Consumer behavior prediction method and system based on psychological information analysis

Abstract

The invention relates to the technical field of data analysis, in particular to a consumer behavior prediction method and system based on psychological information analysis. The method comprises the following steps: performing consumption behavior influence factor analysis on consumer consumption behavior data to obtain consumption behavior influence factor data; performing consumption behavior cognitive load analysis and group structure division according to the consumption behavior influence factor data to obtain consumption behavior cognitive group structure data; performing multi-view structure learning on the consumer behavior cognition group structure data and constructing a consumer consumer behavior cognition prediction model to obtain the consumer behavior cognition prediction model; and according to the consumption behavior cognition prediction model, performing prediction processing on the psychological activity consumption behavior of the consumer to obtain consumer consumption behavior prediction data. According to the method, the consumer behavior prediction technology is optimized, so that the consumer behavior prediction is more accurate.

Description

Technical Field

[0001] The present invention relates to the technical field of data analysis, and in particular to a consumer behavior prediction method and system based on psychological information analysis. Background Art

[0002] The consumer behavior prediction method based on psychological information analysis can help enterprises more accurately predict market trends and formulate effective marketing strategies by deeply studying consumers' psychological characteristics, attitudes and behavior patterns, and exploring potential driving factors and motivations. This method combines psychological theory and market research technology. By analyzing consumers' psychological states, emotions and cognitive processes, it reveals the influencing factors and motivation mechanisms of consumers in their purchase decisions, and provides insight and guidance for enterprises, so as to achieve more accurate market positioning and product positioning and improve market competitiveness. Therefore, the consumer behavior prediction method based on psychological information analysis is of great significance in today's highly competitive market environment, and provides strong support for the development and success of enterprises. However, the traditional consumer behavior prediction method based on psychological information analysis has the problems of inaccurate consumer behavior prediction and low prediction processing efficiency. Summary of the invention

[0003] Based on this, it is necessary to provide a consumer behavior prediction method and system based on psychological information analysis to solve at least one of the above technical problems.

[0004] To achieve the above purpose, a method for predicting consumer behavior based on psychological information analysis is provided, the method comprising the following steps:

[0005] Step S1: acquiring consumer behavior data; performing consumer behavior cluster analysis based on the consumer behavior data to obtain consumer behavior cluster data; performing consumer behavior influencing factor analysis on the consumer behavior cluster data to obtain consumer behavior influencing factor data;

[0006] Step S2: performing cognitive load analysis on consumer behavior based on consumer behavior influencing factor data to obtain consumer behavior cognitive load data; performing group structure division on the consumer behavior cognitive load data to obtain consumer behavior cognitive group structure data;

[0007] Step S3: Performing a difference analysis between different cognitive group structures on the consumer behavior cognitive group structure data to obtain cognitive group structure difference data; performing multi-view structure learning on the consumer behavior cognitive group structure data based on the cognitive group structure difference data to obtain cognitive group multi-view structure data; constructing a consumer consumption behavior cognitive prediction model on the cognitive group multi-view structure data based on a support vector machine to obtain a consumer behavior cognitive prediction model;

[0008] Step S4: Predict and process the consumer's psychological activity consumption behavior according to the consumption behavior cognitive prediction model to obtain consumer consumption behavior prediction data.

[0009] The present invention can divide consumers into different behavior groups by acquiring consumer behavior data and performing cluster analysis. This helps to understand the purchase preferences, frequencies and purchase habits of different groups, so as to better design marketing strategies and services for different groups. Through the analysis of consumer behavior influencing factors, the key factors affecting consumer purchase decisions, such as price, brand awareness, product characteristics, etc., can be determined. This helps companies understand consumers' purchase motivations and preferences, and then adjust product pricing and marketing strategies. Consumer behavior cognitive load data can reveal the cognitive pressure and load level that consumers bear in the purchase decision process. This helps companies design simpler and easier-to-understand purchase processes and information presentation methods, improve consumers' purchase experience, divide consumers into different group structures according to their cognitive load and behavior patterns, help companies more accurately locate target customer groups, and formulate personalized marketing strategies and service plans for different groups. By deeply understanding consumer behavior and cognitive characteristics, companies can more accurately predict consumer behavior and needs, thereby optimizing marketing strategies, product design and service processes, and improving marketing efficiency and market competitiveness. By performing differential analysis on cognitive group structure data, differences in cognitive characteristics, behavioral patterns, etc. between different cognitive groups can be found. This helps companies understand the characteristics of consumer behavior of different groups more finely and provide a basis for personalized marketing and services. Multi-view structural learning can integrate information from different data sources to more comprehensively understand the characteristics and behavioral patterns of cognitive groups. This helps improve the accuracy and robustness of the model and enhance the ability to understand and predict consumer behavior. Based on machine learning algorithms such as support vector machines, a consumer behavior cognitive prediction model is constructed using multi-view structural data of cognitive groups. This model can more accurately predict consumer behavior and preferences and provide companies with more targeted marketing strategies and service designs. Through the consumer behavior cognitive prediction model, companies can more quickly and accurately identify consumer needs and behavioral trends, thereby optimizing product design, marketing activities and service processes, improving decision-making efficiency and market response speed, and establishing an effective consumer behavior cognitive prediction model can help companies better meet consumer needs and enhance product competitiveness and market share. At the same time, by continuously optimizing models and strategies, competitive advantages can be maintained and market changes can be responded to.

[0010] Preferably, step S1 comprises the following steps:

[0011] Step S11: Obtain consumer behavior data;

[0012] Step S12: performing consumption environment analysis based on the consumer behavior data to obtain consumption environment data; performing consumption behavior cluster analysis based on the consumer consumption behavior data to obtain consumption behavior cluster data;

[0013] Step S13: performing a consumer behavior influencing factor analysis on the consumer behavior clustering data according to the consumer environment data to obtain consumer behavior influencing factor data.

[0014] The present invention obtains extensive and comprehensive consumer behavior data, which is the basis for in-depth understanding of consumer behavior patterns and trends. This includes data on purchase records, browsing behavior, and click-through rates; by analyzing consumer behavior data, the scene and environment of consumer activities are understood, including shopping places, time, location and other factors. This helps to grasp the background and context of consumer behavior and provide a basis for subsequent analysis; cluster analysis is performed using consumer behavior data to divide consumers into different groups or categories. This helps to identify the behavior patterns and preferences of different consumer groups and provide guidance for personalized marketing and services. Based on the consumption environment data, the consumer behavior cluster data is analyzed to identify the key factors affecting consumer behavior, such as price, promotional activities, product characteristics, service quality, etc. This helps to understand the driving force and decision-making factors of consumer behavior. By understanding the influencing factors of different consumer groups, enterprises can formulate targeted marketing strategies and service plans to meet the needs and preferences of different consumer groups, improve market competitiveness and customer satisfaction, and the consumer behavior influencing factor data provides an important basis for enterprise decision-making, helping enterprises to better adjust product pricing, promotional strategies, product design and other aspects, so as to more effectively attract consumers and increase sales and profits.

[0015] Preferably, step S13 comprises the following steps:

[0016] Step S131: marking the cluster center of the consumer behavior cluster data to obtain the consumer behavior cluster center data; calculating the inter-cluster distance of the consumer behavior cluster data according to the consumer behavior cluster center data to obtain the consumer behavior inter-cluster distance data;

[0017] Step S132: performing clustering radius calculation on the distance data between consumer behavior groups according to the consumer behavior group center data to obtain consumer behavior clustering radius data; performing consumer behavior distribution dispersion analysis according to the consumer behavior clustering radius data to obtain consumer behavior distribution dispersion data;

[0018] Step S133: performing environment-related mapping on the consumption behavior distribution discreteness data according to the consumption environment data to obtain environment-related behavior distribution discreteness data;

[0019] Step S134: performing environment-consumption behavior factor analysis on the environment-related behavior distribution discrete data to obtain environment-consumption behavior factor data;

[0020] Step S135: constructing an environment-consumption behavior structural factor correlation matrix for the environment-related behavior distribution discrete data according to the environment-consumption behavior factor data to obtain an environment-consumption structural factor matrix; performing matrix structure orthogonal rotation on the environment-consumption structural factor matrix to obtain factor matrix structure simplified data;

[0021] Step S136: performing factor loading analysis on the environment-consumption structure factor matrix according to the simplified data of the factor matrix structure to obtain the environment-consumption structure factor loading data;

[0022] Step S137: Perform consumption behavior influencing factor analysis based on the environment-consumption structure factor load data and the environment-related behavior distribution discrete data to obtain consumption behavior influencing factor data.

[0023] The present invention can identify the central features or representative behaviors of each consumer behavior group by marking the cluster center of consumer behavior cluster data. This helps to understand the core features of different consumer groups, and by calculating the distance between the centers of different consumer behavior groups, the similarities or differences between different groups can be quantified. This helps to determine the relationship between consumer behavior groups and further guide personalized marketing and product positioning; by calculating the cluster radius based on the consumer behavior group center data, the distribution range and aggregation degree of each consumer behavior group can be quantified. This helps to evaluate the spatial distribution characteristics of consumer behavior and provide a basis for regional market strategies; by analyzing the distribution dispersion of consumer behavior, the degree of dispersion of consumer behavior in space can be understood. This helps to identify the concentrated and dispersed areas of consumer behavior, and provides a reference for refined positioning and resource allocation. The consumer behavior distribution dispersion data is associated with the consumer environment data and mapped, which can reveal the differences and characteristics of consumer behavior under different environmental conditions. This helps to understand the degree of influence of the environment on consumer behavior and provide guidance for scenario-based marketing and service design; by factor analysis of the discrete data of environment-related behavior distribution, the key factors between the environment and consumer behavior can be identified. This helps to understand the mechanism and path of environmental factors' impact on consumer behavior, and provides a basis for enterprises to formulate targeted strategies. Combining environmental-consumption behavior factor data and consumer behavior group load data, we can deeply analyze the degree and method of different factors' impact on consumer behavior. This helps to accurately identify the key factors that affect consumer decisions and behaviors, and provide strategic recommendations for marketing and service optimization.

