AI-based cultural product creative customization data acquisition and analysis method

Abstract

The invention relates to the technical field of data processing, in particular to an AI-based cultural product creative customization data acquisition and analysis method, which comprises the steps of data analysis target determination, data collection, data preprocessing, feature engineering, model selection and training, result analysis and result application. Compared with the prior art in which a single data analysis method is mainly adopted, user preferences and market trends are difficult to comprehensively capture, and creative customization is lack of accuracy and individuation, the scheme adopts a multi-dimensional analysis and application strategy integrating user portrait construction, market demand analysis and creative inspiration excitation, so that the creative customization accuracy is improved. By integrating basic information, behavior data and market trends of users, user preferences can be deeply understood, market demand changes can be accurately predicted, creative elements are effectively mined, a highly-personalized creative scheme is generated, and the creative scheme remarkably improves the precision and market adaptability of cultural product creative customization and has a good market prospect. Powerful support is provided for innovative development of the cultural industry.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to an AI-based method for collecting and analyzing data on creative customization of cultural products. Background Art

[0002] In traditional methods, the creative customization of cultural products mainly relies on the designer's personal experience, market intuition or simple data analysis. However, these methods have obvious limitations: First, a single data analysis method often fails to fully capture user preferences and market dynamics. User preferences are diverse, encompassing basic information, behavioral habits, interests, and hobbies. Traditional data analysis methods may focus on only one or a few of these aspects, resulting in a one-sided understanding of user needs. Furthermore, market dynamics are complex and ever-changing, encompassing multiple factors such as historical sales data, market trends, and competitor strategies. These factors interact and collectively influence market demand for cultural products. A single data analysis method often struggles to accurately reflect these complex relationships, resulting in a lack of precision and personalization in creative customization.

[0003] Secondly, the lack of a systematic data integration and analysis strategy also hinders the development of creative customization of cultural products. Traditionally, user data, sales data, and market trend data are often scattered across disparate systems, lacking a unified data management and integration mechanism. This makes it difficult to effectively connect and interact with data, limiting the in-depth exploration and utilization of data value. Furthermore, due to a lack of scientific data analysis tools and methods, designers often struggle to extract valuable information from massive amounts of data to guide the creative customization process. Summary of the Invention

[0004] In order to overcome the problems raised in the above background technology, the present invention proposes an AI-based method for collecting and analyzing data on creative customization of cultural products.

[0005] The technical solution of the present invention is: an AI-based method for collecting and analyzing data on creative customization of cultural products, comprising the following steps: S11: Clarify the data analysis objectives and determine the specific goals of data analysis, including understanding market demand, consumer preferences, competitor situations, or popular trends of cultural products; S12: Data collection: collect data on creative customization of cultural products through the set data collection methods; S13: Data preprocessing, preprocessing the collected data, wherein the preprocessing methods used include data cleaning and data normalization; S14: Feature engineering, extracting and creating required feature data from preprocessed data, specifically including feature selection, feature extraction and feature conversion; S15: Model selection and training: select appropriate machine learning algorithms and conduct training based on data characteristics and analysis objectives; S16: Result analysis: input the extracted and created feature data into the trained model to analyze the data; S17: Result application, analyze and apply the analysis results of the model.

[0006] Preferably, when collecting data on creative customization of cultural products through the specified data collection means, the data collection means used include: A11: Questionnaire survey, design questionnaires, and understand consumer information, including consumer preferences and purchasing intentions for different types of cultural products; A12: Social media analysis: By monitoring the discussion volume, likes, and reposts of relevant keywords, we can understand hot topics and trends. A13: Sales data: collect and analyze sales data, including sales volume, product sales, and customer purchasing behavior; A14: Competitor analysis: collect competitors’ product information, pricing strategies, and market activity data for competitive analysis.

