect of advertising for use in advertising effect evaluation.

[0011] Furthermore, the user cognitive state acquisition module includes the following submodules:

[0012] Multi-source data collection submodule, used to collect multi-source behavioral data before and after users come into contact with advertisements;

[0013] Multimodal feature extraction submodule, used to convert the collected multimodal data into feature vectors;

[0014] The cognitive state vector construction submodule is used to fuse multimodal features to construct the user's cognitive state vector.

[0015] Furthermore, the multimodal feature extraction submodule is used to extract features from web page behavior data, language feedback data and visual emotion data respectively, and uniformly convert the extracted features into high-dimensional embedding representations in the same vector space for use by the cognitive state vector construction submodule.

[0016] Furthermore, the advertising intervention modeling module includes the following submodules:

[0017] The original state change extraction submodule is used to calculate the original state change amount based on the cognitive state vector;

[0018] The intervention attribution modeling submodule is used to model the state change quantity with advertising and user characteristics to generate an intervention vector;

[0019] The intervention vector regularization and label annotation submodule is used to regularize the intervention vector and add semantic labels.

[0020] Furthermore, the intervention vector regularization and label annotation submodule is used to unify the scale of the advertising intervention vector and bind corresponding cognitive semantic labels to each vector dimension. The labels include attention activation, emotional resonance, interest enhancement and intention enhancement, which are used to enhance the interpretability and traceability of the intervention vector.

[0021] Furthermore, the multimodal consistency analysis module includes the following submodules:

[0022] Multimodal behavioral feature extraction submodule, used to extract behavioral features from language, vision and emotion data;

[0023] The behavior vector unified projection submodule is used to uniformly project multimodal features into the behavior response vector;

[0024] The consistency calculation submodule is used to calculate the consistency score between the behavioral response vector and the cognitive intervention vector.

[0025] Furthermore, the conversion chain tracking module includes the following submodules:

[0026] Behavior sequence capture and node structuring submodule, used to record and structure the user's behavior node sequence;

[0027] The conversion chain feature extraction submodule is used to extract path length, node type and response delay features from the behavior sequence;

[0028] The conversion path classification and labeling submodule is used to classify the conversion paths and label the paths;

[0029] The conversion chain structure encoding and output submodule is used to output the structured conversion path feature vector.

[0030] Furthermore, the conversion path classification and labeling submodule is used to divide user paths into quick response paths, deep engagement paths, hesitant paths and bounce paths based on the behavioral node type, node sequence, response time distribution and path density in the conversion path, and assign labels representing the level of behavioral intention to each type of path, which is used to evaluate the structural performance of the advertising conversion chain.

[0031] Furthermore, the effect score generation module includes the following submodules:

[0032] Feature normalization module, used to normalize intervention intensity, consistency score and conversion features;

[0033] The feature weighted fusion module is used to perform weighted combination of normalized features to generate a comprehensive score;

[0034] The threshold grading and interpretation module is used to grade the comprehensive scores and output the scoring results and impact factors.

[0035] The present invention also provides an advertising delivery effect evaluation and analysis method based on data analysis, which uses any of the aforementioned advertising delivery effect evaluation and analysis systems based on data analysis to perform advertising delivery effect evaluation and analysis.

[0036] The present invention provides a data-based advertising effectiveness evaluation and analysis system and method. Focusing on the psychological cognitive changes of users during advertising exposure, and combining user behavioral data with multimodal response signals, this system constructs an organic collaborative framework for advertising intervention modeling, multimodal consistency scoring, and conversion path analysis. This framework can achieve the following beneficial effects:

[0037] By constructing the user's cognitive state vector before and after the advertisement and establishing an advertising intervention model, we can quantify the specific impact of advertising on psychological dimensions such as user attention, interest, emotional arousal, and intention formation, thereby breaking through the traditional click-centered evaluation method.

[0038] The system introduces multimodal data sources such as users' language feedback, visual behavior, and emotional trajectories, and verifies whether cognitive intervention has been accepted by users and converted into actual responses through a behavioral consistency scoring mechanism, thereby enhancing the reliability and explanatory power of the evaluation model.

[0039] Through the conversion chain tracking module, the system can fully record and analyze the user's path behavior from ad exposure to conversion, extract structural features such as path length, node type and response delay, and provide a basis for identifying high-potential users and path optimization.

[0040] The effect score generation module integrates three key factors: cognitive intervention intensity, multimodal response consistency, and conversion path structure, and outputs a comprehensive score of cognitive effects for advertising attribution and strategy optimization. It has the advantages of strong quantitative ability, high interpretability, and strong adaptability.

