A database interactive management method and system
The method addresses the lack of personalized recommendations and data security in traditional database interactions by analyzing user history and blockchain stability, providing secure and efficient resource recommendations.
Patent Information
- Application Number
- CN202410763132.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-06-13
- Publication Date
- 2024-12-06
- Estimated Expiration
- 2044-06-13
AI Technical Summary
Traditional database interaction management methods fail to utilize user historical interaction information for personalized resource recommendations and do not adequately integrate blockchain technology for data security and stability.
A database interaction management method that obtains user interaction requests, analyzes historical resource tags, recommends resources based on these tags, locates the resources on a blockchain, and assesses blockchain stability before retrieval to ensure secure and stable data transfer.
Enables personalized resource recommendations while ensuring data security and stability by integrating blockchain technology, enhancing interaction efficiency and security.
Smart Images

Figure CN118484548B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of database interaction, and in particular to a database interaction management method and system. Background Art
[0002] With the rapid development of information technology, database applications are becoming more and more widespread, and user interactions with databases are becoming more and more frequent. However, traditional database interaction management methods often ignore the importance of user historical interaction information and are unable to provide users with personalized resource recommendations. At the same time, with the rise of blockchain technology, how to combine blockchain technology to ensure data security and stability has also become a new challenge in the field of database interaction management. Summary of the invention
[0003] The purpose of the present invention is to provide a database interaction management method and system.
[0004] In a first aspect, an embodiment of the present invention provides a database interaction management method, comprising:
[0005] Obtain the current user's interaction request for the preset database;
[0006] According to the interaction request, obtaining a historical resource tag set of the current user for the preset database;
[0007] Performing resource content interactive recommendation for the interactive request according to the historical resource tag set;
[0008] Determine, according to the recommendation result of the interactive recommendation of the resource content, the target blockchain where the recommended resource corresponding to the recommendation result is located;
[0009] Performing a stability assessment on the target blockchain to obtain a blockchain stability assessment result corresponding to the target blockchain;
[0010] On the basis that the blockchain stability evaluation result indicates that the blockchain is stable, the recommended resources are obtained from the target blockchain.
[0011] In a second aspect, an embodiment of the present invention provides a server system, including a server, wherein the server is used to execute the method described in the first aspect.
[0012] Compared with the prior art, the beneficial effects provided by the present invention include: adopting a database interaction management method and system disclosed in the present invention, firstly, by obtaining and analyzing the user's interaction request and historical resource tags for the preset database, personalized resource content recommendation is realized. After determining the recommended resource, the target blockchain where the resource is located is further located, and its stability is evaluated. Only when the blockchain is stable, the recommended resource is obtained from the target blockchain, thereby ensuring the security of the data and the stability of the interaction. Such a design integrates personalized recommendation and blockchain technology, and improves the efficiency and security of database interaction. BRIEF DESCRIPTION OF THE DRAWINGS
[0013] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only illustrate certain embodiments of the present invention and should not be regarded as limiting the scope. For those of ordinary skill in the art, other related drawings can also be obtained based on these drawings without creative work.
[0014] Figure 1 A schematic diagram of the steps of the database interaction management method provided by an embodiment of the present invention;
[0015] Figure 2 A schematic block diagram of the structure of a computer device provided in an embodiment of the present invention. DETAILED DESCRIPTION
[0016] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Generally, the components of the embodiments of the present invention described and shown in the drawings here can be arranged and designed in various different configurations.
[0017] The specific implementation modes of the present invention are described in detail below in conjunction with the accompanying drawings.
[0018] In order to solve the technical problems in the aforementioned background technology, Figure 1 This is a flow chart of a database interaction management method provided in an embodiment of the present disclosure. The database interaction management method is introduced in detail below.
[0019] Step S201, obtaining the current user's interaction request for a preset database;
[0020] Step S202, obtaining a historical resource tag set of the current user for the preset database according to the interaction request;
[0021] Step S203, performing resource content interactive recommendation for the interactive request according to the historical resource tag set;
[0022] Step S204, determining the target blockchain where the recommended resource corresponding to the recommendation result is located according to the recommendation result of the resource content interactive recommendation;
[0023] Step S205, performing stability assessment on the target blockchain to obtain a blockchain stability assessment result corresponding to the target blockchain;
[0024] Step S206: on the basis that the blockchain stability assessment result indicates that the blockchain is stable, obtaining the recommended resources from the target blockchain.
[0025] In an embodiment of the present invention, illustratively, the server receives an interaction request from user A, which is initiated through the front-end interface of the database management system. User A wants to query relevant information about "development of artificial intelligence technology". The server accurately captures this request and starts processing. The server queries the historical interaction records of user A and finds that user A has repeatedly queried resources related to tags such as "artificial intelligence", "machine learning", and "deep learning" in the past few months. The server classifies these tags into a historical resource tag set of user A. Based on the historical resource tag set of user A, the server recommends the use of two articles, "Application of Neural Networks in Artificial Intelligence" and "Deep Learning Algorithm Optimization". These recommended contents are closely related to the topics that user A has previously paid attention to. The server determines that both recommended articles are stored on a blockchain named "TechXXX". TechXXX is a blockchain dedicated to storing science and technology related literature. The server conducts a stability assessment on TechXXX, checking its key indicators such as block generation speed, transaction confirmation time, and network latency. After the assessment, the server concludes that TechXXX is currently running stably without obvious performance fluctuations or security risks. Based on the stability evaluation results of TechXXX, the server successfully retrieved and obtained the contents of two articles, "Application of Neural Networks in Artificial Intelligence" and "Deep Learning Algorithm Optimization", from TechXXX. The server returned these contents to user A, completing the processing of this database interaction request. Through the above steps, the server not only made accurate content recommendations based on the user's historical preferences, but also ensured the authenticity and security of the recommended resources through blockchain technology.
[0026] In the embodiment of the present invention, the aforementioned step S203 can be implemented through the following example.
[0027] Acquire the historical resource tag set, wherein the historical resource tag set includes each historical resource tag corresponding to the current user;
[0028] Acquire a historical resource feature set and a user historical operation feature set according to the historical resource tag set, wherein the historical resource feature set is obtained by performing a feature extraction operation on the resource content of each historical resource tag, and the user historical operation feature set is obtained by performing a feature extraction operation on the user operation record of each historical resource tag, and the user operation record is extracted from the business scenario to which the historical resource tag belongs;
[0029] Integrate the historical resource feature set with the user historical operation feature set to obtain an integrated feature set, and extract resource association information of the integrated features in the integrated feature set to obtain a resource association feature set;
[0030] Identifying multiple business scenario optimization information corresponding to the resource-associated feature set from each preset business scenario optimization information, and performing feature optimization processing on the resource-associated feature set according to the multiple business scenario optimization information to obtain an optimized feature set;
[0031] The optimized feature set is used to evaluate and obtain the adaptation coefficient associated with each preset pending resource tag, and to perform resource content interactive recommendation according to the adaptation coefficient associated with each preset pending resource tag.
[0032] In an embodiment of the present invention, illustratively, the server retrieves a historical resource tag set of user A from the database. These tags include "artificial intelligence", "machine learning", "data science", "programming", etc., which are the interests of user A when interacting with the database in the past. For the historical resource tag "artificial intelligence", the server performs a feature extraction operation to obtain resource content features related to the tag, such as "algorithm application", "model training", etc., to form a historical resource feature set. At the same time, the server also extracts user A's operation records from the business scenarios to which the historical resource tags belong, such as "browsing time", "likes", "comments", etc., to form a user historical operation feature set. The server integrates the historical resource feature set with the user historical operation feature set to form a more comprehensive integrated feature set. For example, for the "artificial intelligence" tag, the integrated features may include "algorithm application + high-frequency browsing", "model training + likes", etc. Then, the server extracts the associated features between resources from the integrated feature set, such as "browsing 'algorithm application' and often browsing 'model training'", etc., to form a resource associated feature set. The server identifies the optimization information related to the resource associated feature set from the preset business scenario optimization information. For example, for the associated feature "after browsing 'algorithm application', 'model training' is often browsed", the business scenario optimization information can prompt "after recommending 'algorithm application', resources related to 'model training' should be recommended first". Based on these optimization information, the server performs feature optimization processing on the resource-associated feature set to obtain an optimized feature set. Finally, the server uses the optimized feature set to evaluate the adaptation coefficient of each preset pending resource tag. For example, the pending resource tag "Basic Tutorial on Deep Learning" can obtain a higher adaptation coefficient due to its high correlation with "algorithm application" and "model training". The server makes interactive recommendations for resource content based on these adaptation coefficients and recommends the most relevant resources to user A. Through the above steps, the server can more accurately recommend resource content based on the user's historical preferences and business scenario optimization information, thereby improving user satisfaction and recommendation efficiency.
