Method for dynamically managing innovation point elements of science and technology enterprises in multiple parks

Abstract

The invention discloses a method for dynamically managing science and technology type enterprise innovation point elements in multiple parks, which relates to the technical field of data services, and comprises the following steps: establishing a structured enterprise innovation point element meta-model and a data model, and providing visual management of the meta-model and the data model; a model conversion relation and traceability connection are established between the innovative integral element data model and the display model, and the display model can be automatically and adaptively adjusted according to changes after the innovative integral element data model is changed; the model and data access interface declaration and implementation are unchanged by defining an extensible parameter object and a model conversion technology; enterprise innovation integral element multi-model parallel operation is supported, and multi-park domain labeling is realized by using a clustering algorithm. The problems that when an enterprise innovation point element data model changes, page layout and elements need to be redeveloped, and service interface declaration and specific implementation change accordingly are solved, and dynamic management of multi-park enterprise innovation point elements is achieved through visual configuration.

Description

Technical Field

[0001] The present invention relates to the field of data service technology, and in particular to a method for dynamically managing innovation score elements of technology-based enterprises in multiple parks. Background Art

[0002] The Ministry of Science and Technology has launched a new technology policy tool, the "Enterprise Innovation Points System". The implementation of the policy requires the ability to accurately identify and effectively discover technology-based enterprises with strong R&D capabilities and great growth potential. It needs to be able to reflect the degree of innovation and development of technology companies in a quantitative manner, and objectively reflect the innovation and development gap between different companies through the level of points. This requires good management of enterprise innovation points elements. The core innovation points elements formulated by the Ministry of Science and Technology include operating income, R&D share, and scientific and technological personnel share. Different cities and high-tech parks will also have their own development directions and evaluation dimensions. Obviously, this new policy tool requires a flexible and elastic enterprise innovation points element management method to help cities and parks freely define and manage park-specific innovation points elements, facilitate precise policy implementation, and effectively guide technology, capital, talents, public services and other resources to gather in innovative enterprises.

[0003] In the prior art, the data model is basically defined by enumerating the innovation score elements in advance. There are technical problems such as the inability to dynamically expand the innovation score elements, resulting in high maintenance costs for innovation capability evaluation, and the difficulty in adjusting the innovation score elements as the business is deeply applied after the definition is completed. Summary of the invention

[0004] The purpose of the present invention is to provide a method for dynamically managing innovation points elements of technology-based enterprises in multiple parks, establish a structured metamodel and data model of enterprise innovation points elements, provide visualization model and data management, dynamically expand innovation points elements horizontally, support multi-model parallel operation of innovation points elements of multiple parks enterprises, dynamically adjust the metamodel, dynamically update the corresponding page elements, and automatically complete adaptive adjustment of the front-end display page as the innovation points elements are adjusted. At the same time, the service interface remains unchanged and the specific implementation of the service interface does not need to be changed, so that data service personnel can achieve dynamic management of enterprise innovation points elements by adjusting parameters or changing configurations, so as to solve the problems raised in the above background technology.

[0005] In order to solve the above technical problems, the present invention provides the following solution, a method for dynamically managing innovation score elements of technology-based enterprises in multiple parks, the method comprising the following steps: S1: Define the enterprise innovation score element meta-model, whose attributes include classification attributes, basic attributes and display attributes. The classification attributes include classification ID, classification name, classification description, classification order, whether to display, column style, and number of columns. The basic attributes include element ID, element name, element meaning, data type, and category. The display attributes include element sorting, whether to display, whether required, page controls, and validation rules. S2: Establishing an enterprise innovation score factor data model, which is a meta-model instance obtained based on the innovation score factor meta-model, defining and collecting specific enterprise innovation score factors based on the innovation capabilities of the park enterprises, instantiating discrete innovation score factor items, organizing them into a structured data model, and providing a configurable management page for the innovation score factor data model; S3: Establishing a model conversion relationship and a traceability connection between the innovation credit element data model and the display model, so that the display model automatically adapts to the change after the innovation credit element is changed. The display model includes a page layout and page components. The model conversion relationship is to establish a mapping relationship between the data model and the display model. The traceability connection is to record the connection between the source and target elements during model conversion. S4: Provide interface declaration and interface implementation for access services of innovation credit element metamodel and data model. The interface declaration uses extensible data objects as parameters. The interface implementation uses model conversion and data mapping to complete data storage and data conversion, so as to ensure that the corresponding access services remain unchanged after the innovation credit element metamodel and data model are changed. S5: Based on the tenant model, multiple models of enterprise innovation score elements in multiple parks can be operated in parallel. The tenant model allocates independent memory space and independent database for enterprise innovation score element data models and data storage in different parks, uses data routing technology to establish access channels for different parks to achieve independent access to memory and databases in different parks, and uses clustering algorithm machine learning to complete domain labels and horizontal data comparisons for multi-park data.