[0007] Preferably, the preprocessing of the collected data specifically includes: S21: Process missing values ​​by filling in missing values ​​and deleting missing values. The data filling methods include filling in missing values ​​with statistics, interpolation, and regression prediction. S22: Remove duplicate data by using the deduplication function in the database to remove duplicate data in the data; S23: Remove outliers, use statistics to filter and remove outliers in the data; S24: Data conversion, identifying and recording the data types of all data, and converting all data into the set data types, including converting text data into numerical data; S25: Data normalization, converting data of different dimensions to a unified scale. The data normalization methods used include the minimum-maximum normalization method and the Z-score normalization method. The principle formula of the minimum-maximum normalization method is: ; in, is the original data, is the normalized data, and are the minimum and maximum values ​​of the data respectively; The principle formula of the score standardization method is: ; in, is the original data, is the standardized data, is the mean of the data, is the standard deviation of the data.

[0008] Preferably, when extracting and creating required feature data from the preprocessed data, the following steps are specifically included: S31: Feature selection, using one of the chi-square test, correlation coefficient and mutual information method to evaluate the correlation between features and target variables and perform feature selection; S32: Extract required feature data from raw data through text analysis technology, image processing technology, and AI recognition technology; S33: Feature transformation, using normalization and standardization methods to transform feature data so that the feature values ​​of all feature data are on the same scale; S34: Feature construction, constructing new features by combining existing features, applying mathematical transformations, and introducing external data, and selecting features to remove redundant features.

[0009] As a priority, when selecting and training an appropriate machine learning algorithm based on the characteristics of the data and the analysis objectives, specifically including: S41: Select appropriate machine learning algorithms based on data characteristics and analysis objectives. Types of machine learning algorithms include linear regression, decision trees, support vector machines, and neural networks. S42: Divide the dataset into a training set and a test set, and use the training set to train the model; S43: Adjust model parameters and use techniques such as cross-validation to avoid overfitting.

[0010] Preferably, when parsing and applying the analysis results of the model, the following steps are specifically included: S51: User portrait construction, based on the user's basic information and behavior data, build user portrait model and interest portrait model; S52: Market demand analysis: using AI algorithms to analyze historical sales data and market trends, predict future changes in demand for cultural products, and identify preference differences among different user groups by combining user profiles and interest profiles; S53: Creative inspiration stimulation and mining, extracting creative elements from the database based on the results of user portrait model, interest portrait model and market demand analysis, and generating creative solutions based on the creative elements.

[0011] Preferably, when constructing a user portrait model and an interest portrait model based on the user's basic information and behavior data, the following steps are specifically included: S61: Model construction and training: Based on the analysis results, use data mining to build user portrait models and interest portrait models, and train the constructed models; S62: Profile generation: Input the analysis results into the constructed user profile model and interest profile model to generate user profiles and interest profiles; S63: Continuous updates: regularly collect new data, retrain models, and update user and interest profiles.

[0012] As a priority, when using AI algorithms to analyze historical sales data and market trends, predict future changes in demand for cultural products, and combine user profiles and interest profiles to identify preference differences among different user groups, specifically including: S71: Historical data analysis: AI algorithms are used to conduct in-depth analysis of historical sales data to identify sales trends, popular product categories, and seasonal changes. S72: Market trend forecasting: using a time series-based forecasting model to predict future market trends for cultural products. Market trend indicators include market demand, market share, and market growth rate. The market demand is calculated as follows: ; Among them, D is the market demand, A is the number of potential users, B is the average consumption frequency, and C is the average consumption amount; S73: User preference analysis, through cluster analysis and association rule mining methods, based on user portraits and interest portraits, the preferences of different user groups are segmented and analyzed.

[0013] Preferably, when extracting creative elements from a database based on the user portrait model, interest portrait model, and market demand analysis results, and generating a creative solution based on the creative elements, the process specifically includes: S81: Data integration and analysis: integrating the user portrait model, interest portrait model, and market demand analysis results to form comprehensive user and market insights; S82: Creative element extraction: Based on the analysis results, creative elements related to user needs, market trends and interest preferences are screened from the database; S83: Fusion and innovation of creative elements: fusion and innovation of extracted creative elements to form new creative concepts and design solutions; S84: Creative solution generation, combining the integrated creative elements and creative design methods to generate multiple creative solutions.

[0014] Preferably, when screening out creative elements related to user needs, market trends and interest preferences from the database based on the analysis results, the following are specifically included: S91: Collect and organize data on user preferences, differences among user groups, and consumption by different groups; S92: Use statistical methods and data analysis tools to calculate the user preference weight, user group distinction weight, and consumption weight of different groups for each creative element. S93: Sort and filter the creative elements based on these weights, and select the creative elements that best meet market demands and user preferences.