[0033] In the embodiment of the present invention, the acquisition of the historical resource feature set and the user historical operation feature set according to the historical resource tag set may be implemented through the following examples.
[0034] According to each historical resource tag in the historical resource tag set, searching each pre-acquired historical resource feature for the historical resource feature associated with each historical resource tag, to obtain the historical resource feature set;
[0035] Acquire multiple user operation records of each historical resource tag from multiple business scenarios to which each historical resource tag belongs;
[0036] For each historical resource tag, feature vectors of multiple user operation records are extracted to obtain multiple user operation vectors, and the multiple user operation vectors are integrated to obtain the user historical operation feature set.
[0037] In an embodiment of the present invention, illustratively, the server has a historical resource tag set of user A, including tags such as "artificial intelligence" and "data science". In order to obtain the historical resource features associated with these tags, the server first accesses a pre-acquired historical resource feature library. In this library, each resource feature is associated with the corresponding resource tag. For example, for the tag "artificial intelligence", the server searches for features associated with the tag in the historical resource feature library. These features may include "using neural networks", "natural language processing technology applications", etc. The server classifies these found features into the historical resource feature set. Similarly, for the "data science" tag, the server searches and finds associated features such as "data analysis methods" and "data visualization technology", and adds them to the historical resource feature set. Next, the server obtains the operation records of user A from multiple business scenarios to which each historical resource tag belongs. These business scenarios may include online course learning platforms, technical forum discussion areas, e-book reading applications, etc. Taking the "artificial intelligence" tag as an example, the server extracts relevant information from the learning records of user A on the online course learning platform, such as course viewing time, completion, number of notes, etc. At the same time, user A's posting, replying, and liking behavior records on the topic of artificial intelligence will also be collected from the technical forum discussion area. For each user operation record related to the historical resource tag, the server will further extract feature vectors. These feature vectors can include quantitative indicators such as viewing time, interaction frequency, and number of likes. For example, under the "artificial intelligence" tag, user A's viewing time is an important feature vector. The server will record the total viewing time of user A on artificial intelligence-related courses and use it as a feature value. The server will integrate multiple user operation vectors under each historical resource tag. This means that it will combine these vectors to form a more comprehensive user historical operation feature set. This set not only includes user A's operation behavior in various business scenarios, but also reflects the user's interest and preference for these resource tags. Through the above steps, the server can deeply understand the user's behavioral characteristics in various business scenarios based on the user's historical resource tag set, thereby providing more accurate data support for subsequent resource content recommendations.
[0038] In the embodiment of the present invention, the historical resource feature set and the user historical operation feature set are integrated to obtain an integrated feature set, which can be implemented through the following examples.
[0039] The historical resource features in the historical resource feature set are integrated with the corresponding user historical operation features in the user historical operation feature set to obtain the integrated feature set.
[0040] In the embodiment of the present invention, illustratively, the server first identifies each historical resource feature in the historical resource feature set. Taking the historical resource tag "artificial intelligence" as an example, its associated historical resource features may include "algorithm complexity", "model training efficiency", etc. Then, the server will find the user historical operation features corresponding to these historical resource features from the user historical operation feature set. For "algorithm complexity", the user historical operation features may include the browsing time and repeated browsing times of the user when reading or learning the relevant algorithm; and for "model training efficiency", the user historical operation features involve the operation frequency and the number of successful trainings of the user when practicing model training. Then, the server integrates each historical resource feature with its corresponding user historical operation feature. This can be achieved in many ways, for example, the value of the historical resource feature can be weighted averaged with the value of the user historical operation feature, or the two can be combined according to a certain algorithm model to form a new integrated feature value. Finally, the server generates a new integrated feature set by integrating all historical resource features and user historical operation features in this way. This set not only contains the inherent attributes of the resource, but also integrates the user's actual operation behavior and preferences for these resources, so it can more comprehensively reflect the user's real needs and interests in the resource. Through the above integration processing steps, the server obtains an integrated feature set that combines resource content and user behavior, which will provide a more accurate and personalized data foundation for subsequent resource recommendations.
[0041] In the embodiment of the present invention, the extraction of resource association information of the integrated features in the integrated feature set to obtain the resource association feature set can be implemented through the following examples.
[0042] Projecting the integrated feature set into a feature space to obtain a retrieval vector set, an index vector set and a data representation vector set;
[0043] The correlation coefficient between the retrieval vector set and the index vector set is calculated to obtain a correlation coefficient set, and a weighted adjustment is applied to the data representation vector set according to the correlation coefficient set to obtain the resource correlation feature set.
[0044] In an embodiment of the present invention, illustratively, first, the server performs feature space projection on the integrated feature set. The purpose of this step is to convert complex integrated features into a vector form that is easier to process and analyze. Through specific mathematical transformations, the server projects the integrated feature set into three different vector spaces, generating a retrieval vector set, an index vector set, and a data representation vector set, respectively. For a specific example, assume that the integrated feature set contains multiple features about the "artificial intelligence" course, such as course difficulty, learning time, user interaction frequency, etc. The server converts these features into retrieval vectors (for rapid retrieval of similar resources), index vectors (for establishing an index structure of resources), and data representation vectors (for describing the core features of resources) through projection. Next, the server calculates the correlation coefficient between the retrieval vector set and the index vector set. The purpose of this step is to measure the similarity or correlation between different resources. The server uses a suitable algorithm (such as cosine similarity, Pearson correlation coefficient, etc.) to calculate the correlation coefficient between each pair of retrieval vectors and index vectors, thereby forming a set of correlation coefficients. The server calculates the correlation coefficient between the retrieval vector and the index vector related to the "artificial intelligence" course. These coefficients reflect the similarity between different courses. For example, the "Artificial Intelligence" course and the "Machine Learning" course may have a high correlation coefficient because they have a close content connection. Finally, the server will weight the data representation vector set according to the set of correlation coefficients. The purpose of this step is to strengthen the features of highly relevant resources and weaken the features of low-relevant resources. In this way, the server can capture the association relationship between resources more accurately. After weighted adjustment, the server will obtain the final resource association feature set. This set not only contains the core features of the resources, but also reflects the relationship between resources. Taking the "Artificial Intelligence" course as an example, the weighted adjusted resource association feature set will emphasize the common features of highly relevant courses such as "Machine Learning" and "Deep Learning", thereby providing stronger support for subsequent resource recommendations.
[0045] In an embodiment of the present invention, the step of identifying multiple business scenario optimization information corresponding to the resource-associated feature set from each preset business scenario optimization information, and performing feature optimization processing on the resource-associated feature set according to the multiple business scenario optimization information to obtain an optimized feature set includes:
[0046] Calculate the feature matching degree according to the resource-related feature set to obtain the matching coefficient corresponding to each preset business scenario optimization information;
[0047] Selecting each preset business scenario optimization information according to the matching coefficient corresponding to each preset business scenario optimization information to obtain a plurality of business scenario optimization information corresponding to the resource association feature set;
[0048] According to the plurality of business scenario optimization information, feature optimization processing is performed on the resource-associated feature sets respectively to obtain a plurality of current optimized feature sets;
[0049] The multiple current optimization feature sets are integrated according to the matching coefficients associated with the optimization information of multiple business scenarios to obtain the optimization feature set.
