Data access method and system based on multi-dimensional data analysis and storable medium

Abstract

The invention relates to the technical field of database data access, and particularly discloses a data access method and system based on multi-dimensional data analysis and a storable storage.The method comprises the steps that S1, after a user accesses a database, a management recommendation module automatically allocates a tourist ID to the user, and an acquisition module acquires user information of different dimensions generated in the database access process of the tourist ID and sends the user information to the management recommendation module; the management recommendation module provides personalized content recommendation for the tourist ID; s2, the tourist ID logs in an account in the process of accessing the database, an acquisition module acquires user information of different dimensions of the account, and a management recommendation module judges whether a corresponding account permission is granted or not and provides personalized content recommendation for the user; according to the invention, the management recommendation module analyzes according to the multi-dimensional information collected by the collection module, and provides corresponding personalized content recommendation information for the user accessing the database, so as to achieve the purpose of further improving the search performance, accuracy and user experience of the user.

Description

Technical Field

[0001] The present invention relates to the technical field of database data access, and in particular to a data access method, system and storable medium based on multi-dimensional data analysis. Background Art

[0002] With the popularization and rapid development of the Internet, the number of Internet users continues to increase, and more and more companies and organizations rely on the Internet for daily operations and customer service, which has led to a sharp increase in the demand for data access. In the process of data access, users mostly search by keywords to query the data they need in the database.

[0003] In the prior art, the process of searching for keywords in a database by users is mostly to perform multiple algorithm screening queries based on the keywords input by the users in the database process, and recommend content related to the keywords to the customer query end. This method can indeed significantly improve the search performance, accuracy and user experience, but this method mostly only collects data such as the keywords input by the users, the search time, etc. to recommend content to the users, and cannot combine the user's operating habits, browsing feedback and other data for further intelligent analysis to provide personalized content recommendations to the users. Summary of the invention

[0004] The purpose of the present invention is to provide a data access method, system and storage medium based on multi-dimensional data analysis to solve the following technical problems: How to provide personalized content recommendations to users accessing the database based on multi-dimensional data analysis.

[0005] The purpose of the present invention can be achieved through the following technical solutions: A data access method based on multi-dimensional data analysis, the method comprising: S1. After a user accesses the database, the management and recommendation module automatically assigns a visitor ID to the user. The collection module collects user information of different dimensions generated by the visitor ID during the database access process, and the management and recommendation module provides personalized content recommendations for the visitor ID. S2. The visitor ID logs in to the account during the process of accessing the database. The collection module collects user information of different dimensions of the account. The management and recommendation module determines whether to grant the corresponding account authority and provides personalized content recommendations to the user.

[0006] Furthermore, the user information includes: Search keywords, search terms entered by users and their frequency; Search time, the time period during which the user conducts the search; User attributes, including age, gender, region and occupation as shown in the user's login account information; Browsing feedback: users’ browsing time, browsing speed, and conversion rate during content browsing; User login account data, including the number of times the user enters their account and password, IP address, and input speed during the account login process.

[0007] In one embodiment, in step S1, the process of providing personalized content recommendation for the visitor ID includes: Extract the content features browsed and clicked by the tourist ID, build the interest model of the tourist ID, record the corresponding features of the content preferred by the tourist ID, compare the features of the candidate content in the database with the features of the content preferred by the tourist ID, obtain the scoring parameters corresponding to each feature, and perform weighted summation on the scoring parameters of each feature to obtain the preference recommendation index of the candidate content; According to the behavior habits of the visitor ID in browsing and clicking on the content, other users in the database with similar behavior habits as the visitor ID are matched, and the content liked by other users is analyzed and calculated in turn to obtain the recommendation index of each content; A comprehensive analysis is performed on the preference recommendation index and the like recommendation index of each content to obtain the tourist recommendation index of each content, and each content is sorted in descending order according to the tourist recommendation index, and personalized content recommendations are provided for the tourist ID.

