Recommendation method based on enterprise service bus

Abstract

The invention discloses a recommendation method based on an enterprise service bus, and relates to the technical field of data processing. The method comprises the following steps: collecting business data of each business subsystem in an enterprise system; performing cleaning processing, association integration and partition storage on the data to form an enterprise data sharing center; the sharing center establishes a data exchange and sharing mechanism; a service requester requests an instruction from the sharing center through the service bus; the sharing center performs recommendation algorithm analysis on the received instruction and feeds back an analysis result to the service requester; and the recommended items are introduced into the knowledge graph model for recommendation. The business data of each business subsystem in an enterprise system is collected, an enterprise data sharing center is established, a service requester requests an instruction from the sharing center through a service bus, and the sharing center performs recommendation algorithm analysis on the received instruction and feeds back an analysis result to the service requester. And the information sharing efficiency and the data utilization rate of departments among enterprises are improved.

Description

Technical Field

[0001] The invention belongs to the technical field of data processing, and in particular relates to a recommendation method based on an enterprise service bus. Background Art

[0002] Enterprise Service Bus (ESB) is the product of the combination of traditional middleware technology with XML, Web services and other technologies. Enterprise Service Bus is generally used in enterprises with many internal business systems, such as energy, government, hospitals, electricity and other industries. It has the characteristics of low coupling, real-time updates, and providing Web Service services to end users.

[0003] With the development of information technology and the increasing informatization of society, enterprises are seeing a proliferation of diverse application systems, increasing the amount of information and data between these systems, and the interplay between them. This leads to information isolation and redundancy, making information communication and sharing increasingly important. In this process, Web services have gradually become a key technology in information technology. Service providers use Web services to encapsulate existing or under-development services and publish them as services. These services can also be used to find and consume other services, exchanging data and information both within and outside the enterprise, thus flexibly connecting diverse systems. Summary of the Invention

[0004] The purpose of the present invention is to provide a recommendation method based on the enterprise service bus, which builds an enterprise data sharing center by collecting business data and matches the instructions of the service requester according to the recommendation algorithm, thereby solving the problems of difficulty in information sharing between departments of the existing enterprise and low data utilization.

[0005] To solve the above technical problems, the present invention is achieved through the following technical solutions:

[0006] The present invention is a recommendation method based on an enterprise service bus, comprising the following steps:

[0007] Step S1: Collect business data of each business subsystem in the enterprise system;

[0008] Step S2: Clean, process, associate, integrate, and partition the data to form an enterprise data sharing center;

[0009] Step S3: The sharing center establishes a data exchange and sharing mechanism;

[0010] Step S4: The service requester requests instructions from the sharing center through the service bus;

[0011] Step S5: The sharing center performs a recommendation algorithm analysis on the received instruction;

[0012] Step S6: The recommendation algorithm feeds back the analysis results to the service requester;

[0013] Step S7: The recommended items are introduced into the knowledge graph model for recommendation.

[0014] As a preferred technical solution, in step S1, the enterprise system architecture includes a business service module, a business interface module and a basic management module; the business service module is a module for the enterprise to customize the business logic according to business needs; the business interface module is the entrance to the business service module, which is used for data verification and message queue; the basic management module provides a basic framework for the business service module, which is used for logging, manpower scheduling and interface management; the content of each business system request is processed by the message interface module. Generally, the request is a message based on the Http protocol. After receiving the message, the message content is verified, and the data format and content are verified for legitimacy. After successful verification, it enters the message queue and waits for the business service module to call.

[0015] After extracting the corresponding information from the message queue, the business service module calls the corresponding pre-developed and configured custom business service based on the information content, passes the relevant information to the corresponding custom business service, and executes the specific logic of the custom business service. The custom business service mainly performs operations such as data content verification, data conversion, and business processing. After completion, the operation is passed to the basic management module, which records and distributes the relevant information.