[0050] In an embodiment of the present invention, illustratively, the server first calculates the matching degree between each feature in the resource-associated feature set and the preset business scenario optimization information. This usually involves complex algorithms, such as cosine similarity, Euclidean distance, etc., to measure the similarity or correlation between the feature and the optimization information. For example, assuming that there is a feature in the resource-associated feature set that is "user preference for night learning", the server will calculate the matching degree between this feature and each preset business scenario optimization information. If a business scenario optimization information is about "providing a night learning mode to reduce eye fatigue", then the matching coefficient between this optimization information and the feature will be very high. According to the calculated matching coefficient, the server will select the business scenario optimization information that is most relevant to the resource-associated feature set. Usually, the server will set a threshold and only select the optimization information whose matching coefficient exceeds this threshold. If the matching coefficient between the optimization information of "providing a night learning mode to reduce eye fatigue" and the feature of "user preference for night learning" exceeds the set threshold, then the server will select this optimization information. For each selected business scenario optimization information, the server will optimize its corresponding resource-associated feature. This involves operations such as weighting, filtering, and conversion of features, with the goal of making the features more in line with the needs of business scenarios. Taking "user preference for night learning" as an example, the server will add a new feature "interface brightness suitable for night learning" based on the optimization information of "providing a night learning mode to reduce eye fatigue", or weight the original features to emphasize the importance of night learning. After such optimization processing, the server will obtain multiple current optimized feature sets, each of which is obtained based on a specific business scenario optimization information. Finally, the server will integrate multiple current optimized feature sets based on the matching coefficient associated with each business scenario optimization information. This usually means weighted averaging the features in different sets or weighted superposition based on the matching coefficient. For example, if the matching coefficient of the optimization information of "providing a night learning mode to reduce eye fatigue" is high, then the current optimized feature set associated with it will occupy a greater weight in the integrated processing. The final optimized feature set will be a feature set that integrates multiple business scenario optimization information, which is used to guide the server to make more accurate resource recommendations and service provision.
[0051] In the embodiment of the present invention, the multiple current optimization feature sets are integrated and processed according to the matching coefficients associated with each of the multiple business scenario optimization information to obtain the optimization feature set, which can be implemented through the following examples.
[0052] Acquire the number of optimization information items of the multiple business scenario optimization information, adjust the matching coefficient associated with each preset business scenario optimization information according to the number of optimization information items, and obtain the adjustment coefficient associated with each preset business scenario optimization information;
[0053] Weight-normalize the adjustment coefficients respectively associated with each preset business scenario optimization information to obtain the target matching coefficients respectively associated with each preset business scenario optimization information;
[0054] Determining the target matching coefficients associated with each of the plurality of business scenario optimization information from the target matching coefficients associated with each of the preset business scenario optimization information;
[0055] The multiple current optimization feature sets are integrated and processed according to the target matching coefficients associated with each of the multiple business scenario optimization information to obtain the optimization feature set.
[0056] In an embodiment of the present invention, exemplarily, when the server is preparing to perform integration processing on multiple current optimization feature sets, it will first adjust and optimize the matching coefficient according to the characteristics of the business scenario optimization information to ensure the accuracy and effectiveness of the integration processing. The server will first obtain the number of optimization information items in each business scenario optimization information. This number of items reflects the complexity and detail of the optimization information. For example, a business scenario optimization information may contain optimization suggestions on multiple aspects such as interface design, interaction logic, content recommendation, etc., while another may only focus on improving loading speed. According to these number of items, the server will adjust the matching coefficient associated with each preset business scenario optimization information. Optimization information with a large number of items will obtain a lower adjustment coefficient because they involve a wider range of aspects, and a single optimization point is more important and specific. This adjustment helps to balance the impact of different optimization information in the integration process. The adjusted matching coefficient (i.e., adjustment coefficient) will be weighted and standardized. This means that the server will divide each adjustment coefficient by the sum of all adjustment coefficients to ensure that their sum is 1. In this way, each business scenario optimization information will obtain a standardized target matching coefficient, which will be used for subsequent integration processing. For example, if the adjustment coefficient of a business scenario optimization information is 0.3, and the sum of the adjustment coefficients of all optimization information is 1.5, then its target matching coefficient is 0.3 divided by 1.5, which is equal to 0.2. The server will determine the target matching coefficients associated with each of the multiple business scenario optimization information from the target matching coefficients associated with each preset business scenario optimization information. This step is actually to confirm the calculation results of the previous step to ensure that each optimization information has a corresponding target matching coefficient. Finally, the server will integrate multiple current optimization feature sets according to the target matching coefficients associated with each of the multiple business scenario optimization information. Specifically, it is to perform weighted averaging or linear combination of the features in the current optimization feature set according to each target matching coefficient. For example, if there are two current optimization feature sets A and B, and their target matching coefficients are 0.6 and 0.4 respectively, then each feature value in the integrated optimization feature set will be the sum of 60% of the corresponding feature value in set A and 40% of the corresponding feature value in set B. Through such integrated processing, the server can obtain an optimized feature set that integrates optimization information from multiple business scenarios. This set will more comprehensively reflect user needs and preferences, thereby guiding the server to provide more accurate and personalized services.
[0057] In an embodiment of the present invention, adjusting the matching coefficient associated with each preset business scenario optimization information according to the number of optimization information entries to obtain the adjustment coefficient associated with each preset business scenario optimization information can be implemented through the following examples.
[0058] Determine each matching coefficient of a high priority and each matching coefficient of a low priority from the matching coefficients respectively associated with each preset business scenario optimization information according to the number of optimization information items;
[0059] Each matching coefficient of the high priority is maintained as it is, and each matching coefficient of the low priority is adjusted to a corresponding value of the preset data characterization vector, so as to obtain an adjustment coefficient associated with each preset business scenario optimization information.
[0060] In an embodiment of the present invention, exemplarily, when processing multiple preset business scenario optimization information, the server will adjust the associated matching coefficient according to the number of entries of the optimization information to ensure that the optimization information of different complexity gets a reasonable weight in the integrated processing. The server will first analyze the number of entries of each preset business scenario optimization information. Optimization information with a large number of entries usually means that it covers more details and aspects, and can therefore be regarded as a more complex or comprehensive optimization suggestion. On the contrary, optimization information with a small number of entries can be more focused on a specific aspect. Based on this analysis, the server will determine which are high-priority matching coefficients and which are low-priority matching coefficients from the matching coefficients associated with each preset business scenario optimization information. For example, a threshold number of entries can be set, and optimization information exceeding the threshold is considered to be high priority, and those below the threshold are considered to be low priority. For high-priority matching coefficients, the server will maintain them as they are, because these optimization information are more important or comprehensive, and need to maintain their influence in the integrated processing. For low-priority matching coefficients, the server will adjust them to the corresponding values of the preset data representation vector. This preset data characterization vector can be a fixed value set based on historical data or expert experience to ensure that low-priority optimization information can also be properly considered in the integrated processing, but will not excessively affect the final result. For example, suppose that the matching coefficient originally associated with a low-priority optimization information is 0.2, and the corresponding value of the preset data characterization vector is 0.1. During the adjustment process, the server will adjust the low-priority matching coefficient from 0.2 to 0.1. Through such adjustments, the server can ensure that optimization information of different complexity and importance can be reasonably considered in the integrated processing, thereby improving the accuracy and effectiveness of the final optimization feature set. This adjustment method also helps to balance the influence of different optimization information and avoid some overly complex or overly simple optimization information from having too much or too little impact on the final result.
[0061] In the embodiment of the present invention, the multiple current optimization feature sets are integrated and processed according to the target matching coefficients associated with each of the multiple business scenario optimization information to obtain the optimization feature set, which can be implemented through the following examples.
[0062] For each business scenario optimization information, the corresponding current optimization feature set is modified according to the target matching coefficient to obtain each modified feature set;
[0063] Each of the modified feature sets is integrated to obtain the optimized feature set.
