Intelligent recommendation method and system based on big data

Abstract

The invention provides an intelligent recommendation method and system based on big data, and relates to the technical field of big data analysis. The method comprises the following steps: collecting user historical behavior data, and carrying out behavior characterization analysis to form user historical feature data; historical product data are collected, product characterization analysis is carried out, and historical product feature data are formed; performing collaboration analysis according to the user historical feature data and the historical product feature data to form user-product collaboration feature data; and obtaining current behavior data of the user, and performing preference recommendation analysis in combination with the user-product collaborative feature data to form user preference recommendation data. The method can reasonably and accurately carry out product recommendation analysis.

Description

Technical Field

[0001] This invention relates to the field of big data analytics, and more specifically, to a method and system for intelligent recommendation based on big data. Background Technology

[0002] With the development of information technology, e-commerce has gradually become the mainstream of product marketing. Users determine whether a product meets their needs by obtaining product information on e-commerce platforms.

[0003] Currently, product recommendation methods based on user behavior are gradually emerging, but they still cannot meet users' needs in terms of accuracy and rationality.

[0004] Therefore, designing intelligent recommendation methods and systems based on big data, capable of conducting reasonable and accurate product recommendation analysis, is an urgent problem to be solved. Summary of the Invention

[0005] The purpose of this invention is to provide an intelligent recommendation method based on big data. This method extracts corresponding feature information from historical user behavior data and related product data, and performs collaborative analysis to establish a feature relationship between products and user behavior. Based on the collaborative features extracted from big data, it performs correlation processing on user behavior for the same product, establishing feature data information of the same product mapped to different user behaviors. In subsequent analysis, it can determine potentially relevant products based on the user's current behavior, providing a series of most likely product data for precise product recommendations. On one hand, by using correlation analysis of user and related product feature information, products and user behavior are deeply bound together. Simultaneously, based on big data, it can fully acquire different user behavior characteristics for the same product, providing data reference for diversified product positioning and ensuring that the connection between products and user behavior is diverse. This achieves a more flexible and reasonable mapping between products and user behavior, improving the diversity of product recommendations. On the other hand, the data formed after feature extraction and collaborative analysis can provide users with more accurate behavior-based product recommendation data, greatly improving the satisfaction of user needs and benefiting product marketing and promotion.

[0006] The present invention also aims to provide an intelligent recommendation system based on big data. This system acquires big data for recommendation analysis through a data acquisition unit, and utilizes a feature analysis unit and a collaborative analysis unit to perform collaborative analysis of user behavior and product information. It extracts collaborative feature data of user behavior and product information, thereby enabling the recommendation analysis unit to analyze the current user behavior, determine the products the user is currently interested in, and provide accurate and reasonable recommendations. The coordination between different functional units forms a tightly integrated recommendation analysis system, which is a crucial material foundation for realizing intelligent recommendations.

[0007] In a first aspect, the present invention provides an intelligent recommendation method based on big data, comprising: collecting historical user behavior data, performing behavioral feature analysis to form historical user feature data; collecting historical product data, performing product feature analysis to form historical product feature data; performing collaborative analysis based on historical user feature data and historical product feature data to form user-product collaborative feature data; and obtaining current user behavior data, combining it with user-product collaborative feature data to perform preference recommendation analysis to form user preference recommendation data.

[0008] In this invention, the method extracts corresponding feature information from historical user behavior data and related product data, and performs collaborative analysis to establish a feature relationship between products and user behavior. Based on the collaborative features extracted from big data, it performs correlation processing on user behavior of the same product, establishing feature data information of the same product mapped to different user behaviors. In subsequent analysis, it can determine potentially relevant products based on the user's current behavior, providing a series of most likely product data for precise product recommendations. On one hand, by using correlation analysis of user and related product feature information, products and user behavior are deeply bound together. Simultaneously, based on big data, it can fully acquire different user behavior characteristics of the same product, providing data references for diversified product positioning and ensuring that the connection between products and user behavior is diverse. This achieves a more flexible and reasonable mapping between products and user behavior, improving the diversity of product recommendations. On the other hand, the data formed after feature extraction and collaborative analysis can provide users with more accurate behavior-based product recommendation data, greatly improving the satisfaction of user needs and facilitating product marketing and promotion.

[0009] One possible approach is to collect historical user behavior data, perform behavioral feature analysis, and form historical user feature data. This includes: extracting behavioral data for different products for different users to form different user-product historical behavior data; and quantifying the different user-product historical behavior data to form corresponding historical product behavior parameter matrices. ,in, , m represents the product number, and n represents the parameter number of different behaviors. This represents the parameter value of the behavior parameter with number n for a user under product number m; a matrix of behavior parameters for different users across different historical products. Normalization is performed to form a corresponding historical product behavior parameter normalization matrix. A normalized matrix of all historical product behavior parameters from different users. This forms user historical characteristic data.

[0010] In this invention, different users focus on different products, and even at the same time, different users may simultaneously focus on multiple products. To obtain user-specific recommendations, both users and products need to be used as data extraction and filtering criteria for data clustering. This ensures that the acquired data is clearly targeted to specific users and products. Of course, most of the data is raw record data, requiring proper feature extraction to ensure the acquired data is representative of specific users and products. Furthermore, the extracted feature information can be statistically analyzed in subsequent analyses to improve data processing capabilities and more accurately reflect user product interest. It's important to note that parameterization of product data for different users can be extracted from various dimensions, including but not limited to browsing time, average browsing speed, product clicks, purchase frequency, number of favorites / follows, and browsing frequency. However, to ensure cross-sectional comparison of parameterized data across different users and products, the parameters must be standardized across all users and products. This means the parameter types in the established parameter matrix must have a one-to-one correspondence across different users and products. Considering that extracted parameters may not have inherent minimum and maximum values, the parameter matrix needs to be properly normalized after acquisition to facilitate subsequent comparative analysis. AI can be leveraged for parameter extraction and the establishment of one-to-one corresponding parameter matrices, significantly improving the efficiency of data collection and processing.

[0011] As one possible implementation, a parameter matrix of different users' historical product behaviors is used. Normalization is performed to form a corresponding historical product behavior parameter normalization matrix. This includes: for the same product, based on the historical product behavior parameter matrix of different users. Determine the maximum parameter value of each behavior parameter. and minimum parameter value of behavior parameters Historical product behavior parameter matrix for different users under the same product Each behavior parameter in the process is normalized to determine the corresponding behavior parameter normalization parameter value. ,in, For different parameter matrices The normalized parameter values ​​of the corresponding different behavioral parameters are set to form a normalized matrix of behavioral parameters. .