[0041] In summary, the system provided by the present invention constructs a cognition-driven advertising effect analysis framework from the perspective of user cognition, filling the gap in the existing technology in the lack of advertising intervention mechanism modeling and multimodal response verification, and has high practical value and promotion prospects. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0043] Figure 1 It is a system block diagram of the present invention. DETAILED DESCRIPTION

[0044] Below, the invention is preferably described with reference to the accompanying drawings and specific embodiments.

[0045] This embodiment solves the above problem through the following steps:

[0046] In one embodiment, reference Figure 1The present invention provides an advertising effectiveness evaluation and analysis system based on data analysis. Specifically, it is a comprehensive system that quantitatively evaluates the influence of advertising, conversion paths, and cognitive response levels based on multi-source user behavior data analysis, combined with user cognitive state modeling and multimodal consistency calculation. The system constructs a user's cognitive state vector before and after advertising exposure, models the cognitive offset caused by advertising intervention, and integrates multimodal data such as user language feedback, visual behavior, and emotional trajectory to form a cognitive behavior consistency score. Combined with the path chain characteristics from user clicks to conversions, it calculates a comprehensive score of the cognitive effect of the advertisement, thereby achieving interpretable evaluation and refined optimization of the advertising effectiveness.

[0047] The user cognitive state acquisition module is used to construct the user's cognitive state vector based on the multi-source behavioral data before and after the user is exposed to the advertisement.

[0048] The core concept behind the user cognitive state acquisition module is that the true purpose of advertising isn't simply to prompt users to click, but rather to induce a shift in their cognitive state, thereby driving subsequent behavioral changes. Cognitive psychology posits that individuals process information through an inherent mechanism of "acceptance-comprehension-evaluation-response." The effectiveness of advertising is reflected in the gradual cognitive evolution of users, from "indifference" to "interest" and even "purchase intent." Therefore, to scientifically assess the effectiveness of advertising, we shouldn't rely solely on behavioral metrics like click-through rate or dwell time. Instead, we should construct a cognitive state vector model that captures user psychological changes.

[0049] By analyzing multi-source behavioral data before and after ad exposure (such as clickstreams, voice feedback, eye movements, and emotion recognition), we can extract a series of characteristic indicators that collectively represent the user's cognitive state. Mathematically, this representation can be modeled as a high-dimensional vector, called the "cognitive state vector," which represents the user's current performance across multiple cognitive dimensions (attention, interest, emotional activation, purchase intent, etc.). By comparing the cognitive state vectors before and after an ad, we can calculate cognitive offsets, thereby determining whether the ad has effectively interfered with the user's cognitive system.

[0050] The user cognitive status acquisition module includes the following four core sub-modules:

[0051] 1. Multi-source data acquisition submodule

[0052] Function: Collect user behavior and physiological performance from multiple data sources in real time or near real time for subsequent cognitive state modeling.

[0053] enter:

[0054] User page browsing data and click behavior data before and after ad exposure;

[0055] User text or voice interaction content;

[0056] Optional visual and physiological data (such as facial expressions collected by cameras, heart rate collected by wearable devices, etc.);

[0057] User historical profile data (such as gender, occupation, and interest tags).

[0058] Processing logic:

[0059] Call the website or app tracking system to obtain page access traces;

[0060] Access ASR / NLP system to process user language content;

[0061] Connect to the CV module (optional) to extract facial expression recognition features or eye movement data;

[0062] Access a unified timestamp and extract data within the advertisement's "pre-contact window" and "post-contact window" in segments.

[0063] Output:

[0064] The original multimodal data set D after time alignment

[0065] 2. Multimodal feature extraction submodule

[0066] Function: Convert raw behavior and perception data into standardized feature vectors for subsequent modeling.

[0067] enter:

[0068] The data set D includes the user's web page behavior, language content, visual trajectory, etc.

[0069] Processing logic:

[0070] Web page behavior characteristics: extract page dwell time, scroll depth, mouse hot spots, and click frequency;

[0071] Language / text features: Use pre-trained language models (such as BERT) to extract semantic embeddings and perform sentiment analysis.

[0072] Visual features: Extract facial muscle activation, micro-expression change frequency, etc. through facial expression recognition models (such as FER and OpenFace);

[0073] All features are normalized and standardized.

[0074] Output:

[0075] Eigenvectors F of different modes behavior ,F text ,F visual; Optionally perform Early Fusion or LateFusion to form a unified input vector F all .

