A method and system for realizing data recommendation based on cloud computing

Abstract

The present invention relates to the technical field of data recommendation, and discloses a method and system for realizing data recommendation based on cloud computing, wherein the method comprises: using the historical recommendation information and browsing information of the user to be recommended to construct a browsing data set, and performing standard processing to obtain a standard browsing data set; analyzing the distribution characteristics of the standard data set to construct a data feature matrix of the standard data set, and analyzing the data features of the standard data set; mining the associated users of the user to be recommended, clustering the users to obtain clustered users, calculating the similarity between the user to be recommended and the clustered users, and using an optimization matrix to perform user rating on the similar users to obtain a user rating result; collecting the current browsing information of the user to be recommended to construct an initial recommendation list of the user to be recommended, using the user rating result to update the initial recommendation list to obtain a target recommendation list, and using the target recommendation list to perform data recommendation on the user to be recommended. The present invention can improve the accuracy of data recommendation.

Description

Technical Field

[0001] The present invention relates to the technical field of data recommendation, and in particular to a method and system for implementing data recommendation based on cloud computing. Background Art

[0002] Data recommendation is a technology that uses the user's historical behavior data and preference information to predict and recommend content, products or services that may be of interest. Data recommendation can provide personalized suggestions and recommendations based on the user's past behavior and preferences. This can help users find content or products of interest more quickly, by showing users related products or services, and can improve sales conversion rates, increase transaction volume, and promote cross-selling and add-on sales. In today's era of information explosion, data recommendation can help users filter and obtain the most relevant information, thereby reducing information overload, improving user experience, and making it easier for users to find what they want, thereby increasing user satisfaction and loyalty.

[0003] At present, data recommendation is generally based on big data recommendation algorithms. By collecting information such as users' personal interests, preferences, and behavioral history, the internal network calculates the content that users are interested in and pushes it to users. However, when using this method for calculation, the system always recommends based on users' browsing history, and does not take into account the user's preference for certain information, resulting in frequent recommendations of information that users are not interested in. Summary of the invention

[0004] In order to solve the above problems, the present invention provides a method for implementing data recommendation based on cloud computing, which can improve the accuracy of data recommendation.

[0005] In a first aspect, the present invention provides a method for implementing data recommendation based on cloud computing, comprising:

[0006] Using a cloud computing database to obtain historical recommendation information and browsing information of the user to be recommended, constructing a browsing data set of the user to be recommended based on the historical recommendation information and the browsing information, and performing data standardization processing on the browsing data set to obtain a standard browsing data set;

[0007] Analyzing data distribution characteristics of the standard browsing data set, constructing a data characteristic matrix of the standard browsing data set based on the data distribution characteristics, and analyzing data characteristics of the standard browsing data set using the data characteristic matrix;

[0008] Based on the data features, using the secure network of the cloud computing to mine the associated users of the user to be recommended, clustering the associated users to obtain clustered users, calculating the similarity between the user to be recommended and the clustered users to obtain similarity users, constructing a rating matrix for the similarity users, optimizing the rating matrix to obtain an optimized matrix, and using the optimized matrix to perform user ratings on the similarity users to obtain user rating results;

[0009] The streaming processing technology of cloud computing is used to collect the current browsing information of the user to be recommended data, and based on the current browsing information, an initial recommendation list for the user to be recommended data is constructed, and the initial recommendation list is updated with the user rating results to obtain a target recommendation list, and data is recommended to the user to be recommended data according to the target recommendation list.

[0010] In a possible implementation manner of the first aspect, performing data standardization processing on the browsing data set to obtain a standard browsing data set includes:

[0011] Performing data cleaning on the browsing data set to obtain cleaned data;

[0012] Interpolating missing values ​​on the cleaned data to obtain interpolated data;

[0013] The interpolation data are interval-integrated using the following formula to obtain a standard browsing data set:

[0014]

[0015] in, represents standardized data, Q represents interpolated data, represents the minimum data for interpolation data, Indicates the maximum data of the interpolation data.

[0016] In a possible implementation manner of the first aspect, analyzing data distribution characteristics of the standard browsing data set includes:

[0017] constructing a density map of the standard browsing data set;

[0018] Analyzing the data concentration and data dispersion of the standard browsing data set using the density map, and identifying the graph of the density map;

[0019] Analyzing the data skew state of the standard browsing data set based on the graph;

[0020] The data distribution characteristics of the standard browsing data set are determined based on the data centrality, the data dispersion and the data skew state.

