Method and system for evaluating matching degree of production research and development project and park

Abstract

The invention relates to the technical field of park investment attraction, in particular to a production research and development project and park matching degree evaluation method and system, and the method comprises the steps: obtaining the target demand data of a production research and development project, and constructing a production research and development project demand map; obtaining park space feature data, and constructing a park space feature map; obtaining the number of isomorphic sub-graphs by adopting a sub-graph isomorphism algorithm; and according to the number of the isomorphic sub-graphs and a matching degree scoring formula, evaluating the matching degree between the park spatial features and the production research and development project requirements. According to the method, accurate matching is carried out by utilizing a subgraph isomorphism algorithm through the constructed production research and development project demand map and the park space feature map, a quantitative matching degree score is provided, and the matching degree between park space features and production research and development project demands is evaluated, so that an enterprise is helped to efficiently match a proper park space, and the production research and development project demand matching efficiency is improved. And more scientific and reasonable decision support is provided for enterprises.

Description

Technical Field

[0001] The present invention relates to the technical field of park investment promotion, and in particular to a method and system for evaluating the matching degree between a production and research and development project and a park. Background Art

[0002] With the rapid development of the global economy, the complexity and diversity of production and R&D projects are increasing, and companies face many challenges in choosing suitable park spaces. At present, many companies often rely on experience and intuition when selecting project sites, lacking systematic analysis tools, resulting in waste of resources and low efficiency. For example, some companies may ignore the convenience of transportation, resulting in increased subsequent logistics costs; or there is a mismatch between energy demand and park energy supply, resulting in reduced production efficiency. In addition, with increasingly stringent environmental regulations, companies also need to consider factors such as environmental tolerance and safety tolerance when choosing a park, which makes traditional matching methods seem inadequate.

[0003] Therefore, there is an urgent need for an evaluation method and system for the matching degree between production and R&D projects and the soft and hard environment of the park space based on knowledge graphs. Summary of the invention

[0004] Based on this, it is necessary to provide an evaluation method and system for the matching degree between production R&D projects and the park in response to the above technical issues.

[0005] According to a first aspect of the present invention, there is provided a method for evaluating the matching degree between a production R&D project and a park, comprising: obtaining target demand data of the production R&D project, and constructing a production R&D project demand map based on target demand attribute information of the target demand data of the production R&D project, wherein the production R&D project demand map is generated through a plurality of production R&D project demand types and correlation relationships between the production R&D project demand types, and the correlation relationships are mutual influence relationships and dependency relationships; obtaining park space feature data, and constructing a park space feature map based on park space feature attribute information of the park space feature data, wherein the park space feature map is generated through a plurality of park space feature types and correlation relationships between the park space feature types, and the correlation relationships are mutual influence relationships and dependency relationships; using a subgraph isomorphism algorithm to search the park space feature map for subgraphs isomorphic to the production R&D project demand map, and obtaining the number of isomorphic subgraphs; and evaluating the matching degree between the park space features and the production R&D project requirements based on the number of isomorphic subgraphs and a matching degree scoring formula.

[0006] Optionally, the method of obtaining target demand data of the production and R&D projects and constructing a production and R&D project demand map based on the target demand attribute information of the target demand data of the production and R&D projects includes: obtaining the target demand data of the production and R&D projects, standardizing the target demand data to obtain target demand attribute information; identifying the production and R&D project demand type to which the target demand attribute information belongs based on the defined production and R&D project demand types; creating each production and R&D project demand type as a first-class first-level node; if there is a mutual influence relationship between any two production and R&D project demand types, creating a first-class first-level edge between the first-class first-level nodes corresponding to the two production and R&D project demand types, wherein the first-class first-level edge is directional and is determined based on the dependency relationship; and constructing a production and R&D project demand map based on the created first-class first-level nodes and the first-class first-level edges.

[0007] Optionally, constructing a production R&D project demand map further includes: creating the target demand attribute information under the first-type first-level node as a first-type second-level node; and constructing a first-type second-level edge between the first-type first-level node and the first-type second-level node based on the attribute relationship.

[0008] Optionally, the obtaining of park spatial feature data and constructing a park spatial feature map according to the park spatial feature attribute information of the park spatial feature data include: obtaining the park spatial feature data, standardizing the park spatial feature data to obtain the park spatial feature attribute information; identifying the park spatial feature type to which the park spatial feature attribute information belongs according to the defined park spatial feature types; creating each park spatial feature type as a second-class first-level node; if there is a mutual influence relationship between any two park spatial feature types, creating a second-class first-level edge between the second-class first-level nodes corresponding to the two park spatial feature types, wherein the second-class first-level edge is directional and is determined according to the dependency relationship; and constructing a park spatial feature map based on the created second-class first-level nodes and the second-class first-level edges.

[0009] Optionally, constructing the park space feature map also includes: creating the park space feature attribute information under the second-type first-level node as a second-type second-level node; and constructing a second-type second-level edge between the second-type first-level node and the second-type second-level node according to the attribute relationship.