[0064] In an embodiment of the present invention, exemplarily, after having multiple business scenario optimization information and their respective associated target matching coefficients, the server will perform integration processing to obtain the final optimized feature set. This process can be divided into two main steps. First, for each business scenario optimization information, the server will correct the corresponding current optimized feature set according to its associated target matching coefficient. This correction process is to ensure that each feature set can be properly weighted according to its importance when integrated. Assume that the server has three business scenario optimization information, namely A, B and C, and their target matching coefficients are 0.5, 0.3 and 0.2 respectively. Each optimization information corresponds to a current optimized feature set. For optimization information A, its target matching coefficient is 0.5, which means that it should occupy a larger weight in the integration process. Therefore, the server will correct the current optimized feature set corresponding to A according to a coefficient of 0.5 to strengthen the feature values in this set. For optimization information B and C, the server will make similar corrections to their corresponding current optimized feature sets according to coefficients of 0.3 and 0.2 respectively. In this way, each corrected feature set reflects the importance of its corresponding business scenario optimization information in the integration process. Next, the server will integrate each modified feature set. This process usually involves weighted averaging or other forms of combination of the features in the modified feature set. The server now has three modified feature sets, corresponding to the optimization information A, B, and C. The server will first align the same features in each modified feature set. Then, these features are weighted averaged according to the target matching coefficient of each modified feature set. For example, for a specific feature, its value in the final optimized feature set will be the sum of 0.5 times the feature value in set A, 0.3 times in set B, and 0.2 times in set C. This process is performed on all features, and finally an optimized feature set that integrates the optimization information of multiple business scenarios is obtained. Through this integrated processing, the server can comprehensively consider the needs and characteristics of multiple business scenarios to generate a more comprehensive and refined optimized feature set to support subsequent service provision and decision making.
[0065] In the embodiment of the present invention, after performing feature optimization processing on the resource-associated feature set according to the plurality of business scenario optimization information to obtain an optimized feature set, the following implementation manner is also provided.
[0066] The optimized feature set is used as the integrated feature set, and the step of extracting resource association information of the integrated features in the integrated feature set to obtain a resource association feature set is repeatedly performed until a feature enhancement termination criterion is met, thereby obtaining an advanced optimized feature set;
[0067] The advanced optimization feature set is used to evaluate and obtain the advanced adaptation coefficient associated with each preset pending resource tag, and to perform resource content interactive recommendation according to the advanced adaptation coefficient associated with each preset pending resource tag.
[0068] In an embodiment of the present invention, exemplarily, after the server completes the feature optimization processing of the resource-associated feature set and obtains the optimized feature set, it will enter a new cycle process, aiming to further improve the quality of the feature set in an iterative manner to obtain an advanced optimized feature set. The server first sets the obtained optimized feature set as the current integrated feature set. This integrated feature set will serve as the starting point for the next cycle. Assume that the server has obtained an optimized feature set containing features such as user behavior, content attributes, and interaction history through a series of optimization processes. Now, this set is given a new role, that is, as the basis for the subsequent feature extraction and optimization. Next, the server will repeat the step of extracting resource association information of the integrated features in the integrated feature set. This means that the server will analyze these features again to find out the deep connection between them and the resources. The server will conduct an in-depth analysis of the user behavior features in the integrated feature set, such as the user's browsing habits, purchase history, etc., to find the connection between these behaviors and specific resources. Such analysis helps the server to understand user needs more accurately and optimize resource recommendations accordingly. The server will continue to repeat the above process until the feature enhancement termination criteria are met. These criteria may include reaching a predetermined number of iterations, the quality improvement of the feature set reaching a certain threshold, or other custom conditions. Suppose the server sets 10 iterations as the termination criteria for feature enhancement. In each iteration, the server extracts resource association information based on the current integrated feature set, and then optimizes this information to obtain a new optimized feature set. This process is repeated 10 times, and the final optimized feature set is the advanced optimized feature set. Finally, the server uses the advanced optimized feature set to evaluate the advanced adaptation coefficient of each preset pending resource tag. These adaptation coefficients reflect the degree of match between resources and user needs, and the server will make interactive recommendations for resource content based on these coefficients. Suppose the server has a movie recommendation system that uses the advanced optimized feature set to evaluate the advanced adaptation coefficient of each movie. These coefficients can be based on multiple factors such as the user's viewing history, movie type preferences, actor and director preferences, etc. The server recommends the movie that best suits the user's taste based on these adaptation coefficients.
[0069] In an embodiment of the present invention, after identifying multiple business scenario optimization information corresponding to the resource-associated feature set from each preset business scenario optimization information, and performing feature optimization processing on the resource-associated feature set according to the multiple business scenario optimization information to obtain the optimized feature set, the following implementation method is also provided.
[0070] Mapping the optimization feature set according to each preset pending resource tag to obtain a mapping feature associated with each preset pending resource tag;
[0071] Performing adaptation coefficient conversion on the mapping processing feature associated with each of the preset pending resource tags to obtain the adaptation coefficient associated with each of the preset pending resource tags;
[0072] Selecting each of the preset pending resource tags according to the adaptation coefficients associated with each of the preset pending resource tags to obtain a target resource tag;
[0073] Interactively recommend the resource content of the target resource tag.
[0074] In an embodiment of the present invention, illustratively, after the server completes the feature optimization processing of the resource-related feature set and obtains the optimized feature set, it will further use these optimized features to make more accurate resource content recommendations for users. This process includes several key steps. First, the server will map the optimized feature set according to each preset pending resource tag. This means that the server will associate the features in the optimized feature set with each preset pending resource tag to determine which features can best represent or reflect these tags. Assume that the server has a product recommendation system for an e-commerce platform, which contains multiple preset pending resource tags, such as "fashion clothing", "home furnishings", etc. The server will map the user's purchase history, browsing behavior and other features in the optimized feature set with these tags to find the most relevant features for each tag. Next, the server will convert the mapping processing features associated with each preset pending resource tag into an adaptation coefficient. This conversion process is to quantify the degree of match between each tag and its features, so as to obtain a specific adaptation coefficient. The server will analyze the degree of association between the "fashion clothing" tag and features such as user purchase history and browsing behavior, and convert it into a specific adaptation coefficient. This coefficient is a value between 0 and 1, indicating the degree of match between the tag and the feature. With the adaptation coefficient, the server will select each preset pending resource tag according to these coefficients to obtain the target resource tag. This usually means selecting the tag with the highest adaptation coefficient or that meets specific conditions as the basis for recommendation. In the scenario of the e-commerce platform, the server will select the "fashion clothing" tag with the highest adaptation coefficient as the target resource tag, because it means that the user's purchase history and browsing behavior are highly related to this tag. Finally, the server will interactively recommend the resource content related to the target resource tag to the user. This is based on the analysis results of the previous steps to provide users with resource recommendations that best suit their interests and needs. Based on the target resource tag "fashion clothing", the server will select related fashion clothing products from the product library and display them to users through the platform's recommendation system. In this way, users can more easily find clothing products that suit their tastes and improve their shopping experience and satisfaction.
[0075] In the embodiments of the present invention, the following implementation modes are also provided.
[0076] Loading the historical resource tag set into an interactive recommendation model, wherein the interactive recommendation model includes an encoding unit, a resource association encoding unit, a feature enhancement unit, and an interactive recommendation unit;
[0077] Acquire a historical resource feature set and a user historical operation feature set corresponding to the historical resource tag set through the encoding unit, and integrate the historical resource feature set with the user historical operation feature set to obtain an integrated feature set;
[0078] Extracting resource association information of the integrated features in the integrated feature set by the resource association coding unit to obtain a resource association feature set;
[0079] Identifying, by the feature enhancement unit, a plurality of business scenario optimization information corresponding to the resource-associated feature set from each preset business scenario optimization information, and performing feature optimization processing on the resource-associated feature set according to the plurality of business scenario optimization information to obtain an optimized feature set;
[0080] The interactive recommendation unit obtains the adaptation coefficient associated with each preset pending resource tag according to the optimization feature set evaluation, and interactively recommends resource content according to the adaptation coefficient associated with each preset pending resource tag.