[0024] Preferably, step S2 comprises the following steps:

[0025] Step S21: performing consumption orientation analysis on the consumption behavior clustering data to obtain consumption behavior orientation data;

[0026] Step S22: performing consumer behavior cognitive load analysis based on consumer behavior orientation data and consumer behavior influencing factor data to obtain consumer behavior cognitive load data;

[0027] Step S23: performing cognitive load bias analysis on the consumer behavior cognitive load data to obtain cognitive load bias data;

[0028] Step S24: dividing the consumer behavior cognitive load data into groups according to the cognitive load bias data to obtain consumer behavior cognitive group structure data.

[0029] The present invention can reveal the consumption preferences, behavioral tendencies and purchasing motivations of different consumer groups by analyzing the consumer behavior clustering data. This helps to understand the dominant factors of consumers in purchasing decisions and provides a basis for customizing personalized marketing strategies; combined with consumer behavior orientation data and consumer behavior influencing factor data, the degree of cognitive load that consumers bear in the decision-making process is analyzed. This helps to understand the cognitive effort required by consumers in the purchasing process, and provides a reference for simplifying the shopping experience and improving purchase satisfaction. By analyzing the consumer behavior cognitive load data, the consumer's cognitive load bias can be determined, that is, which aspects or factors are more concerned in the decision-making process. This helps to understand the attention allocation and decision-making focus of different consumer groups in purchasing behavior. Based on the cognitive load bias data, consumers are divided into different cognitive group structures to identify common cognitive characteristics. This helps personalized marketing and service design, and provides targeted product information and purchasing experience for different cognitive groups.

[0030] Preferably, step S22 includes the following steps:

[0031] Step S221: performing linear structure analysis on the consumption behavior influencing factor data to obtain the influencing factor linear structure data;

[0032] Step S222: constructing a linear matrix of consumer behavior influencing factors based on the linear structure data of influencing factors, and obtaining a linear matrix of influencing factors;

[0033] Step S223: Calculate the factor loads of the impact factor linear matrix to obtain the impact factor load data; perform factor structure rotation adjustment on the impact factor linear matrix according to the impact factor load data to obtain the impact factor linear adjustment matrix;

[0034] Step S224: using the impact factor contribution calculation formula to calculate the impact factor contribution of the impact factor linear adjustment matrix and the impact factor load data to obtain the impact factor contribution data;

[0035] Step S225: Perform consumer behavior cognitive load analysis based on consumer behavior orientation data, influencing factor load data, and influencing factor co-contribution data to obtain consumer behavior cognitive load data.

[0036] The present invention analyzes the influencing factors of consumer behavior through linear structure, and can identify the linear relationship between these factors. This helps to understand the degree of influence of different factors on consumer behavior and the interaction between them. Based on the linear structure data of the influencing factors, a linear matrix of the influencing factors of consumer behavior is constructed. This step can provide a basis for the subsequent factor load calculation, help quantify the degree of influence of different factors in consumer decision-making, and obtain the load degree of each influencing factor on consumer behavior by calculating the factor load of the influencing factor linear matrix. Subsequently, the factor structure rotation adjustment is performed to make the relationship between the factors clearer and more explainable. The influencing factor contribution calculation formula is used to calculate the linear adjustment matrix of the influencing factors and the influencing factor load data to obtain the contribution of each factor to the overall cognitive load. This helps to determine the relative importance of each factor in the decision-making process, and combines the consumer behavior orientation data, the influencing factor load data and the influencing factor co-contribution data to perform consumer behavior cognitive load analysis. This step aims to comprehensively consider the consumer's behavior orientation, factor load and the contribution of each factor to evaluate the cognitive load level in the consumer decision-making process.

[0037] Preferably, the impact factor contribution calculation formula is as follows:

[0038]

[0039] In the formula, C represents the result value of the co-contribution of the influencing factors, t represents the extreme value of the calculation time, X represents the structural complexity of the linear adjustment matrix of the influencing factors, Y represents the linear structural correlation coefficient of the linear adjustment matrix of the influencing factors, Z represents the load difference between different influencing factors in the influencing factor load data, β represents the dimensionality coefficient of the linear adjustment matrix of the influencing factors, γ represents the load range interval of the influencing factor load data, δ represents the cross-load intensity value of the influencing factor load data; φ represents the element correlation of the linear adjustment matrix of the influencing factors, and R represents the error correction value of the calculation formula of the influencing factor contribution.

[0040] The present invention constructs a formula for calculating the contribution of impact factors, which comprehensively evaluates the contribution of impact factors by taking into account factors such as the structure, correlation, difference, range and interaction of impact factors. The formula fully considers the extreme value of the calculation time t, which is used to determine the upper limit of the integral, that is, the time range for calculating the contribution of the impact factor, which is determined according to the time span of the study or the time period to be evaluated, and can ensure that the factor changes and influences within the time range of interest are taken into account when calculating the contribution of the impact factor; the structural complexity X of the linear adjustment matrix of the impact factor, this parameter represents the structural complexity of the matrix, if the structure of the matrix is ​​more complex, it can capture more interactions and correlations between different factors, so as to more accurately evaluate the contribution of the impact factor; the linear structure correlation coefficient Y of the linear adjustment matrix of the impact factor, which represents the correlation of the linear structure in the matrix, if the linear structure correlation is higher, it means that the linear relationship between the impact factors is stronger, so the contribution of the impact factor will be higher; the load difference Z between different impact factors in the impact factor load data, which represents the load difference between different factors, if the load difference between different factors is larger, it means that their contribution to the research topic is different, so the contribution This difference will be given more attention during calculation; the dimension coefficient β of the linear adjustment matrix of the influencing factors, this parameter represents the dimension of the matrix. The higher the dimension, the more factors can be taken into account, so as to more comprehensively evaluate the contribution of the influencing factors; the load range interval γ of the influencing factor load data, represents the range interval of the load. If the range is wider, it can include more loads of different strengths, so it can more comprehensively consider the impact of different loads on the contribution; the cross-load intensity value δ of the influencing factor load data, represents the cross-load intensity between different factors. If the cross-load intensity is higher, it means that the interaction between different factors is stronger, so it will have a greater impact on the contribution; the element correlation φ of the linear adjustment matrix of the influencing factors, represents the degree of correlation between the elements in the matrix. If the element correlation is higher, it means that the elements in the matrix are more interdependent, so that the contribution of the influencing factors can be more accurately evaluated; the error correction value R of the calculation formula of the influencing factor contribution is used to correct the possible errors in the calculation formula to ensure that the calculation results are more accurate.

[0041] Preferably, step S24 comprises the following steps:

[0042] Step S241: performing cognitive load Bayesian statistical analysis on the cognitive load data of consumer behavior to obtain cognitive load statistical data; and drawing a load statistical curve on the cognitive load statistical data to obtain a cognitive load statistical curve;

[0043] Step S242: extracting inflection points of the cognitive load statistical curve to obtain curve inflection point data; performing time series trend decomposition on the cognitive load statistical curve according to the curve inflection point data to obtain cognitive load time series trend data;

[0044] Step S243: calculating the autocorrelation coefficient of the cognitive load statistical curve according to the curve inflection point data and the cognitive load time series trend data to obtain the cognitive load autocorrelation coefficient;

[0045] Step S244: performing a hysteresis effect evaluation on the cognitive load statistical curve according to the cognitive load autocorrelation coefficient to obtain cognitive load hysteresis effect data;

[0046] Step S245: performing biased hysteresis effect mapping on the cognitive load hysteresis effect data according to the cognitive load bias data to obtain cognitive load biased hysteresis effect data;

[0047] Step S246: Divide the consumer behavior cognitive load data into group structures according to the cognitive load bias lag effect data and the cognitive load autocorrelation coefficient to obtain consumer behavior cognitive group structure data.

[0048] Through the Bayesian statistical analysis of cognitive load, the present invention can help you understand the cognitive load characteristics of consumer behavior and convert them into statistical data. This helps to identify the level and pattern of consumers' cognitive load. By drawing the load statistical curve, you can more intuitively understand the distribution and change trend of cognitive load; inflection point extraction can help you determine the change points in the curve, which represent the changes or turning points of cognitive load in different time periods. Time series trend decomposition can further decompose the curve into different trend components, such as long-term trends, seasonal trends, and cyclical trends, so as to have a deeper understanding of the evolution of cognitive load; autocorrelation coefficient calculation can help you evaluate the correlation and delay effect between cognitive load data. This helps to understand the temporal correlation between cognitive loads and provide a basis for subsequent analysis; by evaluating the lag effect of cognitive load, you can determine the extent to which changes in cognitive load at different time points affect subsequent behavior. This helps to understand the dynamic characteristics of cognitive load and take corresponding strategies to deal with these effects. Through biased lag effect mapping, you can combine the lag effect with the bias of cognitive load to more accurately evaluate the impact of cognitive load on subsequent behavior. This helps to identify the differences between different consumer groups and develop targeted strategies. Through group structure division, consumers can be divided into different groups or categories according to their cognitive load characteristics and behavior patterns. This helps to personalize marketing, product positioning and service design to meet the needs and preferences of different groups.