[0006] The S1 defines the enterprise innovation element metadata model, which specifically includes the following: Define the categories to which the innovation score elements belong, including category names and descriptions, and provide different display schemes for different categories, including: whether to display attributes for each category, the order of display, the number of columns displayed in each row of the category, and the column style of each category, such as attribute key-value pair mode, data list mode, and file attachment mode; Define the names and meanings of the innovation score elements, define a computer-recognizable name for all innovation score elements, and provide a user-friendly meaning description; Define the data type of the innovation score element. All innovation score elements need to define data types, such as numeric type, character type, date type, and enumeration type, to facilitate the storage and display of innovation score element instances; Using the element name as the traceability link, a mapping relationship between the basic metamodel and the display model is established. Validation rules for binding metamodel data are supported to ensure the correctness of the data. On this basis, display rules for innovation points elements are defined to help users better understand and apply the metamodel after instantiation. The display rule attributes provided include whether it is required, whether to display, display order, display controls, and legality verification rules for binding innovation points elements.

[0007] The S2 establishes a structured enterprise innovation integral factor data model, which specifically includes the following contents: The data model of innovation score elements is an instance of the metamodel. The expansion of the data model of enterprise innovation score elements is reflected in the increase of metamodel instance data. Conversely, the deletion or modification of metamodel instance data is associated with the deletion or modification of the data model of enterprise innovation score elements, thus realizing the dynamic horizontal expansion of enterprise innovation score elements. After the data model is defined, specific enterprise innovation score elements are defined and collected according to the innovation capabilities of the park enterprises. Each innovation score element further instantiates the sorting attributes of each data item, whether it is required, whether it is displayed, the verification rules, the field type mapped to the database, and the corresponding visual components when the binding page is presented according to the display rules defined in the metamodel; After the innovation points elements are defined, a configuration management page for the innovation points element data model is provided.

[0008] The S3 described above establishes a model conversion relationship and a traceability connection between the innovation score element data model and the display model. After the innovation score element is changed, the display model is adaptively adjusted according to the change. The specific contents are as follows: The classification management of enterprise innovation score elements defines the attributes of classification attributes when displayed on the page, and realizes the meta-model attribute configuration capability through visual page management, including: each category determines whether to display or close the display according to parameterized configuration; each category defines the display number of elements in each row; each category defines the display style, including attribute key-value pair mode, data list mode and file attachment mode; the display parameters of the category are increased or adjusted according to the actual application, and the change of classification parameters automatically adjusts the display of page components through model conversion, completing the adaptive management of classification page components; When instantiating the innovation score element data model, define the sorting order between the elements, define the required attributes of each element, and inject or bind different verification rules to different elements; the configuration attributes of the elements are added or adjusted according to the actual application, and the management page of the innovation score element data is generated through the defined display rules. The data content of the operation is stored by establishing a mapping relationship between the basic metamodel and the display model. The change of the innovation score element data model automatically completes the adjustment of the corresponding page components of the innovation score element through model conversion; The model conversion relationship and traceability connection between the data model and the display model are realized. The traceability connection is based on the same unique identification of the elements. The model conversion is that each innovation integral element in the data model is mapped to the display model. Through the traceability connection, there is a unique display control corresponding to it, and vice versa. The unique display control refers to the control that is bound to the display attribute and the model verification rule and has the same name identification. The model conversion specifically includes: The basic attributes of the innovation score elements are recorded as ; The display attribute of the innovation score element is recorded as ; The classification of the innovation points elements is recorded as follows: ; The page components are denoted as ,in Represents the basic attribute set of innovation integral elements , Display attribute collection and classification collection To the page component collection The mapping can also be written as ; For any , for any set , for the set The solving process is called image mapping, which completes the mapping from the innovative integral element data model to the display model; and for any set , for the set The solution process is called inverse image mapping, which completes the inverse mapping of data storage from the innovation integral element display model to the data model.