[0008] Furthermore, the process of obtaining the tourist recommendation index analysis includes:

[0009]

[0010]

[0011] The tourist recommendation index of the i-th content in the database is obtained by analyzing and calculating formulas (1)-(3) ; in, is the preference recommendation index of the i-th content in the database relative to the visitor ID, is the recommendation index of the i-th content in the database relative to the visitor ID, and is the comprehensive recommendation weight coefficient of the i-th content in the database relative to the visitor ID, M is the number of features preset for each content in the database, , is the preference score parameter of the jth feature of the i-th content relative to the visitor ID, is the weight coefficient corresponding to the jth feature in the analysis of tourist preference rating parameters, P is the number of other users in the database with similar behavior habits to the tourist ID, , is the degree of popularity of the i-th content by the k-th other user, is the similarity between the behavior habits of the kth other user and the tourist ID.

[0012] In one embodiment, in step S2, the process of providing personalized content recommendation to the user includes: Read the browsing history of the user account in the database record, extract the content features of the user's browsing and clicking in a preset time period, build the user's interest model in this time period, record the corresponding features of the user's preferred content, compare the features of the candidate content in the database with the features of the user's preferred content, obtain the scoring parameters corresponding to each feature, and perform weighted summation on the scoring parameters of each feature to obtain the preference recommendation index of the candidate content; According to the behavior habits of the user account in browsing and clicking on the content, other users in the database with similar behavior habits as the user account are matched, and the content liked by other users is analyzed and calculated in turn to obtain the like recommendation index of each content; Read the user attributes of the user account, match other accounts in the database with similar user attributes to the account, and analyze and calculate the content liked by other accounts in turn to obtain the user attribute recommendation index of each content; A comprehensive analysis is conducted on the preference recommendation index, liking recommendation index and user attribute recommendation index of each content to obtain the account recommendation index of each content. Based on the account recommendation index, the content in the database is filtered and sorted from large to small to provide personalized content recommendations to the user.

[0013] Furthermore, the process of obtaining the account recommendation index analysis includes:

[0014]

[0015]

[0016]

[0017] The account recommendation index of the i-th content in the database is obtained by analyzing and calculating formulas (4)-(7) ; in, is the preference recommendation index of the i-th content in the database relative to the user account, is the recommendation index of the i-th content in the database relative to the user account, is the user attribute recommendation index of the i-th content in the database relative to the user account, , and is the comprehensive recommendation weight coefficient of the i-th content in the database that is equivalent to the user account, is the user preference rating parameter for the jth feature of the i-th content, is the weight coefficient corresponding to the j-th user preference feature scoring parameter, is the degree of popularity of the i-th content by the k-th other user, is the similarity between the behavior habits of the kth other user and the user account, is the number of features of user attributes for each user account in the database, , is the degree of popularity of the i-th content by the user account with the l-th user attribute feature, is the influence weight coefficient of the lth user attribute feature on the user attribute recommendation index.

[0018] Furthermore, in step S2, the process in which the management recommendation module determines whether to grant the corresponding account authority includes: S21, read the permission category of the corresponding account, select the corresponding authorization permission judgment method according to the permission category of the account, if the account is a common user category, execute step S22; if the account is an administrator category, execute step S23; S22, analyzing the user login account data to determine whether the account login is abnormal. If abnormal, the account is not granted the permission to modify the data created by itself; S23. Analyze the user's login account data and browsing feedback to determine whether the account login is abnormal. If abnormal, close the account's permission to modify data in the database, and restore all data modified after the account login.

[0019] A data access system based on multi-dimensional data analysis, the system is used for a data access method based on multi-dimensional data analysis, the system comprises: Database, for users to browse, search, store data, modify and delete data in the database; The collection module is used to collect multi-dimensional data from users accessing the database; The management recommendation module analyzes multi-dimensional data based on the user status of the database access and provides personalized content recommendations to users; it manages the account permissions for logging into the database based on the account permission category and multi-dimensional data analysis.