[0021] In a possible implementation manner of the first aspect, constructing a data feature matrix of the standard browsing data set based on the data distribution feature includes:

[0022] Performing data conversion on the standard browsing data set according to the data distribution characteristics to obtain converted data;

[0023] Performing data encoding on the converted data to obtain encoded data;

[0024] Performing data dimensionality reduction on the encoded data to obtain dimensionality-reduced data;

[0025] constructing an initial matrix of the standard browsing data set;

[0026] After integrating the dimension-reduced data using the initial matrix, a data feature matrix is ​​obtained.

[0027] In a possible implementation manner of the first aspect, analyzing the data features of the standard browsing data set by using the data feature matrix includes:

[0028] Using the data feature matrix to count the descriptive information of the standard browsing data set;

[0029] Constructing a visualization chart of the standard browsing data set according to the descriptive information;

[0030] The visual chart is used to analyze data features of the standard browsing data set.

[0031] In a possible implementation manner of the first aspect, clustering the associated users to obtain clustered users includes:

[0032] Performing vector transformation on the associated user to obtain a vector user;

[0033] Constructing a vector centroid of the vector user;

[0034] The vector distance between the vector user and the vector centroid is calculated using the following formula:

[0035]

[0036] in, represents the vector distance, represents the spatial dimension of the vector user, represents the abscissa of the centroid of the i-th vector, represents the ordinate of the centroid of the i-th vector;

[0037] The associated users are clustered based on the vector distance to obtain clustered users.

[0038] In a possible implementation manner of the first aspect, calculating the similarity between the user to be recommended data and the clustered users to obtain similarity users includes:

[0039] Constructing vector data of the user to be recommended data and the clustered users to obtain a first vector and a second vector;

[0040] The vector similarity between the first vector and the second vector is calculated based on the first vector and the second vector in combination with the following formula:

[0041]

[0042] in, represents the vector similarity, represents the first vector, represents the second vector, represents the magnitude of the first vector, represents the magnitude of the second vector, represents the inclination of the first vector and the corresponding spatial coordinate point i, represents the inclination of the second vector and the corresponding spatial coordinate point i;

[0043] Based on the vector similarity, the user to be recommended data and the clustered users are divided into similarity groups to obtain similarity users.

[0044] In a possible implementation manner of the first aspect, using the optimization matrix to perform user scoring on the similarity users to obtain a user scoring result includes:

[0045] Constructing the interest set of the similarity user based on the optimization matrix;

[0046] Based on the interest set, the following formula is used to calculate the interest preference degree of the similarity user to the interest set:

[0047]

[0048] in, Indicates the degree of interest preference. represents the vector similarity, represents the set of users who have rated the interest set, represents the rating of user C on interest i, represents the average value of user C's interest in his rated represents the set of similar users, represents the rating vector of user C;

[0049] According to the interest preference degree, the similarity users are rated to obtain a user rating result.

[0050] In a possible implementation manner of the first aspect, updating the initial recommendation list by using the user rating result to obtain a target recommendation list includes:

[0051] Determine the preference information of the similarity user based on the user rating result;

[0052] Using the preference information, weighting the initial recommendation list to obtain an updated list;

[0053] Collaborative filtering is performed on the update list to obtain a target recommendation list.

[0054] In a second aspect, the present invention provides a system for implementing data recommendation based on cloud computing, the system comprising:

[0055] A data processing module is used to obtain historical recommendation information and browsing information of the user to be recommended data by using a cloud computing database, construct a browsing data set of the user to be recommended data according to the historical recommendation information and the browsing information, and perform data standardization processing on the browsing data set to obtain a standard browsing data set;

[0056] A data analysis module, used to analyze data distribution characteristics of the standard browsing data set, construct a data feature matrix of the standard browsing data set based on the data distribution characteristics, and analyze data characteristics of the standard browsing data set using the data feature matrix;

[0057] A user scoring module is used to mine the associated users of the user to be recommended for data based on the data features by using the secure network of the cloud computing, cluster the associated users to obtain clustered users, calculate the similarity between the user to be recommended for data and the clustered users to obtain similarity users, construct a scoring matrix for the similarity users, optimize the scoring matrix to obtain an optimized matrix, and use the optimized matrix to score the similarity users to obtain a user scoring result;

[0058] The data recommendation module is used to use the streaming processing technology of the cloud computing to collect the current browsing information of the user to be recommended data, build an initial recommendation list for the user to be recommended data based on the current browsing information, update the data of the initial recommendation list using the user scoring results to obtain a target recommendation list, and recommend data to the user to be recommended data according to the target recommendation list.