[0010] Optionally, the subgraph isomorphism algorithm is used to search for a subgraph isomorphic to the production and R&D project demand graph in the park space feature map, and obtain the number of isomorphic subgraphs, including: using one-hot encoding technology to encode the first type of nodes and the first type of edges contained in the production and R&D project demand map, respectively, to obtain a feature vector set of the first type of nodes and a feature vector set of the first type of edges, wherein the first type of nodes includes the first type of first-level nodes and the first type of second-level nodes, and the first type of edges includes the first type of first-level edges and the first type of second-level edges; using one-hot encoding technology to encode the second type of nodes and the second type of edges contained in the park space feature map, respectively, to obtain a feature vector set of the second type of nodes and a feature vector set of the second type of edges, wherein the second type The nodes include the second-class first-level nodes and the second-class second-level nodes, and the second-class edges include the second-class first-level edges and the second-class second-level edges; based on the correspondence between the first-class nodes and the first-class edges, the feature vector set of the first-class nodes and the feature vector set of the first-class edges are integrated to generate a project feature vector set; based on the correspondence between the second-class nodes and the second-class edges, the feature vector set of the second-class nodes and the feature vector set of the second-class edges are integrated to generate a park feature vector set; a subgraph isomorphism algorithm is used to compare the project feature vector set and the park feature vector set to obtain a matching result for each subgraph, and the matching result includes a successful match and a failed match; the number of successfully matched subgraphs is counted as the number of isomorphic subgraphs.

[0011] Optionally, the matching degree between the park space characteristics and the production R&D project requirements is evaluated according to the number of isomorphic subgraphs and the matching degree scoring formula, including: using the matching degree scoring formula to output the evaluation value of the park space characteristics and the production R&D project requirements, wherein the size of the evaluation value represents the degree of matching between the park space characteristics and the production R&D project requirements; the matching degree scoring formula is as follows: ; Where: It represents the overall matching score between the park's spatial characteristics and the production and R&D project requirements. Indicates the number of subgraphs that are successfully matched. Represents the i-th matching subgraph Matching subgraph with j The similarity between The calculation of is based on the similarity between node feature vectors and edge feature vectors.

[0012] Optionally, before using the one-hot encoding technology to encode the production and R&D project demand map and the park space characteristic map, the method also includes: normalizing the first type of nodes contained in the production and R&D project demand map and the second type of nodes contained in the park space characteristic map.

[0013] Optionally, a subgraph isomorphism algorithm is used to compare the project feature vector set and the park feature vector set to obtain a matching result for each subgraph, including: constructing an empty matching subgraph set; selecting any feature vector from the project feature vector set as a starting point, traversing each feature vector in the park feature vector set, and outputting a successful matching result if the starting point successfully matches any feature vector in the park feature vectors, and continuing to recursively search for the next feature vector consisting of the first-class nodes and first-class edges adjacent to the starting point; if the match fails, mark the path, which represents the first-class nodes and first-class edges when the starting point fails to match any feature vector in the park feature vectors, and trace back to the last successfully matched feature vector, and continue to recursively search for other unmarked paths based on the last successfully matched feature vector; put all successfully matched subgraphs into the matching subgraph set.

[0014] Optionally, whether two feature vectors are matched successfully is determined based on a calculation result of a cosine similarity formula, where the cosine similarity formula is: ; Where A represents an eigenvector, B represents another eigenvector, · represents the dot product, represents the modulus of the one eigenvector, represents the modulus of the other eigenvector.

[0015] According to a second aspect of the present invention, there is provided a system for evaluating the matching degree between a production R&D project and a park, comprising: a production R&D project demand map construction module, used to obtain target demand data of a production R&D project, and construct a production R&D project demand map according to target demand attribute information of the target demand data of the production R&D project, wherein the production R&D project demand map is generated by a plurality of production R&D project demand types and the association relationship between the production R&D project demand types, and the association relationship is a mutual influence relationship and a dependency relationship; a park space feature map construction module, used to obtain park space feature data, and construct a park space feature map according to park space feature attribute information of the park space feature data, wherein the park space feature map is generated by a plurality of park space feature types and the association relationship between the park space feature types, and the association relationship is a mutual influence relationship and a dependency relationship; a subgraph statistics module, used to search for a subgraph isomorphic to the production R&D project demand map in the park space feature map by using a subgraph isomorphism algorithm, and obtain the number of isomorphic subgraphs; an evaluation and analysis module, used to evaluate the matching degree between the park space feature and the production R&D project demand according to the number of isomorphic subgraphs and a matching degree scoring formula.

[0016] The advantages and beneficial effects of the present invention are as follows: the present invention provides a method and system for evaluating the matching degree between a production R&D project and a park. The method comprehensively analyzes the target demand data of the production R&D project and the spatial feature data of the park, and respectively constructs a production R&D project demand map and a park spatial feature map. The method can not only identify the specific needs of the production R&D project, but also systematically compare and analyze these specific needs with the spatial feature data of the park, so as to help enterprises efficiently match the appropriate park space. Through the constructed production R&D project demand map and park spatial feature map, the subgraph isomorphism algorithm is used for precise matching, and a quantitative matching score is provided to evaluate the matching degree between the spatial characteristics of the park and the production R&D project needs, so as to provide enterprises with more scientific and reasonable decision-making support, thereby improving resource utilization efficiency, reducing operating costs, and promoting sustainable development. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 It is the overall flow chart of the present invention.