[0020] A storable medium stores a program, which, when executed by a processor, implements the steps of a data access method based on multi-dimensional data analysis.

[0021] Beneficial effects of the present invention: (1) The present invention uses the management recommendation module to analyze the multi-dimensional information collected by the collection module and provide corresponding personalized content recommendation information to users accessing the database, so as to further improve the user search performance, accuracy and user experience.

[0022] (2) The present invention reads the permission category of the corresponding account, selects the corresponding authorization permission judgment method according to the account permission category, and provides a permission judgment method with appropriate monitoring intensity for users who log in to the database, thereby avoiding the possibility of providing excessive authorization to the logged-in user and improving the protection of database data. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] The present invention will be further described below in conjunction with the accompanying drawings.

[0024] Figure 1 It is a flowchart of the steps of the data access method based on multi-dimensional data analysis proposed by the present invention; Figure 2 It is a schematic block diagram of the data access system based on multi-dimensional data analysis proposed by the present invention. DETAILED DESCRIPTION

[0025] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0026] See also Figure 1 As shown, in one embodiment, a data access method based on multi-dimensional data analysis is provided, the method comprising: S1. After a user accesses the database, the management and recommendation module automatically assigns a visitor ID to the user. The collection module collects user information of different dimensions generated by the visitor ID during the database access process, and the management and recommendation module provides personalized content recommendations for the visitor ID. S2. The visitor ID logs in to the account during the process of accessing the database. The collection module collects user information of different dimensions of the account. The management and recommendation module determines whether to grant the corresponding account authority and provides personalized content recommendations to the user.

[0027] The user information includes: Search keywords, search terms entered by users and their frequency; Search time, the time period during which the user conducts the search; User attributes, including age, gender, region and occupation as shown in the user's login account information; Browsing feedback: users’ browsing time, browsing speed, and conversion rate during content browsing; User login account data, including the number of times the user enters their account and password, IP address, and input speed during the account login process.

[0028] Through the above technical scheme, this embodiment provides a data access method based on multi-dimensional data analysis. After the user accesses the database, the management recommendation module assigns a visitor ID to the user, and the collection module collects user information of different dimensions generated by the visitor ID in the process of accessing the database. The management recommendation module analyzes the multi-dimensional information collected by the collection module and provides personalized content recommendations for the visitor ID. If the visitor ID logs in to an account in the process of accessing the database, the collection module collects user information of different dimensions of the account. The management recommendation module provides personalized content recommendations to the user based on the multi-dimensional information and determines whether to grant corresponding account permissions. The multi-dimensional user information includes but is not limited to search keywords, search time, user attributes, browsing feedback, and user login account data. In this way, based on the multi-dimensional data collected during the user's access to the database, personalized content recommendations can be provided to the visiting user, so as to further improve the user's search performance, accuracy and user experience.

[0029] In one embodiment, in step S1, the process of providing personalized content recommendation for the visitor ID includes: Extract the content features browsed and clicked by the tourist ID, such as collecting all the interactive data such as browsing, clicking, dwelling time, commenting, liking, sharing, etc. of the tourist ID in the database, removing invalid, duplicate or abnormal data records, and extracting the content features preferred by tourists from the data, such as content type (articles, videos, pictures, etc.), content theme (travel guide, food recommendation, scenic spot introduction, etc.), browsing time, number of clicks, etc., using machine learning or deep learning methods (such as collaborative filtering, matrix decomposition, neural network, etc.) to build an interest model for the tourist ID, record the corresponding features in the interest model of the tourist ID, compare the features of the candidate content in the database with the features of the content preferred by the tourist ID, calculate the similarity, set scoring parameters for each feature, these parameters can be adjusted according to the importance of the feature or the historical behavior feedback of the tourist, and perform weighted summation on the scoring parameters of each feature to obtain the preference recommendation index of each candidate content; Based on the behavior habits of the visitor ID during browsing and clicking on the content, such as browsing time, click frequency, jump mode, etc., other users in the database with similar behavior habits as the visitor ID are matched, the content liked by these similar users is analyzed, and the recommendation index of each content is calculated, which may include indicators such as the average rating, number of likes, and number of shares of similar users; A comprehensive analysis is performed on the preference recommendation index and liking recommendation index of each content to obtain the tourist recommendation index of each content. A weighted average or other algorithm may be used to obtain a comprehensive tourist recommendation index. Each content is then sorted in descending order according to the tourist recommendation index to provide personalized content recommendations for the tourist ID.