[0081] In an embodiment of the present invention, exemplarily, in an intelligent recommendation system, the server adopts an advanced interactive recommendation model, which is composed of an encoding unit, a resource association encoding unit, a feature enhancement unit and an interactive recommendation unit, and is used to provide personalized resource recommendations for users. First, the server loads the historical resource tag set into the interactive recommendation model. These historical resource tag sets record the user's past behavior and preferences and are important inputs to the recommendation system. Assuming that the server manages a music recommendation platform, the historical resource tag set may contain information such as the type of music, singers, and tracks that the user has listened to in the past. Next, the encoding unit begins to process these historical resource tag sets. It first obtains the historical resource feature set and the user's historical operation feature set, which represent the attributes of the resource and the user's interactive behavior with the resource, respectively. In the scenario of the music recommendation platform, the historical resource feature set may include the melody, rhythm, style and other attributes of the song, while the user's historical operation feature set may record the number of times the user has played each song, the collection behavior, etc. The encoding unit then integrates the two feature sets to generate a more comprehensive integrated feature set, which takes into account both the attributes of the resource and the user's behavior. The resource association encoding unit further extracts resource association information from the integrated feature set, which reveals the relationship between resources and the user's preference for different resources. In music recommendation, the resource association encoding unit finds that users like to listen to a certain type of music and also tend to listen to other music similar to the type. The feature enhancement unit identifies the optimization information related to the resource association feature set from the preset business scenario optimization information and optimizes these features. If the system knows that the user usually listens to classical music to relax at night, the feature enhancement unit will use this information to optimize the feature set to make the recommendation of classical music in the evening more accurate. Finally, the interactive recommendation unit evaluates the fitness coefficient of each preset pending resource label based on the optimized feature set. These fitness coefficients reflect the degree of match between resources and user needs. In music recommendation, the interactive recommendation unit calculates the fitness coefficient of each song based on the optimized feature set, and then recommends the song with the highest fitness coefficient to the user. Based on these fitness coefficients, the server finally makes interactive recommendations of resource content and provides personalized recommended content for users. The whole process ensures that the recommendation system can fully understand the user's needs and provide resources that best meet the user's interests.
[0082] In the embodiment of the present invention, the training of the interactive recommendation model includes the following steps.
[0083] Obtaining a preset historical resource tag set, and determining a training historical resource tag set and a resource tag target value according to the preset historical resource tag set;
[0084] Loading the training history resource tag set into the original interactive recommendation model to obtain the training adaptation coefficient associated with each preset pending resource tag output;
[0085] Performing cost calculation according to the training adaptation coefficient associated with each preset pending resource tag and the target value of the resource tag to obtain a cost parameter;
[0086] A training process is performed on the original interactive recommendation model according to the cost parameter, and the interactive recommendation model is obtained when a preset training termination condition is met.
[0087] In an embodiment of the present invention, exemplarily, the server prepares to train an interactive recommendation model to improve the accuracy of resource recommendation and user satisfaction. The training process involves multiple key steps. First, the server obtains a set of historical resource tags from historical data. These historical resource tag sets record the user's past behavior and preferences and are the basic data for training the recommendation model. Assuming that the server manages a movie recommendation system, the historical resource tag set may contain information such as the user's past viewing records, ratings, and comments. The server determines the training historical resource tag set and the resource tag target value based on these historical resource tag sets. The training historical resource tag set is a data subset for model training, and the resource tag target value is the goal that needs to be achieved or optimized during the model training process. Next, the server loads the training historical resource tag set into the original interactive recommendation model. This model is an untrained initial version and needs to be trained to optimize its parameters and performance. In the scenario of the movie recommendation system, the original interactive recommendation model is a neural network model based on deep learning, which is used to predict the user's preference for movies. The server processes the training historical resource tag set through the model to obtain the training adaptation coefficients associated with each preset pending resource tag output. These training adaptation coefficients reflect the degree of fit of the model to the current data. Then, the server performs cost calculation based on the training adaptation coefficient and resource tag target value associated with each preset pending resource tag. Cost calculation is the process of measuring the gap between the model prediction result and the actual target value. In the movie recommendation system, cost calculation involves calculating the difference between the user preference predicted by the model and the actual user rating. The larger the difference, the higher the cost, indicating that the prediction accuracy of the model is lower. Through cost calculation, the server obtains a cost parameter, which quantifies the accuracy of the model prediction. Finally, the server performs a training process on the original interactive recommendation model according to the cost parameter. The purpose of training is to minimize the cost parameter by adjusting the parameters of the model, thereby improving the prediction accuracy of the model. In the scenario of the movie recommendation system, the server may use optimization algorithms such as gradient descent to adjust the weights and biases of the neural network to minimize the cost parameter. The training process continues until the preset training termination conditions are met. These conditions may include reaching a predetermined number of training rounds, the cost parameter is reduced below a certain threshold, etc. Once the termination conditions are met, the server obtains a trained interactive recommendation model, which has improved the prediction accuracy and recommendation effect by optimizing parameters.
[0088] In an embodiment of the present invention, the historical resource tag set includes a game material tag set, the game material tag set includes each game material tag corresponding to the current user, and each preset pending resource tag includes each preset pending game material tag. The embodiment of the present invention also provides the following implementation method.
[0089] Acquire an optimization feature set corresponding to the game material tag set, and obtain an adaptation coefficient associated with each of the preset pending game material tags according to the optimization feature set corresponding to the game material tag set;
[0090] Selecting each of the preset pending game material tags according to the adaptation coefficient associated with each of the preset pending game material tags to obtain a target game material tag;
[0091] Interactively recommend the game material content of the target game material tag.
[0092] In an embodiment of the present invention, exemplarily, when processing game-related recommendation tasks, the server pays special attention to game material tag sets, which record the user's preferences and behaviors for game materials in detail. The server uses these game material tags to provide personalized game material recommendations for users through a series of steps. First, the server obtains the optimization feature set corresponding to the game material tag set. These optimization feature sets are extracted by analyzing various information such as user historical behavior and game material attributes, and can more accurately reflect the user's interests and needs. Assume that the server manages a game platform, and user A often plays shooting games on the platform and prefers science fiction themes. The server extracts optimization features such as "shooting games" and "science fiction themes" by analyzing user A's game history records. Next, the server evaluates the adaptation coefficient of each preset pending game material tag based on these optimization feature sets. The adaptation coefficient reflects the degree of match between the game material tag and the user's interests. Based on the adaptation coefficient of each preset pending game material tag, the server selects tags to determine which game material tags are most suitable for recommendation to the user. In the scenario of user A, the server will find that the adaptation coefficients of the two game material tags "Future War" and "Interstellar Adventure" are relatively high. Therefore, the server selects these two tags as target game material tags. Finally, the server recommends relevant game material content to the user based on the selected target game material tags. In this way, users can receive game material recommendations that are more in line with their interests and needs. For user A, the server will recommend game materials related to "Future War" and "Interstellar Adventure", such as science fiction-themed shooting games, game screenshots and videos of future war scenes, etc. Such recommendations not only improve user satisfaction, but also increase user activity and stickiness on the platform. The entire process is user-centric, making full use of the game material tag set and optimized feature set to achieve personalized and accurate game material content recommendations.
[0093] In the embodiment of the present invention, the aforementioned step S205 can be implemented through the following example execution.