[0059] Compared with the prior art, the technical principle and beneficial effects of this solution are:

[0060] The embodiment of the present invention can understand the interests and preferences of users by using the cloud computing database to obtain the historical recommendation information and browsing information of the user to be recommended, thereby providing personalized recommendation content and enhancing user experience, and analyzing the data distribution characteristics of the standard browsing data set can help users understand the structure, shape and change rules of the data, thereby better performing data analysis and modeling. Further, the embodiment of the present invention can select, extract and transform features in the data feature matrix of the standard browsing data set based on the data distribution characteristics to improve the accuracy of data analysis, and based on the data characteristics, the cloud computing security network is used to mine the associated users of the user to be recommended, so as to identify and analyze the user group related to the data to be recommended, so as to provide more accurate and personalized recommendation services. Furthermore, the embodiment of the present invention can collect the current browsing information of the user to be recommended by using the streaming processing technology of the cloud computing to collect the current browsing information of the user to be recommended, so as to realize the real-time or near-real-time collection of the user's current browsing information, thereby analyzing the user's interests and needs in real time, providing the user with personalized recommendation content, and improving user experience and satisfaction. The method and system for realizing data recommendation based on cloud computing proposed in the embodiment of the present invention can improve the accuracy of data recommendation. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0062] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0063] Figure 1 A schematic diagram of a process for implementing data recommendation based on cloud computing provided by an embodiment of the present invention;

[0064] Figure 2 A schematic diagram of modules of a system for implementing data recommendation based on cloud computing provided by one embodiment of the present invention. DETAILED DESCRIPTION

[0065] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0066] The embodiment of the present invention provides a method for implementing data recommendation based on cloud computing, and the execution subject of the method for implementing data recommendation based on cloud computing includes but is not limited to at least one of the electronic devices such as a server and a terminal that can be configured to execute the method provided by the embodiment of the present invention. In other words, the method for implementing data recommendation based on cloud computing can be executed by software or hardware installed on a terminal device or a server device, and the software can be a blockchain platform. The server includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, etc. The server can be an independent server, or it can be a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (CDN), and big data and artificial intelligence platforms.

[0067] See also Figure 1 FIG. 1 is a flow chart of a method for implementing data recommendation based on cloud computing provided by an embodiment of the present invention. Figure 1 A method for implementing data recommendation based on cloud computing described in the specification includes:

[0068] S1. Utilize a cloud computing database to obtain historical recommendation information and browsing information of the user to be recommended, construct a browsing data set of the user to be recommended based on the historical recommendation information and the browsing information, and perform data standardization on the browsing data set to obtain a standard browsing data set.

[0069] The embodiments of the present invention can understand the interests and preferences of users by using the cloud computing database to obtain the historical recommendation information and browsing information of the user to be recommended, thereby providing personalized recommendation content and enhancing the user experience. The cloud computing database refers to a database service deployed and operated in a cloud computing environment, the historical recommendation information refers to the content actively recommended to the user by the system, such as entertainment, real-time news, travel and other information, and the browsing information refers to the information actively browsed by the user on the Internet or in an app, such as shopping information, sports information and financial information. The cloud computing database can store a large amount of information, and can identify user information through cloud computing, and then query the historical recommendation information and browsing information of the user to be recommended based on the user information.

[0070] Furthermore, the embodiments of the present invention can provide users with a practical and reliable data basis by constructing a browsing data set of the user to be recommended data based on the historical recommendation information and the browsing information, and support users to conduct more in-depth and accurate data analysis and mining.

[0071] Optionally, the browsing data set may be obtained by inputting the historical recommendation information and the browsing information into a pre-constructed SQL database.

[0072] The embodiment of the present invention performs data standardization processing on the browsing data set to obtain a standard browsing data set, which can convert data of different scales and ranges into a unified standard format for easy comparison, analysis and processing.

[0073] As an embodiment of the present invention, the data standardization processing of the browsing data set to obtain a standard browsing data set includes: performing data cleaning on the browsing data set to obtain cleaned data, interpolating missing values ​​of the cleaned data to obtain interpolated data, and interval integrating the interpolated data using the following formula to obtain a standard browsing data set:

[0074]

[0075] in, represents standardized data, Q represents interpolated data, represents the minimum data for interpolation data, Indicates the maximum data of the interpolation data.

[0076] Optionally, performing data cleaning on the browsing data set to obtain cleaned data refers to identifying and correcting errors in the data set, which can be achieved through Python's Pandas library; performing missing value interpolation on the cleaned data to obtain interpolated data means that when the numerical variables in the data set contain missing values, various techniques can be used to estimate, predict or fill these missing values ​​to maintain the integrity and accuracy of the data, which can be achieved through nearest neighbor interpolation.

[0077] S2. Analyze data distribution characteristics of the standard browsing data set, construct a data feature matrix of the standard browsing data set based on the data distribution characteristics, and analyze data characteristics of the standard browsing data set using the data feature matrix.

[0078] The embodiment of the present invention can help users understand the structure, shape and change rules of data by analyzing the data distribution characteristics of the standard browsing data set, so as to better perform data analysis and modeling. The data distribution characteristics refer to the attributes or characteristics that describe how data values ​​are distributed in the entire data set, such as the discreteness of the data, the symmetry of the data, and the distribution shape.