[0030] The process of obtaining the tourist recommendation index analysis includes:

[0031]

[0032]

[0033] The tourist recommendation index of the i-th content in the database is obtained by analyzing and calculating formulas (1)-(3) ; in, is the preference recommendation index of the i-th content in the database relative to the visitor ID, is the recommendation index of the i-th content in the database relative to the visitor ID, and is the comprehensive recommendation weight coefficient of the i-th content in the database relative to the visitor ID, which is preset based on experience. M is the preset feature number for each content in the database, which is obtained by data processing and feature extraction of each content in the database. , is the preference scoring parameter of the jth feature of the i-th content relative to the tourist ID. It is obtained by comparing the jth feature of the i-th content with the feature of the preferred content of the tourist ID and setting the scoring parameters. These parameters can be adjusted according to the importance of the feature or the historical behavior feedback of the tourist. is the weight coefficient corresponding to the jth feature in the analysis of tourist preference scoring parameters, which is preset based on experience. P is the number of other users in the database with similar behavior habits to the tourist ID. This number can be preset based on experience, such as other users whose similarity reaches a certain threshold or a certain number of other users whose similarity values ​​range from large to small, etc. , is the popularity of the i-th content by the k-th other user, which can be obtained by weighted summing up the multiple behavior records of the K-th other user on the i-th content, including ratings, clicks, browsing time, likes, comments, and shares. To represent the similarity between the behavior habits of the kth other user and the tourist ID, a behavior vector can be constructed based on the behavior data of other users and the tourist DI. Each element in this vector corresponds to a behavior type (such as click, purchase, etc.), and the value of the element can be the frequency, total duration, total amount or other indicators that can reflect the intensity of the behavior. The behavior vectors of the two users are regarded as two vectors in space, and the cosine angle between them is calculated. The smaller the angle, the higher the similarity.

[0034] Through the above technical method, this embodiment provides a method for making personalized recommendations to tourists who visit a database. The method can analyze the content features and behavioral habits clicked by tourists during their browsing of the database to obtain the tourist recommendation index of each content in the database relative to the tourist, and then provide personalized content recommendations to the tourist based on the size of the tourist recommendation index, thereby improving the search accuracy and access experience of tourists during their access to the database.

[0035] In one embodiment, in step S2, the process of providing personalized content recommendation to the user includes: Read the browsing history of the user account in the database record, such as collecting all browsing, clicking, dwelling time, comments, likes, sharing and other interactive data of the user account in the database within a preset time period, removing invalid, duplicate or abnormal data records, and extracting the content features preferred by the account from the data, such as content type (articles, videos, pictures, etc.), content theme (travel guides, food recommendations, scenic spot introductions, etc.), browsing time, number of clicks, etc., using machine learning or deep learning methods (such as collaborative filtering, matrix decomposition, neural networks, etc.) to build an interest model of the account, record the corresponding features in the interest model of the account, compare the features of the candidate content in the database with the features of the account's preferred content, calculate the similarity, set scoring parameters for each feature, these parameters can be adjusted according to the importance of the feature or the historical behavior feedback of the account, and perform weighted summation on the scoring parameters of each feature to obtain the preference recommendation index of each candidate content; Based on the user account's behavior habits when browsing and clicking on content, such as browsing time, click frequency, jump mode, etc., match other users in the database with similar behavior habits to the user account, analyze the content liked by these similar users, and calculate the recommendation index for each content, which may include indicators such as the average rating, number of likes, and number of shares of similar users; Read the user attributes of the user account, such as age, gender, region, occupation, etc., match other accounts in the database with similar user attributes to the account, analyze and calculate the content liked by other accounts in turn, and obtain the user attribute recommendation index of each content; A comprehensive analysis is conducted on the preference recommendation index, liking recommendation index and user attribute recommendation index of each content to obtain the account recommendation index of each content. These indexes can be simply weighted summed, or a more complex model can be used to merge them. The account recommendation index is used as a standard to filter and sort the content in the database from large to small, and provide personalized content recommendations to the user.