[0094] On the basis of obtaining a server detection request for a target blockchain, according to the target blockchain identity tag in the server detection request, a server security information extraction request is transmitted to a third-party security platform, so that the third-party security platform can feedback target blockchain information corresponding to the target blockchain, wherein the target blockchain information includes target server identity tags corresponding to at least two target servers corresponding to the target blockchain and target server security information of each target server; the third-party security platform is used to update the blockchain information corresponding to each of the at least two blockchains;
[0095] According to the target server identity tags of the target servers, obtaining target detection strategies corresponding to at least two servers to be detected corresponding to the target blockchain and target detection result verification strategies corresponding to the at least two servers to be detected;
[0096] According to the target detection strategy of each server to be detected and the target server security information of each server to be detected, the server detection indication corresponding to each server to be detected is transmitted to each server to be detected, so that each server to be detected feeds back the server detection result corresponding to each server to be detected according to the server detection indication;
[0097] According to the target detection result verification strategy of each server to be detected and the server detection result corresponding to each server to be detected, a verification operation is performed on each server to be detected to obtain the server detection result corresponding to each server to be detected;
[0098] According to the server detection results corresponding to each of the servers to be detected, a blockchain stability assessment result corresponding to the target blockchain is generated.
[0099] In an embodiment of the present invention, exemplarily, after receiving a server detection request for a target blockchain, the server will send a server security information extraction request to a third-party security platform according to the target blockchain identity tag in the request. This third-party security platform is responsible for updating and maintaining the blockchain information of multiple blockchains. Assuming that there is a target blockchain named "XXXA", after receiving a stability assessment request for XXXA, the server will send a request to the third-party security platform to extract server security information related to XXXA. The third-party security platform responds to the server's request and feeds back the target blockchain information corresponding to the target blockchain. This information includes the identity tags and security information of at least two target servers. The third-party security platform returns information about XXXA to the server, including the identity tags of XXXA's two main servers ServerA and ServerB and their security configuration, vulnerability repair records and other security information. The server obtains the corresponding target detection strategy and detection result verification strategy from its own policy library according to the identity tags of each target server. The server retrieves specific detection strategies for these two servers based on the identity tags of ServerA and ServerB, such as port scanning for ServerA and vulnerability detection for ServerB. At the same time, the corresponding detection result verification strategy is also obtained to ensure the accuracy of the detection results. The server sends detection instructions to each server to be detected according to the detection strategy and server security information of each server to be detected, and waits to receive the detection results. The server sends a port scan instruction to ServerA and a vulnerability detection instruction to ServerB. ServerA and ServerB perform corresponding detection tasks according to these instructions and return the detection results to the server. The server verifies the detection results according to the verification strategy of each server to be detected and the returned detection results, and finally generates the stability evaluation results of the target blockchain. After receiving the detection results of ServerA and ServerB, the server verifies these results using the verification strategy. For example, for the port scan results of ServerA, the server will perform a second scan to confirm the accuracy of the results. Finally, the server generates an evaluation report on the stability of the XXXA blockchain based on all verified detection results. This evaluation report can include information such as which servers have security risks and which servers need to be optimized, thereby providing an important reference for the maintenance and management of the blockchain.
[0100] In an embodiment of the present invention, according to the target blockchain identity tag in the server detection request, a server security information extraction request is transmitted to a third-party security platform so that the third-party security platform can feedback the target blockchain information corresponding to the target blockchain, which can be implemented through the following example.
[0101] According to the target blockchain identity tag, the server security information extraction request is transmitted to the third-party security platform, so that the third-party security platform obtains the target blockchain information from the dynamic block information according to the target blockchain identity tag, and feeds back the target blockchain information; the dynamic block information is obtained by real-time monitoring based on the real-time block status corresponding to the at least two blockchains.
[0102] In an embodiment of the present invention, exemplarily, in a database interaction management scenario, the detailed process of the server requesting the third-party security platform to extract server security information according to the target blockchain identity tag is as follows: When the server receives a stability assessment request for a specific target blockchain, it will first identify the target blockchain identity tag in the request. This identity tag is a string or code that uniquely identifies a specific blockchain. Assume that there is an identity tag of a target blockchain called "XXXZ". After identifying this tag, the server will prepare to send a server security information extraction request to the third-party security platform. The server sends a request containing the target blockchain identity tag "XXXZ" to the third-party security platform. After receiving this request, the third-party security platform starts to process this request. The third-party security platform has a dynamic block information system that monitors and updates the real-time block status of at least two blockchains in real time. These dynamic block information include key data such as the operating status, security, and server configuration of the blockchain. The third-party security platform searches in its dynamic block information system based on the identity tag "XXXZ". The system monitors and records the status, security configuration, and latest block data of each server of the XXXZ blockchain in real time. The third-party security platform will feed back the relevant information of the XXXZ blockchain retrieved from the dynamic block information to the server that initiated the request. The third-party security platform will send the server list of the XXXZ blockchain, the security status of each server, the latest block update information, etc. to the server. This information will help the server to conduct subsequent stability assessments. Through the above steps, the server can obtain the latest and most comprehensive information of the target blockchain in real time, thereby providing an accurate data basis for its stability assessment.
[0103] In an embodiment of the present invention, according to the target detection result verification strategy of each server to be detected and the server detection result corresponding to each server to be detected, a verification operation is performed on each server to be detected, and after obtaining the server detection result corresponding to each server to be detected, the embodiment of the present invention may also include the following implementation methods.
[0104] Determining at least one faulty server corresponding to the at least one server detection result based on determining that the to-be-detected server corresponding to the at least one server detection result is abnormal;
[0105] According to the target detection strategy corresponding to each faulty server and the target server security information corresponding to each faulty server, the self-repair capability verification indication corresponding to each faulty server is transmitted to each faulty server, so that each faulty server feeds back the indication call status corresponding to each faulty server according to the self-repair capability verification indication;
[0106] According to the indicated call status corresponding to each faulty server and the target detection result verification strategy corresponding to each faulty server, a verification operation is performed on each faulty server to obtain a self-repair result corresponding to each faulty server; the self-repair result corresponding to each faulty server is used to indicate the self-repair status of each faulty server;
[0107] The generating, according to the server detection results corresponding to the servers to be detected, a blockchain stability assessment result corresponding to the target blockchain, comprises:
[0108] According to the server detection results corresponding to each of the servers to be detected and the self-repair results corresponding to each of the faulty servers, a blockchain stability assessment result corresponding to the target blockchain is generated.
[0109] In an embodiment of the present invention, exemplarily, in a database interaction management scenario, the detailed process of the server performing verification operations on each server to be detected and handling a faulty server is as follows: After the server performs verification operations according to the target detection result verification strategy of each server to be detected and the server detection result, it will determine which servers' detection result indications are abnormal. Assume that during the verification process, the server finds that the detection results of the servers to be detected ServerC and ServerD are abnormal, which indicates that these two servers may have faults or safety hazards. Therefore, the server determines ServerC and ServerD as faulty servers. For the determined faulty servers ServerC and ServerD, the server will transmit self-repair capability verification instructions to these two servers according to their respective target detection strategies and target server security information. The server sends an instruction to ServerC to verify its self-repair capability, such as requiring ServerC to try to automatically repair a known security vulnerability. Similarly, the server also sends a similar verification instruction to ServerD. After receiving the self-repair capability verification instruction, the faulty servers ServerC and ServerD will try to perform self-repair and feedback the instruction call status during the repair process to the server. After ServerC attempts to repair the security vulnerability, it will feed back the status information of the repair process (such as the execution status of the repair steps, whether it is successful, etc.) to the server. After receiving this information, the server will verify it according to the target detection result verification strategy of ServerC to determine the self-repair result of ServerC. Similarly, ServerD will also feedback its self-repair status information, which will be verified by the server. Finally, the server will generate the stability assessment results of the target blockchain based on the detection results of all servers to be detected and the self-repair results of the faulty servers. After comprehensively considering the status, security of all servers to be detected and the self-repair capabilities of the faulty servers, the server will obtain a comprehensive stability assessment report on the target blockchain. This report may point out which servers are operating normally, which have security risks but have successfully self-repaired, and which still require manual intervention, thereby providing an important reference for the maintenance and management of the blockchain.
[0110] In an embodiment of the present invention, according to the target detection result verification strategy of each server to be detected and the server detection result corresponding to each server to be detected, a verification operation is performed on each server to be detected. After obtaining the server detection result corresponding to each server to be detected, the embodiment of the present invention provides the following implementation method.