[0018] Figure 2 It is a schematic diagram of the production and R&D project demand map of the present invention.

[0019] Figure 3 It is a schematic diagram of the park space characteristic map of the present invention.

[0020] Figure 4 It is a system principle block diagram of the present invention.

[0021] Figure 5 A schematic diagram of an electronic device. DETAILED DESCRIPTION

[0022] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below through specific implementation methods in conjunction with the accompanying drawings. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0023] Based on the technical problem that existing enterprises often rely on experience and intuition when selecting project sites and lack systematic analysis tools, resulting in waste of resources and low efficiency, this application proposes a method for evaluating the matching degree between production and R&D projects and parks. This method can obtain the target demand data of the production and R&D projects and the spatial feature data of the park, respectively construct a production and R&D project demand map and a park spatial feature map, and use a subgraph isomorphism algorithm and a quantitative matching score to evaluate the matching degree between the park spatial characteristics and the production and R&D project needs, thereby systematically comparing and analyzing the specific needs of the production and R&D projects with the park spatial feature data, so as to help enterprises efficiently match suitable park spaces, and then provide enterprises with more scientific and reasonable decision-making support, thereby improving resource utilization efficiency, reducing operating costs, and promoting sustainable development.

[0024] Embodiment 1

[0025] like Figure 1 As shown, a flow chart of a method for evaluating the matching degree between a production R&D project and a park according to the present application is shown, and the evaluation method includes the following steps.

[0026] S1. Obtain target demand data of production R&D projects, and construct a production R&D project demand map based on the target demand attribute information of the target demand data of the production R&D projects, wherein the production R&D project demand map is generated through multiple production R&D project demand types and the association relationships between the production R&D project demand types, and the association relationships are mutual influence relationships and dependency relationships.

[0027] In this step, we first need to collect and organize various types of demand data for production and R&D projects, and then standardize the collected demand data to ensure the comparability between different demand characteristics. Then, we use the various types of production and R&D project demand collected as nodes, and the relationship between the types of production and R&D project demand as the edge connecting the two nodes to construct the production and R&D project demand map.

[0028] In some embodiments, obtaining target demand data for a production R&D project and constructing a production R&D project demand map may include the following steps.

[0029] S101. Obtain target demand data for a production R&D project, standardize the target demand data, and obtain target demand attribute information.

[0030] Standardization processing can be understood as format processing. Since the information may be obtained through questionnaires, expert interviews, market research, etc., there will be different formats, voice needs to be converted into text, and demand needs to be identified. All of the above problems are handled in a unified manner to form a unified format to ensure the comparability between different demand characteristics.

[0031] S102. Define the types of production and R&D project requirements.

[0032] The above-mentioned types of demand can be dynamically added according to the current market situation. In a specific embodiment, the types of demand may include: space site demand, transportation demand, parts supply demand, energy demand, manpower demand, environmental tolerance demand and safety tolerance demand. Among them, space site demand refers to the physical space required for the project, and its demand attribute information includes building area, layout requirements, etc.; transportation demand refers to the project's requirements for traffic convenience, and its demand attribute information includes the distance to access major traffic arteries, accessibility of public transportation, etc.; parts supply demand refers to the project's supply chain requirements for raw materials and parts, and its demand attribute information includes the supplier's geographical location, delivery time, etc.; energy demand refers to the type of energy required by the project and the stability of its supply, and its demand attribute information includes electricity, natural gas, etc.; manpower demand refers to the human resources required by the project, and its demand attribute information includes skill requirements, number of people and its availability; environmental tolerance demand refers to the project's adaptability to environmental factors, and its demand attribute information includes noise, pollution, etc.; safety tolerance demand refers to the project's requirements for safety standards and risk management, and its demand attribute information includes fire safety, etc.

[0033] S103. According to the defined production and research and development project demand types, identify the production and research and development project demand type to which the target demand attribute information belongs.

[0034] According to the production R&D project demand type provided in S102, the acquired target demand attribute information is identified and processed.

[0035] Assume that the target demand attribute information obtained is "Area: 5000 square meters", "Layout: Open", "Transportation: Public transportation available", "Environment: No noise or less noise". According to "Area: 5000 square meters", it can be understood as a space demand in the demand category, according to "Layout: Open", it can be understood as a space demand in the demand category, according to "Transportation: Public transportation available", it can be understood as a transportation demand in the demand category, and according to "Environment: No noise or less noise", it can be understood as an environmental tolerance demand in the demand category. Based on the target demand attribute information obtained above, the space demand, transportation demand, and environmental tolerance demand are identified.

[0036] S104. Create each production R&D project demand type as a first-class level node.