[0036] The process of obtaining the account recommendation index analysis includes:

[0037]

[0038]

[0039]

[0040] The account recommendation index of the i-th content in the database is obtained by analyzing and calculating formulas (4)-(7) ; in, is the preference recommendation index of the i-th content in the database relative to the user account, is the recommendation index of the i-th content in the database relative to the user account, is the user attribute recommendation index of the i-th content in the database relative to the user account, , and is the comprehensive recommendation weight coefficient of the i-th content in the database, which is equivalent to the user account, and is preset based on experience. The user preference scoring parameters for the jth feature of the i-th content are obtained by comparing the jth feature of the i-th content with the features of the historical preferred content of the account and setting the scoring parameters. These parameters can be adjusted according to the importance of the features or the historical behavior feedback of visitors. is the weight coefficient corresponding to the j-th user preference feature scoring parameter, which is preset based on experience. is the popularity of the i-th content by the k-th other user, which can be obtained by weighted summing up the multiple behavior records of the K-th other user on the i-th content, including ratings, clicks, browsing time, likes, comments, and shares. is the similarity between the behavior habits of the kth other user and the user account. A behavior vector can be constructed based on the behavior data of other users and the visitor DI. Each element in this vector corresponds to a behavior type (such as click, purchase, etc.). The value of the element can be the frequency, total duration, total amount or other indicators that can reflect the intensity of the behavior. The behavior vectors of the two users are regarded as two vectors in space, and the cosine angle between them is calculated. The smaller the angle, the higher the similarity. is the characteristic number of user attributes of each user account in the database, which is obtained based on experience. , To calculate the degree of liking of the i-th content by the user account having the l-th user attribute characteristic, the user account having the l-th user attribute characteristic can be screened out from the database. These users constitute a specific user group. For the screened user group, their behavior data related to the i-th content is collected. These data may include: number of views, number of clicks, dwell time, interactive behavior and conversion behavior, etc. A liking model is constructed using the collected user behavior data. This model can be based on statistics or machine learning, and is used to predict the liking of the i-th content by the users in this group according to their behavior. is the influence weight coefficient of the lth user attribute feature on the user attribute recommendation index, which is obtained based on experience.

[0041] Through the above technical solution, this embodiment provides a method for personalized recommendation for accounts accessing a database. The account recommendation index of each content in the database relative to the account can be obtained based on the content features, behavioral habits and account user attributes clicked by the account during browsing the database within a preset time period, and then personalized content recommendations can be provided to the user who logs in to the account based on the size of the account recommendation index, thereby achieving the search accuracy and access experience of the user who logs in to the account during accessing the database.