[0016] The message interface module is the data entry point and one of the triggering channels for data transfer calls from external systems. It is primarily responsible for processing message requests from external business systems, primarily transmitting content in a RESTful manner, primarily in JSON or XML via HTTP. Technically, it is written using ASP.NET Web API. By comparing the RequestHeaders content with the relevant information stored in the database of the basic module, the detailed information and key values of the relevant interface are obtained. The message interface module then records, verifies, and parses the data. Based on the configured data information, it converts the data into the corresponding model and places it into the message queue.

[0017] The business service module is the core of the entire framework. It implements the specific operations of the entire business logic and serves as the carrier for all customized business services. Customized services interact with data by calling the APIs provided by the business service module. This module provides loose coupling and an effective means of achieving data connectivity without modifying existing systems through non-invasive data exchange. By orchestrating different hardware carriers for customized services, deploying multiple customized business services across multiple servers, and scheduling their execution through the business service module, a distributed control system with high scalability, high concurrency, high throughput, and failover can be achieved.

[0018] As a preferred technical solution, in step S2, during data cleaning, the source data is sequentially subjected to format verification, gap filling, duplicate removal, and code conversion cleaning operations, and the presentation of data fields in different business systems is unified; during the association integration, the data is sequentially subjected to field mapping, field splitting, field merging, and pre-calculation; during the partition storage, the different types of data areas after integration are distinguished, and corresponding security and access policies are set according to the different data characteristics of different data partitions.

[0019] As a preferred technical solution, the partitioned storage is divided into a five-layer architecture, namely the front-end area, data buffer area, data core area, extended subject area and data sharing area; the front-end area is used to define the data area of the external application business system, storing the database of each business system of the enterprise, which can be the query view of each business system; the data buffer area is used to store the data collected by the external business system and perform data analysis, which is first implemented in the data buffer area. The buffer area is designed to reduce the pressure brought by data collection to the server and data processing. The tables collected by the buffer area are tables that contain data value information of each business system and can perform data analysis, excluding the data tables and log tables required for the construction of the business system itself; the data core area regularly calls pre-set functions, and the data processed again by the buffer area includes the authority of the data, business relationships, and abnormal data, where abnormalities include business abnormalities and data abnormalities; the extended subject area generates data sets based on the data processed by the data buffer area and the data core area, and stores each data set according to different database segments, which are independent of each other; the data sharing area is used for use by other departments or other projects. If a business department needs data, the business department shall propose data sharing requirements. The storage method is the same as that of the expansion area, which is physically unified storage. The data requirements of each business system are independent of each other.

[0020] As a preferred technical solution, in step S3, the information sent by the sharing center needs to be encrypted, the message is hashed and verified using a digital signature method, and the encryption and digital signature are combined as a key. The specific encryption process is as follows:

[0021] Step J1: Randomly set the initial value x0 and the chaos coefficient parameter α;

[0022] Step J2: Use the formula x n+1 =μ×x n ×(1-x n ) for 1000 iterations;

[0023] Step J3: Determine whether encryption or decryption is completed;

[0024] If completed, terminate;

[0025] If not completed, proceed to step J4;

[0026] Step J4: Read plaintext or ciphertext;

[0027] Step J5: Use the formula x n+1 =μ×x n ×(1-x n ) for 5 iterations;

[0028] Step J6: Take x n 4, 5, or 6 digits after the decimal point, and take the remainder of 256 to get the key K;

[0029] Step J7: Get the XOR result and output it.

[0030] As a preferred technical solution, in step S5, the recommendation algorithm analysis process is as follows:

[0031] Step S51: Calculate the similarity between the instruction and the data using the cosine similarity formula. The specific calculation formula is as follows:

[0032]

[0033] In the formula, m represents the similarity between the user instruction and the request data, N represents the set of common ratings between the user instruction and the request data, r represents the set of items for the user instruction, y represents the rating of the request data for item t, h represents the average of all rated items of the user instruction, and g represents the average of all rated items of the request data. Based on this result, a similarity table is established, which facilitates the subsequent direct selection of similarities from the similarity table for the next step of calculation;

[0034] Step S52: Calculate the similarity between the user instruction μ and the request data i. The calculation formula is as follows:

[0035]

[0036] Where S(μ,K) represents the top K request data most similar to user instruction μ, N(i) represents the user instruction set of request data i, m μv represents the similarity between user instruction μ and user instruction v, R vi Indicates the interest of user instruction v in requesting data i;

[0037] Step S53: For the first K request data that are most similar to the user instruction μ, calculate the set l of the user instruction μ and these K request data respectively. n The similarity between them is obtained, and the similarity list is recommended to the service requester.