[0111] Loading the server detection results corresponding to the at least two servers to be detected into a pre-trained fault analysis model for fault identification, and obtaining a fault identification result corresponding to the target blockchain;
[0112] The generating, according to the server detection results corresponding to the servers to be detected, a blockchain stability assessment result corresponding to the target blockchain, comprises:
[0113] The blockchain stability assessment result is generated according to the server detection results corresponding to the at least two servers to be detected and the fault identification result corresponding to the target blockchain.
[0114] In an embodiment of the present invention, exemplarily, in a database interaction management scenario, the server uses a pre-trained fault analysis model to identify faults, and the process of generating blockchain stability assessment results based on this can be refined into the following steps, and is explained in conjunction with actual examples: After obtaining the detection results of each server to be detected, the server will load these results into a pre-trained fault analysis model. This model is trained with a large amount of historical fault data and can identify various potential fault modes. Assume that the server has completed the detection of multiple servers to be detected of the target blockchain and collected a series of data on server performance, security configuration, network connection, etc. Next, the server inputs these data into a pre-trained fault analysis model, which can identify possible faults or abnormal conditions based on these data. After receiving the server detection results, the fault analysis model will perform complex calculations and analysis to identify possible hidden faults or abnormal modes. This process involves a variety of advanced technologies such as data mining and pattern recognition. When processing the data provided by the server, the fault analysis model will find that the response time of some servers is abnormally long, or that some security configurations have potential risks. These will be marked as possible fault points or security risks by the model. After being processed by the fault analysis model, the server will obtain a detailed fault identification result. This result will clearly indicate which servers or aspects have faults or anomalies, as well as the possible causes and severity of these faults or anomalies. The fault analysis model will output a report indicating that the long response time of ServerE is caused by hardware aging or network congestion, and the security configuration problem of ServerF is caused by the failure to update the security patch in time. These fault identification results provide an important basis for subsequent stability evaluation. Finally, the server will comprehensively generate a stability evaluation result of the target blockchain based on the detection results of each server to be detected and the fault identification results provided by the fault analysis model. This result will comprehensively reflect the stability and potential risks of the target blockchain in all aspects. After comprehensively considering the performance data, security configuration and identification results of the fault analysis model of all servers to be detected, the server will generate a comprehensive blockchain stability evaluation report. This report will list in detail the status of each server, the existing fault points and security risks, as well as possible solutions and suggestions for these problems. In this way, managers can optimize and improve the target blockchain in a targeted manner based on this report.
[0115] In an embodiment of the present invention, the server detection request includes a blockchain verification token; based on obtaining the server detection request for the target blockchain, the server security information extraction request is transmitted to the third-party security platform according to the target blockchain identity tag in the server detection request, which can be implemented through the following example.
[0116] Based on obtaining a server detection request for the target blockchain, performing a token verification operation on the blockchain verification token to obtain a token verification result;
[0117] On the basis that the token verification result indicates that the blockchain verification token verification is successful, the server security information extraction request is transmitted to the third-party security platform according to the target blockchain identity tag.
[0118] In an embodiment of the present invention, exemplarily, in a database interaction management scenario, after receiving a server detection request containing a blockchain verification token, the server will perform a series of verifications and operations. The following is an example of a detailed scenario of this process: The server first receives a server detection request for a target blockchain, which contains an important element-the blockchain verification token. Assume that a blockchain system named "XXXSecure" sends a detection request to the server, and the request contains a specific blockchain verification token, such as "TokenABC123". After receiving the request, the server will first perform a token verification operation on the blockchain verification token "TokenABC123". This step is to ensure the legitimacy and security of the request. The server verifies "TokenABC123" through a built-in token verification mechanism or by connecting to a dedicated token verification service. In this process, the server checks the format, validity period, and whether the token matches a known legal token. After the verification operation, the server will obtain a token verification result, which will clearly indicate whether the token has been successfully verified. After verification, "TokenABC123" is confirmed as a valid blockchain verification token, and the server therefore obtains a token verification success result. After confirming that the blockchain verification token has been successfully verified, the server will pass a server security information extraction request to the third-party security platform based on the target blockchain identity tag in the request. Since "TokenABC123" has been successfully verified, the server recognizes that the target blockchain identity tag in the request is "XXXSecure". Next, the server will construct a server security information extraction request for "XXXSecure" and send it to the third-party security platform to obtain relevant security information of the blockchain system. Through the above steps, the server can ensure that only requests carrying valid blockchain verification tokens will be processed, thereby greatly improving the security of the system and the accuracy of the data.
[0119] The embodiment of the present invention provides a computer device 100, which includes a processor and a non-volatile memory storing computer instructions. When the computer instructions are executed by the processor, the computer device 100 executes the aforementioned database interaction management method and system. Figure 2 As shown, Figure 2 The block diagram of the computer device 100 provided in the embodiment of the present invention. The computer device 100 includes a memory 111, a processor 112 and a communication unit 113. To achieve data transmission or interaction, the memory 111, the processor 112 and the communication unit 113 are directly or indirectly electrically connected to each other. For example, these components can be electrically connected to each other through one or more communication buses or signal lines.
[0120] For illustrative purposes, the foregoing description is made with reference to specific embodiments. However, the above illustrative discussion is not intended to be exhaustive or to limit the present disclosure to the precise form disclosed. Numerous modifications and variations are possible in accordance with the above teachings. These embodiments are selected and described in order to best illustrate the principles of the present disclosure and its practical application, so that those skilled in the art can best utilize the present disclosure and utilize various embodiments with different modifications to suit the intended specific application.
Claims
1. A database interaction management method, It is characterized in that include: Obtain the current user's interaction request for the preset database; According to the interaction request, obtaining a historical resource tag set of the current user for the preset database; Performing resource content interactive recommendation for the interactive request according to the historical resource tag set; Determine, according to the recommendation result of the interactive recommendation of the resource content, the target blockchain where the recommended resource corresponding to the recommendation result is located; Performing a stability assessment on the target blockchain to obtain a blockchain stability assessment result corresponding to the target blockchain; On the basis that the blockchain stability assessment result indicates that the blockchain is stable, obtaining the recommended resources from the target blockchain; The performing, according to the historical resource tag set, interactive recommendation of resource content for the interactive request includes: Acquire the historical resource tag set, wherein the historical resource tag set includes each historical resource tag corresponding to the current user; According to each historical resource tag in the historical resource tag set, searching each pre-acquired historical resource feature for the historical resource feature associated with each historical resource tag, to obtain the historical resource feature set; Acquire multiple user operation records of each historical resource tag from multiple business scenarios to which each historical resource tag belongs; For each historical resource tag, extract feature vectors of multiple user operation records to obtain multiple user operation vectors, and integrate the multiple user operation vectors to obtain a user historical operation feature set, wherein the historical resource feature set is obtained by performing a feature extraction operation on the resource content of each historical resource tag, and the user historical operation feature set is obtained by performing a feature extraction operation on the user operation record of each historical resource tag, and the user operation record is extracted from the business scenario to which the historical resource tag belongs; Integrate the historical resource features in the historical resource feature set with the corresponding user historical operation features in the user historical operation feature set to obtain an integrated feature set, and extract resource association information of the integrated features in the integrated feature set to obtain a resource association feature set; Calculate the feature matching degree according to the resource-related feature set to obtain the matching coefficient corresponding to each preset business scenario optimization information; Selecting each preset business scenario optimization information according to the matching coefficient corresponding to each preset business scenario optimization information to obtain a plurality of business scenario optimization information corresponding to the resource association feature set; According to the plurality of business scenario optimization information, feature optimization processing is performed on the resource-associated feature sets respectively to obtain a plurality of current optimized feature sets; Acquire the number of optimization information items of the multiple business scenario optimization information, and determine each matching coefficient of a high priority and each matching coefficient of a low priority from the matching coefficients respectively associated with each preset business scenario optimization information according to the number of optimization information items; Maintaining each matching coefficient of the high priority as it is, and adjusting each matching coefficient of the low priority to a value corresponding to a preset data characterization vector, to obtain an adjustment coefficient associated with each preset business scenario optimization information; Weight-normalize the adjustment coefficients respectively associated with each preset business scenario optimization information to obtain the target matching coefficients respectively associated with each preset business scenario optimization information; Determining the target matching coefficients associated with each of the plurality of business scenario optimization information from the target matching coefficients associated with each of the preset business scenario optimization information; For each business scenario optimization information, the corresponding current optimization feature set is modified according to the target matching coefficient to obtain each modified feature set; Integrate each of the modified feature sets to obtain the optimized feature set; The optimized feature set is used to evaluate and obtain the adaptation coefficient associated with each preset pending resource tag, and to perform resource content interactive recommendation according to the adaptation coefficient associated with each preset pending resource tag.