[0042] In one embodiment, in step S2, the process in which the management recommendation module determines whether to grant the corresponding account permission includes: S21, read the permission category of the corresponding account, select the corresponding authorization permission judgment method according to the permission category of the account, if the account is a common user category, the common user has the permission to access the database and modify and delete the data created and stored in the database, and execute step S22; if the account is an administrator category, the administrator has the permission to access the database and modify and delete all stored data in the database, and execute step S23; S22. Analyze the user login account data, including but not limited to login time, location, device and other information, to determine whether there is abnormal login behavior (such as frequent changes in login location, use of unknown devices to log in, excessive number of incorrect passwords during login, large deviation between password input speed and historical data during login, etc.), and determine whether the account login is abnormal. If abnormal, do not grant the account the right to modify the data created by itself to prevent potential security risks; S23. Analyze the user's login account data and browsing feedback. The user's login account data includes but is not limited to information such as login time, location, and device, to determine whether there is abnormal login behavior (such as frequent changes in login location, use of unknown devices to log in, excessive number of incorrect password attempts during login, and password input speed during login deviating too much from historical data, etc.). Browsing feedback includes the frequency of access, duration of access, browsing path, browsing content, click operation habits, and operation habits of modifying and deleting data in the database during the account's access to the database, to determine whether the account login is abnormal. If abnormal, close the account's permission to modify data in the database to prevent data from being maliciously tampered with. At the same time, in order to ensure data security, the system will also restore all data modified by the account after this login to ensure that the data is restored to the state before the abnormal operation.

[0043] See also Figure 2 As shown, in one embodiment, a data access system based on multi-dimensional data analysis is provided, the system is used for a data access method based on multi-dimensional data analysis, and the system includes: Database, for users to browse, search, store data, modify and delete data in the database; The collection module is used to collect multi-dimensional data from users accessing the database; The management recommendation module collects different multi-dimensional data according to the user status of accessing the database, analyzes the multi-dimensional data, and provides personalized content recommendations to users; it manages the account permissions for logging into the database according to the account permission category and multi-dimensional data analysis.

[0044] A storable medium stores a program, which, when executed by a processor, implements the steps of a data access method based on multi-dimensional data analysis.

[0045] The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and cannot be considered to limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. A data access method based on multi-dimensional data analysis, characterized in that: The method comprises: S1. After a user accesses the database, the management and recommendation module automatically assigns a visitor ID to the user. The collection module collects user information of different dimensions generated by the visitor ID during the database access process, and the management and recommendation module provides personalized content recommendations for the visitor ID. S2. The visitor ID logs in to the account during the process of accessing the database. The collection module collects user information of different dimensions of the account. The management and recommendation module determines whether to grant the corresponding account authority and provides personalized content recommendations to the user.

2. The data access method based on multi-dimensional data analysis according to claim 1, characterized in that: The user information includes: Search keywords, search terms entered by users and their frequency; Search time, the time period during which the user conducts the search; User attributes, including age, gender, region and occupation as shown in the user's login account information; Browsing feedback: users’ browsing time, browsing speed, and conversion rate during content browsing; User login account data, including the number of times the user enters their account and password, IP address, and input speed during the account login process.

3. The data access method based on multi-dimensional data analysis according to claim 2, characterized in that: In step S1, the process of providing personalized content recommendation for the visitor ID includes: Extract the content features browsed and clicked by the tourist ID, build the interest model of the tourist ID, record the corresponding features of the content preferred by the tourist ID, compare the features of the candidate content in the database with the features of the content preferred by the tourist ID, obtain the scoring parameters corresponding to each feature, and perform weighted summation on the scoring parameters of each feature to obtain the preference recommendation index of the candidate content; According to the behavior habits of the visitor ID in browsing and clicking on the content, other users in the database with similar behavior habits as the visitor ID are matched, and the content liked by other users is analyzed and calculated in turn to obtain the recommendation index of each content; A comprehensive analysis is performed on the preference recommendation index and the like recommendation index of each content to obtain the tourist recommendation index of each content, and each content is sorted in descending order according to the tourist recommendation index, and personalized content recommendations are provided for the tourist ID.