[0038] As a preferred technical solution, in step S7, the process of introducing the recommended items into the knowledge graph model is as follows:

[0039] Step S71: extract knowledge from the original project data set, express it in triple form, standardize the triples and import them into the database for storage, and construct a project knowledge graph;

[0040] Step S72: Randomly sample the project data, represent each entity and relationship with a multi-dimensional feature vector, and train the TransH model;

[0041] Step S73: Sampling from the project dataset, randomly replacing the head entity or tail entity to generate a test dataset, and importing it into the TransH model;

[0042] Step S74: Obtain the embedded feature vector of the entity in the project knowledge graph through the TransH model, expressed as In the formula, z represents any item and h represents the number of dimensions;

[0043] Step S75: Calculate the similarity between the two items. The formula is: Where, represents the Euclidean distance between item i and item j;

[0044] Step S76: storing multiple items with the highest similarity to item i in the database.

[0045] The present invention has the following beneficial effects:

[0046] The present invention establishes an enterprise data sharing center by collecting business data from various business subsystems in the enterprise system. Service requesters request instructions from the sharing center through the service bus. The sharing center performs recommendation algorithm analysis on the received instructions and feeds back the analysis results to the service requester, thereby improving the information sharing efficiency and data utilization rate among departments of the enterprise.

[0047] Of course, any product implementing the present invention does not necessarily need to achieve all of the advantages described above at the same time. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0049] Figure 1 A flow chart of a recommended method based on an enterprise service bus according to the present invention;

[0050] Figure 2 This is a diagram of the enterprise system architecture. DETAILED DESCRIPTION

[0051] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making any creative efforts shall fall within the scope of protection of the present invention.

[0052] In addition, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0053] To make the purpose, technical solutions and advantages of this application clearer, Figure 1 The implementation methods of this application are described in further detail.

[0054] Before introducing the embodiments of the present application, the enterprise service bus is first described.

[0055] The Enterprise Service Bus (ESB) evolved from the Service-Oriented Architecture (SOA) and is the product of combining traditional middleware technology with technologies such as XML and Web services. The emergence of the ESB has transformed traditional software architectures, providing solutions that are more affordable than traditional middleware products. It can also eliminate technical differences between different applications, allowing different application servers to operate in coordination and enabling communication and integration between different services. Functionally, the ESB provides event-driven and document-oriented processing models, as well as a distributed operation and management mechanism. It supports content-based routing and filtering, has the ability to transmit complex data, and provides a series of standard interfaces.

[0056] The company's current business information systems are complex and rapidly evolving, encompassing design, procurement, manufacturing, construction, and other related businesses. These application systems, built over more than a decade, incorporate diverse technologies and a complex technical history. As the foundation of the company's entire system, stability is paramount. With the company's continued expansion, further integration of all system data is inevitable. Failure to quickly and effectively integrate this data will hinder effective data transfer across many businesses, creating a disconnect between new businesses and systems and existing platforms, impacting the company's general contracting operations. Rapid information system integration will be a key focus for the company's next steps in its informatization efforts. The company has a large number of existing information systems, covering complex business needs. Transforming them is difficult and costly. New businesses and upgrades to existing businesses typically involve replacing them with new business systems, gradually phasing out older systems. Therefore, the new integration platform must be compatible with existing information systems without changing existing system interfaces, while also supporting data integration for new systems.

[0057] In order to make the purpose, technical solutions and advantages of this application more clear, the following Figure 1-2 It should be understood that the specific embodiments described herein are only used to explain the present application and are not intended to limit the present application.