[0037] The above-listed production and R&D project demand types can all be represented as first-class first-level nodes, and each production and R&D project demand type can be represented by a first-class first-level node. For example, the three production and R&D project demand types of space site demand, transportation demand, and environmental tolerance demand are each set with a first-class first-level node.

[0038] S105. If there is a mutual influence relationship between any two production and R&D project demand types, a first-class first-class edge is created between the first-class first-class nodes corresponding to the two production and R&D project demand types.

[0039] Specifically, first determine whether there is a mutual influence relationship between the two types of production and R&D project requirements. If there is a mutual influence relationship, a first-class level edge is established between the two types of production and R&D project requirements. If there is no mutual influence relationship, it is not established. The influence relationship can be defined as the relationship strength, weight value, etc. between the two types of production and R&D project requirements. For example, there may be a direct relationship between space requirements and manpower requirements, because the size of the space may affect the number of human resources required. For another example, space requirements depend on transportation requirements, because the convenience of transportation or the distance to the main road will bring a large amount of manpower or vehicle traffic, which requires more space to accommodate.

[0040] Furthermore, the first-class primary edge can also be a directional edge, and the directional arrow is determined by the dependency relationship. For example, a first-class primary edge points from the space site demand to the manpower demand, which means that if there is no space site as support, there will be no manpower demand; for another example, a first-class primary edge points from the space site demand to the transportation demand, which means that if there is no space site as support, there will be no transportation demand.

[0041] S106. Based on the created first-class first-level nodes and first-class first-level edges, a production and R&D project demand map is constructed.

[0042] Use a graph database (such as Neo4j) to build a production R&D project demand graph. In the graph database, create first-class first-level nodes and add the identified demand types to each first-class first-level node, and then define the first-class first-level edge between two first-class first-level nodes.

[0043] In order to improve the comparison between the production and R&D project demand map and the park space feature map, in some embodiments, constructing the production and R&D project demand map also includes: creating the target demand attribute information under the first-class first-level node as a first-class second-level node; and constructing the first-class second-level edge between the first-class first-level node and the first-class second-level node according to the attribute relationship.

[0044] It can be understood that the first-class secondary node belongs to the child node of the first-class primary node, that is, each demand type node can include secondary expansion content, and the secondary expansion content as the demand attribute information of the demand type can include information such as the specific value, unit and importance weight of the demand. Considering that a demand may correspond to multiple target demand attribute information, for example, the secondary expansion of the space site demand node can include demand attribute information such as "area: 5000 square meters" and "layout: open". The secondary expanded demand attribute information can be represented as a first-class secondary node, and the relationship between the first-class primary node and the first-class secondary node will be represented as a first-class secondary edge.

[0045] This step can also be achieved by using a graph database (such as Neo4j) to further improve the production and R&D project demand map.

[0046] S2. Obtain park space feature data, and construct a park space feature map based on the park space feature attribute information of the park space feature data, wherein the park space feature map is generated through multiple park space feature types and the association relationship between the park space feature types, and the association relationship is a mutual influence relationship and a dependency relationship.

[0047] In this step, we first need to collect and organize various park space feature data, and then standardize the collected feature data to ensure the comparability between different feature data. Then, we use the collected park space feature types as nodes, and the association between the park space feature types as the edge connecting the two nodes to construct the production park space feature map.

[0048] In some embodiments, obtaining park space feature data and constructing a park space feature map includes the following steps.

[0049] S201, obtaining park space characteristic data, and performing standardization processing on the park space characteristic data to obtain park space characteristic attribute information.

[0050] Standardization processing can be understood as format processing. Since the data may be obtained through geographic information systems (GIS), market research, government statistical data, etc., there will be different formats, voice needs to be converted into text, and demand needs to be identified. All of the above problems are handled in a unified manner to form a unified format to ensure the comparability between different demand characteristics.

[0051] S202. Define the types of park space characteristics.

[0052] It is necessary to define that the types of park spatial characteristics may include: location characteristics, traffic network connectivity characteristics, supply chain coverage characteristics, energy supply characteristics, manpower supply characteristics, environmental tolerance characteristics and safety tolerance characteristics, among which, location characteristics refer to the geographical location of the park, and its park spatial characteristic attribute information includes coordinates, surrounding environment, adjacent facilities, etc.; traffic network connectivity characteristics refer to the connection between the park and major traffic arteries and public transportation stations, and its park spatial characteristic attribute information includes road type, traffic capacity, etc.; supply chain coverage characteristics refer to the supply chain network inside and outside the park, and its park spatial characteristic attribute information includes the distribution of major suppliers and logistics centers and their distances; energy supply characteristics refer to the types of energy in the park, supply stability and its sources, and its park spatial characteristic attribute information includes electricity, natural gas, etc.; manpower supply characteristics refer to the available human resources in the park, and its park spatial characteristic attribute information includes the skill level, quantity and accessibility of the labor force; environmental tolerance characteristics refer to the park's adaptability to environmental factors, and its park spatial characteristic attribute information includes noise, pollution, climate, etc.; safety tolerance characteristics refer to the park's requirements for safety standards and risk management, and its park spatial characteristic attribute information includes fire safety, public security conditions, etc.