[0015] Beneficial effects of the present invention: 1. Compared with existing technologies that primarily use a single data analysis method, which makes it difficult to fully capture user preferences and market dynamics, resulting in a lack of precision and personalization in creative customization, this solution adopts a multi-dimensional analysis and application strategy that integrates user portrait construction, market demand analysis, and creative inspiration. By integrating basic user information, behavioral data, historical sales data, and market trends, this solution can more deeply understand user preferences, accurately predict changes in market demand, and effectively explore creative elements to generate highly personalized creative solutions. This innovative solution significantly improves the precision and market adaptability of creative customization of cultural products, providing strong support for the innovative development of the cultural industry; 2. Compared with existing technologies that mainly use static user portrait construction methods, lack a real-time update mechanism, and the analysis of historical sales data and market trends is relatively superficial, making it difficult to accurately capture the dynamic changes in user preferences. This solution adopts a dynamically updated user portrait and interest portrait construction strategy, combined with in-depth historical sales data analysis and market trend forecasting technology. By regularly collecting new data and retraining the model, this solution ensures the timeliness and accuracy of user portraits and interest portraits. At the same time, it uses AI algorithms to deeply mine historical sales data, combines time series forecasting models to accurately predict future market trends, and uses cluster analysis and association rule mining methods to carefully analyze the preference differences of different user groups. This innovative solution not only improves the accuracy and practicality of user portraits, but also enhances the reliability of market forecasts and the depth of understanding of user preferences, providing more scientific and comprehensive data support for the creative customization of cultural products. 3. Compared with the existing technology that mainly uses intuitive judgment or simple data matching to screen creative elements, which lacks systematicness and precision and makes it difficult to accurately capture the deep connection between users and market needs, this solution adopts a comprehensive data integration and analysis strategy based on user portraits, interest portraits and market demand analysis. By deeply mining user preferences, user group differences and consumption data, and using statistical methods and data analysis tools to accurately calculate the weights of creative elements, this solution realizes the scientific sorting and precise screening of creative elements. Furthermore, the screened creative elements are integrated and innovated to generate multiple creative solutions that are highly in line with market needs and user preferences. This innovative solution not only significantly improves the accuracy and efficiency of creative element screening, but also greatly enhances the pertinence and market competitiveness of creative solutions, providing a more scientific and efficient design path for the creative customization of cultural products. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 Shown is a workflow diagram of the AI-based cultural product creative customization data collection and analysis method of the present invention; Figure 2 What is shown is a flow chart of the application of the results in the AI-based cultural product creative customization data collection and analysis method of the present invention. DETAILED DESCRIPTION

[0017] The present invention will be further described below with reference to the accompanying drawings and examples.

[0018] See also Figure 1-2 The present invention provides an embodiment: an AI-based method for collecting and analyzing data on creative customization of cultural products, comprising the following steps: S11: Clarify the data analysis objectives and determine the specific goals of data analysis, including understanding market demand, consumer preferences, competitor situations, or popular trends of cultural products; S12: Data collection: collect data on creative customization of cultural products through the set data collection methods; S13: Data preprocessing, preprocessing the collected data, wherein the preprocessing methods used include data cleaning and data normalization; S14: Feature engineering, extracting and creating required feature data from preprocessed data, specifically including feature selection, feature extraction and feature conversion; S15: Model selection and training: select appropriate machine learning algorithms and conduct training based on data characteristics and analysis objectives; S16: Result analysis: input the extracted and created feature data into the trained model to analyze the data; S17: Result application, analyze and apply the analysis results of the model.

[0019] As described above, the present invention effectively improves the accuracy and efficiency of creative customization of cultural products through a systematic AI data collection and analysis process. From clarifying the analysis objectives to data collection, preprocessing, feature engineering, and then to model selection and training, each step is closely centered around core elements such as market demand, consumer preferences, competitor conditions, and popular trends, ensuring the pertinence and practicality of data analysis. Ultimately, through the analysis and application of results, it not only provides a scientific basis for the creative customization of cultural products, but also promotes differentiated competition and the improvement of market adaptability of products, realizing the innovation and development of cultural products driven by data.