[0049] Preferably, step S3 comprises the following steps:

[0050] Step S31: performing a difference analysis between different cognitive group structures on the consumer behavior cognitive group structure data to obtain cognitive group structure difference data;

[0051] Step S32: performing topic modeling between the consumer behavior cognitive group structure data and the cognitive group structure data according to the cognitive group structure difference data to obtain cognitive group topic difference data;

[0052] Step S33: performing multi-view structural learning on the consumer behavior cognitive group structural data based on the cognitive group subject difference data and the cognitive group structural difference data to obtain cognitive group multi-view structural data;

[0053] Step S34: constructing a consumer consumption behavior cognitive prediction model for the cognitive group multi-view structure data based on a support vector machine to obtain a consumer behavior cognitive prediction model.

[0054] The present invention performs differential analysis on cognitive group structure data, which helps to understand the differences between different cognitive groups. Such differences may involve differences in consumers' cognition, attitudes, preferences, etc. of products or services. By analyzing these differences, the behavior patterns and decision-making processes of different groups can be better understood; by performing topic modeling on cognitive group structure difference data, the topic differences between different cognitive groups can be found. These topic differences may reflect the differences in different groups' focus on products or services, importance evaluation, etc. The identification of such topic differences helps to accurately understand the needs and expectations of different groups; by performing topic modeling on cognitive group structure difference data, the topic differences between different cognitive groups can be found. These topic differences may reflect the differences in different groups' focus on products or services, importance evaluation, etc. The identification of such topic differences helps to accurately understand the needs and expectations of different groups; by performing multi-view structural learning based on cognitive group theme difference data and cognitive group structure difference data, multiple views (such as cognitive themes, structural features, etc.) of different groups can be combined for learning and analysis. Such multi-view learning can provide a more comprehensive and accurate consumer behavior cognitive model; using support vector machine to construct a consumer behavior cognitive prediction model for cognitive group multi-view structural data is a key step in applying the analysis results of the previous steps to model construction. Support vector machine is a powerful machine learning algorithm that can effectively process multi-dimensional data and complex patterns to build accurate consumer behavior cognitive prediction models.

[0055] Preferably, step S34 includes the following steps:

[0056] Step S341: extracting multi-view structural parameters from the cognitive group multi-view structural data to obtain cognitive group multi-view structural parameters;

[0057] Step S342: performing recursive feature elimination processing on the cognitive group multi-view structure parameters to obtain cognitive group recursive feature data;

[0058] Step S343: performing recursive feature multidimensional scaling transformation on the cognitive group multi-view structure parameters according to the cognitive group recursive feature data to obtain cognitive group multidimensional scaling transformation parameters;

[0059] Step S344: constructing a consumer behavior cognitive prediction model based on the multidimensional scaling transformation parameters of the cognitive group based on the support vector machine to obtain a consumer behavior cognitive prediction model.

[0060] The present invention can integrate and abstract the information of multiple views into specific parameter forms by extracting parameters from the multi-view structure data of cognitive groups. These parameters can capture the associations and features between different views and provide basic data for subsequent model construction; recursive feature elimination is a feature selection method that helps to screen out the most influential features for the consumer behavior cognitive prediction model. This step can help exclude those features that contribute less to the prediction model, thereby improving the accuracy and interpretability of the model; by performing multidimensional scaling transformation on the recursive feature data of cognitive groups, the original data can be converted into a more efficient and informative representation. This transformation helps to reduce data dimensions and information redundancy while retaining important feature information, providing better input data for model construction; the last step is to use support vector machine (SVM) to model the multidimensional scaling transformation parameters of cognitive groups that have been feature processed. SVM is a powerful supervised learning algorithm that is suitable for processing complex nonlinear relationships and can effectively construct a consumer behavior cognitive prediction model.

[0061] Preferably, the present invention further provides a consumer behavior prediction system based on psychological information analysis, which is used to execute the consumer behavior prediction method based on psychological information analysis as described above, and the consumer behavior prediction system based on psychological information analysis comprises:

[0062] The consumption behavior influencing factor analysis module is used to obtain consumer behavior data; perform consumption behavior cluster analysis based on the consumer behavior data to obtain consumption behavior cluster data; perform consumption behavior influencing factor analysis on the consumption behavior cluster data to obtain consumption behavior influencing factor data;

[0063] The consumer behavior cognitive load division module is used to analyze the consumer behavior cognitive load according to the consumer behavior influencing factor data to obtain the consumer behavior cognitive load data; divide the consumer behavior cognitive load data into group structures to obtain the consumer behavior cognitive group structure data;

[0064] The consumer behavior cognitive prediction model construction module is used to perform a difference analysis between different cognitive group structures on the consumer behavior cognitive group structure data to obtain cognitive group structure difference data; perform multi-view structure learning on the consumer behavior cognitive group structure data based on the cognitive group structure difference data to obtain cognitive group multi-view structure data; and construct a consumer consumption behavior cognitive prediction model based on the cognitive group multi-view structure data based on the support vector machine to obtain a consumer behavior cognitive prediction model;

[0065] The consumer consumption behavior prediction module is used to predict and process the consumer psychological activity consumption behavior according to the consumer behavior cognitive prediction model to obtain the consumer consumption behavior prediction data.

[0066] The beneficial effects of the present invention are as follows: the present invention provides a consumer behavior prediction system based on psychological information analysis, which is an optimization process of the traditional consumer behavior prediction system based on psychological information analysis, and solves the problems of inaccurate consumer behavior prediction and low prediction processing efficiency in the traditional consumer behavior prediction method based on psychological information analysis, thereby improving the accuracy of the prediction and the efficiency of the prediction processing capability. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] Figure 1 A schematic diagram of the steps of a consumer behavior prediction method based on psychological information analysis;

[0068] Figure 2 for Figure 1 Detailed implementation steps of step S2 in the flowchart;

[0069] Figure 3 for Figure 1 Detailed implementation steps of step S3 in FIG.

[0070] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0071] The following is a clear and complete description of the technical method of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by technicians in this field without creative work are within the scope of protection of the present invention.

[0072] In addition, the accompanying drawings are only schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor methods and / or microcontroller methods.

[0073] It should be understood that, although the terms "first", "second", etc. may be used herein to describe various units, these units should not be limited by these terms. These terms are used only to distinguish one unit from another unit. For example, without departing from the scope of the exemplary embodiments, the first unit may be referred to as the second unit, and similarly the second unit may be referred to as the first unit. The term "and / or" used herein includes any and all combinations of one or more of the listed associated items.

[0074] To achieve this, please refer to Figures 1 to 3 , a consumer behavior prediction method based on psychological information analysis, the method comprising the following steps:

[0075] Step S1: acquiring consumer behavior data; performing consumer behavior cluster analysis based on the consumer behavior data to obtain consumer behavior cluster data; performing consumer behavior influencing factor analysis on the consumer behavior cluster data to obtain consumer behavior influencing factor data;

[0076] Step S2: performing cognitive load analysis on consumer behavior based on consumer behavior influencing factor data to obtain consumer behavior cognitive load data; performing group structure division on the consumer behavior cognitive load data to obtain consumer behavior cognitive group structure data;

[0077] Step S3: Performing a difference analysis between different cognitive group structures on the consumer behavior cognitive group structure data to obtain cognitive group structure difference data; performing multi-view structure learning on the consumer behavior cognitive group structure data based on the cognitive group structure difference data to obtain cognitive group multi-view structure data; constructing a consumer consumption behavior cognitive prediction model on the cognitive group multi-view structure data based on a support vector machine to obtain a consumer behavior cognitive prediction model;

[0078] Step S4: Predict and process the consumer's psychological activity consumption behavior according to the consumption behavior cognitive prediction model to obtain consumer consumption behavior prediction data.

[0079] In the embodiment of the present invention, reference Figure 1The above is a schematic diagram of the steps of a method for predicting consumer behavior based on psychological information analysis of the present invention. In this example, the method for predicting consumer behavior based on psychological information analysis includes the following steps:

[0080] Step S1: acquiring consumer behavior data; performing consumer behavior cluster analysis based on the consumer behavior data to obtain consumer behavior cluster data; performing consumer behavior influencing factor analysis on the consumer behavior cluster data to obtain consumer behavior influencing factor data;

[0081] In the embodiment of the present invention, consumer behavior data can be collected from public channels, such as consumer behavior papers, social media interactions, etc. Then, cluster these data using cluster analysis methods (such as K-means algorithm) to divide consumers into different groups, each group representing a type of consumer behavior pattern. Next, through statistical methods such as factor analysis, key factors that affect consumer behavior, such as price sensitivity and brand loyalty, are determined.

[0082] Step S2: performing cognitive load analysis on consumer behavior based on consumer behavior influencing factor data to obtain consumer behavior cognitive load data; performing group structure division on the consumer behavior cognitive load data to obtain consumer behavior cognitive group structure data;

[0083] In the embodiment of the present invention, based on the obtained consumption behavior influencing factor data, the load theory in cognitive psychology is used to analyze the cognitive load that consumers bear when making purchase decisions, and then statistical analysis is used to determine the cognitive load levels of different consumer groups.