[0012] In this invention, parameter values ​​are normalized to ensure a unified numerical measurement basis for all parameters. This guarantees the consistency and rationality of parameter comparisons, ensuring the correctness and reasonableness of the analysis. Furthermore, using the maximum and minimum values ​​as reference benchmarks for normalization eliminates dimensions, achieving data consistency across different parameters and making the analysis of synergistic correlations more reasonable and effective. The maximum and minimum values ​​of each parameter can be obtained by comparing the parameter values ​​of the same parameter in the parameter matrix corresponding to different users under the same product.

[0013] One possible approach is to collect historical product data, perform product feature analysis, and generate historical product feature data. This includes: extracting historical product information for different products based on the historical product data; and quantifying parameters based on the different historical product information to form a corresponding product information parameter matrix. ,in, , where i represents the number of different product parameters. This represents the parameter value of the product parameter corresponding to product number i, where product number m is the product; for different product information parameter matrices Normalization is performed to form the corresponding product information parameter normalization matrix. A normalized matrix of different product information parameters. This generates historical product characteristic data.

[0014] In this invention, the extraction of product feature data primarily involves obtaining product information that users are interested in, including but not limited to product price, shelf life, weight, volume, and delivery date. Any information that influences customer needs regarding product usage can be extracted as feature information. However, it's important to note that the product output for different products should be consistent in terms of parameter type. This ensures horizontal comparability between different products in subsequent feature analysis, improving the rationality and accuracy of product recommendation content analysis. Of course, after obtaining the feature data, since the parameter values ​​for different product outputs may not have reasonable ranges, data planning and processing are necessary to ensure horizontal comparability. Normalization of product output should facilitate the normalization of user behavior data for different products. Normalization primarily determines the level of the product's parameter value among all products. This normalized data also provides a reasonable reference for the specific positioning of different products. Similarly, the extraction of product parameter data and the establishment of a matching parameter matrix can be achieved through efficient and accurate analysis using AI.

[0015] As one possible implementation method, different product information parameter matrices Normalization is performed to form the corresponding product information parameter normalization matrix. This includes: determining the information parameter matrix for different products based on different product parameters. Maximum parameter value of product parameters and minimum parameter value of product parameters For different product information parameter matrices Each product parameter in the data is normalized to determine the corresponding normalized parameter value. ,in, For different product information parameter matrices The product information parameter normalization matrix is ​​formed by collecting the normalized parameter values ​​of different product parameters. .

[0016] In this invention, the normalization of product parameters involves determining the maximum and minimum values ​​of a unified parameter across different product parameter matrices. These maximum and minimum values ​​are then used as constraints to obtain the corresponding normalized parameters. It's important to clarify that the normalization matrices established for different products essentially define them based on the different parameter values ​​within the matrices. For example, products with different prices are considered different products because their corresponding normalization matrix parameter values ​​are different. This distinction between similar and different products based on planning parameters aligns perfectly with the fact that different users are interested in the same product, since different users consider different parameter values.

[0017] As one possible approach, collaborative analysis is performed based on historical user characteristic data and historical product characteristic data to form user-product collaborative characteristic data. This includes: using different products in the historical product characteristic data as a benchmark, clustering the normalized matrix of behavioral parameters of different historical products corresponding to the same products in the historical user characteristic data. This forms a set of product clustering behavior parameter matrices corresponding to different products; and then, the different product clustering behavior parameter matrix sets are normalized to the corresponding product information parameter matrix. The collaborative analysis forms user-product collaborative information corresponding to the product; the different user-product collaborative information is combined to form user-product collaborative feature data.

[0018] In this invention, reasonable data collaboration can only be achieved when user behavior is associated with the product. Therefore, the prerequisite for collaborative analysis is to cluster all behavioral data related to the product. Collaborative analysis mainly involves performing feature analysis based on numerical analysis on the quantified behavioral data to obtain the numerical relationship between behavioral features and the product. AI can be used for efficient and accurate processing and analysis of clustering the parameter data of user behavior with the parameter information of the product.

[0019] As one possible implementation, different sets of product clustering behavior parameter matrices are normalized to the corresponding product information parameter matrix. The collaborative analysis generates user-product collaborative information corresponding to the product, including: a normalized matrix of comparative product information parameters. and historical product behavior parameter normalization matrix The total number of parameters in the matrix is ​​determined, and the larger total number of parameters is used as the total number of variables in the multidimensional coordinate system to be established. All matrices with smaller total numbers of parameters are expanded by using unit values ​​as parameter values ​​for fixed-position parameter expansion, so that the total number of parameters in all matrices is consistent with the total number of variables in the multidimensional coordinate system, forming different expanded-dimensional historical product behavior parameter normalization matrices. And the normalized matrix of product information parameters after dimensional expansion u represents the normalized matrix of different historical product behavior parameters in the product clustering behavior parameter matrix set. k represents the normalization matrix of behavior parameters of different historical products. The numbering of different parameters after dimension expansion, where r represents the product information parameter normalization matrix. The numbering of different parameters after dimension expansion, where the maximum number of numbers for r is equal to the maximum number of numbers for k; the normalization matrix of product information parameters after dimension expansion. Normalized matrix of historical product behavior parameters after different dimension expansion Mapped in the same multidimensional coordinate system, the product information parameter normalization matrix is ​​obtained after dimensional expansion. The collaborative envelope range centered on the user is extracted to form corresponding user-product collaborative information.

[0020] In this invention, the collaborative analysis of the same product information after clustering all corresponding user behavior information mainly considers two aspects. First, it involves collaboratively analyzing product information and user behavior information using matrices as numerical references, ensuring consistency across the matrix dimensions. Inconsistent matrix parameters necessitate more complex data processing, which is extremely demanding for big data analysis. Therefore, maintaining a consistent matrix parameter count is essential to reduce processing difficulty and improve efficiency while ensuring analytical rationality and accuracy. Here, we consider expanding matrices with fewer parameters from those with the most parameters. Alternatively, we could use matrices with fewer parameters as a benchmark to reduce the dimensionality of matrices with more parameters, but this would increase analytical complexity. Therefore, dimensionality expansion is preferable. If the product information parameter matrix has a large number of parameters, the parameters of all corresponding product behavior parameter matrices must be expanded to match the number of parameters in the product information parameter matrix. It's important to note that the expanded parameters must maintain their original positions within the matrix. Furthermore, considering that all parameters are percentage data, using a unit value of 1 as a supplementary value is reasonable. Similarly, if the product behavior parameter matrix has a large number of parameters, the corresponding product information parameter matrix should be expanded to the same number of parameters as the product behavior parameter matrix, with the expanded parameter positions also filled with a unit value of 1. After completing the consistency expansion processing of the matrix dimensions, the matrix data can be transformed into a coordinate system for collaborative correlation analysis to obtain collaborative feature data.