[0020] Preferably, when collecting data on creative customization of cultural products through the specified data collection means, the data collection means used include: A11: Questionnaire survey, design questionnaires, and understand consumer information, including consumer preferences and purchasing intentions for different types of cultural products; A12: Social media analysis: By monitoring the discussion volume, likes, and reposts of relevant keywords, we can understand hot topics and trends. A13: Sales data: collect and analyze sales data, including sales volume, product sales, and customer purchasing behavior; A14: Competitor analysis: collect competitors’ product information, pricing strategies, and market activity data for competitive analysis.

[0021] As described above, this invention utilizes a diverse set of data collection methods, including questionnaires to gain a deep understanding of consumer preferences, social media analysis to capture trending topics and trends, sales data analysis to gain insights into market feedback and purchasing behavior, and competitor analysis to understand market dynamics and competitive landscapes. These comprehensive data collection methods not only ensure comprehensiveness and accuracy but also significantly enhance the market sensitivity and responsiveness of creative customization of cultural products, providing strong data support for precise product positioning, optimized design, and marketing strategies.

[0022] Preferably, the preprocessing of the collected data specifically includes: S21: Process missing values ​​by filling in missing values ​​and deleting missing values. The data filling methods include filling in missing values ​​with statistics, interpolation, and regression prediction. S22: Remove duplicate data by using the deduplication function in the database to remove duplicate data in the data; S23: Remove outliers, use statistics to filter and remove outliers in the data; S24: Data conversion, identifying and recording the data types of all data, and converting all data into the set data types, including converting text data into numerical data; S25: Data normalization, converting data of different dimensions to a unified scale. The data normalization methods used include the minimum-maximum normalization method and the Z-score normalization method. The principle formula of the minimum-maximum normalization method is: ; in, is the original data, is the normalized data, and are the minimum and maximum values ​​of the data respectively; The principle formula of the score standardization method is: ; in, is the original data, is the standardized data, is the mean of the data, is the standard deviation of the data.

[0023] As described above, during the data preprocessing stage, the present invention effectively improves the accuracy and comparability of the data by carefully processing missing values ​​(using data filling and deletion strategies to ensure data integrity), removing duplicate data (using database functions to improve data quality), screening and removing outliers (enhancing data reliability), unifying data types (to facilitate subsequent analysis and processing), and data normalization (using minimum-maximum normalization and Z-score standardization methods to eliminate dimensional effects), laying a solid foundation for subsequent feature engineering, model training, and result analysis, thereby ensuring the accuracy and effectiveness of the creative customization data analysis of cultural products.

[0024] Preferably, when extracting and creating required feature data from the preprocessed data, the following steps are specifically included: S31: Feature selection, using one of the chi-square test, correlation coefficient and mutual information method to evaluate the correlation between features and target variables and perform feature selection; S32: Extract required feature data from raw data through text analysis technology, image processing technology, and AI recognition technology; S33: Feature transformation, using normalization and standardization methods to transform feature data so that the feature values ​​of all feature data are on the same scale; S34: Feature construction, constructing new features by combining existing features, applying mathematical transformations, and introducing external data, and selecting features to remove redundant features.

[0025] As described above, the present invention accurately screens features that are highly correlated with the target variable through methods such as the chi-square test, correlation coefficient, and mutual information, and uses text analysis, image processing, and AI recognition technology to efficiently extract key feature data. It further implements feature conversion to ensure that all features are comparable on the same scale, and innovatively combines existing features through feature construction, applies mathematical transformations, and introduces external data to generate new features and eliminate redundancy. This series of sophisticated operations greatly enriches the feature set, improves the effectiveness of the features and the predictive performance of the model, and provides more accurate data insights for the creative customization of cultural products.

[0026] As a priority, when selecting and training an appropriate machine learning algorithm based on the characteristics of the data and the analysis objectives, specifically including: S41: Select appropriate machine learning algorithms based on data characteristics and analysis objectives. Types of machine learning algorithms include linear regression, decision tree, support vector machine, and neural network. S42: Divide the dataset into a training set and a test set, and use the training set to train the model; S43: Adjust model parameters and use techniques such as cross-validation to avoid overfitting.