[0009] The S4 provides an interface declaration and interface implementation for access services of the innovation credit element metamodel and data model. The interface declaration uses an extensible data object as a parameter. The interface implementation uses model conversion and data mapping to complete data storage and data conversion, so as to achieve that the corresponding access services remain unchanged after the innovation credit element metamodel and data model are changed. The specific contents are as follows: Provides management pages and data services for the enterprise innovation integral element metamodel, allowing data managers to manage the metamodel in a visual way; Provide management pages and data services for the enterprise innovation score element data model, allowing business managers to maintain enterprise innovation score elements in a visual way; It provides operation services for instantiated model data, provides data services and access interface declarations externally, defines interface parameters as extensible data objects, completes metamodel data abstract encapsulation management and database model conversion and storage at the internal implementation level, uses the image mapping when accessing and generating pages, and uses inverse image mapping when page operations are completed to save model data. When adding, reducing or changing enterprise innovation score elements, the interface declaration prototype and interface implementation remain unchanged, and can adapt to enterprise innovation score element domain models in different parks and regions; Provide enterprise innovation data management page and business services, through the design of the underlying metamodel, complete the mapping management between the innovation score data object and the score element metamodel data instance when the interface is implemented, so that the data operator does not need to perceive the metamodel, and operates the innovation score data just like ordinary data. All user interaction methods are the same as ordinary data model interaction methods. The biggest difference between the ordinary enterprise innovation score element model and the innovation score element model generated based on the present invention is that the present invention supports model adjustment and configuration at any time according to business needs without the intervention of programmers.

[0010] The S5 described above implements the parallel operation of multiple models of enterprise innovation integral elements based on the tenant model, and specifically includes the following contents: Based on the hierarchical management of innovation points elements in the tenant model, the hierarchical management of the tenant model is to divide managers into park managers, park superior managers and superior superior managers, among which the park managers manage the data and models of the park, and the superior managers or superior superior managers of the park manage and view the innovation points elements and innovation points data of the parks under their jurisdiction, and complete machine learning through the cluster analysis algorithm K-means to realize horizontal data comparison between parks of the same level; The K-means clustering algorithm is an algorithm used in machine learning to achieve automatic field classification of each park, including 1) collecting enterprise data for field identification, 2) counting the number of enterprises in different fields of the park, 3) creating an initial cluster center, 4) repeatedly traversing the distance from each park to the cluster center and finding the minimum distance, 5) continuously iterating steps 3 and 4 until the position of the cluster center no longer changes significantly or reaches a preset iteration frequency, as follows: The aforementioned collection of enterprise data provides a field identification for each enterprise, including electronic information, biology and new medicine, aerospace, new materials, high-tech services, new energy and energy conservation, resources and environment, advanced manufacturing and automation, and others; The number of enterprises in each park area is recorded as ; The different park field categories are recorded as Indicate point The category, use Represents the cluster center ,use express The cluster center to which it belongs; The initial cluster center Use random scheme to generate , the number of cluster centers and the number of territories are consistent; Traverse each point , calculate the points To cluster center Distances are compared, and the points with the smallest distance in the cluster are continuously marked and screened out; Traverse all points of the same class separately, calculate the geometric mean position of these points, and move the cluster center to this position. The cluster center during the traversal process is recorded as: ,In the process of machine learning, the system continuously seeks to minimize the sum of the squares of the distances between all points and their cluster centers, which can be expressed as: ; The K-means algorithm requires multiple iterations and repeated tuning, and the cluster center is randomly initialized in each iteration until the position of the cluster center no longer changes significantly or reaches the preset iteration frequency; It provides horizontal data comparison of innovation score elements among enterprises in the same industry, horizontal data comparison of innovation score elements among parks of the same level, and data training in different fields of enterprise innovation score element meta-model. The clustering K-means algorithm and iterative learning process are also applicable to the classification and training of other innovation score element data, forming a complete innovation score element training library.

[0011] Compared with the prior art, the method for dynamically managing innovation points elements of technology-based enterprises in multiple parks provided by the present invention has the following beneficial effects: establishing a set of scientific and standard innovation points metamodels and data models for dynamically managing enterprise innovation points elements; realizing visual management of metamodels and data models, and display models and components can perceive changes and adaptively adjust after the innovation points element data model is changed; the innovation points elements can be expanded and changed at any time according to the needs of the innovation points business without the need for redevelopment and programmer intervention; the stability of the service interfaces provided externally and internally by the enterprise innovation points system is guaranteed through the underlying model conversion technology; and supporting the parallel operation of multiple models of innovation points elements for enterprises in multiple parks. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings: Figure 1 It is a flow chart of a method for dynamically managing innovation score elements of technology-based enterprises in multiple parks of the present invention; Figure 2 This is a management page of the enterprise innovation integral element metamodel according to an embodiment of the present invention; Figure 3 This is a management page of the enterprise innovation score element data model according to an embodiment of the present invention; Figure 4 This is a storage example diagram of the enterprise innovation score element data model according to an embodiment of the present invention; Figure 5 This is a diagram showing an example of storage of enterprise innovation score element data according to an embodiment of the present invention; DETAILED DESCRIPTION

[0013] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0014] The goal of the embodiment of the present invention is to first define the meta-model of innovation points elements of technology-based enterprises, then establish a structured data model of enterprise innovation points elements, and then explain how page components and data access interfaces can maintain automatic adaptation and interface stability after the innovation points element model is changed. Finally, it explains how to realize parallel operation of multiple models of enterprise innovation points elements based on the tenant model.