[0084] Step S3: Performing a difference analysis between different cognitive group structures on the consumer behavior cognitive group structure data to obtain cognitive group structure difference data; performing multi-view structure learning on the consumer behavior cognitive group structure data based on the cognitive group structure difference data to obtain cognitive group multi-view structure data; constructing a consumer consumption behavior cognitive prediction model on the cognitive group multi-view structure data based on a support vector machine to obtain a consumer behavior cognitive prediction model;

[0085] In an embodiment of the present invention, a statistical analysis method, such as analysis of variance or a chi-square test, is used to study the differences between different cognitive group structures. This can help determine the significant differences in cognitive characteristics between different groups. Then, through a multi-view structural learning method, such as multi-core learning or graph embedding technology, different cognitive group structural data are integrated to form a more comprehensive cognitive group multi-view structural data. At the same time, a machine learning method such as a support vector machine (SVM) can be used to build a consumer consumption behavior cognitive prediction model. Including, a prediction model can be trained using the consumer's cognitive group multi-view structural data as input and consumption behavior as output. Such a model can predict the consumer's future consumption behavior based on his or her cognitive characteristics.

[0086] Step S4: Predict and process the consumer's psychological activity consumption behavior according to the consumption behavior cognitive prediction model to obtain consumer consumption behavior prediction data.

[0087] In the embodiment of the present invention, the constructed consumer behavior cognitive prediction model is used to predict the psychological activities and consumer behavior of consumers. Including, assuming that a support vector machine model has been trained, it can predict the consumer's purchase behavior based on the multi-view structure data of the consumer's cognitive group. Then, when new consumer data is input, these data can be input into the model, and the model will predict the consumer's purchase behavior based on these data.

[0088] The present invention can divide consumers into different behavior groups by acquiring consumer behavior data and performing cluster analysis. This helps to understand the purchase preferences, frequencies and purchase habits of different groups, so as to better design marketing strategies and services for different groups. Through the analysis of consumer behavior influencing factors, the key factors affecting consumer purchase decisions, such as price, brand awareness, product characteristics, etc., can be determined. This helps companies understand consumers' purchase motivations and preferences, and then adjust product pricing and marketing strategies. Consumer behavior cognitive load data can reveal the cognitive pressure and load level that consumers bear in the purchase decision process. This helps companies design simpler and easier-to-understand purchase processes and information presentation methods, improve consumers' purchase experience, divide consumers into different group structures according to their cognitive load and behavior patterns, help companies more accurately locate target customer groups, and formulate personalized marketing strategies and service plans for different groups. By deeply understanding consumer behavior and cognitive characteristics, companies can more accurately predict consumer behavior and needs, thereby optimizing marketing strategies, product design and service processes, and improving marketing efficiency and market competitiveness. By performing differential analysis on cognitive group structure data, differences in cognitive characteristics, behavioral patterns, etc. between different cognitive groups can be found. This helps companies understand the characteristics of consumer behavior of different groups more finely and provide a basis for personalized marketing and services. Multi-view structural learning can integrate information from different data sources to more comprehensively understand the characteristics and behavioral patterns of cognitive groups. This helps improve the accuracy and robustness of the model and enhance the ability to understand and predict consumer behavior. Based on machine learning algorithms such as support vector machines, a consumer behavior cognitive prediction model is constructed using multi-view structural data of cognitive groups. This model can more accurately predict consumer behavior and preferences and provide companies with more targeted marketing strategies and service designs. Through the consumer behavior cognitive prediction model, companies can more quickly and accurately identify consumer needs and behavioral trends, thereby optimizing product design, marketing activities and service processes, improving decision-making efficiency and market response speed, and establishing an effective consumer behavior cognitive prediction model can help companies better meet consumer needs and enhance product competitiveness and market share. At the same time, by continuously optimizing models and strategies, competitive advantages can be maintained and market changes can be responded to.

[0089] Preferably, step S1 comprises the following steps:

[0090] Step S11: Obtain consumer behavior data;

[0091] Step S12: performing consumption environment analysis based on the consumer behavior data to obtain consumption environment data; performing consumption behavior cluster analysis based on the consumer consumption behavior data to obtain consumption behavior cluster data;

[0092] Step S13: performing a consumer behavior influencing factor analysis on the consumer behavior clustering data according to the consumer environment data to obtain consumer behavior influencing factor data.

[0093] In an embodiment of the present invention, consumer behavior data may be collected from public channels, such as consumer behavior papers and social media interactions, and consumption environment analysis may be performed based on the consumer behavior data to obtain consumption environment data. This may include information such as the consumer's geographical location, purchase channel, and purchase time. For example, consumption patterns in different regions or time periods may be analyzed to understand consumers' purchasing preferences and habits. At the same time, cluster analysis may be performed on consumer behavior data to divide consumers into different groups in order to better understand consumer behavior, and consumer behavior influencing factor analysis may be performed on consumer behavior cluster data based on consumption environment data to obtain consumer behavior influencing factor data. This may be performed by using statistical analysis methods, such as regression analysis or factor analysis, to determine the key factors that affect consumer behavior. For example, the degree to which consumer purchasing behavior is affected by factors such as price, promotional activities, and product characteristics may be analyzed, thereby providing a basis for subsequent decision-making.

[0094] The present invention obtains extensive and comprehensive consumer behavior data, which is the basis for in-depth understanding of consumer behavior patterns and trends. This includes data on purchase records, browsing behavior, and click-through rates; by analyzing consumer behavior data, the scene and environment of consumer activities are understood, including shopping places, time, location and other factors. This helps to grasp the background and context of consumer behavior and provide a basis for subsequent analysis; cluster analysis is performed using consumer behavior data to divide consumers into different groups or categories. This helps to identify the behavior patterns and preferences of different consumer groups and provide guidance for personalized marketing and services. Based on the consumption environment data, the consumer behavior cluster data is analyzed to identify the key factors affecting consumer behavior, such as price, promotional activities, product characteristics, service quality, etc. This helps to understand the driving force and decision-making factors of consumer behavior. By understanding the influencing factors of different consumer groups, enterprises can formulate targeted marketing strategies and service plans to meet the needs and preferences of different consumer groups, improve market competitiveness and customer satisfaction, and the consumer behavior influencing factor data provides an important basis for enterprise decision-making, helping enterprises to better adjust product pricing, promotional strategies, product design and other aspects, so as to more effectively attract consumers and increase sales and profits.

[0095] Preferably, step S13 comprises the following steps:

[0096] Step S131: marking the cluster center of the consumer behavior cluster data to obtain the consumer behavior cluster center data; calculating the inter-cluster distance of the consumer behavior cluster data according to the consumer behavior cluster center data to obtain the consumer behavior inter-cluster distance data;

[0097] Step S132: performing clustering radius calculation on the distance data between consumer behavior groups according to the consumer behavior group center data to obtain consumer behavior clustering radius data; performing consumer behavior distribution dispersion analysis according to the consumer behavior clustering radius data to obtain consumer behavior distribution dispersion data;

[0098] Step S133: performing environment-related mapping on the consumption behavior distribution discreteness data according to the consumption environment data to obtain environment-related behavior distribution discreteness data;

[0099] Step S134: performing environment-consumption behavior factor analysis on the environment-related behavior distribution discrete data to obtain environment-consumption behavior factor data;

[0100] Step S135: constructing an environment-consumption behavior structural factor correlation matrix for the environment-related behavior distribution discrete data according to the environment-consumption behavior factor data to obtain an environment-consumption structural factor matrix; performing matrix structure orthogonal rotation on the environment-consumption structural factor matrix to obtain factor matrix structure simplified data;

[0101] Step S136: performing factor loading analysis on the environment-consumption structure factor matrix according to the simplified data of the factor matrix structure to obtain the environment-consumption structure factor loading data;

[0102] Step S137: Perform consumption behavior influencing factor analysis based on the environment-consumption structure factor load data and the environment-related behavior distribution discrete data to obtain consumption behavior influencing factor data.