[0076] 3. Cognitive State Vector Construction Submodule

[0077] Function: Model the user's cognitive state vector based on multimodal feature fusion, which serves as an expression of the user's psychological state before and after advertising intervention.

[0078] enter:

[0079] The fused feature vector F all ;

[0080] Feature label dictionary and cognitive dimension mapping rules (used to explain the psychological meaning of each dimension).

[0081] Processing logic:

[0082] Use multimodal fusion models, such as:

[0083] Attention-based Transformer encoder; multi-channel CNN fusion model; graph neural network (if considering the user's historical behavior chain graph);

[0084] Output cognitive state vector C=[C1,C2,…,C d ], each dimension represents a psychological cognitive indicator, such as attention index, interest, emotional activation, depth of information understanding, etc.

[0085] Output:

[0086] Vector C before advertising exposure pre ;

[0087] Vector C after advertising exposure post .

[0088] Specifically and illustratively:

[0089] A real estate company placed an advertisement for its new property on a short video platform. The system recorded the following data about a user before and after being exposed to the advertisement:

[0090] Before the ad: The average page view time was 5 seconds, no clicks were made, and the facial expression was stable.

[0091] After the ad: the user's page dwell time increases to 25 seconds. Click "View Details" and the facial recognition function shows a smile and concentration.

[0092] The key words “feel good” and “hadn’t considered it before” appeared in the text feedback.

[0093] After processing, the system generates:

[0094] Cpre =[0.3, 0.1, 0.2, 0.1] (attention, emotional activation, interest matching, purchase intention)

[0095] C post =[0.7,0.6,0.8,0.5].

[0096] The advertising intervention modeling module is used to model the changes in the user's cognitive state vector caused by advertising delivery and generate an advertising cognitive intervention vector, which represents the change in the user's cognitive state after being exposed to the advertisement.

[0097] The design concept of the advertising intervention modeling module is based on the "stimulus-response" mechanism and causal modeling in cognitive psychology. In the field of advertising evaluation, traditional methods focus on statistical metrics such as click-through rate (CTR) and conversion rate (CVR). However, these behavioral metrics cannot reveal how advertising actually changes users' attitudes and decisions at a psychological level. To understand the true value of advertising, it is necessary to quantify its direct impact on users' cognitive states—that is, whether the advertising triggers increased user attention, generated interest, activated emotions, or enhanced purchase intention.

[0098] Therefore, the role of the advertising intervention modeling module is to treat advertising as an input perturbation to the cognitive system and model and analyze the differences in users' cognitive states before and after the advertisement. By outputting a cognitive intervention vector, the system can capture the psychological intensity, direction, and multi-dimensional components of the advertising effect. This vector can be used in subsequent advertising attribution analysis and provide a quantitative basis for creative optimization and strategy adjustment.

[0099] This module is implemented in a modular structure and includes the following three sub-modules:

[0100] Submodule 1: Original state change extraction module

[0101] Function: Extract the difference in users' original cognitive states before and after exposure to advertisements as the basis for subsequent modeling.

[0102] Implementation steps:

[0103] Receive vector data C provided by the front module pre ,C post ;

[0104] Calculate the original change vector: ΔC raw =C post -C pre ;

[0105] This vector preliminarily reflects the direction and magnitude of changes in cognitive state, but does not perform causal attribution or noise removal.

[0106] Output: Original change vector ΔCraw .

[0107] Submodule 2: Intervention Attribution Modeling Module

[0108] Function: Introducing the ad's own attributes, user portrait information, and external context, conducting causal attribution analysis on changes in cognitive states and extracting the dominant impact of the ad.

[0109] Implementation steps:

[0110] enter:

[0111] Original change vector ΔC raw ;

[0112] Advertisement feature vector A∈R k (such as advertising type, distribution channel, copywriting style, image emotion, etc.);

[0113] User portrait vector U∈R m (e.g. age, interests, preferences, historical behavior);

[0114] Contextual features X∈R n (e.g. time, geographical location, parallel activities, etc.).

[0115] Construct the intervention modeling function:

[0116] I ad =f intv (ΔC raw ,A,U,X)

[0117] In the function f intv Can be built based on any of the following methods:

[0118] Multi-task neural network modeling;

[0119] Heterogeneous graph neural network modeling (user-advertisement-context triples);

[0120] Interpretable causal modeling based on structural equation modeling (SEM);

[0121] Bayesian network combined with attention mechanism for modeling.