4. The data access method based on multi-dimensional data analysis according to claim 3, characterized in that: The process of obtaining the tourist recommendation index analysis includes: The tourist recommendation index of the i-th content in the database is obtained by analyzing and calculating formulas (1)-(3) ; in, is the preference recommendation index of the i-th content in the database relative to the visitor ID, is the recommendation index of the i-th content in the database relative to the visitor ID, and is the comprehensive recommendation weight coefficient of the i-th content in the database relative to the visitor ID, M is the number of features preset for each content in the database, , is the preference score parameter of the jth feature of the i-th content relative to the visitor ID, is the weight coefficient corresponding to the jth feature in the analysis of tourist preference rating parameters, P is the number of other users in the database with similar behavior habits to the tourist ID, , is the degree of popularity of the i-th content by the k-th other user, is the similarity between the behavior habits of the kth other user and the tourist ID.

5. The data access method based on multi-dimensional data analysis according to claim 4, characterized in that: In step S2, the process of providing personalized content recommendation to the user includes: Read the browsing history of the user account in the database record, extract the content features of the user's browsing and clicking in a preset time period, build the user's interest model in this time period, record the corresponding features of the user's preferred content, compare the features of the candidate content in the database with the features of the user's preferred content, obtain the scoring parameters corresponding to each feature, and perform weighted summation on the scoring parameters of each feature to obtain the preference recommendation index of the candidate content; According to the behavior habits of the user account in browsing and clicking on the content, other users in the database with similar behavior habits to the user account are matched, and the content liked by other users is analyzed and calculated in turn to obtain the like recommendation index of each content; Read the user attributes of the user account, match other accounts in the database with similar user attributes to the account, and analyze and calculate the content liked by other accounts in turn to obtain the user attribute recommendation index of each content; A comprehensive analysis is conducted on the preference recommendation index, liking recommendation index and user attribute recommendation index of each content to obtain the account recommendation index of each content. Based on the account recommendation index, the content in the database is filtered and sorted from large to small to provide personalized content recommendations to the user.

6. The data access method based on multi-dimensional data analysis according to claim 5, characterized in that: The process of obtaining the account recommendation index analysis includes: The account recommendation index of the i-th content in the database is obtained by analyzing and calculating formulas (4)-(7) ; in, is the preference recommendation index of the i-th content in the database relative to the user account, is the recommendation index of the i-th content in the database relative to the user account, is the user attribute recommendation index of the i-th content in the database relative to the user account, , and is the comprehensive recommendation weight coefficient of the i-th content in the database that is equivalent to the user account, is the user preference rating parameter for the jth feature of the i-th content, is the weight coefficient corresponding to the j-th user preference feature scoring parameter, is the degree of popularity of the i-th content by the k-th other user, is the similarity between the behavior habits of the kth other user and the user account, is the number of features of user attributes for each user account in the database, , is the degree of popularity of the i-th content by the user account with the l-th user attribute feature, is the influence weight coefficient of the lth user attribute feature on the user attribute recommendation index.

7. The data access method based on multi-dimensional data analysis according to claim 5, characterized in that: In step S2, the process of the management recommendation module determining whether to grant the corresponding account authority includes: S21, read the permission category of the corresponding account, select the corresponding authorization permission judgment method according to the permission category of the account, if the account is a common user category, execute step S22; if the account is an administrator category, execute step S23; S22, analyzing the user login account data to determine whether the account login is abnormal. If abnormal, the account is not granted the permission to modify the data created by itself; S23. Analyze the user's login account data and browsing feedback to determine whether the account login is abnormal. If abnormal, close the account's permission to modify data in the database, and restore all data modified after the account login.

8. A data access system based on multi-dimensional data analysis, characterized in that: The system is used for the data access method based on multi-dimensional data analysis according to any one of claims 1 to 7, and the system comprises: Database, for users to browse, search, store data, modify and delete data in the database; The collection module is used to collect multi-dimensional data from users accessing the database; The management recommendation module analyzes multi-dimensional data based on the user status of the database access and provides personalized content recommendations to users; it manages the account permissions for logging into the database based on the account permission category and multi-dimensional data analysis.

9. A storable medium having a program stored thereon, characterized in that: When the program is executed by a processor, the steps of the data access method based on multi-dimensional data analysis as described in any one of claims 1 to 7 are implemented.

Images