[0103] In an embodiment of the present invention, the consumer behavior cluster data is marked as a cluster center to obtain consumer behavior cluster center data. This can be achieved by calculating the center point of each cluster group, for example, calculating the average or median of all consumer behavior data in each group as the cluster center; performing inter-group distance calculation on the consumer behavior cluster data according to the consumer behavior group center data to obtain consumer behavior inter-group distance data. This can be achieved by calculating the distance between different groups. Common methods include Euclidean distance, Manhattan distance or cosine similarity. The cluster radius is calculated for the consumer behavior inter-group distance data according to the consumer behavior group center data to obtain consumer behavior cluster radius data. This can be achieved by calculating the average or standard deviation of the distance between all data points in each group and the group center, which is used to measure the distribution of data points within the group; performing consumer behavior distribution dispersion analysis according to the consumer behavior cluster radius data to obtain consumer behavior distribution dispersion data. This can be achieved by comparing the cluster radius between different groups to evaluate the closeness and dispersion of the groups, thereby understanding the distribution of consumer behavior; performing environmental association mapping on the consumer behavior distribution dispersion data according to the consumer environment data to obtain environmental association behavior distribution discrete data. This can be achieved by performing correlation analysis on the discrete distribution data of consumer behavior and the consumer environment data to understand the distribution of consumer behavior under different environmental conditions, and performing environmental-consumption behavior factor analysis on the discrete distribution data of environmental-related behavior to obtain environmental-consumption behavior factor data. This can be done by using statistical analysis methods, such as factor analysis or correlation analysis, to determine the correlation and degree of influence between environmental factors and consumer behavior, thereby revealing the influencing factors of the environment on consumer behavior. Based on the environmental-consumption behavior factor data, the environmental-consumption behavior structural factor correlation matrix is ​​constructed for the discrete distribution data of environmental-related behavior to obtain the environmental-consumption structural factor matrix. This can be achieved by calculating the correlation coefficient matrix between factors to reflect the degree of correlation between environmental factors and consumer behavior factors; orthogonal rotation of the matrix structure of the environmental-consumption structural factor matrix is ​​performed to obtain simplified factor matrix structure data. This can be done by performing orthogonal rotation using the factor analysis method to make the new factor matrix more explanatory and simplified. Based on the simplified data of the factor matrix structure, the factor loading analysis of the environmental-consumption structural factor matrix is ​​performed to obtain the environmental-consumption structural factor loading data. This can be done by measuring the strength of the relationship between each factor and the original variable through the factor loading matrix, and then understanding the degree of influence of environmental factors and consumer behavior factors on consumer behavior. According to the environmental-consumption structural factor loading data and the discrete data of environmental-related behavior distribution, the consumer behavior influencing factor analysis is carried out to obtain the consumer behavior influencing factor data. This can be done through statistical analysis methods, such as regression analysis or path analysis, to determine the degree of influence of different environmental factors on consumer behavior, thereby revealing the mechanism of the influence of environmental factors on consumer behavior.

[0104] The present invention can identify the central features or representative behaviors of each consumer behavior group by marking the cluster center of consumer behavior cluster data. This helps to understand the core features of different consumer groups, and by calculating the distance between the centers of different consumer behavior groups, the similarities or differences between different groups can be quantified. This helps to determine the relationship between consumer behavior groups and further guide personalized marketing and product positioning; by calculating the cluster radius based on the consumer behavior group center data, the distribution range and aggregation degree of each consumer behavior group can be quantified. This helps to evaluate the spatial distribution characteristics of consumer behavior and provide a basis for regional market strategies; by analyzing the distribution dispersion of consumer behavior, the degree of dispersion of consumer behavior in space can be understood. This helps to identify the concentrated and dispersed areas of consumer behavior, and provides a reference for refined positioning and resource allocation. The consumer behavior distribution dispersion data is associated with the consumer environment data and mapped, which can reveal the differences and characteristics of consumer behavior under different environmental conditions. This helps to understand the degree of influence of the environment on consumer behavior and provide guidance for scenario-based marketing and service design; by factor analysis of the discrete data of environment-related behavior distribution, the key factors between the environment and consumer behavior can be identified. This helps to understand the mechanism and path of environmental factors' impact on consumer behavior, and provides a basis for enterprises to formulate targeted strategies. Combining environmental-consumption behavior factor data and consumer behavior group load data, we can deeply analyze the degree and method of different factors' impact on consumer behavior. This helps to accurately identify the key factors that affect consumer decisions and behaviors, and provide strategic recommendations for marketing and service optimization.

[0105] Preferably, step S2 comprises the following steps:

[0106] Step S21: performing consumption orientation analysis on the consumption behavior clustering data to obtain consumption behavior orientation data;

[0107] Step S22: performing consumer behavior cognitive load analysis based on consumer behavior orientation data and consumer behavior influencing factor data to obtain consumer behavior cognitive load data;

[0108] Step S23: performing cognitive load bias analysis on the consumer behavior cognitive load data to obtain cognitive load bias data;

[0109] Step S24: dividing the consumer behavior cognitive load data into groups according to the cognitive load bias data to obtain consumer behavior cognitive group structure data.

[0110] As an example of the present invention, refer to Figure 2 As shown, in this example, step S2 includes:

[0111] Step S21: performing consumption orientation analysis on the consumption behavior clustering data to obtain consumption behavior orientation data;

[0112] In the embodiment of the present invention, consumption behavior clustering data is subjected to consumption orientation analysis to obtain consumption behavior orientation data. This can be achieved by analyzing the purchasing tendencies and preferences of consumers in different situations. For example, the purchase proportions or preferences of consumers in each cluster group for different product categories can be counted to understand the consumption orientation of consumers.

[0113] Step S22: performing consumer behavior cognitive load analysis based on consumer behavior orientation data and consumer behavior influencing factor data to obtain consumer behavior cognitive load data;

[0114] In the embodiment of the present invention, the cognitive load analysis of consumer behavior is performed based on the consumer behavior orientation data and the consumer behavior influencing factor data to obtain the consumer behavior cognitive load data. This can be done by using statistical analysis methods, such as regression analysis or factor analysis, combined with cognitive load theory to determine the degree of cognitive load of consumers in different situations and the degree of influence of influencing factors on cognitive load.

[0115] Step S23: performing cognitive load bias analysis on the consumer behavior cognitive load data to obtain cognitive load bias data;

[0116] In the embodiment of the present invention, cognitive load bias analysis is performed on the cognitive load data of consumer behavior to obtain cognitive load bias data. This can be achieved by comparing the performance of different consumers or consumer groups in terms of cognitive load. For example, the amount of information, decision complexity and cognitive load level faced by different groups in the purchase process can be counted to determine the bias of their cognitive load.

[0117] Step S24: dividing the consumer behavior cognitive load data into groups according to the cognitive load bias data to obtain consumer behavior cognitive group structure data.

[0118] In the embodiment of the present invention, the consumer behavior cognitive load data is divided into groups according to the cognitive load bias data to obtain consumer behavior cognitive group structure data. This can be achieved by cluster analysis or factor analysis to divide consumers or consumer groups into different cognitive load bias groups.

[0119] The present invention can reveal the consumption preferences, behavioral tendencies and purchasing motivations of different consumer groups by analyzing the consumer behavior clustering data. This helps to understand the dominant factors of consumers in purchasing decisions and provides a basis for customizing personalized marketing strategies; combined with consumer behavior orientation data and consumer behavior influencing factor data, the degree of cognitive load that consumers bear in the decision-making process is analyzed. This helps to understand the cognitive effort required by consumers in the purchasing process, and provides a reference for simplifying the shopping experience and improving purchase satisfaction. By analyzing the consumer behavior cognitive load data, the consumer's cognitive load bias can be determined, that is, which aspects or factors are more concerned in the decision-making process. This helps to understand the attention allocation and decision-making focus of different consumer groups in purchasing behavior. Based on the cognitive load bias data, consumers are divided into different cognitive group structures to identify common cognitive characteristics. This helps personalized marketing and service design, and provides targeted product information and purchasing experience for different cognitive groups.

[0120] Preferably, step S22 includes the following steps:

[0121] Step S221: performing linear structure analysis on the consumption behavior influencing factor data to obtain the influencing factor linear structure data;

[0122] Step S222: constructing a linear matrix of consumer behavior influencing factors based on the linear structure data of influencing factors, and obtaining a linear matrix of influencing factors;

[0123] Step S223: Calculate the factor loads of the impact factor linear matrix to obtain the impact factor load data; perform factor structure rotation adjustment on the impact factor linear matrix according to the impact factor load data to obtain the impact factor linear adjustment matrix;

[0124] Step S224: using the impact factor contribution calculation formula to calculate the impact factor contribution of the impact factor linear adjustment matrix and the impact factor load data to obtain the impact factor contribution data;

[0125] Step S225: Perform consumer behavior cognitive load analysis based on consumer behavior orientation data, influencing factor load data, and influencing factor co-contribution data to obtain consumer behavior cognitive load data.

[0126] In an embodiment of the present invention, a linear structure analysis is performed on the consumer behavior influencing factor data to obtain the influencing factor linear structure data. This can be done by using statistical analysis methods, such as factor analysis or regression analysis, to determine the linear structure relationship between the consumer behavior influencing factors. For example, factor analysis can be used to determine the common factors between the influencing factors, and the consumer behavior influencing factor linear matrix is ​​constructed based on the influencing factor linear structure data to obtain the influencing factor linear matrix. The influencing factor data can be represented as a matrix form, in which each row represents an influencing factor and each column represents a sample, and the factor loading is calculated on the influencing factor linear matrix to obtain the influencing factor loading data. This can be done by using methods such as principal component analysis or factor analysis to calculate the factor loading value of each influencing factor on each factor, reflecting the degree of association with each factor, and the factor structure rotation adjustment is performed on the influencing factor linear matrix according to the influencing factor loading data to obtain the influencing factor linear adjustment matrix. This can be done by adjusting the factor structure through methods such as orthogonal rotation or oblique rotation, so that the relationship between factors is clearer and more explanatory; the influence factor contribution calculation formula is used to calculate the influence factor linear adjustment matrix and the influence factor load data to obtain the influence factor co-contribution data, and the consumer behavior cognitive load analysis is performed based on the consumer behavior orientation data, the influence factor load data, and the influence factor co-contribution data to obtain the consumer behavior cognitive load data. This can be done through statistical analysis methods, such as regression analysis or path analysis, to determine the degree of cognitive load of consumers in different situations. For example, the degree of influence of consumers' consumption orientation on cognitive load and the degree of contribution of different influencing factors to cognitive load can be analyzed.