[0015] Reference Figure 1 , the method of dynamically managing the innovation points of technology-based enterprises in multiple parks is as follows: S1: Define the enterprise innovation score element meta-model, whose attributes include classification attributes, basic attributes and display attributes. The classification attributes include classification ID, classification name, classification description, classification order, whether to display, column style, and number of columns. The basic attributes include element ID, element name, element meaning, data type, and category. The display attributes include element sorting, whether to display, whether required, page controls, and validation rules. S2: Establishing an enterprise innovation score factor data model, which is a meta-model instance obtained based on the innovation score factor meta-model, defining and collecting specific enterprise innovation score factors based on the innovation capabilities of the park enterprises, instantiating discrete innovation score factor items, organizing them into a structured data model, and providing a configurable management page for the innovation score factor data model; S3: Establishing a model conversion relationship and a traceability connection between the innovation credit element data model and the display model, so that the display model automatically adapts to the change after the innovation credit element is changed. The display model includes a page layout and page components. The model conversion relationship is to establish a mapping relationship between the data model and the display model. The traceability connection is to record the connection between the source and target elements during model conversion. S4: Provide interface declaration and interface implementation for access services of innovation credit element metamodel and data model. The interface declaration uses extensible data objects as parameters. The interface implementation uses model conversion and data mapping to complete data storage and data conversion, so as to ensure that the corresponding access services remain unchanged after the innovation credit element metamodel and data model are changed. S5: Based on the tenant model, multiple models of enterprise innovation score elements in multiple parks can be operated in parallel. The tenant model allocates independent memory space and independent database for enterprise innovation score element data models and data storage in different parks, uses data routing technology to establish access channels for different parks to achieve independent access to memory and databases in different parks, and uses clustering algorithm machine learning to complete domain labels and horizontal data comparisons for multi-park data.

[0016] In a specific embodiment of the present invention, the S1 defines the enterprise innovation element metadata model, and the specific steps are as follows: Step 1: Determine the relational database mysql to complete the data storage of the example of the present invention; Step 2: Define the classification table to which the innovation points elements belong, where the core attribute fields include classification id, classification name title, classification description remark, whether the classification is displayed is_show, display order sort, the number of columns displayed in each row view_columns, and the column style view_type. The column style corresponds to a drop-down list or a subtable. The classification content includes attribute key-value pair mode, data list mode, and file attachment mode. In addition to the above core fields, add or adjust classification attributes according to specific business; Step 3: Define the ID column of the category of the innovation score element table as menu_id, store it in long type, and associate it with the category table defined in step 2; Step 4: Define the name column of the innovation score element table as name, which stores the computer-recognizable innovation score element identifiers and uses the varchar type for storage; Step 5: Define the column name of the innovation score element table as remark, which stores the meaning description that is easy for users to understand, and uses the varchar type for storage; Step 6: Define the data type column name of the innovation score element table as data_type, use varchar type for storage, and support numeric, character, date, and enumeration data types. The data types are not limited to the above and can be expanded as needed; Step 7: Define the column name for sorting the elements in the innovation score element table, which is stored in int type; Step 8: Define the display column name of the innovation score element table as is_show, and use char type for storage; Step 9: Define the column name of the innovation score element table as is_required, and use the char type for storage; Step 10: Define the data type column name of the innovation score element table as view_type, and use varchar type for storage; Step 11: Define the validation rule column name of the innovation score element table as view_validate, which is stored in varchar type. The validation rule script or rule code can be directly stored according to the implementation situation. For the definition and description of the relevant fields in the innovation score element table, see Figure 2 In addition to the above core fields, feature attributes are added or adjusted according to specific business.