[0058] See also Figure 1 As shown, the present invention is a recommendation method based on an enterprise service bus, comprising the following steps:

[0059] Step S1: Collect business data of each business subsystem in the enterprise system;

[0060] Step S2: Clean, process, associate, integrate, and partition the data to form an enterprise data sharing center;

[0061] Step S3: The sharing center establishes a data exchange and sharing mechanism;

[0062] Step S4: The service requester requests instructions from the sharing center through the service bus;

[0063] Step S5: The sharing center performs a recommendation algorithm analysis on the received instruction;

[0064] Step S6: The recommendation algorithm feeds back the analysis results to the service requester;

[0065] Step S7: The recommended items are introduced into the knowledge graph model for recommendation.

[0066] See also Figure 2 As shown, in step S1, the enterprise system architecture includes a business service module, a business interface module and a basic management module; the business service module is a module for enterprises to customize business logic according to business needs; the business interface module is the entrance to the business service module, which is used for data verification and message queue; the basic management module provides a basic framework for the business service module, which is used for log recording, manpower scheduling and interface management; the content of each business system request is processed by the message interface module. Generally, the request is a message based on the Http protocol. After receiving the message, the message content is verified, and the data format and content are verified for legitimacy. After successful verification, it enters the message queue and waits for the business service module to call.

[0067] After extracting the corresponding information from the message queue, the business service module calls the corresponding pre-developed and configured custom business service based on the information content, passes the relevant information to the corresponding custom business service, and executes the specific logic of the custom business service. The custom business service mainly performs operations such as data content verification, data conversion, and business processing. After completion, it passes the information to the basic management module, which records and distributes the relevant information.

[0068] The message interface module is the data entry point and one of the triggering channels for data transfer calls from external systems. It is primarily responsible for processing message requests from external business systems. The primary content transmitted is in a RESTful format, primarily in JSON or XML format via HTTP. Technically, it is written using ASP.NET Web API. By comparing the RequestHeaders content with the relevant information stored in the database of the basic module, detailed information and key values of the relevant interface can be obtained. The message interface module then records, verifies, and parses the data. Based on the configured data information, it converts the data into the corresponding model and places it into the message queue.

[0069] The business service module is the core of the entire framework. It implements the specific operations of the entire business logic and serves as the carrier for all custom business services. Custom businesses interact with data by calling the APIs provided by the business service module. This module provides loose coupling and non-invasive data exchange, offering an effective means of achieving data connectivity without modifying existing systems. By orchestrating different hardware carriers for custom businesses, deploying multiple custom business services across multiple servers, and scheduling their execution through the business service module, a distributed control system with high scalability, high concurrency, high throughput, and failover can be achieved.

[0070] In step S2, during data cleaning, the source data is sequentially subjected to format verification, gap filling, duplicate removal, and code conversion cleaning operations, and the presentation of data fields in different business systems is unified; during association and integration, the data is sequentially subjected to field mapping, field splitting, field merging, and pre-calculation; during partitioned storage, the different types of data areas after integration are distinguished, and corresponding security and access policies are set according to the different data characteristics of different data partitions.

[0071] Partitioned storage is organized into a five-tiered architecture: the frontend, data buffer, data core, extended subject, and data sharing. The frontend defines the data area for external application business systems, storing the databases of each enterprise business system and serving as query views for each business system. The data buffer stores data collected from external business systems and performs data analysis, initially in the data buffer. The buffer is designed to reduce the pressure placed on servers and data processing by data collection. The tables collected in the buffer contain valuable data information from each business system and enable data analysis, excluding data tables and log tables required for the business system itself. The data core periodically processes data after it passes through the buffer by calling pre-defined functions. This processing includes data authority, business relationships, and anomalies, including both business and data anomalies. The extended subject generates data sets based on the data processed by the data buffer and data core, storing each data set in a separate, independent database segment. The data sharing area is used by other departments or projects. Business departments can request data sharing if they require it. Storage is physically unified, similar to the expansion area, and the data requirements of each business system are independent.