[0127] The present invention analyzes the influencing factors of consumer behavior through linear structure, and can identify the linear relationship between these factors. This helps to understand the degree of influence of different factors on consumer behavior and the interaction between them. Based on the linear structure data of the influencing factors, a linear matrix of the influencing factors of consumer behavior is constructed. This step can provide a basis for the subsequent factor load calculation, help quantify the degree of influence of different factors in consumer decision-making, and obtain the load degree of each influencing factor on consumer behavior by calculating the factor load on the linear matrix of the influencing factors. Subsequently, the factor structure rotation adjustment is performed to make the relationship between the factors clearer and more explainable. The linear adjustment matrix of the influencing factors and the influencing factor load data are calculated using the influencing factor contribution calculation formula to obtain the contribution of each factor to the overall cognitive load. This helps to determine the relative importance of each factor in the decision-making process, and combines the consumer behavior orientation data, the influencing factor load data, the influencing factor co-contribution data and the cognitive load theory to perform cognitive load analysis of consumer behavior. This step aims to comprehensively consider the consumer's behavioral orientation, factor load and the contribution of each factor to evaluate the cognitive load level in the consumer decision-making process.

[0128] Preferably, the impact factor contribution calculation formula is as follows:

[0129]

[0130] In the formula, C represents the result value of the co-contribution of the influencing factors, t represents the extreme value of the calculation time, X represents the structural complexity of the linear adjustment matrix of the influencing factors, Y represents the linear structural correlation coefficient of the linear adjustment matrix of the influencing factors, Z represents the load difference between different influencing factors in the influencing factor load data, β represents the dimensionality coefficient of the linear adjustment matrix of the influencing factors, γ represents the load range interval of the influencing factor load data, δ represents the cross-load intensity value of the influencing factor load data; φ represents the element correlation of the linear adjustment matrix of the influencing factors, and R represents the error correction value of the calculation formula of the influencing factor contribution.

[0131] The present invention constructs a formula for calculating the contribution of impact factors, which comprehensively evaluates the contribution of impact factors by taking into account factors such as the structure, correlation, difference, range and interaction of impact factors. The formula fully considers the extreme value of the calculation time t, which is used to determine the upper limit of the integral, that is, the time range for calculating the contribution of the impact factor, which is determined according to the time span of the study or the time period to be evaluated, and can ensure that the factor changes and influences within the time range of interest are taken into account when calculating the contribution of the impact factor; the structural complexity X of the linear adjustment matrix of the impact factor, this parameter represents the structural complexity of the matrix, if the structure of the matrix is ​​more complex, it can capture more interactions and correlations between different factors, so as to more accurately evaluate the contribution of the impact factor; the linear structure correlation coefficient Y of the linear adjustment matrix of the impact factor, which represents the correlation of the linear structure in the matrix, if the linear structure correlation is higher, it means that the linear relationship between the impact factors is stronger, so the contribution of the impact factor will be higher; the load difference Z between different impact factors in the impact factor load data, which represents the load difference between different factors, if the load difference between different factors is larger, it means that their contribution to the research topic is different, so the contribution This difference will be given more attention during calculation; the dimension coefficient β of the linear adjustment matrix of the influencing factors, this parameter represents the dimension of the matrix. The higher the dimension, the more factors can be taken into account, so as to more comprehensively evaluate the contribution of the influencing factors; the load range interval γ of the influencing factor load data, represents the range interval of the load. If the range is wider, it can include more loads of different strengths, so it can more comprehensively consider the impact of different loads on the contribution; the cross-load intensity value δ of the influencing factor load data, represents the cross-load intensity between different factors. If the cross-load intensity is higher, it means that the interaction between different factors is stronger, so it will have a greater impact on the contribution; the element correlation φ of the linear adjustment matrix of the influencing factors, represents the degree of correlation between the elements in the matrix. If the element correlation is higher, it means that the elements in the matrix are more interdependent, so that the contribution of the influencing factors can be more accurately evaluated; the error correction value R of the calculation formula of the influencing factor contribution is used to correct the possible errors in the calculation formula to ensure that the calculation results are more accurate.

[0132] Preferably, step S24 includes the following steps:

[0133] Step S241: performing cognitive load Bayesian statistical analysis on the cognitive load data of consumer behavior to obtain cognitive load statistical data; and drawing a load statistical curve on the cognitive load statistical data to obtain a cognitive load statistical curve;

[0134] Step S242: extracting inflection points of the cognitive load statistical curve to obtain curve inflection point data; performing time series trend decomposition on the cognitive load statistical curve according to the curve inflection point data to obtain cognitive load time series trend data;

[0135] Step S243: calculating the autocorrelation coefficient of the cognitive load statistical curve according to the curve inflection point data and the cognitive load time series trend data to obtain the cognitive load autocorrelation coefficient;

[0136] Step S244: performing a hysteresis effect evaluation on the cognitive load statistical curve according to the cognitive load autocorrelation coefficient to obtain cognitive load hysteresis effect data;

[0137] Step S245: performing biased hysteresis effect mapping on the cognitive load hysteresis effect data according to the cognitive load bias data to obtain cognitive load biased hysteresis effect data;

[0138] Step S246: Divide the consumer behavior cognitive load data into group structures according to the cognitive load bias lag effect data and the cognitive load autocorrelation coefficient to obtain consumer behavior cognitive group structure data.

[0139] In an embodiment of the present invention, cognitive load Bayesian statistical analysis is performed on cognitive load data of consumer behavior to obtain cognitive load statistical data. This can be achieved by probabilistically modeling and analyzing cognitive load data using a Bayesian statistical method, thereby obtaining statistical characteristics of cognitive load, and drawing a load statistical curve for cognitive load statistical data to obtain a cognitive load statistical curve. This can be achieved by drawing a frequency distribution curve or a cumulative distribution curve of cognitive load data to show the overall distribution of cognitive load data, extracting inflection points from the cognitive load statistical curve, and obtaining curve inflection point data. This can be achieved by identifying the extreme value points or points where the rate of change suddenly changes in the curve to determine the inflection point position, thereby finding the inflection point of the curve, and performing time series trend decomposition on the cognitive load statistical curve according to the curve inflection point data to obtain cognitive load time series trend data. This can be achieved by taking the curve inflection point as a time node and decomposing the curve into time series data of different stages to show the trend of cognitive load changes over time; the autocorrelation coefficient of the cognitive load statistical curve is calculated according to the curve inflection point data and the cognitive load time series trend data to obtain the cognitive load autocorrelation coefficient. This can be done by calculating the autocorrelation coefficient of the time series data at each stage of the curve to determine the autocorrelation of the cognitive load data, so as to understand the time series characteristics of the cognitive load, and the hysteresis effect evaluation of the cognitive load statistical curve according to the cognitive load autocorrelation coefficient to obtain the cognitive load hysteresis effect data. This can be done by calculating the hysteresis order of the autocorrelation coefficient to evaluate the hysteresis effect between cognitive load data. Higher autocorrelation coefficients and hysteresis orders indicate that the change in cognitive load has a certain delay effect in time; cognitive load hysteresis effect data can be mapped with biased hysteresis effect according to cognitive load bias data to obtain cognitive load biased hysteresis effect data. This can be done by performing correlation analysis on cognitive load bias data and hysteresis effect data to determine the differences in hysteresis effects between different biased groups; consumer behavior cognitive load data can be divided into group structures according to cognitive load biased hysteresis effect data and cognitive load autocorrelation coefficients to obtain consumer behavior cognitive group structure data. This can be done by dividing consumers into different cognitive load biased hysteresis effect groups through cluster analysis or factor analysis, so as to understand the cognitive load characteristics and behavioral preferences of different groups.

[0140] Through the Bayesian statistical analysis of cognitive load, the present invention can help you understand the cognitive load characteristics of consumer behavior and convert them into statistical data. This helps to identify the level and pattern of consumers' cognitive load. By drawing the load statistical curve, you can more intuitively understand the distribution and change trend of cognitive load; inflection point extraction can help you determine the change points in the curve, which represent the changes or turning points of cognitive load in different time periods. Time series trend decomposition can further decompose the curve into different trend components, such as long-term trends, seasonal trends, and cyclical trends, so as to have a deeper understanding of the evolution of cognitive load; autocorrelation coefficient calculation can help you evaluate the correlation and delay effect between cognitive load data. This helps to understand the temporal correlation between cognitive loads and provide a basis for subsequent analysis; by evaluating the lag effect of cognitive load, you can determine the extent to which changes in cognitive load at different time points affect subsequent behavior. This helps to understand the dynamic characteristics of cognitive load and take corresponding strategies to deal with these effects. Through biased lag effect mapping, you can combine the lag effect with the bias of cognitive load to more accurately evaluate the impact of cognitive load on subsequent behavior. This helps to identify the differences between different consumer groups and develop targeted strategies. Through group structure division, consumers can be divided into different groups or categories according to their cognitive load characteristics and behavior patterns. This helps to personalize marketing, product positioning and service design to meet the needs and preferences of different groups.

[0141] Preferably, step S3 comprises the following steps:

[0142] Step S31: performing a difference analysis between different cognitive group structures on the consumer behavior cognitive group structure data to obtain cognitive group structure difference data;

[0143] Step S32: performing topic modeling between the consumer behavior cognitive group structure data and the cognitive group structure data according to the cognitive group structure difference data to obtain cognitive group topic difference data;

[0144] Step S33: performing multi-view structural learning on the consumer behavior cognitive group structural data based on the cognitive group subject difference data and the cognitive group structural difference data to obtain cognitive group multi-view structural data;

[0145] Step S34: constructing a consumer consumption behavior cognitive prediction model for the cognitive group multi-view structure data based on a support vector machine to obtain a consumer behavior cognitive prediction model.