[0079] As an embodiment of the present invention, the analysis of data distribution characteristics of the standard browsing data set includes: constructing a density map of the standard browsing data set, using the density map to analyze data concentration and data dispersion of the standard browsing data set, identifying a graph of the density map, analyzing a data skew state of the standard browsing data set based on the graph, and determining the data distribution characteristics of the standard browsing data set based on the data concentration, the data dispersion and the data skew state.

[0080] The density map refers to a data visualization chart used to display the distribution of continuous variables. It shows the relative concentration of data points by creating a kernel density estimate in the data distribution. The data concentration refers to the observation of data concentration.

[0081] Optionally, the density map of the standard browsing data set can be constructed by Java, and the data concentration and data dispersion of the standard browsing data set analyzed by the density map can be determined by observing the peak and tail of the density map. The data skew state of the standard browsing data set analyzed based on the graph can be analyzed by observing the graph state. If the graph state is a bell shape, it indicates a normal distribution. If the graph state is a skewed distribution, the data will show a shape tilted to the left or right in the density map. The process of determining the data distribution characteristics of the standard browsing data set based on the data concentration, the data dispersion and the data skew state is: using the dispersion to determine the difference and fluctuation characteristics of the standard browsing data set. The greater the dispersion, the greater the data fluctuation or difference of the standard browsing data set. For example, in a set of entertainment data, if there are tourism and food, it means that the data fluctuation or difference is large. The data skew state can be used to understand the rules and trends of the standard browsing data set. For example, a city center business district is prosperous and lively, attracting a large number of businesses and tourists, resulting in a higher population density in the area than the surrounding areas. In this case, the development of the business district may exceed other areas, forming a positively skewed distribution.

[0082] Furthermore, in this embodiment, by constructing a data feature matrix of the standard browsing data set based on the data distribution characteristics, feature selection, extraction and conversion can be performed therein to improve the accuracy of data analysis. The data feature matrix refers to a data set containing all samples.

[0083] As an embodiment of the present invention, constructing a data feature matrix of the standard browsing data set based on the data distribution characteristics includes: performing data conversion on the standard browsing data set according to the data distribution characteristics to obtain conversion data, performing data encoding on the conversion data to obtain encoded data, performing data dimension reduction on the encoded data to obtain reduced dimension data, constructing an initial matrix of the standard browsing data set, and obtaining a data feature matrix after integrating the reduced dimension data using the initial matrix.

[0084] Optionally, performing data conversion on the standard browsing data set according to the data distribution characteristics, and converting the converted data through a data conversion matrix constructed in python, performing data encoding on the converted data to obtain encoded data can be achieved through feature encoding, performing data dimension reduction on the encoded data to obtain reduced dimension data can be achieved through PCA technology, and integrating the reduced dimension data using the initial matrix refers to the process of merging, cleaning and integrating data from different data sources, formats or systems into a unified data storage, which can be achieved through data synchronization tools such as Apache Kafka.

[0085] The embodiment of the present invention can help users extract useful information from the data, simplify the data processing process, optimize model performance, and better understand the patterns and relationships in the data set by analyzing the data features of the standard browsing data set using the data feature matrix.

[0086] As an embodiment of the present invention, the use of the data feature matrix to analyze the data features of the standard browsing data set includes: using the data feature matrix to count the descriptive information of the standard browsing data set, constructing a visualization chart of the standard browsing data set based on the descriptive information, and using the visualization chart to analyze the data features of the standard browsing data set.

[0087] Among them, the descriptive information refers to the process of organizing, summarizing and displaying the collected data in order to better understand the characteristics and changes of the data. The visual chart refers to the use of graphics, images or other interactive methods to present data in order to more intuitively understand the meaning and trends of the data. The descriptive information refers to the information that summarizes and describes the characteristics, variables or samples in the data set.

[0088] Optionally, the descriptive information of the standard browsing data set obtained by using the data feature matrix statistics can be obtained by calculating quartile statistics, and the visualization chart of the standard browsing data set is constructed based on the descriptive information through a script constructed in java, based on the first quartile, median, and third quartile of the data. The process of analyzing the data features of the standard browsing data set using the visualization chart is as follows: using the visualization chart to display the standard browsing data set in a scatter plot format to show the relationship between different variables, such as correlation, trend, etc., and using the visualization chart to display the standard browsing data set in a heat map format to analyze the correlation between data, such as when processing a large amount of data, the degree of mutual influence of the data can be clearly presented, and using the visualization chart to display the standard browsing data set in a pie chart format to analyze the proportion of each part of the standard browsing data set to the whole, so as to compare the proportion of different categories or groups of data.