[0027] As described above, during the model selection and training stage, the present invention carefully selects suitable machine learning algorithms (such as linear regression, decision trees, support vector machines, and neural networks) based on data characteristics and analysis objectives, ensures the adequacy of model training and the effectiveness of verification by scientifically dividing the training set and test set, and cleverly uses cross-validation and other techniques to fine-tune model parameters to effectively prevent the risk of overfitting. This series of measures significantly improves the generalization ability and prediction accuracy of the model, and provides strong technical support for data analysis for creative customization of cultural products.

[0028] Preferably, when parsing and applying the analysis results of the model, the following steps are specifically included: S51: User portrait construction, based on the user's basic information and behavior data, build user portrait model and interest portrait model; S52: Market demand analysis: using AI algorithms to analyze historical sales data and market trends, predict future changes in demand for cultural products, and identify preference differences among different user groups by combining user profiles and interest profiles; S53: Creative inspiration stimulation and mining, extracting creative elements from the database based on the results of user portrait model, interest portrait model and market demand analysis, and generating creative solutions based on the creative elements.

[0029] As described above, compared with the existing technology, which mainly adopts a single data analysis method, it is difficult to fully capture user preferences and market dynamics, resulting in a lack of accuracy and personalization in creative customization. This solution adopts a multi-dimensional analysis and application strategy that integrates user portrait construction, market demand analysis, and creative inspiration. By integrating basic user information, behavioral data, historical sales data, and market trends, this solution can more deeply understand user preferences, accurately predict changes in market demand, and effectively tap into creative elements to generate highly personalized creative solutions. This innovative solution significantly improves the accuracy and market adaptability of creative customization of cultural products, providing strong support for the innovative development of the cultural industry.

[0030] Preferably, when constructing a user portrait model and an interest portrait model based on the user's basic information and behavior data, the following steps are specifically included: S61: Model construction and training: Based on the analysis results, use data mining to build user portrait models and interest portrait models, and train the constructed models; S62: Profile generation: Input the analysis results into the constructed user profile model and interest profile model to generate user profiles and interest profiles; S63: Continuous updates: regularly collect new data, retrain models, and update user and interest profiles.

[0031] As a priority, when using AI algorithms to analyze historical sales data and market trends, predict future changes in demand for cultural products, and combine user profiles and interest profiles to identify preference differences among different user groups, specifically including: S71: Historical data analysis: AI algorithms are used to conduct in-depth analysis of historical sales data to identify sales trends, popular product categories, and seasonal changes. S72: Market trend forecasting: using a time series-based forecasting model to predict future market trends for cultural products. Market trend indicators include market demand, market share, and market growth rate. The market demand is calculated as follows: ; Among them, D is the market demand, A is the number of potential users, B is the average consumption frequency, and C is the average consumption amount; S73: User preference analysis, through cluster analysis and association rule mining methods, based on user portraits and interest portraits, the preferences of different user groups are segmented and analyzed.

[0032] As described above, compared with the prior art, the present invention mainly adopts a static user portrait construction method, lacks a real-time update mechanism, and the analysis of historical sales data and market trends is relatively superficial, making it difficult to accurately capture the dynamic changes in user preferences. This solution adopts a dynamically updated user portrait and interest portrait construction strategy, combined with in-depth historical sales data analysis and market trend forecasting technology. By regularly collecting new data and retraining the model, this solution ensures the timeliness and accuracy of user portraits and interest portraits; at the same time, it uses AI algorithms to deeply mine historical sales data, combines time series forecasting models to accurately predict future market trends, and uses cluster analysis and association rule mining methods to carefully analyze the preference differences of different user groups. This innovative solution not only improves the accuracy and practicality of user portraits, but also enhances the reliability of market forecasts and the depth of understanding of user preferences, providing more scientific and comprehensive data support for the creative customization of cultural products.

[0033] Preferably, when extracting creative elements from a database based on the user portrait model, interest portrait model, and market demand analysis results, and generating a creative solution based on the creative elements, the process specifically includes: S81: Data integration and analysis: integrating the user portrait model, interest portrait model, and market demand analysis results to form comprehensive user and market insights; S82: Creative element extraction: Based on the analysis results, creative elements related to user needs, market trends and interest preferences are screened from the database; S83: Fusion and innovation of creative elements: fusion and innovation of extracted creative elements to form new creative concepts and design solutions; S84: Creative solution generation, combining the integrated creative elements and creative design methods to generate multiple creative solutions.