[0146] As an example of the present invention, refer to Figure 3 As shown, in this example, step S3 includes:

[0147] Step S31: performing a difference analysis between different cognitive group structures on the consumer behavior cognitive group structure data to obtain cognitive group structure difference data;

[0148] In the embodiment of the present invention, the difference analysis between different cognitive group structures is performed on the consumer behavior cognitive group structure data to obtain cognitive group structure difference data. This can be done by comparing the characteristics, behavioral preferences or other indicators between different groups to evaluate the differences between them. Including, statistical methods such as variance analysis, chi-square test or t-test can be used to determine the differences between different groups.

[0149] Step S32: performing topic modeling between the consumer behavior cognitive group structure data and the cognitive group structure data according to the cognitive group structure difference data to obtain cognitive group topic difference data;

[0150] In an embodiment of the present invention, the cognitive group structure data of consumer behavior is subjected to topic modeling between the same cognitive group structure according to the cognitive group structure difference data, so as to obtain cognitive group topic difference data. This can be done by using a topic modeling method, such as a topic model (such as Latent Dirichlet Allocation) to discover topic differences between different cognitive groups. The topic model can convert the group structure data into topic distribution, thereby discovering topic differences between different groups, that is, their preferences or attention levels on specific topics.

[0151] Step S33: performing multi-view structural learning on the consumer behavior cognitive group structural data based on the cognitive group subject difference data and the cognitive group structural difference data to obtain cognitive group multi-view structural data;

[0152] In the embodiment of the present invention, multi-view structural learning is performed on the cognitive group structure data of consumer behavior based on the cognitive group topic difference data and the cognitive group structure difference data. This can be achieved by combining data from different views through a multi-view learning method, such as multi-core learning or joint learning, to obtain more comprehensive group structure information. This includes taking the topic difference data and the structure difference data as different views, and obtaining more accurate cognitive group structure data by learning the correlation between them.

[0153] Step S34: constructing a consumer consumption behavior cognitive prediction model for the cognitive group multi-view structure data based on a support vector machine to obtain a consumer behavior cognitive prediction model.

[0154] In an embodiment of the present invention, a consumer consumption behavior cognitive prediction model is constructed based on the multi-view structure data of the cognitive group based on a support vector machine. Support vector machine (SVM) is a commonly used machine learning method that can be used for classification and regression tasks. Here, we can use the support vector machine algorithm to train the multi-view structure data to construct a consumer behavior cognitive prediction model. Through this model, we can predict consumers' consumption behavior cognition, that is, their purchasing preferences or behavioral characteristics.

[0155] The present invention performs differential analysis on cognitive group structure data, which helps to understand the differences between different cognitive groups. Such differences may involve differences in consumers' cognition, attitudes, preferences, etc. of products or services. By analyzing these differences, the behavior patterns and decision-making processes of different groups can be better understood; by performing topic modeling on cognitive group structure difference data, the topic differences between different cognitive groups can be found. These topic differences may reflect the differences in different groups' focus on products or services, importance evaluation, etc. The identification of such topic differences helps to accurately understand the needs and expectations of different groups; by performing topic modeling on cognitive group structure difference data, the topic differences between different cognitive groups can be found. These topic differences may reflect the differences in different groups' focus on products or services, importance evaluation, etc. The identification of such topic differences helps to accurately understand the needs and expectations of different groups; by performing multi-view structural learning based on cognitive group theme difference data and cognitive group structure difference data, multiple views (such as cognitive themes, structural features, etc.) of different groups can be combined for learning and analysis. Such multi-view learning can provide a more comprehensive and accurate consumer behavior cognitive model; using support vector machine to construct a consumer behavior cognitive prediction model for cognitive group multi-view structural data is a key step in applying the analysis results of the previous steps to model construction. Support vector machine is a powerful machine learning algorithm that can effectively process multi-dimensional data and complex patterns to build accurate consumer behavior cognitive prediction models.

[0156] Preferably, step S34 includes the following steps:

[0157] Step S341: extracting multi-view structural parameters from the cognitive group multi-view structural data to obtain cognitive group multi-view structural parameters;

[0158] Step S342: performing recursive feature elimination processing on the cognitive group multi-view structure parameters to obtain cognitive group recursive feature data;

[0159] Step S343: performing recursive feature multidimensional scaling transformation on the cognitive group multi-view structure parameters according to the cognitive group recursive feature data to obtain cognitive group multidimensional scaling transformation parameters;

[0160] Step S344: constructing a consumer behavior cognitive prediction model based on the multidimensional scaling transformation parameters of the cognitive group based on the support vector machine to obtain a consumer behavior cognitive prediction model.

[0161] In an embodiment of the present invention, multi-view structural parameters are extracted from the multi-view structural data of the cognitive group to obtain the multi-view structural parameters of the cognitive group. This can represent the structural information of the group by extracting key features or parameters from different views. Including, statistics such as the mean, variance, correlation coefficient, etc. in each view can be calculated as the multi-view structural parameters of the cognitive group; recursive feature elimination is performed on the multi-view structural parameters of the cognitive group to obtain recursive feature data of the cognitive group. Recursive feature elimination is a feature selection method that repeatedly trains the model and eliminates features that have little effect on the prediction results to obtain more representative feature data; recursive feature multi-dimensional scaling is performed on the multi-view structural parameters of the cognitive group according to the recursive feature data of the cognitive group to obtain multi-dimensional scaling transformation parameters of the cognitive group. This can be achieved by using methods such as multidimensional scaling analysis (MDS) to map high-dimensional feature data to low-dimensional space in order to better visualize and understand the structure of the data; based on the support vector machine, the multi-dimensional scaling transformation parameters of the cognitive group are used to construct a cognitive prediction model for consumer behavior, and a cognitive prediction model for consumer behavior is obtained. In this step, we use the support vector machine algorithm to train the parameters after feature selection and scaling to build a consumer behavior cognition prediction model. This model can be used to predict consumers' behavioral cognition, such as their purchasing preferences or behavioral characteristics.

[0162] The present invention can integrate and abstract the information of multiple views into specific parameter forms by extracting parameters from the multi-view structure data of cognitive groups. These parameters can capture the associations and features between different views and provide basic data for subsequent model construction; recursive feature elimination is a feature selection method that helps to screen out the most influential features for the consumer behavior cognitive prediction model. This step can help exclude those features that contribute less to the prediction model, thereby improving the accuracy and interpretability of the model; by performing multidimensional scaling transformation on the recursive feature data of cognitive groups, the original data can be converted into a more efficient and informative representation. This transformation helps to reduce data dimensions and information redundancy while retaining important feature information, providing better input data for model construction; the last step is to use support vector machine (SVM) to model the multidimensional scaling transformation parameters of cognitive groups that have been feature processed. SVM is a powerful supervised learning algorithm that is suitable for processing complex nonlinear relationships and can effectively construct a consumer behavior cognitive prediction model.

[0163] Preferably, the present invention further provides a consumer behavior prediction system based on psychological information analysis, which is used to execute the consumer behavior prediction method based on psychological information analysis as described above, and the consumer behavior prediction system based on psychological information analysis comprises:

[0164] The consumption behavior influencing factor analysis module is used to obtain consumer behavior data; perform consumption behavior cluster analysis based on the consumer behavior data to obtain consumption behavior cluster data; perform consumption behavior influencing factor analysis on the consumption behavior cluster data to obtain consumption behavior influencing factor data;

[0165] The consumer behavior cognitive load division module is used to analyze the consumer behavior cognitive load according to the consumer behavior influencing factor data to obtain the consumer behavior cognitive load data; divide the consumer behavior cognitive load data into group structures to obtain the consumer behavior cognitive group structure data;

[0166] The consumer behavior cognitive prediction model construction module is used to perform a difference analysis between different cognitive group structures on the consumer behavior cognitive group structure data to obtain cognitive group structure difference data; perform multi-view structure learning on the consumer behavior cognitive group structure data based on the cognitive group structure difference data to obtain cognitive group multi-view structure data; and construct a consumer consumption behavior cognitive prediction model based on the cognitive group multi-view structure data based on the support vector machine to obtain a consumer behavior cognitive prediction model;

[0167] The consumer consumption behavior prediction module is used to predict and process the consumer psychological activity consumption behavior according to the consumer behavior cognitive prediction model to obtain the consumer consumption behavior prediction data.

[0168] The beneficial effects of the present invention are as follows: the present invention provides a consumer behavior prediction system based on psychological information analysis, which is an optimization process of the traditional consumer behavior prediction system based on psychological information analysis, and solves the problems of inaccurate consumer behavior prediction and low prediction processing efficiency in the traditional consumer behavior prediction method based on psychological information analysis, thereby improving the accuracy of the prediction and the efficiency of the prediction processing capability.

[0169] Therefore, the embodiments should be regarded as illustrative and non-restrictive from all points, and the scope of the present invention is limited by the appended claims rather than the above description, and it is therefore intended that all changes falling within the meaning and range of equivalent elements of the application documents are included in the present invention.