2. The method according to claim 1, It is characterized in that The step of extracting resource association information of the integrated features in the integrated feature set to obtain the resource association feature set includes: Projecting the integrated feature set into a feature space to obtain a retrieval vector set, an index vector set and a data representation vector set; The correlation coefficient between the retrieval vector set and the index vector set is calculated to obtain a correlation coefficient set, and a weighted adjustment is applied to the data representation vector set according to the correlation coefficient set to obtain the resource correlation feature set.
3. The method according to claim 1, It is characterized in that After performing feature optimization processing on the resource-associated feature set according to the plurality of business scenario optimization information to obtain an optimized feature set, the method further includes: The optimized feature set is used as the integrated feature set, and the step of extracting resource association information of the integrated features in the integrated feature set to obtain a resource association feature set is repeatedly performed until a feature enhancement termination criterion is met, thereby obtaining an advanced optimized feature set; The advanced optimization feature set is used to evaluate and obtain the advanced adaptation coefficient associated with each preset pending resource tag, and to perform resource content interactive recommendation according to the advanced adaptation coefficient associated with each preset pending resource tag.
4. The method according to claim 1, It is characterized in that After integrating each of the modified feature sets to obtain the optimized feature set, the method further includes: Mapping the optimization feature set according to each preset pending resource tag to obtain a mapping feature associated with each preset pending resource tag; Performing adaptation coefficient conversion on the mapping processing feature associated with each of the preset pending resource tags to obtain the adaptation coefficient associated with each of the preset pending resource tags; Selecting each of the preset pending resource tags according to the adaptation coefficients associated with each of the preset pending resource tags to obtain a target resource tag; Interactively recommend the resource content of the target resource tag.
5. The method according to claim 1, It is characterized in that The method further comprises: Loading the historical resource tag set into an interactive recommendation model, wherein the interactive recommendation model includes an encoding unit, a resource association encoding unit, a feature enhancement unit, and an interactive recommendation unit; Acquire a historical resource feature set and a user historical operation feature set corresponding to the historical resource tag set through the encoding unit, and integrate the historical resource feature set with the user historical operation feature set to obtain an integrated feature set; Extracting resource association information of the integrated features in the integrated feature set by the resource association coding unit to obtain a resource association feature set; Identifying, by the feature enhancement unit, a plurality of business scenario optimization information corresponding to the resource-associated feature set from each preset business scenario optimization information, and performing feature optimization processing on the resource-associated feature set according to the plurality of business scenario optimization information to obtain an optimized feature set; The interactive recommendation unit obtains the adaptation coefficient associated with each preset pending resource tag according to the optimization feature set evaluation, and interactively recommends resource content according to the adaptation coefficient associated with each preset pending resource tag; The training of the interactive recommendation model includes the following steps: Obtaining a preset historical resource tag set, and determining a training historical resource tag set and a resource tag target value according to the preset historical resource tag set; Loading the training history resource tag set into the original interactive recommendation model to obtain the training adaptation coefficient associated with each preset pending resource tag output; Performing cost calculation according to the training adaptation coefficient associated with each preset pending resource tag and the target value of the resource tag to obtain a cost parameter; A training process is performed on the original interactive recommendation model according to the cost parameter, and the interactive recommendation model is obtained when a preset training termination condition is met.
6. The method according to claim 1, It is characterized in that The historical resource tag set includes a game material tag set, the game material tag set includes each game material tag corresponding to the current user, each preset pending resource tag includes each preset pending game material tag, and the method further includes: Acquire an optimization feature set corresponding to the game material tag set, and obtain an adaptation coefficient associated with each of the preset pending game material tags according to the optimization feature set corresponding to the game material tag set; Selecting each of the preset pending game material tags according to the adaptation coefficients associated with each of the preset pending game material tags to obtain a target game material tag; Interactively recommend the game material content of the target game material tag.
7. The method according to claim 1, It is characterized in that The performing stability assessment on the target blockchain to obtain a blockchain stability assessment result corresponding to the target blockchain includes: On the basis of obtaining a server detection request for a target blockchain, a server security information extraction request is transmitted to a third-party security platform according to the identity tag of the target blockchain, so that the third-party security platform obtains the target blockchain information from the dynamic block information according to the identity tag of the target blockchain, and feeds back the target blockchain information; the dynamic block information is obtained by real-time monitoring based on the real-time block status corresponding to at least two blockchains, and the target blockchain information includes the target server identity tags corresponding to at least two target servers corresponding to the target blockchain and the target server security information of each target server; the third-party security platform is used to update the blockchain information corresponding to each of the at least two blockchains; According to the target server identity tags of the target servers, obtaining target detection strategies corresponding to at least two servers to be detected corresponding to the target blockchain and target detection result verification strategies corresponding to the at least two servers to be detected; According to the target detection strategy of each server to be detected and the target server security information of each server to be detected, the server detection indication corresponding to each server to be detected is transmitted to each server to be detected, so that each server to be detected feeds back the server detection result corresponding to each server to be detected according to the server detection indication; According to the target detection result verification strategy of each server to be detected and the server detection result corresponding to each server to be detected, a verification operation is performed on each server to be detected to obtain the server detection result corresponding to each server to be detected; According to the server detection results corresponding to each of the servers to be detected, a blockchain stability assessment result corresponding to the target blockchain is generated.
8. The method according to claim 7, It is characterized in that After performing verification operations on each server to be detected according to the target detection result verification strategy of each server to be detected and the server detection result corresponding to each server to be detected and obtaining the server detection result corresponding to each server to be detected, the method further includes: Determining at least one faulty server corresponding to the at least one server detection result based on determining that the to-be-detected server corresponding to the at least one server detection result is abnormal; According to the target detection strategy corresponding to each faulty server and the target server security information corresponding to each faulty server, the self-repair capability verification indication corresponding to each faulty server is transmitted to each faulty server, so that each faulty server feeds back the indication call status corresponding to each faulty server according to the self-repair capability verification indication; According to the indicated call status corresponding to each faulty server and the target detection result verification strategy corresponding to each faulty server, a verification operation is performed on each faulty server to obtain a self-repair result corresponding to each faulty server; the self-repair result corresponding to each faulty server is used to indicate the self-repair status of each faulty server; Generate a blockchain stability assessment result corresponding to the target blockchain according to the server detection results corresponding to each of the servers to be detected and the self-repair results corresponding to each of the faulty servers; Loading the server detection results corresponding to the at least two servers to be detected into a pre-trained fault analysis model for fault identification, and obtaining a fault identification result corresponding to the target blockchain; The blockchain stability assessment result is generated according to the server detection results corresponding to the at least two servers to be detected and the fault identification result corresponding to the target blockchain.
9. A server system, It is characterized in that The method comprises a server, wherein the server is used to execute the method described in any one of claims 1 to 8.
Citation Information
Patent Citations
Payment mode recommendation method based on big data and block chain and cloud computing server
CN112613871A
Green certificate transaction system based on block chain
CN112712420A
Resource control method and device based on artificial intelligence, electronic equipment and medium
CN114119203A
Scientific and technological service recommendation method and device based on machine learning, equipment and medium
CN115374354A
Commodity recommendation method and system based on big data and artificial intelligence
CN118014684A