[0021] As one possible implementation, the normalized matrix of product information parameters after dimensional expansion is used. Normalized matrix of historical product behavior parameters after different dimension expansion Mapped in the same multidimensional coordinate system, the product information parameter normalization matrix is ​​obtained after dimensional expansion. The collaborative envelope range centered on the user is extracted to form corresponding user-product collaborative information, including: normalization matrices of historical product behavior parameters after different dimension expansions. Determine the normalized matrix of product information parameters after expansion of dimensions. Effective collaborative distance between and effective collaboration perspective ,in, , This represents the expanded-dimensional normalized matrix of historical product behavior parameters, with the number u as the reference. The normalized parameter value of the expanded product behavior parameter with dimension number k in the middle. This represents the normalized parameter value of the expanded product information parameter corresponding to the normalized parameter value of the expanded product behavior parameter with number r and number k. , This indicates that in a multidimensional coordinate system, coordinates are represented by points. To coordinate point directional vector, This represents the unit vector synthesized by each variable taking the value 1 in a multidimensional coordinate system. Represents two direction vectors and The included angle; based on different effective coordination angles Determine the effective range of collaborative angles And within the scope of effective collaboration Internally based on different effective coordination distances Perform angle-based fitting to generate user-product collaborative information. ,in, .

[0022] In this invention, the collaborative envelope feature extraction centered on the normalized matrix of product information parameters after dimensional expansion mainly involves two aspects. Firstly, it examines the numerical relationships among all user behavior parameter data related to product information. This application matrixifies the data and spatializes it using a coordinate system to establish vector relationships. These vector relationships include the magnitude and angle relationships. The magnitude relationship is decisive, while the angle relationship is relative. This relative relationship is absolutized by establishing a relationship with unit vectors, ensuring the uniformity and analyzable comparability of the angle relationships. The magnitude and angle relationships clearly and accurately reflect the characteristic connections between behavioral and product information. Secondly, while the relationship between behavioral and product information identified by big data is discrete, it essentially encompasses the range of these characteristic relationships. Therefore, this application uses the range defined by the vector angle to fit the change in vector magnitude with the angle value, clearly defining the variation law of the vector magnitude formed by behavioral information within the range defined by big data. This accurately and reasonably expresses the collaborative characteristic information between product and behavioral information.

[0023] One possible implementation involves acquiring current user behavior data and combining it with user-product collaborative feature data to perform preference recommendation analysis, forming user preference recommendation data. This includes: extracting current product behavior information for different products based on current user behavior data, performing parameter normalization and dimensionality expansion processing to form a current product behavior dimension-expanded normalized matrix for each product; acquiring product information corresponding to the current product behavior dimension-expanded normalized matrix, performing parameter normalization and dimensionality expansion processing to form a corresponding current product dimension-expanded normalized matrix; and determining the current effective collaborative angle based on the current product behavior dimension-expanded normalized matrix and the corresponding current product dimension-expanded normalized matrix. and current effective collaborative distance And based on user-product collaboration information matched with the current product Perform preference recommendation analysis using the following methods: compare current effective collaboration perspectives. and the corresponding effective range of collaborative angles ,like Not belonging to If not, then no product will be recommended or labeled; if belong Then: based on user-product collaboration information Determine the current effective collaboration angle The corresponding formula is the effective collaborative distance. ,like Less than If so, no product will be recommended or labeled accordingly. Not less than Then, the corresponding products will be recommended and calibrated.

[0024] In this invention, the main purpose of using collaborative feature data to perform preference recommendation analysis based on current user behavior is to determine whether the relationship between behavioral information and product information in terms of the magnitude and angle in numerical vectorization falls within the range defined by the corresponding collaborative features. Of course, the first step is to determine whether the range falls within the corresponding defined range; if it does, then it is confirmed whether it falls within the mapped magnitude range. Only when both the magnitude and angle conditions are met can it be determined that the user is interested in the product, and only then can the product be recommended to the user, providing more product-related advertisements and other corresponding content.

[0025] Secondly, this invention provides an intelligent recommendation system based on big data, comprising: a data acquisition unit for acquiring user historical behavior data, user historical product data, and user current behavior data; a feature analysis unit for performing feature analysis on the user historical behavior data acquired by the data acquisition unit to form user historical feature data, and performing product feature analysis on the user historical product data to form user historical product feature data; a collaborative analysis unit for performing collaborative analysis on the user historical feature data and user historical product feature data formed by the feature analysis unit to form user-product collaborative feature data; and a recommendation analysis unit for performing preference recommendation analysis on the user current behavior data acquired by the data acquisition unit, combined with the user-product collaborative feature data formed by the collaborative analysis unit, to form user preference recommendation data.

[0026] In this invention, the system acquires big data for recommendation analysis through a data acquisition unit, and performs collaborative analysis of user behavior and product information using a feature analysis unit and a collaborative analysis unit. This extracts collaborative feature data between user behavior and product information, which is then used in the recommendation analysis unit to analyze the current user behavior, determine the products the user is currently interested in, and provide accurate and reasonable recommendations. The coordination between different functional units forms a tightly integrated recommendation analysis system, which is a crucial material foundation for achieving intelligent recommendations.