[0017] In a specific embodiment of the present invention, the S2 establishes a structured enterprise innovation score element data model, and the specific steps are as follows: Step 1: Determine that the example of the present invention is implemented using JAVA language; Step 2: Define the categories of innovation score elements of the present invention example, such as Figure 4 In the storage example shown, a category "business information" is defined, and its identification ID is 10000, and a category "growth management indicators" is defined, and its identification ID is 100101. The enterprise innovation score elements of the example of the present invention can be respectively classified into corresponding categories, for example Figure 4 The establishment date belongs to the business information with category ID 10000. Figure 4 Net asset profit margin, which belongs to the growth operating indicator with category ID 100101; Step 3: Collect the innovation score elements of the enterprise, such as Figure 4 As shown, there are the date of establishment, whether it is a high-tech enterprise, income from high-tech products, operating income, the proportion of employees with postgraduate degrees or above, tax exemptions for additional deductions for R&D expenses, net asset profit rate, and operating income growth rate. Figure 4 The innovation score elements listed as examples of data model storage are only for the purpose of explaining the invention examples and are not limited to these elements. In fact, the present invention can horizontally expand any innovation score element as needed; Step 4: Figure 4 As shown, according to the display rules defined in the metamodel, the specific sorting attributes of each data item, whether it is required, whether it is displayed, the verification rules, the field type mapped to the database, and the corresponding visual components when the binding page is presented are further instantiated. For example Figure 4 The operating income in this record has a sort value of 2 and a menu_id value of 100101, indicating that the operating income innovation score element ranks second among all elements of the growth management indicator with the category ID 100101. The is_required value in this record is 1, indicating that this element is required. The view_type value in this record is WANYUAN_INPUT, indicating that it will be bound to the 10,000 yuan input control WANYUAN_INPUT when the page is displayed. The data_type value in this record is decimal, indicating that the type stored in the data instance is a numeric decimal type. Step 5: After defining the enterprise innovation integral element meta-model and the innovation integral element model instance, further define a data table to store the specific data of each innovation integral element, such as Figure 5 As shown, the field names of this data table correspond one to one with the value of the attribute name defined in the metamodel in step 4. For example, the operating income mentioned in step 4 has a name value of idx009 in the metamodel. Therefore, the field name in this data table should be set to idx009. In this way, the idx009 field in the data table will store the operating income. Through the definition, transformation and field matching mapping of steps 4 and 5, all innovation score element data can be stored on demand and expanded freely.

[0018] In a specific embodiment of the present invention, the S3 establishes a model conversion relationship and a traceability connection between the innovation score element data model and the display model, so that the display model automatically adapts to the change after the innovation score element is changed. The specific steps are as follows: Step 1: Define the classification management of enterprise innovation score elements. Different categories are displayed or closed and the display order is determined according to parameterized configuration; different categories define the number of elements displayed in each row; different categories define display styles, including attribute key-value pair mode, data list mode and file attachment mode; the display parameters of the category are also increased or adjusted according to the actual application. The classification parameterization scheme managed by the metamodel can complete the adaptive management of the classification page components. For example, define a category name as "Technology Innovation Index", its is_show=true, sort=4, such as Figure 3As shown, "Technology Innovation Index" is displayed in the 4th position in the category menu list on the right side of the figure. The attribute is_show is also used to decide whether to use or not use a certain type of innovation score element during model debugging. If this attribute is set to false, the display will be turned off and it will not be displayed in the category menu. At the same time, all innovation score elements under this category will not only not be displayed, but also will not participate in the input and output of the model and data. Step 2: Figure 3 It is a management page of the instantiated data model of the present invention. Drag each innovation score element on the page to complete the sorting and positioning of the element. In the small menu on the right side of the innovation score element, complete the setting of required or canceled required fields and bind different verification rules to the innovation score element; click [Add Field] to add a new innovation score element; select the delete button on the right side of a row with the mouse to delete the corresponding innovation score element; click the edit icon on the right side of each category to set the category attributes and modify the name; click [Add Category] below the category list on the right to add a new category; Step 3: Implement the model conversion relationship and traceability connection between the data model and the display model. The traceability connection is based on the same unique identifier of the elements. The model conversion is to map each innovation score element in the data model to the display model. Through the traceability connection, there is a unique display control corresponding to it. The unique display control refers to a control with the same name identifier, and the control is bound to the display properties and model verification rules of the innovation score element with the same name. The specific content of the model conversion includes: The basic attributes of the innovation score elements are recorded as ; The display attribute of the innovation score element is recorded as ; The classification of the innovation points elements is recorded as follows: ; The page components are denoted as ,in Represents the basic attribute set of innovation integral elements , Display attribute collection and classification collection To the page component collection The mapping can also be written as ; For any , for any set , for the set The solving process is called image mapping, which completes the mapping from the innovative integral element data model to the display model; and for any set , for the set The solution process is called inverse image mapping, which completes the inverse mapping of data storage from the innovation integral element display model to the data model.