[0072] In step S3, the information sent by the sharing center needs to be encrypted, and the message is hashed and verified using a digital signature. The encryption and digital signature are then combined as a key. The specific encryption process is as follows:

[0073] Step J1: Randomly set the initial value x0 and the chaos coefficient parameter α;

[0074] Step J2: Use the formula x n+1 =μ×x n ×(1-x n ) for 1000 iterations;

[0075] Step J3: Determine whether encryption or decryption is completed;

[0076] If completed, terminate;

[0077] If not completed, proceed to step J4;

[0078] Step J4: Read plaintext or ciphertext;

[0079] Step J5: Use the formula x n+1 =μ×x n ×(1-x n ) for 5 iterations;

[0080] Step J6: Take x n 4, 5, or 6 digits after the decimal point, and take the remainder of 256 to get the key K;

[0081] Step J7: Get the XOR result and output it.

[0082] In step S5, the recommendation algorithm analysis process is as follows:

[0083] Step S51: Calculate the similarity between the instruction and the data using the cosine similarity formula. The specific calculation formula is as follows:

[0084]

[0085] In the formula, m represents the similarity between the user instruction and the request data, N represents the set of common ratings between the user instruction and the request data, r represents the set of items for the user instruction, y represents the rating of the request data for item t, h represents the average of all rated items of the user instruction, and g represents the average of all rated items of the request data. Based on this result, a similarity table is established, which facilitates the subsequent direct selection of similarities from the similarity table for the next step of calculation;

[0086] Step S52: Calculate the similarity between the user instruction μ and the request data i. The calculation formula is as follows:

[0087]

[0088] Where S(μ,K) represents the top K request data most similar to user instruction μ, N(i) represents the user instruction set of request data i, m μv represents the similarity between user instruction μ and user instruction v, R viIndicates the interest of user instruction v in requesting data i;

[0089] Step S53: For the first K request data that are most similar to the user instruction μ, calculate the set l of the user instruction μ and these K request data respectively. n The similarity between them is obtained, and the similarity list is recommended to the service requester.

[0090] In step S7, the process of introducing the recommended items into the knowledge graph model is as follows:

[0091] Step S71: extract knowledge from the original project dataset, express it in triple form, standardize the triples and import them into the database for storage, and construct a project knowledge graph;

[0092] Step S72: Randomly sample the project data, represent each entity and relationship with a multi-dimensional feature vector, and train the TransH model;

[0093] Step S73: Sampling from the project dataset, randomly replacing the head entity or tail entity to generate a test dataset, and importing it into the TransH model;

[0094] Step S74: Obtain the embedded feature vector of the entity in the project knowledge graph through the TransH model, expressed as In the formula, z represents any item and h represents the number of dimensions;

[0095] Step S75: Calculate the similarity between the two items. The formula is: Where, represents the Euclidean distance between item i and item j;

[0096] Step S76: storing multiple items with the highest similarity to item i in the database.

[0097] It is worth noting that in the above system embodiment, the various units included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of the present invention.

[0098] In addition, those skilled in the art will appreciate that all or part of the steps in the above-mentioned embodiments can be accomplished by instructing related hardware through a program, and the corresponding program can be stored in a computer-readable storage medium.

[0099] The preferred embodiments of the present invention disclosed above are intended only to help illustrate the present invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the present invention to the specific embodiments described. Obviously, many modifications and variations are possible based on the content of this specification. These embodiments are selected and described in detail in this specification to better explain the principles and practical applications of the present invention, thereby enabling those skilled in the art to better understand and utilize the present invention. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. A recommendation method based on enterprise service bus, characterized in that: The steps include: Step S1: Collect business data of each business subsystem in the enterprise system; Step S2: Clean, process, associate, integrate, and partition the data to form an enterprise data sharing center; Step S3: The sharing center establishes a data exchange and sharing mechanism; Step S4: The service requester requests instructions from the sharing center through the service bus; Step S5: The sharing center performs a recommendation algorithm analysis on the received instruction; Step S6: The recommendation algorithm feeds back the analysis results to the service requester; Step S7: The recommended items are introduced into the knowledge graph model for recommendation.