[0089] S3. Based on the data features, the secure network of the cloud computing is used to mine the associated users of the user to be recommended, and the associated users are clustered to obtain clustered users. The similarity between the user to be recommended and the clustered users is calculated to obtain similar users, and a rating matrix of the similar users is constructed. After optimizing the rating matrix, an optimized matrix is ​​obtained, and the optimized matrix is ​​used to perform user rating on the similar users to obtain user rating results.

[0090] The embodiment of the present invention can identify and analyze the user groups related to the data by mining the associated users of the data recommendation users based on the data features and utilizing the cloud computing security network to provide more accurate and personalized recommendation services. The cloud computing security network refers to a comprehensive security architecture that protects the cloud computing system, network and data by taking various security measures and mechanisms in the cloud computing environment, such as identity authentication and access control systems, network security systems and data encryption systems.

[0091] Optionally, based on the data features, the secure network of the cloud computing is used to mine associated users of the user to be recommended data. The powerful computing resources of the cloud computing can be used to build an association model in a secure network environment, and collaborative filtering, graph neural networks or other machine learning techniques can be used to discover the association relationship between users.

[0092] Furthermore, the embodiment of the present invention clusters the associated users to obtain clustered users, and can analyze the characteristics and commonalities of each user cluster group, and understand the differences and connections between different user groups.

[0093] As an embodiment of the present invention, clustering the associated users to obtain clustered users includes: performing vector transformation on the associated users to obtain vector users, constructing the vector centroid of the vector users, and calculating the vector distance between the vector users and the vector centroid using the following formula:

[0094]

[0095] in, represents the vector distance, represents the spatial dimension of the vector user, represents the abscissa of the centroid of the i-th vector, represents the ordinate of the centroid of the i-th vector;

[0096] The associated users are clustered based on the vector distance to obtain clustered users.

[0097] Optionally, the vector centroid of the vector user is constructed by constructing a small sample set of initial points according to user needs, i.e., the initial vector centroid, such as a small sample of shopping, a small sample of news, or a sports class. The associated users are clustered based on the vector distance to obtain clustered users, by dividing the associated users into categories, such as dividing distances 0-1 into one category and 1-2 into another category.

[0098] The embodiment of the present invention calculates the similarity between the user to be recommended and the clustered users, and obtains the similarity user, which can conveniently recommend content with similar interests and preferences to the user to be recommended, thereby improving the accuracy of the recommendation system and user satisfaction.

[0099] As an embodiment of the present invention, the calculating the similarity between the user to be recommended and the clustered user to obtain the similarity user includes: constructing vector data of the user to be recommended and the clustered user to obtain a first vector and a second vector, and calculating the vector similarity between the first vector and the second vector based on the first vector and the second vector in combination with the following formula:

[0100]

[0101] in, represents the vector similarity, represents the first vector, represents the second vector, represents the magnitude of the first vector, represents the magnitude of the second vector, represents the inclination of the first vector and the corresponding spatial coordinate point i, represents the inclination of the second vector and the corresponding spatial coordinate point i;

[0102] Based on the vector similarity, the user to be recommended data and the clustered users are divided into similarity groups to obtain similarity users.

[0103] Optionally, based on the vector similarity, the user to be recommended and the clustered user are divided by similarity, and the similar user is obtained by the similarity calculation result of the first vector and the second vector. If the similarity calculation result is not less than a preset similarity value, it is represented as a similar user. The preset similarity value is 0.8, and can also be set according to the actual application scenario.

[0104] In the embodiment of the present invention, by constructing the rating matrix of the similar users, after optimizing the rating matrix, an optimized matrix is ​​obtained, which can be used to find out which items the users are more interested in, thereby improving the accuracy and efficiency of the recommendation system.

[0105] Optionally, the constructing of the rating matrix of the similarity users, after optimizing the rating matrix, the optimized matrix can be obtained by performing matrix optimization on the rating matrix through the PAC technology.

[0106] Furthermore, in the embodiment of the present invention, by using the optimization matrix to rate the similarity users, the user rating result can be used to understand the degree of preference of the similarity users for different interests in the data they are interested in. The user rating refers to the rating or preference of the similarity users for the information they are interested in.

[0107] As an embodiment of the present invention, the method of using the optimization matrix to score the similarity users to obtain the user scoring result includes: constructing an interest set of the similarity users based on the optimization matrix, and calculating the interest preference degree of the similarity users for the interest set based on the interest set using the following formula:

[0108]

[0109] in, Indicates the degree of interest preference. represents the vector similarity, represents the set of users who have rated the interest set, represents the rating of user C on interest i, represents the average value of user C's interest in his rated represents the set of similar users, represents the rating vector of user C;

[0110] According to the interest preference degree, the similarity users are rated to obtain a user rating result.