[0122] Output cognitive intervention vector:

[0123] I ad ∈R d , each dimension represents the intervention intensity of advertising on a certain cognitive indicator (such as attention, interest, emotion);

[0124] Optionally, output intervention intensity index S intensity =||I ad ||2.

[0125] Submodule 3: Intervention vector regularization and label annotation module

[0126] Function: Normalize intervention vectors and assign semantic labels by dimension for subsequent interpretable output and decision support.

[0127] step:

[0128] Vector normalization processing improves comparability between different users;

[0129] Bind a cognitive dimension label to each dimension component, such as:

[0130] Dimension 1 → Attention Activation

[0131] Dimension 2 → Degree of emotional resonance

[0132] Dimension 3 → Intention Enhancement Factor

[0133] If the intervention model is an interpretable model (such as attention weight or regression coefficient), it can further output the importance score of advertising factors in each dimension of intervention;

[0134] Output structured intervention vector results and annotation information for use by visualization modules and optimization engines.

[0135] Output: Final cognitive intervention vector I ad and semantic label mapping table.

[0136] In a further implementation, the intervention modeling function is constructed as follows:

[0137] The model input is:

[0138] Original change vector ΔC raw ;

[0139] Advertisement feature vector A∈R k ;

[0140] User portrait vector U∈R m ;

[0141] Contextual features X∈R n .

[0142] Project each modal input into a unified dimensional space:

[0143] h ΔC =W1·ΔC raw +b1

[0144] h A =W2·A+b2

[0145] h U =W3·U+b3

[0146] h X =W4·X+b4

[0147] in,

[0148] W i and b i is the training parameter.

[0149] Construct a self-attention module to perform weighted fusion of the importance of the four inputs:

[0150]

[0151] The Score function is a simple linear score:

[0152] Score(h i )=v T tanh(W a h i +b a )

[0153] Where v is the learnable parameter vector, h i Unified dimensional space projection, W a and b a are the learnable parameters of the corresponding uniform dimensional space projection.

[0154] The final fusion output is:

[0155]

[0156] Use a multi-layer perceptron (MLP) to output the final intervention vector:

[0157]

[0158] Among them, I ad is the intervention vector; σ is a nonlinear activation function (such as ReLU or Swish); the rest are learnable parameters.

[0159] If the system already has a conversion label y∈{0,1} (such as whether to purchase), the following loss function is jointly trained:

[0160] Intervention prediction loss:

[0161]

[0162] Conversion prediction loss:

[0163]

[0164] Among them, y is the sample label, is the predicted label.

[0165] Joint losses:

[0166]

[0167] λ1,λ2 are preset weight parameters.

[0168] For example:

[0169] The cognitive status of a user before and after being exposed to the coffee machine advertisement is as follows: C pre =[0.3,0.1,0.2,0.1](attention, emotional activation, interest, purchase intention)C post =[0.7,0.6,0.8,0.5]

[0170] Calculate the original state change vector:

[0171] ΔC raw =[0.4,0.5,0.6,0.4]

[0172] The ad feature vector includes:

[0173] Writing style: lighthearted and cheerful;

[0174] Main image color: warm colors;

[0175] CTA: Highlight the emotional incentive of "reward yourself";

[0176] Delivery time: 20:00 in the evening, when users are active.

[0177] User portrait:

[0178] Aged 35, prefers a high quality of life;

[0179] Historical browsing preferences: small appliances, lifestyle content;

[0180] Geographical location: first-tier cities.

[0181] The intervention modeling module integrates the above information and outputs:

[0182] I ad =[0.35,0.47,0.55,0.33]

[0183] This vector shows that advertising has obvious advantages in the two dimensions of "emotional activation" and "interest arousal".

[0184] The multimodal consistency analysis module is used to evaluate the consistency between the user's behavioral response and the direction of the cognitive intervention vector based on the user's language feedback, visual behavior and emotional trajectory, and output a multimodal behavior consistency score.

[0185] The ultimate goal of advertising is to induce users to produce perceptible behavioral responses, such as clicks, purchases, and inquiries. However, simply relying on the impact of an ad on a user's cognitive state (i.e., the intervention vector) cannot guarantee that this impact is truly effective. A shift in user cognition may occur, but this may not translate into observable behavior, and even spurious responses (such as habitual clicks) may exist. Therefore, it is necessary to introduce a verification mechanism to assess whether the user's actual behavior is consistent with the cognitive intervention direction predicted by the system.