[0053] S203: Identify the park space feature type to which the park space feature attribute information belongs according to the defined park space feature type.

[0054] According to the park space feature type provided in S202, the acquired park space feature attribute information is identified and processed.

[0055] For example, for location feature nodes, the acquired park space feature attribute information may include "coordinates: latitude, longitude", "surrounding facilities: schools, hospitals", etc.; for traffic network connectivity feature nodes, it may include "distance to main roads: 500 meters", "number of public transportation stations: 3", etc.; for supply chain coverage feature nodes, it may include "number of main suppliers: 10", "distance to logistics center: 2 kilometers", etc.; for energy supply feature nodes, it may include "power supply stability: high", "natural gas supply source: local", etc.; for manpower supply feature nodes, it may include "number of available labor: 200 people", "skill level: medium", etc.; for environmental tolerance feature nodes, it may include "noise tolerance: 60 decibels", "pollutant emission standards: in line with national standards", etc.; for safety tolerance feature nodes, it may include "fire safety level: A", "public security situation: good", etc.

[0056] S204. Create each park space feature type as a second-class first-level node.

[0057] The above-mentioned park spatial characteristics are represented as second-class first-level nodes, and each park spatial characteristic can be represented by a second-class first-level node. For example, a second-class first-level node is set for each of the location characteristics, traffic network connectivity characteristics, supply chain coverage characteristics, energy supply characteristics, human supply characteristics, environmental tolerance characteristics, and safety tolerance characteristics.

[0058] S205. If there is a mutual influence relationship between any two park space feature types, a second-type first-level edge is created between the second-type first-level nodes corresponding to the two park space feature types.

[0059] Specifically, first determine whether there is a mutual influence relationship between the two park spatial feature types. If there is a mutual influence relationship, a second-class first-level edge is established between the two park spatial feature types. If there is no mutual influence relationship, it is not established. For example, the correlation between the location feature and the traffic network connectivity feature is characterized as "traffic convenience"; the correlation between the supply chain coverage feature and the human resource supply feature is characterized as "human resource accessibility"; the correlation between the energy supply feature and the environmental tolerance feature is characterized as "environmental impact". The influence relationship can be defined as the relationship strength, weight value, etc. of the two park spatial feature types. For example, the edge between the traffic network connectivity feature and the location feature can be defined as "influence intensity: high".

[0060] Furthermore, the second-class first-level edge is also directional, and the arrow pointing to it is determined by the dependency relationship. For example, a second-class first-level edge points from the location feature to the traffic network connectivity feature, which means that the location feature will affect the traffic network connectivity feature. If the location is poor, the traffic convenience will be poor. For another example, a second-class first-level edge points from the labor supply feature to the supply chain coverage feature, which means that the labor supply feature will support the supply chain coverage feature. If the labor force decreases, the number of suppliers will decrease.

[0061] S206. Construct a park spatial feature map based on the created second-class first-level nodes and second-class first-level edges.

[0062] The construction method of the park space feature map is the same as the construction method of the production and R&D project demand map, that is, the park space feature map can also be constructed through a graph database (such as Neo4j), which will not be explained in detail here.

[0063] Similarly, in order to improve the comparison between the production and R&D project demand map and the park space feature map, in some embodiments, constructing the park space feature map also includes: creating the park space feature attribute information under the second-class first-level node as a second-class second-level node; and constructing a second-class second-level edge between the second-class first-level node and the second-class second-level node according to the attribute relationship.

[0064] The secondary expanded park space feature attribute information can be represented as a second-class secondary node, and the relationship between the second-class primary node and the second-class secondary node will be represented as a second-class secondary edge. It can be understood that the second-class secondary node belongs to the child node of the second-class primary node, that is, each park space feature type can include secondary expansion content, and the secondary expansion content as the park space feature attribute information of each type of park space feature can include specific values, units and importance weights of each type of park space feature.

[0065] Through step S1 and step S2, a complete park space feature map and a complete production and R&D project demand map are finally formed. These two maps can not only clearly display various features and their relationships, but also provide basic data structures for subsequent matching analysis.

[0066] S3. Use the subgraph isomorphism algorithm to search for subgraphs that are isomorphic to the production and R&D project demand graph in the park's spatial feature graph, and obtain the number of isomorphic subgraphs.

[0067] Before using subgraph isomorphism, the production and R&D project demand graph and the park space feature graph need to be processed to facilitate subsequent matching calculations. This can be done by converting all involved nodes and edges into numerical values. This can be achieved by following the steps below.

[0068] S301. Use one-hot encoding technology to encode the first-class nodes and the first-class edges contained in the production and R&D project demand graph respectively, and obtain a set of feature vectors of the first-class nodes and a set of feature vectors of the first-class edges, wherein the first-class nodes include the first-class first-level nodes and the first-class second-level nodes, and the first-class edges include the first-class first-level edges and the first-class second-level edges.