[0019] In a specific embodiment of the present invention, the S4 provides an interface declaration and interface implementation of access services of the innovation integral element metamodel and the data model. The interface declaration uses an extensible data object as a parameter, and the interface implementation uses model conversion and data mapping to complete data storage and data conversion, so as to achieve that the corresponding access services remain unchanged after the innovation integral element metamodel and the data model are changed. The specific steps are as follows: Step 1: Figure 2 The enterprise innovation integral element metamodel shown provides management pages and data services. The data manager configures the management innovation integral element metamodel. On the management page, the configuration parameters of the metamodel attributes and the corresponding display controls of each metamodel attribute after instantiation can be added, modified or deleted as needed. For example, text boxes, drop-down boxes, etc. If it is a drop-down box, the data dictionary corresponding to the drop-down option must also be set. After the definition of the metamodel is completed, the system will provide the enterprise innovation integral element metamodel with add, delete, modify and query services and the mapping service of display control data to JAVA data; Step 2: Provide management pages and data services for the enterprise innovation integration element metamodel instance. Figure 3 It is the management page of the enterprise innovation score element data model. Figure 3 The unique identifier of the display control in is and Figure 4 The name field in the table is consistent. Business managers maintain the enterprise innovation score elements in a visual way. Figure 4 The model storage instance shown provides the enterprise innovation integral element metamodel instance management capability, including the type management of innovation integral elements, the management page for innovation integral element configuration, the corresponding display controls, data dictionary, and storage services for related model instances until completion. Figure 3 The management page elements shown to Figure 4 The model storage instance shown has mapping, transformation and manipulation services; Step 3: Figure 5 The innovation score element data shown provides the innovation score element data management page and data services. The system will Figure 4 The stored metamodel instance content dynamically generates the operation page of the data instance. The innovation score elements on the page will be displayed in the corresponding categories in sequence according to the data results saved in the previous steps. Each innovation score element is displayed on the page according to the designed control binding. When submitted, the data is verified according to the set verification rules. The system completes the mapping management between the innovation score data object and the score element metamodel data instance during interface implementation through the design of the underlying metamodel, so that data operators do not need to perceive the metamodel and can operate the innovation score data just like operating ordinary data. All user interaction methods are the same as ordinary data model interaction methods; Step 4: Provide an operation service interface for instantiated model data, provide external data services and access interfaces to express service interfaces similar to ordinary innovation integral element models, use relatively simple extensible data objects for interface parameter definition, and support lightweight data exchange objects JSON and extensible language XML as interface parameters to construct interfaces; use Java reflection technology to dynamically generate corresponding properties and property set and get methods when converting parameter objects to Java objects, so that interface declarations and interface parameter objects have good flexibility and good extensibility. Based on the implementation method of the present invention, when adding, reducing or changing enterprise innovation integral elements, the interface declaration prototype remains unchanged; Step 5: Next, at the internal implementation level, the unique identifier of the innovation score element is used as the traceability connection. The image mapping is used when accessing and generating pages. The inverse image mapping is used when the page operation is completed to save the model data. The metamodel data abstract encapsulation management and database model conversion and storage are completed. The collected extensible data objects are mapped to the metamodel instance table by name. The stored data format is as follows: Figure 5 As shown, based on the implementation method of the present invention, when adding, reducing or changing the enterprise innovation score elements, the interface implementation content remains unchanged, and the enterprise innovation score element data model in different parks and regions can be adapted.