[0186] The Multimodal Consistency Analysis module is based on this concept. It extracts behavioral features from three modalities: verbal feedback, visual behavior, and emotional change. It then compares the directionality and intensity of these features with the cognitive intervention vector to quantify their degree of consistency. The module outputs a consistency score that reflects the extent to which users actually receive and respond to the advertising intervention, thereby improving the overall reliability and interpretability of the system.

[0187] In a specific implementation, the multimodal consistency analysis module consists of the following four modules:

[0188] Submodule 1: Multimodal behavior feature extraction module

[0189] Function: Extract feature representations of three types of behavioral signals from data after user ad exposure.

[0190] enter:

[0191] Within 5 minutes after user is exposed to the ad:

[0192] Language feedback data: such as customer service conversation texts, comments, and speech-to-text conversions;

[0193] Visual behavior data: eye movement trajectory, page sliding behavior, mouse path;

[0194] Emotional trajectory data: facial expression recognition data, intonation emotion recognition results (such as joy, disgust, etc.).

[0195] Output:

[0196] Behavior feature vector B∈R d , which is composed of the following sub-vectors:

[0197] Language feedback embedding (e.g., via BERT encoding);

[0198] Visual behavior characteristics (e.g., gaze duration, revisit frequency);

[0199] Characteristics of emotional changes (such as the amplitude of emotional fluctuations and the increase in positivity).

[0200] Submodule 2: Behavior vector unified projection module function: Map multimodal behavior features B to intervention vector I ad In the same cognitive space, define the transformation function g: The specific implementation is as follows:

[0201] Use a separate linear projection for each mode:

[0202]

[0203] After splicing, use the fully connected layer for unified fusion:

[0204]

[0205] in:

[0206] W1,W2,W3,W f ,b1,b2,b3,b f is a learnable parameter;

[0207] is a behavioral response representation in the same space as the intervention vector.

[0208] Submodule 3: Consistency calculation module function: Evaluate behavioral response vector With intervention vector I ad degree of consistency.

[0209] Use the fusion of direction consistency and amplitude consistency:

[0210] Directional consistency (cosine similarity):

[0211]

[0212] Amplitude ratio consistency:

[0213]

[0214] Final consistency score (weighted fusion):

[0215] S consistency =α·S dir +(1-α)·S mag

[0216] Where α is a configurable weight parameter.

[0217] For example:

[0218] In the above scenario, the user's post-ad cognitive intervention vector is:

[0219] I ad =[0.35,0.47,0.55,0.33]

[0220] The system collects user behavior data after advertising:

[0221] Verbal feedback: “I hadn’t thought about buying a coffee machine before, this feels like a great value”;

[0222] Browse the page for more than 30 seconds and repeatedly zoom in to view the image;

[0223] Expression recognition: From neutral to positive, smile lasts for 6 seconds.

[0224] After feature extraction and projection, the behavior vector:

[0225]

[0226] Calculate the consistency score:

[0227] S dir =0.996 (direction highly consistent);

[0228] S mag ≈0.94 (behavioral response amplitude close to intervention);

[0229] S consistency =0.7·0.996+0.3·0.94=0.978

[0230] System output:

[0231] Consistency score: 0.978;

[0232] Response label: High consistency;

[0233] Dominant modality: Verbal feedback and visual behavior.

[0234] This means that the advertisement not only has a significant impact on the cognitive level, but also users show strong and real response behavior, which is a combination of high-quality materials and highly adaptable users.

[0235] The conversion chain tracking module is used to track the user's behavioral path from ad click to final conversion, form conversion chain features, and identify path length, path node type, and path response delay.

[0236] In advertising analysis, whether a user ultimately converts isn't determined by a single click or cognitive response, but rather by a sequence of multiple interactive behaviors. This behavioral sequence often includes multiple "node events"—such as clicking an ad, browsing details, adding an item to a shopping cart, contacting customer service, registering an account, and submitting an order. Together, these behaviors constitute the user's conversion chain.

[0237] The end point (conversion) alone cannot fully reflect the value of advertising because:

[0238] Some ads guide long-path conversions (e.g., buying homes, cars, and other products with high decision-making costs);

[0239] Even if some ads do not lead to conversions, their behavioral paths show strong intentions (such as long stay time and frequent consultations);

[0240] Some paths appear active, but there is no logical order between their behaviors, which may be noise.

[0241] Therefore, it is necessary to build a systematic mechanism to track and model the user's complete behavioral path from advertising exposure to final conversion, and extract its key structural features (path length, node type, response delay, etc.), so as to support the structural evaluation, attribution analysis and strategy optimization of advertising effectiveness.