[0069] S302. Use the one-hot encoding technology to encode the second-category nodes and the second-category edges contained in the park space feature map respectively, and obtain the feature vector set of the second-category nodes and the feature vector set of the second-category edges, wherein the second-category nodes include the second-category first-level nodes and the second-category second-level nodes, and the second-category edges include the second-category first-level edges and the second-category second-level edges.

[0070] Step S301 and step S302 both perform encoding processing on respective nodes.

[0071] In one embodiment, before using the one-hot encoding technology to encode the production and R&D project demand map and the park space feature map, the first type of nodes included in the production and R&D project demand map and the second type of nodes included in the park space feature map are normalized.

[0072] For each node (including first-class nodes and second-class nodes), its attribute information needs to be obtained first, and data can be processed and converted through data collection and analysis tools (such as Python's Pandas library). Then, the attribute information is converted into numerical form using normalization and encoding techniques. For example, a second-class first-level node represents "location features", and its attribute information includes "coordinates", "surrounding facilities", etc. These attribute information are encoded to form a feature vector, and then the type of surrounding facilities (such as schools, hospitals, etc.) is represented by one-hot encoding. Each facility corresponds to a binary bit. Taking the location feature node as an example, the latitude and longitude are normalized to coordinates in the interval [0, 1], and then the type of surrounding facilities (such as schools, hospitals, etc.) is represented by one-hot encoding. Each facility corresponds to a binary bit, and finally a node feature vector for the location feature node is formed: ,in, , , , ..., They respectively represent the one-hot encoding of the facilities around the coordinate after normalization.

[0073] For each edge (including the first-type and second-type edges), we first need to obtain its attribute information, such as "influence strength" and "relationship type". These attribute information can be converted into numerical form to form the feature vector of the edge. Taking the traffic convenience edge as an example, a scoring system of 1-5 is used to represent the degree of influence, and unique hot encoding is used to represent different relationship types (such as "dependence" and "support"), and finally the edge feature vector for the traffic convenience edge is formed: ,in, Indicates the intensity of the impact, One-hot encoding representing the relationship type.

[0074] It should be noted that the encoding process of the attribute features that have been normalized into coordinates can be understood as first converting the attribute information of the location feature node into longitude and latitude, and then further converting it into a value in the interval [0, 1], and then performing one-hot encoding. In this process, there is no need to distinguish whether the attributes belong to the same feature node and directly encode each attribute, or the feature nodes and attributes can be comprehensively considered to form a one-hot encoding.

[0075] In the specific implementation, it is first necessary to define and classify the characteristic attributes of each node and edge in detail. The data can be processed and converted through data collection and analysis tools (such as Python's Pandas library), and then the features can be converted into numerical form using normalization and encoding techniques. For example, for location feature nodes, Python's MinMaxScaler can be used for normalization, and for the one-hot encoding of surrounding facilities, OneHotEncoder can be used for conversion. After the feature vectorization is completed, the feature vectors of all nodes and edges in each graph are stored in a data structure (such as a NumPy array or Pandas DataFrame) for subsequent matching calculations.

[0076] S303 . Based on the correspondence between the first-type nodes and the first-type edges, the feature vector set of the first-type nodes and the feature vector set of the first-type edges are integrated to generate a project feature vector set.

[0077] S304. Based on the correspondence between the second-type nodes and the second-type edges, the feature vector set of the second-type nodes and the feature vector set of the second-type edges are integrated to generate a park feature vector set.

[0078] After completing step S301 and step S302, the types of production and R&D project requirements and the association between two types of production and R&D project requirements can be combined into an overall project feature vector set, that is, the project feature vector set is constructed based on the first-type node feature vector and the first-type edge feature vector; the types of park space features and the association between two types of park space features can be combined into an overall park feature vector set, that is, the park feature vector set is constructed based on the second-type node feature vector and the second-type edge feature vector.

[0079] Define the park feature vector set as ,in, represents the set of feature vectors of the second type of nodes, Represents the set of second-class edge eigenvectors.

[0080] Define the set of item feature vectors as ,in, represents the first type of node feature vector set, Represents the set of first-class edge eigenvectors, and then uses the subgraph isomorphism algorithm to find isomorphic subgraphs in the two eigenvector sets.

[0081] S305. Use the subgraph isomorphism algorithm to compare the project feature vector set and the park feature vector set to obtain the matching results of each subgraph, which include successful matching and failed matching. Specifically, it includes: constructing an empty matching subgraph set; selecting any feature vector from the project feature vector set as the starting point, traversing each feature vector in the park feature vector set, and outputting the successful matching result if the starting point successfully matches any feature vector in the park feature vector, and continuing to recursively search for the next feature vector composed of the first-class node and the first-class edge adjacent to the starting point, if the match fails, marking the path, and backtracking to the last successfully matched feature vector, and continuing to recursively search for other unmarked paths based on the last successfully matched feature vector; putting all successfully matched subgraphs into the matching subgraph set.

[0082] It is understandable that multiple matching subgraphs that are isomorphic to the production and R&D project demand graph are traversed in the park space feature graph to ensure that the bijective relationship between nodes and the corresponding relationship between edges are satisfied. First, an empty matching set M is constructed, and then Select a feature vector as the starting point. Iterate over each feature vector in and check whether it can be compared with If the match is successful, it will be added to the matching set M and the next recursive search will continue. Feature vector. If the current path cannot find more matches, it will backtrack to the previous one. feature vector, try other possible matches.