[0020] In a specific embodiment of the present invention, the S5 implements the parallel operation of multiple models of enterprise innovation integral elements based on the tenant model, and the specific steps are as follows: Step 1: Support tenant mode to include model data of multiple parks. The system can accommodate data and models of multiple parks at the same time, provide independent park identification and memory space for each park, build a database for each park data independently, and use Java data routing technology to direct to the corresponding data and memory area according to the park identification to complete independent data access; Step 2: The system uses the collected enterprise data training as input data for model training to better identify the innovation fields and development needs of the park or street. The system completes machine learning through the clustering analysis algorithm K-means, and labels the parks or streets with domain labels to facilitate horizontal data observation of different parks or streets. The K-means clustering algorithm is an algorithm used in machine learning to achieve automatic classification of each park, including 1) collecting enterprise data for domain identification, 2) counting the number of enterprises in different fields in the park, 3) creating an initial cluster center, 4) repeatedly traversing the distance from each park to the cluster center and finding the minimum distance, 5) continuously iterating steps 3 and 4 until the position of the cluster center no longer changes significantly or reaches the preset iteration frequency, as follows: The aforementioned collection of enterprise data provides a field identification for each enterprise, including electronic information, biology and new medicine, aerospace, new materials, high-tech services, new energy and energy conservation, resources and environment, advanced manufacturing and automation, and others; The number of enterprises in each park area is recorded as ; The different park field categories are recorded as Indicate point The category, use Represents the cluster center ,use express The cluster center to which it belongs; The initial cluster center Use random scheme to generate , the number of cluster centers and the number of territories are consistent; Traverse each point , calculate the points To cluster center Distances are compared, and the points with the smallest distance in the cluster are continuously marked and screened out; Traverse all points of the same class separately, calculate the geometric mean position of these points, and move the cluster center to this position. The cluster center during the traversal process is recorded as: ,In the process of machine learning, the system continuously seeks to minimize the sum of the squares of the distances between all points and their cluster centers, which can be expressed as: ; The K-means algorithm requires multiple iterations and repeated tuning, with the cluster center randomly initialized in each iteration until the position of the cluster center no longer changes significantly or reaches the preset iteration frequency; Step 3: Supporting the model data training of different fields of the enterprise innovation integral element meta-model. The clustering K-means algorithm and iterative learning process are also applicable to the classification and training of other innovation integral element data of the park or street, forming a complete innovation integral element training library; Step 4: Provide different enterprise models of innovation score elements in the same stage and industry for simultaneous operation, provide bar chart and radar chart modes, support horizontal comparison of enterprises in the same industry, and find horizontal advantages and shortcomings of enterprise data.

Claims

1. A method for dynamically managing innovation points of technology-based enterprises in multiple parks, characterized in that: The method comprises: S1: Define the enterprise innovation score element meta-model, whose attributes include classification attributes, basic attributes and display attributes. The classification attributes include classification ID, classification name, classification description, classification order, whether to display, column style, and number of columns. The basic attributes include element ID, element name, element meaning, data type, and category. The display attributes include element sorting, whether to display, whether required, page controls, and validation rules. S2: Establishing an enterprise innovation score factor data model, which is a meta-model instance obtained based on the innovation score factor meta-model, defining and collecting specific enterprise innovation score factors based on the innovation capabilities of the park enterprises, instantiating discrete innovation score factor items, organizing them into a structured data model, and providing a configurable management page for the innovation score factor data model; S3: Establishing a model conversion relationship and a traceability connection between the innovation credit element data model and the display model, so that the display model automatically adapts to the change after the innovation credit element is changed. The display model includes a page layout and page components. The model conversion relationship is to establish a mapping relationship between the data model and the display model. The traceability connection is to record the connection between the source and target elements during model conversion. S4: Provide interface declaration and interface implementation for access services of innovation credit element metamodel and data model. The interface declaration uses extensible data objects as parameters. The interface implementation uses model conversion and data mapping to complete data storage and data conversion, so as to ensure that the corresponding access services remain unchanged after the innovation credit element metamodel and data model are changed. S5: Based on the tenant model, multiple models of enterprise innovation score elements in multiple parks can be operated in parallel. The tenant model allocates independent memory space and independent database for enterprise innovation score element data models and data storage in different parks, uses data routing technology to establish access channels for different parks to achieve independent access to memory and databases in different parks, and uses clustering algorithm machine learning to complete domain labels and horizontal data comparisons for multi-park data.

2. The method for dynamically managing innovation points of technology-based enterprises in multiple parks according to claim 1 is characterized in that Establish a structured enterprise innovation score element metamodel and data model to achieve visual management of the metamodel and data model. The visual management is to achieve metamodel attributes, metamodel data types, and mapping binding relationships between metamodel attributes and page elements through page management. It supports the definition of validation rules for binding metamodel data. The validation rules will be executed when the data is submitted to ensure the legitimacy of the data. The visual management also includes generating a visual innovation score element data management page through the defined metamodel.

3. The method for dynamically managing innovation score elements of technology-based enterprises in multiple parks according to claim 2 is characterized in that: The method also includes: the data model of the innovation points element is an instance of the metamodel, the expansion of the data model of the enterprise innovation points element is reflected in the increase of the metamodel instance data, and conversely, the deletion or modification of the metamodel instance data is associated with the deletion or modification of the enterprise innovation points element data model, thereby realizing the dynamic horizontal expansion of the enterprise innovation points element.