[0027] The beneficial effects of the intelligent recommendation method and system based on big data provided by this invention are as follows: This method extracts corresponding feature information from historical user behavior data and related product data, and performs collaborative analysis to establish a feature relationship between products and user behavior. Based on the collaborative features extracted from big data, it performs correlation processing on user behavior related to the same product, establishing feature data information of the same product mapped to different user behaviors. In subsequent analysis, it can determine potentially relevant products based on the user's current behavior, providing a series of most likely product data for precise product recommendations. On one hand, by using correlation analysis of user and related product feature information, it deeply binds products and user behavior. Simultaneously, based on big data, it can fully acquire different user behavior characteristics of the same product, providing data references for diversified product positioning and ensuring that the connection between products and user behavior is diverse. This achieves a more flexible and reasonable mapping between products and user behavior, improving the diversity of product recommendations. On the other hand, the data formed after feature extraction and collaborative analysis can provide users with more accurate behavior-based product recommendation data, greatly improving the satisfaction of user needs and benefiting product marketing and promotion.

[0028] This system acquires big data for recommendation analysis through a data acquisition unit, and then uses a feature analysis unit and a collaborative analysis unit to perform collaborative analysis on user behavior and product information. It extracts collaborative feature data between user behavior and product information, which is then used in the recommendation analysis unit to analyze the current user behavior, determine the products the user is currently interested in, and provide accurate and reasonable recommendations. The coordination between different functional units forms a tightly integrated recommendation analysis system, which is a crucial material foundation for achieving intelligent recommendations. Attached Figure Description

[0029] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments of the present invention will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0030] Figure 1 A flowchart illustrating the steps of the intelligent recommendation method based on big data provided in an embodiment of the present invention; Figure 2 This is a schematic diagram of the structure of an intelligent recommendation system based on big data provided in an embodiment of the present invention. Detailed Implementation

[0031] The technical solutions of the present invention will now be described with reference to the accompanying drawings in the embodiments of the present invention.

[0032] With the development of information technology, e-commerce has gradually become the mainstream of product marketing. Users determine whether a product meets their needs by obtaining product information on e-commerce platforms.

[0033] Currently, product recommendation methods based on user behavior are gradually emerging, but they still cannot meet users' needs in terms of accuracy and rationality.

[0034] refer to Figures 1-2 This invention provides an intelligent recommendation method based on big data. This method extracts corresponding feature information from historical user behavior data and related product data, and performs collaborative analysis to establish a feature relationship between products and user behavior. Based on the collaborative features extracted from big data, it performs correlation processing on user behavior for the same product, establishing feature data information of the same product mapped to different user behaviors. In subsequent analysis, it can determine potentially relevant products based on the user's current behavior, providing a series of most likely product data for precise product recommendations. On one hand, by using correlation analysis of user and related product feature information, products and user behavior are deeply bound together. Simultaneously, based on big data, it can fully acquire different user behavior characteristics for the same product, providing data reference for diversified product positioning and ensuring that the connection between products and user behavior is diverse. This achieves a more flexible and reasonable mapping between products and user behavior, improving the diversity of product recommendations. On the other hand, the data formed after featureization and collaborative analysis can provide users with more accurate behavior-based product recommendation data, greatly improving the satisfaction of user needs and facilitating product marketing and promotion.

[0035] The intelligent recommendation method based on big data specifically includes the following steps: S1: Collect users' historical behavior data, perform behavioral feature analysis, and form users' historical feature data.

[0036] Collect historical user behavior data, perform behavioral feature analysis, and form historical user feature data. This includes: extracting behavioral data for different products for different users to form different user-product historical behavior data; and quantifying the different user-product historical behavior data to form corresponding historical product behavior parameter matrices. ,in, , m represents the product number, and n represents the parameter number of different behaviors. This represents the parameter value of the behavior parameter with number n for a user under product number m; a matrix of behavior parameters for different users across different historical products. Normalization is performed to form a corresponding historical product behavior parameter normalization matrix. A normalized matrix of all historical product behavior parameters from different users. This forms user historical characteristic data.

[0037] Different users focus on different products, and even at the same time, different users may simultaneously be interested in multiple products. To obtain targeted recommendations for specific users, both users and products need to be used as criteria for data extraction and filtering through data clustering. This ensures that the acquired data is clearly targeted to specific users and products. Of course, most of the data is raw record data, requiring proper feature extraction to ensure the data is representative of specific users and products. This also allows for the extraction of features for statistical analysis in subsequent analysis, improving data processing capabilities and more accurately reflecting user product interest. It's important to note that parameterization of product data for different users can be extracted from various dimensions, including but not limited to browsing time, average browsing speed, product clicks, purchase frequency, number of favorites / follows, and browsing frequency. However, to ensure cross-sectional comparison of parameterized data across different users and products, the parameters must be standardized across all users and products. This means the parameter types in the established parameter matrix must have a one-to-one correspondence across different users and products. Considering that extracted parameters may not have inherent minimum and maximum values, the parameter matrix needs to be properly normalized after acquisition to facilitate subsequent comparative analysis. AI can be leveraged for parameter extraction and the establishment of one-to-one corresponding parameter matrices, significantly improving the efficiency of data collection and processing.

[0038] Parameter matrix of different users' historical product behaviors Normalization is performed to form a corresponding historical product behavior parameter normalization matrix. This includes: for the same product, based on the historical product behavior parameter matrix of different users. Determine the maximum parameter value of each behavior parameter. and minimum parameter value of behavior parameters Historical product behavior parameter matrix for different users under the same product Each behavior parameter in the process is normalized to determine the corresponding behavior parameter normalization parameter value. ,in, For different parameter matrices The normalized parameter values ​​of the corresponding different behavioral parameters are set to form a normalized matrix of behavioral parameters. .

[0039] Normalizing parameter values ​​ensures a consistent numerical basis for all parameters, guaranteeing comparability and rationality in comparisons and ensuring the accuracy and validity of the analysis. Using maximum and minimum values ​​as reference benchmarks for normalization also eliminates dimensions, maintaining data consistency across different parameters and making the analysis of synergistic correlations more reasonable and effective. The maximum and minimum values ​​of each parameter can be obtained by comparing the parameter values ​​of the same parameter in the parameter matrix corresponding to different users within the same product.

[0040] S2: Collect historical product data, perform product feature analysis, and generate historical product feature data.

[0041] Historical product data is collected and analyzed to form historical product feature data, including: extracting historical product information for different products based on historical product data; and quantifying parameters based on the different historical product information to form corresponding product information parameter matrices. ,in, , where i represents the number of different product parameters. This represents the parameter value of the product parameter corresponding to product number i, where product number m is the product; for different product information parameter matrices Normalization is performed to form the corresponding product information parameter normalization matrix. A normalized matrix of different product information parameters. This generates historical product characteristic data.