[0111] Optionally, constructing the interest set of the similarity users based on the optimization matrix is ​​obtained by classifying the interest sets of the similarity users through the optimization matrix, and scoring the similarity users according to the interest preference degrees, and scoring the user scoring results according to preset scoring rules, wherein the preset scoring rules are to divide the interest preference degrees into three levels, and the interest preference degree range between 0.1 and 0.4 indicates a general score, between 0.4 and 0.7 indicates a medium score, and between 0.7 and 1 indicates a high score.

[0112] S4. Utilize the streaming processing technology of the cloud computing to collect the current browsing information of the user to be recommended data, construct an initial recommendation list for the user to be recommended data based on the current browsing information, update the initial recommendation list with the user rating results to obtain a target recommendation list, and recommend data to the user to be recommended data based on the target recommendation list.

[0113] The embodiment of the present invention can collect the current browsing information of the user to be recommended data in real time or near real time by utilizing the streaming processing technology of cloud computing, thereby analyzing the user's interests and needs in real time, providing the user with personalized recommended content, and improving user experience and satisfaction.

[0114] Furthermore, the embodiments of the present invention can help users discover new content, products or services, promote user participation and interaction, and improve user retention and conversion rates by constructing an initial recommendation list for the user to be recommended based on the current browsing information.

[0115] The initial recommendation list is updated with the user rating results to obtain a target recommendation list, which can optimize the user's personalized recommendation list and make the recommended content more in line with the user's interests and preferences.

[0116] As an embodiment of the present invention, the method of using the user rating results to update the data of the initial recommendation list to obtain a target recommendation list includes: determining the preference information of the similarity users based on the user rating results, using the preference information to weight the initial recommendation list to obtain an updated list, and performing collaborative filtering on the updated list to obtain a target recommendation list.

[0117] Optionally, the using of the preference information to weight the initial recommendation list to obtain an updated list is weighted by using the user scores in combination with a weighted average method, and the collaborative filtering of the updated list to obtain a target recommendation list refers to data recommendation based on the interests and hobbies of the similar users combined with the historical behavior habits of the users to be recommended. For example, if a user likes to watch the news in the morning and sports games at night, the recommendation is made based on the behavior habits of the users to be recommended and the interests and hobbies of the similar users.

[0118] In the embodiment of the present invention, by performing data recommendation for the user to be recommended data according to the target recommendation list, content of interest to the user to be recommended data can be recommended more accurately.

[0119] like Figure 2 As shown, it is a system functional module diagram for realizing data recommendation based on cloud computing according to the present invention.

[0120] The system 200 for implementing data recommendation based on cloud computing described in the present invention can be installed in an electronic device. According to the functions implemented, the system for implementing data recommendation based on cloud computing can include a data processing module 201, a data analysis module 202, a user rating module 203, and a data recommendation module 204. The module described in the present invention can also be referred to as a unit, which refers to a series of computer program segments that can be executed by an electronic device processor and can complete fixed functions, which are stored in the memory of the electronic device.

[0121] In the embodiment of the present invention, the functions of each module / unit are as follows:

[0122] The data processing module 201 is used to obtain the historical recommendation information and browsing information of the user to be recommended using the cloud computing database, construct a browsing data set of the user to be recommended based on the historical recommendation information and the browsing information, and perform data standardization processing on the browsing data set to obtain a standard browsing data set;

[0123] The data analysis module 202 is used to analyze the data distribution characteristics of the standard browsing data set, construct a data feature matrix of the standard browsing data set based on the data distribution characteristics, and analyze the data characteristics of the standard browsing data set using the data feature matrix;

[0124] The user scoring module 203 is used to mine the associated users of the user to be recommended based on the data features by using the secure network of the cloud computing, cluster the associated users to obtain clustered users, calculate the similarity between the user to be recommended and the clustered users to obtain similar users, construct a scoring matrix for the similar users, optimize the scoring matrix to obtain an optimized matrix, and use the optimized matrix to score the similar users to obtain a user scoring result;

[0125] The data recommendation module 204 is used to collect the current browsing information of the user to be recommended data by using the streaming processing technology of the cloud computing, build an initial recommendation list for the user to be recommended data based on the current browsing information, update the data of the initial recommendation list by using the user rating results to obtain a target recommendation list, and recommend data to the user to be recommended data according to the target recommendation list.

[0126] In detail, each module in the system 200 for implementing data recommendation based on cloud computing in the embodiment of the present invention is used in the same manner as described above. Figure 1 The same technical means as described in the method for realizing data recommendation based on cloud computing and can produce the same technical effects are not described here.