[0242] In a specific implementation, the conversion chain tracking module is implemented by the following four sub-modules:

[0243] Submodule 1: Behavior sequence capture and node structuring module

[0244] Function: Starting from ad clicks, record all key behavioral events of users and structure them into "path nodes".

[0245] Implementation steps:

[0246] Define a standardized path event type set: E = {click on the ad (e1), enter the details page (e2), slide the page (e3), add the product to your favorites (e4), customer service conversation (e5), submit the form (e6), order completed (e7), ...};

[0247] The system monitors all user interaction events starting from the ad click time t0, recording the timestamp and event type:

[0248] P u ={(e1,t1),(e2,t2),…,(e n ,t n )}

[0249] Sort each path by time to form a user conversion path chain with time attributes.

[0250] Submodule 2: Conversion chain feature extraction module

[0251] Function: Extract multi-dimensional behavior path features based on structured path data.

[0252] Extract dimensions:

[0253] Path length: L = |P u |=n represents the total number of behavior nodes in the path.

[0254] Node type distribution vector (one-hot or frequency): V e =[f1,f2,…,f k ]

[0255] Among them, f i Indicates event type e i The number of times or whether it appears in the path.

[0256] Average response delay (average time interval between actions):

[0257] Total delay from the beginning to the end (from ad to final conversion): T total =t n -t1

[0258] Behavior sparsity index (time density): Indicates the average number of actions performed by a user per unit time.

[0259] Submodule 3: Conversion path classification and labeling module

[0260] Function: Divide paths into typical conversion types based on their behavior patterns to facilitate subsequent interpretation and grouping.

[0261] Classification logic:

[0262] Fast decision path: short path length, fast response time, and contains high-value nodes;

[0263] Deep engagement path: long path, long stay time, including high-intention nodes such as consultation / registration;

[0264] Hesitant Path: Repeated clicks without conversion, commonly seen among price-sensitive users.

[0265] Jump-out path: There are only 1-2 nodes, and the user exits quickly or does not respond.

[0266] Abnormal path: The behavior nodes have no logical order, such as clicking on payment immediately after the details page but not completing the transaction.

[0267] In specific implementation, K-means or HMM (Hidden Markov Model) can be used to cluster path behavior patterns and automatically identify behavior forms.

[0268] Submodule 4: Conversion chain structure encoding and output module

[0269] Function: Encode the extracted structural path features into a unified vector expression for use by subsequent modules.

[0270] Vector structure example:

[0271]

[0272] The first four dimensions are path structure indicators, and the last k dimensions are event distributions;

[0273] It can be used for advertising attribution model input, user segmentation, strategy optimization and other modules;

[0274] Supports comparative analysis (multiple paths of the same user or different user paths of the same ad).

[0275] For example:

[0276] After a user clicks on a coffee machine ad, their behavior path is as follows:

[0277] (e1, t1): Click on the ad (0 seconds);

[0278] (e2, t2): Enter the details page (+1 second);

[0279] (e3, t3): Slide to browse the page (+4 seconds);

[0280] (e4, t4): Collect the item (+12 seconds);

[0281] (e5, t5): Open the comment section (+15 seconds);

[0282] (e6, t6): Enter the customer service window and send a message "Is this durable?" (+24 seconds);

[0283] (e7, t7): Exit the page (+40 seconds);

[0284] The system extracts path features:

[0285] Path length L = 7;

[0286] Node type distribution: including ad clicks, details, favorites, comments, customer service, etc.

[0287]

[0288] T total =40s

[0289] S=7 / 40≈0.175

[0290] The path is classified as a deep engagement path because it contains high-intention behaviors (collection + customer service) but has not yet been converted. The system can mark it as a high-potential user and subsequently push coupons or recall ads.

[0291] The effect score generation module is used to integrate the cognitive intervention vector, multimodal behavior consistency score and conversion chain characteristics to calculate the comprehensive score of the cognitive effect of advertising for use in advertising effect evaluation.

[0292] In ad evaluation tasks, a single click behavior, cognitive response, or conversion event often fails to fully reflect the true effectiveness of an ad. For example, an ad may trigger a strong cognitive shift but fail to elicit a behavioral response; another ad may be quickly clicked but result in no subsequent conversion.

[0293] To address this one-sided evaluation problem, it is necessary to build a comprehensive scoring model that integrates information from different dimensions, including:

[0294] Cognitive intervention intensity: whether the advertisement actually affects the user's cognitive state;

[0295] Multimodal behavioral consistency: whether the user has taken actions consistent with cognitive changes;

[0296] Conversion path structure: Is the user's behavior chain from click to conversion reasonable and high-intention?