[0042] Extracting product feature data primarily involves acquiring product information that users care about, including but not limited to product price, shelf life, weight, volume, and delivery date. Any information that influences customer needs regarding product usage can be extracted as feature information. However, it's important to note that the product output for different products should be consistent in terms of parameter type. This ensures horizontal comparability across products in subsequent feature analysis, improving the rationality and accuracy of product recommendation content analysis. After acquiring feature data, since the parameter values ​​for different product outputs may not have reasonable ranges, data planning and processing are necessary to ensure horizontal comparability. Normalization of product output should facilitate the normalization of user behavior data for different products. Normalization primarily determines the product's parameter value's position within all products. This normalized data provides a reasonable reference for the specific positioning of different products. Similarly, the extraction of product parameter data and the establishment of matching parameter matrices can be achieved efficiently and accurately through AI analysis.

[0043] For different product information parameter matrices Normalization is performed to form the corresponding product information parameter normalization matrix. This includes: determining the information parameter matrix for different products based on different product parameters. Maximum parameter value of product parameters and minimum parameter value of product parameters For different product information parameter matrices Each product parameter in the data is normalized to determine the corresponding normalized parameter value. ,in, For different product information parameter matrices The product information parameter normalization matrix is ​​formed by collecting the normalized parameter values ​​of different product parameters. .

[0044] Normalization of product parameters involves determining the maximum and minimum values ​​of a unified parameter across different product parameter matrices. These maximum and minimum values ​​are then used as constraints to obtain the corresponding normalized parameters. It's important to clarify that the normalization matrices for different products essentially define them based on the different parameter values ​​within the matrix. For example, products with different prices are considered different products because their corresponding normalization matrix parameter values ​​differ. This distinction between similar and different products based on planning parameters aligns perfectly with the different user interests in the same product, as different users consider different parameter values.

[0045] S3: Perform collaborative analysis based on historical user characteristic data and historical product characteristic data to form user-product collaborative characteristic data.

[0046] Based on collaborative analysis of user historical feature data and historical product feature data, user-product collaborative feature data is formed, including: using different products in the historical product feature data as a benchmark, clustering the normalized matrix of behavioral parameters of different historical products corresponding to the same products in the user's historical feature data. This forms a set of product clustering behavior parameter matrices corresponding to different products; and then, the different product clustering behavior parameter matrix sets are normalized to the corresponding product information parameter matrix. The collaborative analysis forms user-product collaborative information corresponding to the product; the different user-product collaborative information is combined to form user-product collaborative feature data.

[0047] User behavior can only achieve reasonable data collaboration when it is linked to the product being acted upon. Therefore, the prerequisite for collaborative analysis is to cluster all behavioral data associated with the product. Collaborative analysis mainly involves performing feature analysis based on numerical analysis on the quantified behavioral data to obtain the numerical relationship between behavioral features and the product. AI can be used for efficient and accurate processing and analysis of clustering user behavior parameter data with product parameter information.

[0048] For different product clustering behavior parameter matrix sets, perform corresponding product information parameter normalization matrix operations. The collaborative analysis generates user-product collaborative information corresponding to the product, including: a normalized matrix of comparative product information parameters. and historical product behavior parameter normalization matrix The total number of parameters in the matrix is ​​determined, and the larger total number of parameters is used as the total number of variables in the multidimensional coordinate system to be established. All matrices with smaller total numbers of parameters are expanded by using unit values ​​as parameter values ​​for fixed-position parameter expansion, so that the total number of parameters in all matrices is consistent with the total number of variables in the multidimensional coordinate system, forming different expanded-dimensional historical product behavior parameter normalization matrices. And the normalized matrix of product information parameters after dimensional expansion u represents the normalized matrix of different historical product behavior parameters in the product clustering behavior parameter matrix set. k represents the normalization matrix of behavior parameters of different historical products. The numbering of different parameters after dimension expansion, where r represents the product information parameter normalization matrix. The numbering of different parameters after dimension expansion, where the maximum number of numbers for r is equal to the maximum number of numbers for k; the normalization matrix of product information parameters after dimension expansion. Normalized matrix of historical product behavior parameters after different dimension expansion Mapped in the same multidimensional coordinate system, the product information parameter normalization matrix is ​​obtained after dimensional expansion. The collaborative envelope range centered on the user is extracted to form corresponding user-product collaborative information.

[0049] For the same product information, the collaborative analysis performed after clustering all corresponding user behavior information mainly considers two aspects. First, it involves collaboratively analyzing product information and user behavior information using matrices as numerical references, ensuring consistency across the matrix dimensions. Inconsistent matrix parameters necessitate more complex data processing, which is extremely demanding for big data analysis. Therefore, maintaining a consistent matrix parameter count is essential to reduce processing difficulty and improve efficiency while ensuring analytical rationality and accuracy. Here, we consider expanding matrices with fewer parameters from those with the most parameters. Alternatively, we could use matrices with fewer parameters as a benchmark to reduce the dimensionality of matrices with more parameters, but this increases analytical complexity. Therefore, dimensionality expansion is preferable. If the product information parameter matrix has a large number of parameters, the parameters of all corresponding product behavior parameter matrices must be expanded to match the number of parameters in the product information parameter matrix. It's important to note that the expanded parameters must maintain their original positions within the matrix. Considering that all parameters are percentage data, using a unit value (1) as a supplementary value is reasonable. Similarly, if the product behavior parameter matrix has a large number of parameters, the corresponding product information parameter matrix must be expanded to the same number of parameters as the product behavior parameter matrix, with the expanded parameter positions filled with a unit value of 1. After completing the consistency expansion of the matrix dimensions, the matrix data can be transformed into a coordinate system for collaborative correlation analysis to obtain collaborative feature data.

[0050] Normalize the product information parameters after dimension expansion. Normalized matrix of historical product behavior parameters after different dimension expansion Mapped in the same multidimensional coordinate system, the product information parameter normalization matrix is ​​obtained after dimensional expansion. The collaborative envelope range centered on the user is extracted to form corresponding user-product collaborative information, including: normalization matrices of historical product behavior parameters after different dimension expansions. Determine the normalized matrix of product information parameters after expansion of dimensions. Effective collaborative distance between and effective collaboration perspective ,in, , This represents the expanded-dimensional normalized matrix of historical product behavior parameters, with the number u as the reference. The normalized parameter value of the expanded product behavior parameter with dimension number k in the middle. This represents the normalized parameter value of the expanded product information parameter corresponding to the normalized parameter value of the expanded product behavior parameter with number r and number k. This indicates that in a multidimensional coordinate system, coordinates are represented by points. To coordinate point directional vector, This represents the unit vector synthesized by each variable taking the value 1 in a multidimensional coordinate system. Represents two direction vectors and The included angle; based on different effective coordination angles Determine the effective range of collaborative angles And within the scope of effective collaboration Internally based on different effective coordination distances Perform angle-based fitting to generate user-product collaborative information. ,in, .