[0170] The above description is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A consumer behavior prediction method based on psychological information analysis, characterized in that: The following steps are involved: Step S1: Obtain consumer behavior data; Conducting consumer behavior cluster analysis based on consumer behavior data to obtain consumer behavior cluster data; conducting consumer behavior influencing factor analysis on consumer behavior cluster data to obtain consumer behavior influencing factor data; Step S2: performing cognitive load analysis on consumer behavior based on consumer behavior influencing factor data to obtain consumer behavior cognitive load data; performing group structure division on the consumer behavior cognitive load data to obtain consumer behavior cognitive group structure data; Step S3: performing a difference analysis between different cognitive group structures on the consumer behavior cognitive group structure data to obtain cognitive group structure difference data; performing multi-view structure learning on the consumer behavior cognitive group structure data based on the cognitive group structure difference data to obtain cognitive group multi-view structure data; Based on the support vector machine, the consumer behavior cognitive prediction model is constructed for the multi-view structure data of the cognitive group to obtain the consumer behavior cognitive prediction model; Step S4: Predict and process the consumer's psychological activity consumer behavior according to the consumer behavior cognitive prediction model to obtain consumer consumer behavior prediction data.

2. The method for predicting consumer behavior based on psychological information analysis according to claim 1, characterized in that: Step S1 includes the following steps: Step S11: Obtaining consumer behavior data; Step S12: performing consumption environment analysis based on the consumer behavior data to obtain consumption environment data; performing consumption behavior cluster analysis based on the consumer consumption behavior data to obtain consumption behavior cluster data; Step S13: performing a consumer behavior influencing factor analysis on the consumer behavior clustering data according to the consumer environment data to obtain consumer behavior influencing factor data.

3. The method for predicting consumer behavior based on psychological information analysis according to claim 2, characterized in that: Step S13 includes the following steps: Step S131: marking the cluster center of the consumer behavior cluster data to obtain the consumer behavior cluster center data; calculating the inter-cluster distance of the consumer behavior cluster data according to the consumer behavior cluster center data to obtain the consumer behavior inter-cluster distance data; Step S132: performing clustering radius calculation on the distance data between consumer behavior groups according to the consumer behavior group center data to obtain consumer behavior clustering radius data; performing consumer behavior distribution dispersion analysis according to the consumer behavior clustering radius data to obtain consumer behavior distribution dispersion data; Step S133: performing environment-related mapping on the consumption behavior distribution discreteness data according to the consumption environment data to obtain environment-related behavior distribution discreteness data; Step S134: performing environment-consumption behavior factor analysis on the environment-related behavior distribution discrete data to obtain environment-consumption behavior factor data; Step S135: constructing an environment-consumption behavior structural factor correlation matrix for the environment-related behavior distribution discrete data according to the environment-consumption behavior factor data to obtain an environment-consumption structural factor matrix; performing matrix structure orthogonal rotation on the environment-consumption structural factor matrix to obtain factor matrix structure simplified data; Step S136: performing factor loading analysis on the environment-consumption structure factor matrix according to the simplified data of the factor matrix structure to obtain the environment-consumption structure factor loading data; Step S137: Perform consumption behavior influencing factor analysis based on the environment-consumption structure factor load data and the environment-related behavior distribution discrete data to obtain consumption behavior influencing factor data.

4. The method for predicting consumer behavior based on psychological information analysis according to claim 1, characterized in that: Step S2 includes the following steps: Step S21: performing consumption orientation analysis on the consumption behavior clustering data to obtain consumption behavior orientation data; Step S22: performing consumer behavior cognitive load analysis based on consumer behavior orientation data and consumer behavior influencing factor data to obtain consumer behavior cognitive load data; Step S23: performing cognitive load bias analysis on the consumer behavior cognitive load data to obtain cognitive load bias data; Step S24: dividing the consumer behavior cognitive load data into groups according to the cognitive load bias data to obtain consumer behavior cognitive group structure data.

5. The method for predicting consumer behavior based on psychological information analysis according to claim 3, characterized in that: Step S22 includes the following steps: Step S221: performing linear structure analysis on the consumption behavior influencing factor data to obtain the influencing factor linear structure data; Step S222: constructing a linear matrix of consumer behavior influencing factors based on the linear structure data of influencing factors, and obtaining a linear matrix of influencing factors; Step S223: Calculate the factor loads of the impact factor linear matrix to obtain the impact factor load data; perform factor structure rotation adjustment on the impact factor linear matrix according to the impact factor load data to obtain the impact factor linear adjustment matrix; Step S224: using the impact factor contribution calculation formula to calculate the impact factor contribution of the impact factor linear adjustment matrix and the impact factor load data to obtain the impact factor contribution data; Step S225: Perform consumer behavior cognitive load analysis based on consumer behavior orientation data, influencing factor load data, and influencing factor co-contribution data to obtain consumer behavior cognitive load data.

6. The method for predicting consumer behavior based on psychological information analysis according to claim 5, characterized in that: The calculation formula of impact factor contribution is as follows: In the formula, C represents the result value of the co-contribution of the influencing factors, t represents the extreme value of the calculation time, X represents the structural complexity of the linear adjustment matrix of the influencing factors, Y represents the linear structural correlation coefficient of the linear adjustment matrix of the influencing factors, Z represents the load difference between different influencing factors in the influencing factor load data, β represents the dimensionality coefficient of the linear adjustment matrix of the influencing factors, γ represents the load range interval of the influencing factor load data, δ represents the cross-load intensity value of the influencing factor load data; φ represents the element correlation of the linear adjustment matrix of the influencing factors, and R represents the error correction value of the calculation formula of the influencing factor contribution.

7. The method for predicting consumer behavior based on psychological information analysis according to claim 4, characterized in that: Step S24 includes the following steps: Step S241: performing cognitive load Bayesian statistical analysis on the cognitive load data of consumer behavior to obtain cognitive load statistical data; and drawing a load statistical curve on the cognitive load statistical data to obtain a cognitive load statistical curve; Step S242: extracting inflection points of the cognitive load statistical curve to obtain curve inflection point data; performing time series trend decomposition on the cognitive load statistical curve according to the curve inflection point data to obtain cognitive load time series trend data; Step S243: calculating the autocorrelation coefficient of the cognitive load statistical curve according to the curve inflection point data and the cognitive load time series trend data to obtain the cognitive load autocorrelation coefficient; Step S244: performing a hysteresis effect evaluation on the cognitive load statistical curve according to the cognitive load autocorrelation coefficient to obtain cognitive load hysteresis effect data; Step S245: performing biased hysteresis effect mapping on the cognitive load hysteresis effect data according to the cognitive load bias data to obtain cognitive load biased hysteresis effect data; Step S246: Divide the consumer behavior cognitive load data into group structures according to the cognitive load bias lag effect data and the cognitive load autocorrelation coefficient to obtain consumer behavior cognitive group structure data.

8. The method for predicting consumer behavior based on psychological information analysis according to claim 1, characterized in that: Step S3 includes the following steps: Step S31: performing a difference analysis between different cognitive group structures on the consumer behavior cognitive group structure data to obtain cognitive group structure difference data; Step S32: performing topic modeling between the consumer behavior cognitive group structure data and the cognitive group structure data according to the cognitive group structure difference data to obtain cognitive group topic difference data; Step S33: performing multi-view structural learning on the consumer behavior cognitive group structural data based on the cognitive group subject difference data and the cognitive group structural difference data to obtain cognitive group multi-view structural data; Step S34: constructing a consumer consumption behavior cognitive prediction model for the cognitive group multi-view structure data based on a support vector machine to obtain a consumer behavior cognitive prediction model.

9. The method for predicting consumer behavior based on psychological information analysis according to claim 8, characterized in that: Step S34 includes the following steps: Step S341: extracting multi-view structural parameters from the cognitive group multi-view structural data to obtain cognitive group multi-view structural parameters; Step S342: performing recursive feature elimination processing on the cognitive group multi-view structure parameters to obtain cognitive group recursive feature data; Step S343: performing recursive feature multidimensional scaling transformation on the cognitive group multi-view structure parameters according to the cognitive group recursive feature data to obtain cognitive group multidimensional scaling transformation parameters; Step S344: constructing a consumer behavior cognitive prediction model based on the multidimensional scaling transformation parameters of the cognitive group based on the support vector machine to obtain a consumer behavior cognitive prediction model.

10. A consumer behavior prediction system based on psychological information analysis, characterized in that: Used to execute the consumer behavior prediction method based on psychological information analysis as claimed in claims 1 to 9, the consumer behavior prediction system based on psychological information analysis comprises: The consumption behavior influencing factor analysis module is used to obtain consumer behavior data; perform consumption behavior cluster analysis based on the consumer behavior data to obtain consumption behavior cluster data; perform consumption behavior influencing factor analysis on the consumption behavior cluster data to obtain consumption behavior influencing factor data; The consumer behavior cognitive load division module is used to analyze the consumer behavior cognitive load according to the consumer behavior influencing factor data to obtain the consumer behavior cognitive load data; divide the consumer behavior cognitive load data into group structures to obtain the consumer behavior cognitive group structure data; The consumer behavior cognitive prediction model construction module is used to perform a difference analysis between different cognitive group structures on the consumer behavior cognitive group structure data to obtain cognitive group structure difference data; perform multi-view structure learning on the consumer behavior cognitive group structure data based on the cognitive group structure difference data to obtain cognitive group multi-view structure data; and construct a consumer consumption behavior cognitive prediction model based on the cognitive group multi-view structure data based on the support vector machine to obtain a consumer behavior cognitive prediction model; The consumer consumption behavior prediction module is used to predict and process the consumer psychological activity consumption behavior according to the consumer behavior cognitive prediction model to obtain the consumer consumption behavior prediction data.

Images