[0297] Therefore, the effect score generation module is committed to integrating the core outputs of these sub-modules and generating a comprehensive cognitive effect score through modeling integration, which serves as a unified quantitative basis for advertising effect evaluation and is used for advertising material selection, delivery budget allocation and user response ranking.

[0298] In a specific implementation, the effect score generation module includes the following three submodules:

[0299] Submodule 1: Feature Normalization Module

[0300] Function: Normalize the scores and vector features from different modules to make them comparable on the same numerical scale.

[0301] enter:

[0302] Intervention Vector I ad ∈R d ;

[0303] Behavioral consistency score S consistency ∈[0,1];

[0304] Transformation chain feature vector F chain ∈R k .

[0305] Processing steps:

[0306] Normalize the modulus of the intervention vector:

[0307]

[0308] Normalization of each indicator in the conversion chain feature:

[0309]

[0310] in:

[0311] u i is the mean of the i-th dimension, σ i is the standard deviation;

[0312] After standardization Submodule 2: Feature Weighted Fusion Module Function: Constructs the final scoring function in a linear weighted manner, with the option of static weights or training optimization. CES calculation:

[0313] CES=α·S intervene +β·S consistency +γ·S chain

[0314] Among them, S intervene represents the advertising cognitive intervention intensity score;

[0315] S consistency represents the multimodal consistency score;

[0316] S chain represents the conversion path structure score;

[0317] The weight coefficients α, β, and γ satisfy α+β+γ=1.

[0318] Submodule 3: Threshold Grading and Interpretation Module Function: Map the score value to different effect level labels for use by the business side.

[0319] Example levels:

[0320] CES ≥ 0.85 → extremely strong cognitive effect

[0321] 0.7≤CES<0.85→effective

[0322] 0.5≤CES<0.7→General

[0323] CES<0.5→weak effect

[0324] Furthermore, it can also output:

[0325] Weight contribution of each indicator (explain the source);

[0326] Weak dimension marking (e.g., path sparsity is too high, or consistency is insufficient);

[0327] Can be used for recall strategies or delivery optimization suggestions.

[0328] For example:

[0329] Continuing with the previous example

[0330] Intervention Vector:

[0331] I ad =[0.35,0.47,0.55,0.33],

[0332] Behavior consistency score: S consistency =0.978

[0333] Conversion path feature F chain =[7,6.67,40,0.175], after normalization S chain =0.81

[0334] Assume weight α = 0.3, β = 0.4, γ = 0.3

[0335] Then we have:

[0336] CES=0.3·0.85+0.4·0.978+0.3·0.81=0.895

[0337] System output:

[0338] CES score: 0.895;

[0339] Rating: Cognitive effect is very strong;

[0340] Dominant factor: The consistency score has the largest weight, and the path structure is an auxiliary factor.

[0341] Strategic recommendations:

[0342] It is recommended that this advertisement continue to be delivered to the high cognition-high consistency group;

[0343] It is recommended to iteratively fine-tune the emotional arousal dimension based on the current material.

[0344] In another embodiment, the present invention further provides an advertising delivery effect evaluation and analysis method based on data analysis, which uses the aforementioned embodiment to perform advertising delivery effect evaluation and analysis.

[0345] It should be noted that the explanation of the aforementioned embodiment of the advertising delivery effect evaluation and analysis system based on data analysis is also applicable to the method of the embodiment of the present application and will not be repeated here.

[0346] Those skilled in the art will appreciate that the various units and algorithm steps described in the embodiments disclosed herein can be implemented using a combination of electronic hardware, computer software, and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0347] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0348] In the several embodiments provided in this application, if any function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the method described in each embodiment of this application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory; hereinafter referred to as: ROM), random access memory (Random Access Memory; hereinafter referred to as: RAM), magnetic disk or optical disk, and other media that can store program code.

[0349] The above is only a specific embodiment of the present application. Any person skilled in the art can easily think of changes or replacements within the technical scope disclosed in this application, which should be included in the scope of protection of this application. For some module structures that are not particularly clear in the present invention, the content recorded in the prior art shall prevail. The prior art mentioned in the above background technology section and the specific embodiment section of the present invention can be regarded as part of the present invention and is used to understand the meaning of some technical features or parameters.