[0051] The collaborative envelope feature extraction centered on the normalized matrix of product information parameters after dimensional expansion mainly involves two aspects. Firstly, it examines the numerical relationships among all user behavior parameters related to product information. This application matrixifies the data and spatializes it using a coordinate system to establish vector relationships. These vector relationships include both magnitude and angle relationships. The magnitude relationship is decisive, while the angle relationship is relative. This relative relationship is absolutized by establishing a relationship with unit vectors, ensuring the uniformity and analyzable comparability of the angle relationships. The magnitude and angle relationships clearly and accurately reflect the characteristic connections between behavioral and product information. Secondly, while the relationship between behavioral and product information identified by big data is discrete, it essentially encompasses the range of these characteristic relationships. Therefore, this application uses the range defined by the vector angle to fit the change in vector magnitude with the angle value, clearly defining the variation law of the vector magnitude formed by behavioral information within the range defined by big data. This accurately and reasonably expresses the collaborative characteristic information between product and behavioral information.

[0052] S4: Obtain current user behavior data, combine it with user-product collaborative feature data to perform preference recommendation analysis, and form user preference recommendation data.

[0053] The process involves acquiring current user behavior data and combining it with user-product collaborative feature data to perform preference recommendation analysis, resulting in user preference recommendation data. This includes: extracting current product behavior information for different products based on current user behavior data, performing parameter normalization and dimensionality expansion processing to form a current product behavior dimension-expanded normalized matrix for each product; acquiring product information corresponding to the current product behavior dimension-expanded normalized matrix, performing parameter normalization and dimensionality expansion processing to form a corresponding current product dimension-expanded normalized matrix; and determining the current effective collaborative angle based on the current product behavior dimension-expanded normalized matrix and the corresponding current product dimension-expanded normalized matrix. and current effective collaborative distance And based on user-product collaboration information matched with the current product Perform preference recommendation analysis using the following methods: compare current effective collaboration perspectives. and the corresponding effective range of collaborative angles ,like Not belonging to If not, then no product will be recommended or labeled; if belong Then: based on user-product collaboration information Determine the current effective collaboration angle The corresponding formula is the effective collaborative distance. ,like Less than If so, no product will be recommended or labeled accordingly. Not less than Then, the corresponding products will be recommended and calibrated.

[0054] The main purpose of using collaborative feature data for user preference recommendation analysis based on current user behavior is to determine whether the relationship between behavioral information and product information in terms of the magnitude and angle in numerical vectorization falls within the range defined by the corresponding collaborative features. Of course, the first step is to determine whether the range falls within the corresponding defined range; if it does, then it is confirmed whether it falls within the mapped magnitude range. Only when both the magnitude and angle conditions are met can it be determined that the user is interested in the product, and only then can the product be recommended to the user, providing more product-related advertisements and other relevant content.

[0055] This invention also provides an intelligent recommendation system based on big data, comprising: a data acquisition unit for collecting user historical behavior data, user historical product data, and user current behavior data; a feature analysis unit for performing feature analysis on the user historical behavior data collected by the data acquisition unit to form user historical feature data, and performing product feature analysis on the user historical product data to form user historical product feature data; a collaborative analysis unit for performing collaborative analysis on the user historical feature data and user historical product feature data formed by the feature analysis unit to form user-product collaborative feature data; and a recommendation analysis unit for performing preference recommendation analysis on the user current behavior data collected by the data acquisition unit, combined with the user-product collaborative feature data formed by the collaborative analysis unit, to form user preference recommendation data.

[0056] This system acquires big data for recommendation analysis through a data acquisition unit, and then uses a feature analysis unit and a collaborative analysis unit to perform collaborative analysis on user behavior and product information. It extracts collaborative feature data between user behavior and product information, which is then used in the recommendation analysis unit to analyze the current user behavior, determine the products the user is currently interested in, and provide accurate and reasonable recommendations. The coordination between different functional units forms a tightly integrated recommendation analysis system, which is a crucial material foundation for achieving intelligent recommendations.

[0057] In summary, the beneficial effects of the intelligent recommendation method and system based on big data provided in the embodiments of the present invention are as follows: This method extracts corresponding feature information from historical user behavior data and related product data, and performs collaborative analysis to establish a feature relationship between products and user behavior. Based on the collaborative features extracted from big data, it performs correlation processing on user behavior related to the same product, establishing feature data information of the same product mapped to different user behaviors. In subsequent analysis, it can determine potentially relevant products based on the user's current behavior, providing a series of most likely product data for precise product recommendations. On one hand, by using correlation analysis of user and related product feature information, it deeply binds products and user behavior. Simultaneously, based on big data, it can fully acquire different user behavior characteristics of the same product, providing data references for diversified product positioning and ensuring that the connection between products and user behavior is diverse. This achieves a more flexible and reasonable mapping between products and user behavior, improving the diversity of product recommendations. On the other hand, the data formed after feature extraction and collaborative analysis can provide users with more accurate behavior-based product recommendation data, greatly improving the satisfaction of user needs and benefiting product marketing and promotion.

[0058] In the embodiments of this application, "instruction" can include direct and indirect instructions, as well as explicit and implicit instructions. The information indicated by a certain piece of information is called the information to be instructed. In the specific implementation process, there are many ways to instruct the information to be instructed, such as, but not limited to, directly instructing the information to be instructed, such as the information to be instructed itself or its index. It can also indirectly instruct the information to be instructed by instructing other information, where there is a relationship between the other information and the information to be instructed. It can also instruct only a part of the information to be instructed, while the other parts are known or pre-agreed upon. For example, the instruction of specific information can be achieved by using a pre-agreed (e.g., protocol-defined) arrangement of various pieces of information, thereby reducing instruction overhead to some extent. At the same time, common parts of various pieces of information can be identified and uniformly indicated to reduce the instruction overhead caused by individually indicating the same information.