[0127] In the several embodiments provided by the present invention, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the system embodiments described above are only illustrative, for example, the division of the modules is only a logical function division, and there may be other division methods in actual implementation.

[0128] The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0129] In addition, each functional module in each embodiment of the present invention may be integrated into one processing unit, each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of hardware plus software functional modules.

[0130] It is obvious to those skilled in the art that the present invention is not limited to the details of the above exemplary embodiments, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0131] It should be noted that, in this article, relational terms such as "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0132] The above description is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A method for implementing data recommendation based on cloud computing, characterized in that: The method comprises: The cloud computing database is used to obtain historical recommendation information and browsing information of the user to be recommended, and a browsing data set of the user to be recommended is constructed according to the historical recommendation information and the browsing information, and the browsing data set is subjected to data standardization to obtain a standard browsing data set. The step of performing data standardization on the browsing data set to obtain a standard browsing data set includes: Performing data cleaning on the browsing data set to obtain cleaned data; Interpolating missing values ​​on the cleaned data to obtain interpolated data; The interpolation data are interval-integrated using the following formula to obtain a standard browsing data set: in, represents standardized data, Q represents interpolated data, represents the minimum data for interpolation data, Indicates the maximum data of the interpolation data; Analyze the data distribution characteristics of the standard browsing data set, construct a data feature matrix of the standard browsing data set based on the data distribution characteristics, and analyze the data characteristics of the standard browsing data set using the data feature matrix. The step of analyzing the data distribution characteristics of the standard browsing data set includes: constructing a density map of the standard browsing data set; Analyzing the data concentration and data dispersion of the standard browsing data set using the density map, and identifying the graph of the density map; Analyzing the data skew status of the standard browsing data set based on the graph; Determining data distribution characteristics of the standard browsing data set based on the data concentration, the data dispersion and the data skew state; The step of constructing a data feature matrix of the standard browsing data set includes: Performing data conversion on the standard browsing data set according to the data distribution characteristics to obtain converted data; Performing data encoding on the converted data to obtain encoded data; Performing data dimensionality reduction on the encoded data to obtain dimensionality-reduced data; constructing an initial matrix of the standard browsing data set; After integrating the dimension-reduced data using the initial matrix, a data feature matrix is ​​obtained; The step of analyzing the data features of the standard browsing data set by using the data feature matrix includes: Using the data feature matrix to count the descriptive information of the standard browsing data set; Constructing a visualization chart of the standard browsing data set according to the descriptive information; Analyzing data features of the standard browsing data set using the visual chart; Based on the data features, the cloud computing security network is used to mine the associated users of the user to be recommended, the associated users are clustered to obtain clustered users, the similarity between the user to be recommended and the clustered users is calculated to obtain similar users, a rating matrix of the similar users is constructed, and after optimizing the rating matrix, an optimized matrix is ​​obtained, and the optimized matrix is ​​used to perform user rating on the similar users to obtain a user rating result. The step of clustering the associated users to obtain clustered users includes: Performing vector transformation on the associated user to obtain a vector user; Constructing a vector centroid of the vector user; The vector distance between the vector user and the vector centroid is calculated using the following formula: in, represents the vector distance, represents the spatial dimension of the vector user, represents the abscissa of the centroid of the i-th vector, represents the ordinate of the centroid of the i-th vector; Perform user clustering on the associated users based on the vector distance to obtain clustered users; The calculating the similarity between the user to be recommended and the clustered users to obtain similarity users includes: constructing vector data of the user to be recommended and the clustered users to obtain a first vector and a second vector; The vector similarity between the first vector and the second vector is calculated based on the first vector and the second vector in combination with the following formula: in, represents the vector similarity, represents the first vector, represents the second vector, represents the magnitude of the first vector, represents the magnitude of the second vector, represents the inclination of the first vector and the corresponding spatial coordinate point i, represents the inclination of the second vector and the corresponding spatial coordinate point i; Based on the vector similarity, performing similarity division on the user to be recommended data and the clustered user to obtain similarity users; The step of using the optimization matrix to perform user scoring on the similarity users to obtain a user scoring result includes: Constructing the interest set of the similarity user based on the optimization matrix; Based on the interest set, the following formula is used to calculate the interest preference degree of the similarity user to the interest set: in, Indicates the degree of interest preference. represents the vector similarity, represents the set of users who have rated the interest set, represents the rating of user C on interest i, represents the average value of user C's interest in his rated represents the set of similar users, represents the rating vector of user C; According to the interest preference degree, the similarity users are rated to obtain a user rating result; The streaming processing technology of cloud computing is used to collect the current browsing information of the user to be recommended data, and based on the current browsing information, an initial recommendation list for the user to be recommended data is constructed, and the initial recommendation list is updated with the user rating results to obtain a target recommendation list, and data is recommended to the user to be recommended data according to the target recommendation list.