[0034] Preferably, when screening out creative elements related to user needs, market trends and interest preferences from the database based on the analysis results, the following are specifically included: S91: Collect and organize data on user preferences, differences among user groups, and consumption by different groups; S92: Use statistical methods and data analysis tools to calculate the user preference weight, user group distinction weight, and consumption weight of different groups for each creative element. S93: Sort and filter the creative elements based on these weights, and select the creative elements that best meet market demands and user preferences.

[0035] As described above, compared with the prior art, the present invention mainly uses intuitive judgment or simple data matching to screen creative elements, which lacks systematicity and precision, and it is difficult to accurately capture the deep connection between users and market needs. This solution adopts a comprehensive data integration and analysis strategy based on user portraits, interest portraits and market demand analysis. By deeply mining user preferences, user group differences and consumption data, and using statistical methods and data analysis tools to accurately calculate the weights of creative elements, this solution achieves scientific sorting and precise screening of creative elements. Furthermore, the screened creative elements are integrated and innovated to generate multiple creative solutions that are highly in line with market needs and user preferences. This innovative solution not only significantly improves the accuracy and efficiency of creative element screening, but also greatly enhances the pertinence and market competitiveness of creative solutions, providing a more scientific and efficient design path for the creative customization of cultural products.

[0036] The embodiments of the present invention are described in detail above with reference to the accompanying drawings. However, the present invention is not limited to the above embodiments. Various changes can be made within the scope of knowledge of those skilled in the art without departing from the spirit of the present invention.

Claims

1. An AI-based method for collecting and analyzing data on creative customization of cultural products, characterized by: The following steps are included: S11: Clarify the data analysis objectives and determine the specific goals of data analysis, including understanding market demand, consumer preferences, competitor situations, or popular trends of cultural products; S12: Data collection: collect data on creative customization of cultural products through the set data collection methods; S13: Data preprocessing, preprocessing the collected data, wherein the preprocessing methods used include data cleaning and data normalization; S14: Feature engineering, extracting and creating required feature data from preprocessed data, specifically including feature selection, feature extraction and feature conversion; S15: Model selection and training: select appropriate machine learning algorithms and conduct training based on data characteristics and analysis objectives; S16: Result analysis: input the extracted and created feature data into the trained model to analyze the data; S17: Result application, analyze and apply the analysis results of the model.

2. The AI-based cultural product creative customization data collection and analysis method according to claim 1 is characterized by: When collecting data on creative customization of cultural products through the established data collection methods, the data collection methods used include: A11: Questionnaire survey, design questionnaires, and understand consumer information, including consumer preferences and purchasing intentions for different types of cultural products; A12: Social media analysis: By monitoring the discussion volume, likes, and reposts of relevant keywords, we can understand hot topics and trends. A13: Sales data: collect and analyze sales data, including sales volume, product sales, and customer purchasing behavior; A14: Competitor analysis: collect competitors’ product information, pricing strategies, and market activity data for competitive analysis.

3. The AI-based cultural product creative customization data collection and analysis method according to claim 2 is characterized by: When preprocessing the collected data, it specifically includes: S21: Process missing values ​​by filling in missing values ​​and deleting missing values. The data filling methods include filling in missing values ​​with statistics, interpolation, and regression prediction. S22: Remove duplicate data by using the deduplication function in the database to remove duplicate data in the data; S23: Remove outliers, use statistics to filter and remove outliers in the data; S24: Data conversion, identifying and recording the data types of all data, and converting all data into the set data types, including converting text data into numerical data; S25: Data normalization, converting data of different dimensions to a unified scale. The data normalization methods used include the minimum-maximum normalization method and the Z-score normalization method. The principle formula of the minimum-maximum normalization method is: ; in, is the original data, is the normalized data, and are the minimum and maximum values ​​of the data respectively; The principle formula of the score standardization method is: ; in, is the original data, is the standardized data, is the mean of the data, is the standard deviation of the data.