[0059] Furthermore, the specific indication method can also be any existing indication method, such as, but not limited to, the above-mentioned indication methods and their various combinations. Specific details of various indication methods can be found in existing technologies, and will not be repeated here. As described above, for example, when multiple pieces of information of the same type need to be indicated, the indication methods for different pieces of information may differ. In the specific implementation process, the required indication method can be selected according to specific needs. This application embodiment does not limit the selected indication method; therefore, the indication methods involved in this application embodiment should be understood to cover various methods that enable the party to be indicated to obtain the information to be indicated.

[0060] It should be understood that the information to be indicated can be sent as a whole or divided into multiple sub-information messages sent separately, and the sending period and / or timing of these sub-information messages can be the same or different. The specific sending method is not limited in this application embodiment. The sending period and / or timing of these sub-information messages can be predefined, for example, according to a protocol, or configured by the sending device by sending configuration information to the receiving device.

[0061] "Predefined" or "pre-configured" can be achieved by pre-saving corresponding codes, tables, or other means that can be used to indicate relevant information in the device. This application does not limit the specific implementation method. "Saving" can refer to saving in one or more memories. These memories can be separate installations or integrated into the encoder, decoder, processor, or communication device. Alternatively, some memories can be separately installed, while others are integrated into the decoder, processor, or communication device. The type of memory can be any form of storage medium, and this application does not limit this.

[0062] The “protocol” mentioned in this application embodiment may refer to a protocol family in the field of communication, a standard protocol with a similar protocol family frame structure, or a related protocol applied to future communication systems. This application embodiment does not specifically limit this.

[0063] In the embodiments of this application, descriptions such as "when," "under the circumstances," "if," and "if" all refer to the device making corresponding processing under certain objective circumstances, and are not limited to a specific time. They do not require the device to make a judgment action during implementation, nor do they imply any other limitations.

[0064] In the description of the embodiments of this application, unless otherwise stated, " / " indicates that the objects before and after are in an "or" relationship. For example, A / B can represent A or B. "And / or" in the embodiments of this application is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, and B alone, where A and B can be singular or plural. Furthermore, in the description of the embodiments of this application, unless otherwise stated, "multiple" refers to two or more. "At least one of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one of a, b, or c can represent: a, b, c, ab, ac, bc, or abc, where a, b, and c can be single or multiple. Additionally, to facilitate a clear description of the technical solutions of the embodiments of this application, the terms "first" and "second" are used in the embodiments of this application to distinguish identical or similar items with essentially the same function and effect. Those skilled in the art will understand that the terms "first," "second," etc., do not limit the quantity or order of execution, and that "first," "second," etc., are not necessarily different. Furthermore, in the embodiments of this application, words such as "exemplary" or "for example" are used to indicate that something is being used as an example, illustration, or description. Any embodiment or design scheme described as "exemplary" or "for example" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or design schemes. Specifically, the use of words such as "exemplary" or "for example" is intended to present the relevant concepts in a concrete manner for ease of understanding.

[0065] It should be understood that the processor in the embodiments of this application can be a central processing unit (CPU), or it can be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor, etc.

[0066] It should also be understood that the memory in the embodiments of this application can be volatile memory or non-volatile memory, or may include both volatile and non-volatile memory. The non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. The volatile memory can be random access memory (RAM), which is used as an external cache. By way of example, but not limitation, many forms of random access memory (RAM) are available, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate synchronous DRAM (DDR SDRAM), enhanced synchronous DRAM (ESDRAM), synchronous linked DRAM (SLDRAM), and direct rambus RAM (DR RAM).

[0067] The above embodiments can be implemented, in whole or in part, by software, hardware (such as circuits), firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more sets of available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. A semiconductor medium can be a solid-state drive.

[0068] It should be understood that the term "and / or" in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. A and B can be singular or plural. Additionally, the character " / " in this article generally indicates an "or" relationship between the preceding and following related objects, but it can also represent an "and / or" relationship. Please refer to the context for a more accurate understanding.

[0069] In this application, "at least one" means one or more, and "more than one" means two or more. "At least one of the following" or similar expressions refer to any combination of these items, including any combination of a single item or a plurality of items. For example, at least one of a, b, or c can mean: a, b, c, ab, ac, bc, or abc, where a, b, and c can be a single item or multiple items.

[0070] It should be understood that in the various embodiments of this application, the order of the above-mentioned processes does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0071] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0072] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0073] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0074] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0075] In addition, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0076] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0077] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A big data-based intelligent recommendation method, characterized in that, include: Collect users' historical behavior data, perform behavioral feature analysis, and form users' historical feature data; Collect historical product data, conduct product feature analysis, and form historical product feature data; Based on the user's historical feature data and the historical product feature data, a collaborative analysis is performed to form user-product collaborative feature data; The system acquires current user behavior data and combines it with the user-product collaborative feature data to perform preference recommendation analysis, thereby generating user preference recommendation data.

2. The intelligent recommendation method based on big data according to claim 1, characterized in that, The process of collecting historical user behavior data, performing behavioral feature analysis, and forming historical user feature data includes: For different users, behavioral data for different products is extracted to form different user product historical behavior data; The historical behavior data of different users are parametrically quantified to form a corresponding historical product behavior parameter matrix. ,in, , m represents the product number, and n represents the parameter number of different behaviors. This represents the parameter value of the action parameter with ID n for the user under the product with ID m. Different historical product behavior parameter matrices for different users Normalization is performed to form a corresponding historical product behavior parameter normalization matrix. ; A normalized matrix of all historical product behavior parameters for different users This forms the user's historical feature data.

3. The intelligent recommendation method based on big data according to claim 2, characterized in that, The parameter matrix for different historical product behaviors of different users Normalization is performed to form a corresponding historical product behavior parameter normalization matrix. ,include: For the same product, based on the historical product behavior parameter matrix of different users Determine the maximum parameter value of each behavior parameter. and minimum parameter value of behavior parameters ; The historical product behavior parameter matrix for different users under the same product Each behavior parameter in the process is normalized to determine the corresponding behavior parameter normalization parameter value. ,in, ; For different parameter matrices The normalized parameter values ​​of the corresponding different behavioral parameters are used to form the behavioral parameter normalization matrix. .