2. The method according to claim 1, characterized in that The step of updating the initial recommendation list by using the user rating result to obtain a target recommendation list includes: Determine the preference information of the similarity user based on the user rating result; Using the preference information, weighting the initial recommendation list to obtain an updated list; Collaborative filtering is performed on the update list to obtain a target recommendation list.

3. A system for implementing data recommendation based on cloud computing, characterized in that: The system comprises: The data processing module is used to obtain the historical recommendation information and browsing information of the user to be recommended using the cloud computing database, construct a browsing data set of the user to be recommended based on the historical recommendation information and the browsing information, and perform data standardization on the browsing data set to obtain a standard browsing data set. The step of performing data standardization on the browsing data set to obtain a standard browsing data set includes: Performing data cleaning on the browsing data set to obtain cleaned data; Interpolating missing values ​​on the cleaned data to obtain interpolated data; The interpolation data are interval-integrated using the following formula to obtain a standard browsing data set: in, represents standardized data, Q represents interpolated data, represents the minimum data for interpolation data, Indicates the maximum data of the interpolation data; A data analysis module is used to analyze the data distribution characteristics of the standard browsing data set, construct a data feature matrix of the standard browsing data set based on the data distribution characteristics, and analyze the data characteristics of the standard browsing data set using the data feature matrix. The step of analyzing the data distribution characteristics of the standard browsing data set includes: constructing a density map of the standard browsing data set; Analyzing the data concentration and data dispersion of the standard browsing data set using the density map, and identifying the graph of the density map; Analyzing the data skew status of the standard browsing data set based on the graph; Determining data distribution characteristics of the standard browsing data set based on the data concentration, the data dispersion and the data skew state; The step of constructing a data feature matrix of the standard browsing data set includes: Performing data conversion on the standard browsing data set according to the data distribution characteristics to obtain converted data; Performing data encoding on the converted data to obtain encoded data; Performing data dimensionality reduction on the encoded data to obtain dimensionality-reduced data; constructing an initial matrix of the standard browsing data set; After integrating the dimension-reduced data using the initial matrix, a data feature matrix is ​​obtained; The step of analyzing the data features of the standard browsing data set by using the data feature matrix includes: Using the data feature matrix to count the descriptive information of the standard browsing data set; Constructing a visualization chart of the standard browsing data set according to the descriptive information; Analyzing data features of the standard browsing data set using the visual chart; The user scoring module is used to mine the associated users of the user to be recommended based on the data features by using the secure network of the cloud computing, cluster the associated users to obtain clustered users, calculate the similarity between the user to be recommended and the clustered users to obtain similar users, construct a scoring matrix for the similar users, optimize the scoring matrix to obtain an optimized matrix, and use the optimized matrix to score the similar users to obtain a user scoring result. The step of clustering the associated users to obtain clustered users includes: Performing vector transformation on the associated user to obtain a vector user; Constructing a vector centroid of the vector user; The vector distance between the vector user and the vector centroid is calculated using the following formula: in, represents the vector distance, represents the spatial dimension of the vector user, represents the abscissa of the centroid of the i-th vector, represents the ordinate of the centroid of the i-th vector; Perform user clustering on the associated users based on the vector distance to obtain clustered users; The calculating the similarity between the user to be recommended and the clustered users to obtain similarity users includes: constructing vector data of the user to be recommended and the clustered users to obtain a first vector and a second vector; The vector similarity between the first vector and the second vector is calculated based on the first vector and the second vector in combination with the following formula: in, represents the vector similarity, represents the first vector, represents the second vector, represents the magnitude of the first vector, represents the magnitude of the second vector, represents the inclination of the first vector and the corresponding spatial coordinate point i, represents the inclination of the second vector and the corresponding spatial coordinate point i; Based on the vector similarity, performing similarity division on the user to be recommended data and the clustered user to obtain similarity users; The step of using the optimization matrix to perform user scoring on the similarity users to obtain a user scoring result includes: Constructing the interest set of the similarity user based on the optimization matrix; Based on the interest set, the following formula is used to calculate the interest preference degree of the similarity user to the interest set: in, Indicates the degree of interest preference. represents the vector similarity, represents the set of users who have rated the interest set, represents the rating of user C on interest i, represents the average value of user C's interest in his rated represents the set of similar users, represents the rating vector of user C; According to the interest preference degree, the similarity users are rated to obtain a user rating result; The data recommendation module is used to use the streaming processing technology of the cloud computing to collect the current browsing information of the user to be recommended data, build an initial recommendation list for the user to be recommended data based on the current browsing information, update the data of the initial recommendation list using the user scoring results to obtain a target recommendation list, and recommend data to the user to be recommended data according to the target recommendation list.

Images