4. The AI-based cultural product creative customization data collection and analysis method according to claim 3 is characterized by: When extracting and creating the required feature data from the preprocessed data, it specifically includes: S31: Feature selection, using one of the chi-square test, correlation coefficient and mutual information method to evaluate the correlation between features and target variables and perform feature selection; S32: Extract required feature data from raw data through text analysis technology, image processing technology, and AI recognition technology; S33: Feature transformation, using normalization and standardization methods to transform feature data so that the feature values ​​of all feature data are on the same scale; S34: Feature construction, constructing new features by combining existing features, applying mathematical transformations, and introducing external data, and selecting features to remove redundant features.

5. The AI-based cultural product creative customization data collection and analysis method according to claim 4 is characterized by: When selecting and training an appropriate machine learning algorithm based on the characteristics of the data and the analysis objectives, the following are the steps: S41: Select appropriate machine learning algorithms based on data characteristics and analysis objectives. Types of machine learning algorithms include linear regression, decision trees, support vector machines, and neural networks. S42: Divide the dataset into a training set and a test set, and use the training set to train the model; S43: Adjust model parameters and use techniques such as cross-validation to avoid overfitting.

6. The AI-based method for collecting and analyzing data on creative customization of cultural products according to claim 5, characterized in that: When analyzing and applying the analysis results of the model, it specifically includes: S51: User portrait construction, based on the user's basic information and behavior data, build user portrait model and interest portrait model; S52: Market demand analysis: using AI algorithms to analyze historical sales data and market trends, predict future changes in demand for cultural products, and identify preference differences among different user groups by combining user profiles and interest profiles; S53: Creative inspiration stimulation and mining, extracting creative elements from the database based on the results of user portrait model, interest portrait model and market demand analysis, and generating creative solutions based on the creative elements.

7. The AI-based method for collecting and analyzing data on creative customization of cultural products according to claim 6, characterized in that: When building user portrait models and interest portrait models based on users' basic information and behavior data, the following are specifically included: S61: Model construction and training: Based on the analysis results, use data mining to build user portrait models and interest portrait models, and train the constructed models; S62: Profile generation: Input the analysis results into the constructed user profile model and interest profile model to generate user profiles and interest profiles; S63: Continuous updates: regularly collect new data, retrain models, and update user and interest profiles.

8. The AI-based cultural product creative customization data collection and analysis method according to claim 7 is characterized by: When using AI algorithms to analyze historical sales data and market trends, predict future changes in demand for cultural products, and combine user profiles and interest profiles to identify preference differences among different user groups, this includes: S71: Historical data analysis: AI algorithms are used to conduct in-depth analysis of historical sales data to identify sales trends, popular product categories, and seasonal changes. S72: Market trend forecasting: using a time series-based forecasting model to predict future market trends for cultural products. Market trend indicators include market demand, market share, and market growth rate. The market demand is calculated as follows: ; Among them, D is the market demand, A is the number of potential users, B is the average consumption frequency, and C is the average consumption amount; S73: User preference analysis, through cluster analysis and association rule mining methods, based on user portraits and interest portraits, the preferences of different user groups are segmented and analyzed.

9. The AI-based cultural product creative customization data collection and analysis method according to claim 8, characterized in that: When extracting creative elements from the database based on the user portrait model, interest portrait model, and market demand analysis results, and generating creative solutions based on the creative elements, specifically including: S81: Data integration and analysis: integrating the user portrait model, interest portrait model, and market demand analysis results to form comprehensive user and market insights; S82: Creative element extraction: Based on the analysis results, creative elements related to user needs, market trends and interest preferences are screened from the database; S83: Fusion and innovation of creative elements: fusion and innovation of extracted creative elements to form new creative concepts and design solutions; S84: Creative solution generation, combining the integrated creative elements and creative design methods to generate multiple creative solutions.

10. The AI-based cultural product creative customization data collection and analysis method according to claim 9 is characterized by: When screening creative elements related to user needs, market trends and interest preferences from the database based on the analysis results, specifically including: S91: Collect and organize data on user preferences, differences among user groups, and consumption by different groups; S92: Use statistical methods and data analysis tools to calculate the user preference weight, user group distinction weight, and consumption weight of different groups for each creative element. S93: Sort and filter the creative elements based on these weights, and select the creative elements that best meet market demands and user preferences.

Images