4. The intelligent recommendation method based on big data according to claim 2, characterized in that, The process of collecting historical product data, performing product feature analysis, and forming historical product feature data includes: Based on the historical product data, extract the product history information for different products; Based on the different historical information of the products, parameters are quantified to form a corresponding product information parameter matrix. ,in, 'i' represents the number of different product parameters. This represents the parameter value of the product parameter corresponding to product number i, which is product number m. For different product information parameter matrices Normalization is performed to form the corresponding product information parameter normalization matrix. ; The set of different product information parameter normalization matrices This forms the historical product feature data.

5. The intelligent recommendation method based on big data according to claim 4, characterized in that, The parameter matrix of different product information Normalization is performed to form the corresponding product information parameter normalization matrix. ,include: For different product parameters, determine the product information parameter matrix for different product parameters. Maximum parameter value of product parameters and minimum parameter value of product parameters ; For different product information parameter matrices Each product parameter in the data is normalized to determine the corresponding normalized parameter value. ,in, ; For different product information parameter matrices The product information parameter normalization matrix is ​​formed by collecting the normalized parameter values ​​of different product parameters. .

6. The intelligent recommendation method based on big data according to claim 5, characterized in that, The step of performing collaborative analysis based on the user's historical feature data and the historical product feature data to form user-product collaborative feature data includes: Based on different products in the historical product feature data, cluster the normalization matrix of different historical product behavior parameters corresponding to the same product in the user's historical feature data. This forms a set of product clustering behavior parameter matrices corresponding to different products; For different sets of product clustering behavior parameter matrices, perform corresponding product information parameter normalization matrix operations. Collaborative analysis is used to generate user-product collaborative information corresponding to the product; The different user-product collaboration information is collected to form the user-product collaboration feature data.

7. The intelligent recommendation method based on big data according to claim 6, characterized in that, The product information parameter normalization matrix is ​​then applied to different sets of product clustering behavior parameter matrices. The collaborative analysis generates user-product collaborative information corresponding to the product, including: Compare the product information parameter normalization matrix and the historical product behavior parameter normalization matrix The total number of parameters in the system is used, and the larger total number of parameters is taken as the total number of variables in the multidimensional coordinate system to be established. For all matrices with a small total number of parameters, parameter expansion is performed using unit values ​​at fixed positions to ensure that the total number of parameters in all matrices matches the total number of variables in the multidimensional coordinate system, thus forming different expanded historical product behavior parameter normalization matrices. And the normalized matrix of product information parameters after dimensional expansion u represents the normalization matrix of different historical product behavior parameters in the product clustering behavior parameter matrix set. k represents the normalization matrix of the behavior parameters of different historical products. The numbering of different parameters after dimension expansion, where r represents the normalization matrix of the product information parameters. The numbering of different parameters after dimension expansion, and the maximum number of numbers for r is equal to the maximum number of numbers for k; The normalization matrix of the expanded product information parameters And the normalized matrix of historical product behavior parameters after dimensional expansion as described above. Mapped within the same multidimensional coordinate system, the product information parameter normalization matrix after dimensional expansion is used. The collaborative envelope range centered on the user is extracted to form corresponding user-product collaborative information.

8. The intelligent recommendation method based on big data according to claim 7, characterized in that, The normalization matrix of the expanded product information parameters is then used. And the normalized matrix of historical product behavior parameters after dimensional expansion as described above. Mapped within the same multidimensional coordinate system, the product information parameter normalization matrix after dimensional expansion is used. The collaborative envelope range centered on the user is extracted to form corresponding user-product collaborative information, including: Normalization matrix of historical product behavior parameters after dimensional expansion for different dimensions The normalized matrix of product information parameters after the dimension expansion is determined. Effective collaborative distance between and effective collaboration perspective ,in, , The expanded-dimensional historical product behavior parameter normalization matrix, numbered u, represents the matrix. The normalized parameter value of the expanded product behavior parameter with dimension number k in the middle. This represents the normalized parameter value of the expanded product information parameter corresponding to the normalized parameter value of the expanded product behavior parameter with the number r and the number k. , This indicates that in a multidimensional coordinate system, coordinates are represented by points. To coordinate point directional vector, This represents the unit vector synthesized by each variable taking the value 1 in a multidimensional coordinate system. Represents two direction vectors and The included angle; According to different effective synergy perspectives Determine the effective range of collaborative angles And within the effective cooperative angle range The effective coordination distance varies depending on the specific circumstances. The user-product collaborative information is formed by fitting angle values. ,in, .

9. The intelligent recommendation method based on big data according to claim 8, characterized in that, The process of acquiring current user behavior data and combining it with user-product collaborative feature data to perform preference recommendation analysis to form user preference recommendation data includes: Based on the user's current behavior data, extract the current product behavior information for different products, and perform parameter normalization and dimension expansion processing to form a dimension-expanded normalized matrix of the current product behavior for each product. Obtain the product information corresponding to the current product behavior dimension-expanded normalization matrix, and perform parameter normalization and dimension expansion processing to form the corresponding current product dimension-expanded normalization matrix; Based on the current product behavior dimension-expanded normalized matrix and the corresponding current product dimension-expanded normalized matrix, the current effective collaboration angle is determined. and current effective collaborative distance And based on the user-product collaboration information matched with the current product. The following are methods for preference recommendation analysis: Comparison of the current effective collaborative perspectives and the corresponding effective range of collaborative angles ,like Not belonging to If not, then no product will be recommended or labeled accordingly; like belong ,but: Based on the user-product collaboration information Determine the current effective collaborative angle The corresponding formula is the effective collaborative distance. ,like Less than If so, no product will be recommended or labeled accordingly. Not less than Then, the corresponding products will be recommended and calibrated.

10. A big data-based intelligent recommendation system, employing the big data-based intelligent recommendation method according to any one of claims 1-9, characterized in that, include: The data acquisition unit is used to collect user historical behavior data, user historical product data, and user current behavior data. The feature analysis unit is used to perform feature analysis on the user historical behavior data collected by the data acquisition unit to form user historical feature data, and to perform product feature analysis on the user historical product data to form user historical product feature data. The collaborative analysis unit is used to perform collaborative analysis based on the user historical feature data and user historical product feature data formed by the feature analysis unit to form user-product collaborative feature data. The recommendation analysis unit is used to perform preference recommendation analysis based on the user's current behavior data collected by the data collection unit and the user-product collaborative feature data formed by the collaborative analysis unit, thereby forming user preference recommendation data.

Images