4. The method for dynamically managing innovation score elements of technology-based enterprises in multiple parks according to claim 1 is characterized in that: The method further includes: realizing a model conversion relationship and a traceability connection between a data model and a display model, wherein the traceability connection is based on the same unique identifier of the elements, wherein the model conversion is to map each innovation score element in the data model to the display model, and to connect a unique display control corresponding thereto through a traceability connection, wherein the unique display control refers to a control with the same name identifier, and the control is bound to display properties and model verification rules of the innovation score element with the same name, and the specific contents of the model conversion include: The basic attributes of the innovation score elements are recorded as ; The display attribute of the innovation score element is recorded as ; The classification of the innovation points elements is recorded as follows: ; The page components are denoted as ,in Represents the basic attribute set of innovation integral elements , Display attribute collection and classification collection To the page component collection The mapping can also be written as ; For any , for any set , for the set The solving process is called image mapping, which completes the mapping from the innovative integral element data model to the display model; , for the set The solution process is called inverse image mapping, which completes the inverse mapping of data storage from the innovation integral element display model to the data model.

5. The method for dynamically managing innovation score elements of technology-based enterprises in multiple parks according to claim 4 is characterized in that: The method also includes: providing a service interface of the enterprise innovation integral element metamodel and the data model to realize model conversion using image mapping and inverse image mapping, the service interface declaration of the enterprise innovation integral element metamodel and the data model uses extensible data objects as data carriers for displaying the model in parameter definition, the extensible data objects include lightweight data exchange objects JSON and extensible language XML, when the enterprise innovation integral element changes, the service interface declaration of the metamodel and the data model is kept stable by changing the data object attributes.

6. The method for dynamically managing innovation score elements of technology-based enterprises in multiple parks according to claim 4 is characterized in that: The method also includes: using the image mapping technology to achieve page layout and control generation and automatic adaptive adjustment when generating a page, using the inverse image mapping technology to save the model and data after the operation is completed, and storing the collected extensible data objects in the data model instance table by name mapping, and when the enterprise innovation score elements change, the interface implementation keeps the content unchanged and adapts to the enterprise innovation score element data models in different parks and regions.

7. The method for dynamically managing innovation score elements of technology-based enterprises in multiple parks according to claim 1 is characterized in that: Innovation score elements are managed hierarchically based on the tenant model. The tenant model hierarchical management divides managers into park managers, park superior managers and superior superior managers. Park managers manage the data and models of their own park, and park superior managers or superior superior managers manage and view the innovation score elements and innovation score element data of each park under their jurisdiction. Machine learning is completed through the cluster analysis algorithm K-means to realize horizontal data comparison between parks of the same level.

8. The method for dynamically managing innovation score elements of technology-based enterprises in multiple parks according to claim 7 is characterized in that: The K-means clustering algorithm is an algorithm used in machine learning to complete the automatic field classification of each park, including 1) collecting enterprise data for field identification, 2) counting the number of enterprises in different fields of the park, 3) creating an initial cluster center, 4) repeatedly traversing the distance from each park to the cluster center and finding the minimum distance, 5) continuously iterating steps 3 and 4 until the position of the cluster center no longer changes significantly or reaches a preset iteration frequency, as follows: The aforementioned collection of enterprise data provides a field identification for each enterprise, including electronic information, biology and new medicine, aerospace, new materials, high-tech services, new energy and energy conservation, resources and environment, advanced manufacturing and automation, and others; The number of enterprises in each park area is recorded as ; The different park field categories are recorded as Indicate point The category, use Represents the cluster center ,use express The cluster center to which it belongs; The initial cluster center Use random scheme to generate , the number of cluster centers and the number of territories are consistent; Traverse each point , calculate the points To cluster center Distances are compared, and the points with the smallest distance in the cluster are continuously marked and screened out; Traverse all points of the same class separately, calculate the geometric mean position of these points, and move the cluster center to this position. The cluster center during the traversal process is recorded as: ,In the process of machine learning, the system continuously seeks to minimize the sum of the squares of the distances between all points and their cluster centers, which can be expressed as: ; The K-means algorithm requires multiple iterations and repeated tuning, and the cluster centers are randomly initialized in each iteration until the positions of the cluster centers no longer change significantly or the preset iteration frequency is reached.

9. The method for dynamically managing innovation score elements of technology-based enterprises in multiple parks according to claim 8 is characterized in that: The method also includes: providing horizontal data comparison of innovation score elements of enterprises in the same industry, providing horizontal data comparison of innovation score elements of parks at the same level, and providing data training in different fields of enterprise innovation score element meta-model. The clustering K-means algorithm and iterative learning process are also applicable to the classification and training of other innovation score element data, forming a complete innovation score element training library.

Images