2. The recommendation method based on enterprise service bus according to claim 1, characterized in that: In step S1, the enterprise system architecture includes a business service module, a business interface module, and a basic management module; the business service module is a module for the enterprise to customize business logic according to business needs; the business interface module is the entrance to the business service module and is used for data verification and message queuing; The basic management module provides a basic framework for the business service module, which is used for log recording, manpower scheduling and interface management.

3. The recommendation method based on enterprise service bus according to claim 1, characterized in that: In step S2, during data cleaning, the source data is sequentially subjected to format verification, gap filling, duplicate removal, and code conversion cleaning operations, and the presentation of data fields in different business systems is unified; during the association integration, the data is sequentially subjected to field mapping, field splitting, field merging, and pre-calculation; during the partition storage, the different types of data areas after integration are distinguished, and corresponding security and access policies are set according to the different data characteristics of different data partitions.

4. The recommendation method based on enterprise service bus according to claim 3, characterized in that: The partitioned storage is divided into a five-layer architecture, namely the front-end area, data buffer area, data core area, extended subject area and data sharing area; the front-end area is used to define the data area of the external application business system and store the database of each business system of the enterprise; the data buffer area is used to store data collected by the external business system and perform data analysis; the data core area processes the data again through the buffer area by regularly calling a pre-set function; the extended subject area generates a data set based on the data processed by the data buffer area and the data core area, and stores each data set according to a different database segment; the data sharing area is used for use by other departments or other projects.

5. The recommendation method based on enterprise service bus according to claim 1, characterized in that: In step S3, the information sent by the sharing center needs to be encrypted, and the message is hashed and verified using a digital signature method. The encryption and digital signature are then combined as a key. The specific encryption process is as follows: Step J1: Randomly set the initial value x0 and the chaos coefficient parameter α; Step J2: Use the formula x n+1 =μ×x n ×(1-x n ) for 1000 iterations; Step J3: Determine whether encryption or decryption is completed; If completed, terminate; If not completed, proceed to step J4; Step J4: Read plaintext or ciphertext; Step J5: Use the formula x n+1 =μ×x n ×(1-x n ) for 5 iterations; Step J6: Take x n 4, 5, or 6 digits after the decimal point, and take the remainder of 256 to get the key K; Step J7: Get the XOR result and output it.

6. The recommendation method based on enterprise service bus according to claim 1, characterized in that: In step S5, the recommendation algorithm analysis process is as follows: Step S51: Calculate the similarity between the instruction and the data using the cosine similarity formula. The specific calculation formula is as follows: Where m represents the similarity between the user instruction and the request data, N represents the set of common ratings between the user instruction and the request data, r represents the set of items for the user instruction, y represents the rating of the request data for item t, h represents the average of all rated items of the user instruction, and g represents the average of all rated items of the request data. Step S52: Calculate the similarity between the user instruction μ and the request data i. The calculation formula is as follows: Where S(μ,K) represents the top K request data most similar to user instruction μ, N(i) represents the user instruction set of request data i, m μv represents the similarity between user instruction μ and user instruction v, R vi Indicates the interest of user instruction v in requesting data i; Step S53: For the first K request data that are most similar to the user instruction μ, calculate the set l of the user instruction μ and these K request data respectively. n The similarity between them is obtained, and the similarity list is recommended to the service requester.

7. The recommendation method based on enterprise service bus according to claim 1, characterized in that: In step S7, the process of introducing the recommended items into the knowledge graph model is as follows: Step S71: extract knowledge from the original project data set, express it in triple form, standardize the triples and import them into the database for storage, and construct a project knowledge graph; Step S72: Randomly sample the project data, represent each entity and relationship with a multi-dimensional feature vector, and train the TransH model; Step S73: Sampling from the project dataset, randomly replacing the head entity or tail entity to generate a test dataset, and importing it into the TransH model; Step S74: Obtain the embedded feature vector of the entity in the project knowledge graph through the TransH model, expressed as In the formula, z represents any item and h represents the number of dimensions; Step S75: Calculate the similarity between the two items. The formula is: Where, represents the Euclidean distance between item i and item j; Step S76: storing multiple items with the highest similarity to item i in the database.

Images