[0083] It should be understood that when matching, the main match is the attribute information under each node, and the attribute information is a supplementary description of each feature node. One situation can be understood as: when the attribute information does not match, the feature vector of the second-class node that was successfully matched is returned; for example, it succeeds after matching the traffic of the park, but fails when matching the park facilities. At this time, it is necessary to return the successfully matched traffic node, and then find other paths from the traffic node except the failed path just now, and try to match again; another situation can be understood as: what is returned is the feature vector of the upper-level second-class node that failed to match, and then try to match other lower-level attribute information of the upper-level second-class node feature vector again; for example, in the matching project of the park area, the park area fails to match 100 square meters, then the other attribute nodes under the park area can be matched again. If none of them succeed under this node, then the last successfully matched node is returned.

[0084] Whether two feature vectors are matched successfully is determined based on the calculation result of the cosine similarity formula, which is: ; Where A represents an eigenvector, B represents another eigenvector, · represents the dot product, represents the modulus of a feature vector, Represents the magnitude of another eigenvector.

[0085] S306: Count the number of successfully matched subgraphs.

[0086] By calculating the matching subgraph set The number of features is calculated to determine the number of successfully matched subgraphs. Then, step S4 is performed to evaluate the matching degree between the park space features and the production and R&D project requirements based on the number of isomorphic subgraphs and the matching degree scoring formula.

[0087] Specifically, a matching score formula is used to output the evaluation value of the park's spatial characteristics and production and R&D project requirements, where the size of the evaluation value represents the degree of matching between the park's spatial characteristics and production and R&D project requirements; the matching score formula is as follows: ; Where: It represents the overall matching score between the park's spatial characteristics and the production and R&D project requirements. Indicates the number of subgraphs that are successfully matched. Represents the i-th matching subgraph Matching subgraph with j The similarity between The calculation of is based on the similarity between node feature vectors and edge feature vectors.

[0088] After finding all possible isomorphic subgraphs, the overall matching degree can be evaluated by calculating the number of successfully matched subgraphs. That is, when calculating the overall matching degree, it is necessary to first traverse the set M of all successfully matched subgraphs, calculate the similarity of each pair of matched subgraphs (that is, calculate the node similarity and edge similarity of each pair of subgraphs), and then sum up all similarity values ​​and divide them by the number of matched subgraphs. , to obtain the final overall matching score.

[0089] Furthermore, when implementing the matching degree calculation, we first need to ensure that the set of successfully matched subgraphs M has been constructed. Next, we use a graph processing library (such as NetworkX) to extract the node feature vectors and edge feature vectors of each subgraph. For each matched subgraph, we calculate the cosine similarity of the node features and average the similarities of all nodes to obtain the node similarity score of the subgraph. Then, we repeat the same process to calculate the similarity of the edge features. For example, suppose there are three subgraphs in the successfully matched subgraph. , and , their node similarities are 0.8, 0.9 and 0.85 respectively, and their edge similarities are 0.7, 0.75 and 0.8 respectively. Then we can calculate the similarity score of each subgraph and substitute it into the overall matching formula. Finally, we can get the overall matching score by calculation.

[0090] Embodiment 2

[0091] Based on the above-mentioned embodiment 1, this embodiment provides an evaluation system 200 for the matching degree between a production R&D project and a park. Figure 4 , implementing a method for evaluating the matching degree between a production R&D project and a park in Example 1, the system includes: a production R&D project demand map construction module 210, a park space feature map construction module 220, a sub-map statistics module 230 and an evaluation and analysis module 240.

[0092] In some embodiments, the production R&D project demand map construction module 210 is used to: obtain the target demand data of the production R&D project, and construct a production R&D project demand map based on the target demand attribute information of the target demand data of the production R&D project, wherein the production R&D project demand map is generated through multiple production R&D project demand types and the association relationships between the production R&D project demand types, and the association relationships are mutual influence relationships and dependency relationships.

[0093] In some embodiments, the park space feature map construction module 220 is used to: obtain park space feature data, and construct a park space feature map based on the park space feature attribute information of the park space feature data, wherein the park space feature map is generated through multiple park space feature types and the association relationship between the park space feature types, and the association relationship is a mutual influence relationship and a dependency relationship.

[0094] In some embodiments, the subgraph statistics module 230 is used to: use a subgraph isomorphism algorithm to search for subgraphs that are isomorphic to the production and R&D project demand graph in the park space feature graph, and obtain the number of isomorphic subgraphs.

[0095] In some embodiments, the evaluation and analysis module 240 is used to evaluate the matching degree between the park space characteristics and the production and R&D project requirements according to the number of isomorphic subgraphs and the matching degree scoring formula.