Claims

1. An advertising effect evaluation and analysis system based on data analysis, characterized in that: The system includes the following modules: A user cognitive state acquisition module is used to construct a user cognitive state vector based on multi-source behavioral data before and after the user is exposed to the advertisement; An advertising intervention modeling module is used to model the change in the user's cognitive state vector caused by advertising delivery and generate an advertising cognitive intervention vector, wherein the cognitive intervention vector represents the change in the user's cognitive state after being exposed to the advertisement; a multimodal consistency analysis module for evaluating the consistency between the user's behavioral response and the direction of the cognitive intervention vector based on the user's language feedback, visual behavior, and emotional trajectory, and outputting a multimodal behavior consistency score; The conversion chain tracking module is used to track the user's behavioral path from ad click to final conversion, form conversion chain features, and identify path length, path node type, and path response delay; The effect score generation module is used to integrate the cognitive intervention vector, multimodal behavior consistency score and conversion chain characteristics to calculate the comprehensive score of the cognitive effect of advertising for use in advertising effect evaluation.

2. The advertising effect evaluation and analysis system based on data analysis according to claim 1 is characterized in that: The user cognitive status acquisition module includes the following submodules: Multi-source data collection submodule, used to collect multi-source behavioral data before and after users come into contact with advertisements; Multimodal feature extraction submodule, used to convert the collected multimodal data into feature vectors; The cognitive state vector construction submodule is used to fuse multimodal features to construct the user's cognitive state vector.

3. The advertising effect evaluation and analysis system based on data analysis according to claim 2 is characterized in that: The multimodal feature extraction submodule is used to extract features from web page behavior data, language feedback data, and visual emotion data respectively, and uniformly convert the extracted features into high-dimensional embedding representations in the same vector space for use by the cognitive state vector construction submodule.

4. The advertising effect evaluation and analysis system based on data analysis according to claim 1 is characterized in that: The advertising intervention modeling module includes the following submodules: The original state change extraction submodule is used to calculate the original state change amount based on the cognitive state vector; The intervention attribution modeling submodule is used to model the state change quantity with advertising and user characteristics to generate an intervention vector; The intervention vector regularization and label annotation submodule is used to regularize the intervention vector and add semantic labels.

5. The advertising effect evaluation and analysis system based on data analysis according to claim 4 is characterized in that: The intervention vector regularization and label annotation submodule is used to unify the scale of the advertising intervention vector and bind corresponding cognitive semantic labels to each vector dimension. The labels include attention activation, emotional resonance, interest enhancement, and intention enhancement, which are used to enhance the interpretability and traceability of the intervention vector.

6. The advertising effect evaluation and analysis system based on data analysis according to claim 1 is characterized in that: The multimodal consistency analysis module includes the following submodules: Multimodal behavioral feature extraction submodule, used to extract behavioral features from language, vision and emotion data; The behavior vector unified projection submodule is used to uniformly project multimodal features into the behavior response vector; The consistency calculation submodule is used to calculate the consistency score between the behavioral response vector and the cognitive intervention vector.

7. The advertising effect evaluation and analysis system based on data analysis according to claim 1 is characterized in that: The conversion chain tracking module includes the following submodules: Behavior sequence capture and node structuring submodule, used to record and structure the user's behavior node sequence; The conversion chain feature extraction submodule is used to extract path length, node type and response delay features from the behavior sequence; The conversion path classification and labeling submodule is used to classify the conversion paths and label the paths; The conversion chain structure encoding and output submodule is used to output the structured conversion path feature vector.

8. The advertising effect evaluation and analysis system based on data analysis according to claim 7 is characterized in that: The conversion path classification and labeling submodule is used to divide user paths into quick response paths, deep engagement paths, hesitant paths and bounce paths based on the behavioral node type, node sequence, response time distribution and path density in the conversion path, and assign labels representing the level of behavioral intention to each type of path, which is used to evaluate the structural performance of the advertising conversion chain.

9. The advertising effect evaluation and analysis system based on data analysis according to claim 1 is characterized in that: The effect score generation module includes the following submodules: Feature normalization module, used to normalize intervention intensity, consistency score and conversion features; The feature weighted fusion module is used to perform weighted combination of normalized features to generate a comprehensive score; The threshold grading and interpretation module is used to grade the comprehensive scores and output the scoring results and impact factors.

10. A method for evaluating and analyzing the effect of advertising based on data analysis, characterized in that: The method uses the advertising delivery effect evaluation and analysis system based on data analysis as described in any one of claims 1 to 9 to perform advertising delivery effect evaluation and analysis.

Images