[0096] Embodiment 3

[0097] This embodiment further provides an electronic device based on the above embodiment 1. Figure 5 , Figure 5 The electronic device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0098] like Figure 5 As shown, the electronic device may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 301, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 302 or a program loaded from a storage device 308 to a random access memory (RAM) 303. Various programs and data required for the operation of the electronic device are also stored in the RAM 303. The processing device 301, the ROM 302, and the RAM 303 are connected to each other via a bus 304. An input / output (I / O) interface 305 is also connected to the bus 304.

[0099] Typically, the following devices may be connected to the I / O interface 305: an input device 306 including, for example, a touch screen, a touch pad, a keyboard, a mouse, a camera, etc., an output device 307 including, for example, a liquid crystal display (LCD), a speaker, etc., a storage device 308 including, for example, a magnetic tape, a hard disk, etc., and a communication device 309. The communication device 309 may allow the electronic device to communicate with other devices wirelessly or by wire to exchange data. Figure 5 An electronic device having various devices is shown, but it should be understood that it is not required to implement or possess all the devices shown. More or fewer devices may be implemented or possessed instead. Figure 5 Each block shown in the figure may represent one device, or may represent multiple devices as required.

[0100] In particular, according to some embodiments of the present disclosure, the process described above with reference to the flowchart can be implemented as a computer software program. For example, some embodiments of the present disclosure include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In some such embodiments, the computer program can be downloaded and installed from the network through the communication device 309, or installed from the storage device 308, or installed from the ROM 302. When the computer program is executed by the processing device 301, the above-mentioned functions defined in the method of some embodiments of the present disclosure are executed.

[0101] Embodiment 4

[0102] Based on the above-mentioned embodiment 1, this embodiment further provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the steps of the above-mentioned method are implemented.

[0103] It should be noted that the computer-readable medium mentioned above in some embodiments of the present disclosure may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In some embodiments of the present disclosure, the computer-readable storage medium may be any tangible medium containing or storing a program, which may be used by or in combination with an instruction execution system, device or device. In some embodiments of the present disclosure, the computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program code. This propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer readable signal media may also be any computer readable medium other than computer readable storage media, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer readable medium may be transmitted using any appropriate medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.

[0104] In some embodiments, the client and the server may communicate using any currently known or future developed network protocol such as HTTP (HyperTextTransferProtocol), and may be interconnected with any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network ("LAN"), a wide area network ("WAN"), an internet (e.g., the Internet), and a peer-to-peer network (e.g., an adhoc peer-to-peer network), as well as any currently known or future developed network.

[0105] The computer-readable medium may be included in the device, or may exist independently without being installed in the electronic device. The computer-readable medium carries one or more programs. When the one or more programs are executed by the electronic device, the electronic device: obtains training data, converts the training data into initial data; determines an initial rule base based on the initial data, and optimizes the parameters of the initial rule base to obtain a target rule base; calculates the rules in the target rule base according to a preset activation weight calculation formula to obtain activation weights; determines abnormal information according to the test data and the activation weights.

[0106] Computer program code for performing the operations of some embodiments of the present disclosure may be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0107] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present disclosure. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some implementations as replacements, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0108] The units described in some embodiments of the present disclosure may be implemented by software or hardware. The units described may also be provided in a processor, for example, may be described as: a processor including a data acquisition unit, a rule determination unit, a weight calculation unit, and an anomaly determination unit. The names of these units do not, in some cases, constitute limitations on the units themselves, for example, the data acquisition unit may also be described as a "unit for acquiring training data".

[0109] The functions described above herein may be performed at least in part by one or more hardware logic components. For example, exemplary types of hardware logic components that may be used include, without limitation, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), etc.

[0110] Obviously, it should be understood by those skilled in the art that the above-mentioned various steps of the present invention can be performed in a manner different from the present invention, and the simulation method and experimental equipment include but are not limited to the above description. The above-mentioned various steps of the present invention can be performed in an order different from that here in some cases, and the steps shown or described above can be performed separately. Therefore, the present invention is not limited to any specific combination of hardware and software.

[0111] The above contents are further detailed descriptions of the present invention in combination with specific implementation methods, and it cannot be determined that the specific implementation of the present invention is limited to these descriptions. For ordinary technicians in the technical field to which the present invention belongs, several simple deductions or substitutions can be made without departing from the concept of the present invention, which should be regarded as falling within the scope of protection of the present invention.

Claims

1. A method for evaluating the matching degree between a production R&D project and a park, characterized in that: include: Obtain target demand data of the production R&D project, and construct a production R&D project demand map according to target demand attribute information of the target demand data of the production R&D project, wherein the production R&D project demand map is generated by multiple production R&D project demand types and associations between the production R&D project demand types, and the associations are mutual influence relationships and dependency relationships; Acquire park space feature data, and construct a park space feature map according to park space feature attribute information of the park space feature data, wherein the park space feature map is generated by a plurality of park space feature types and associations between the park space feature types, wherein the associations are mutual influence relationships and dependency relationships; Using a subgraph isomorphism algorithm to search for a subgraph isomorphic to the production and R&D project demand graph in the park space feature graph, and obtaining the number of isomorphic subgraphs; According to the number of isomorphic subgraphs and the matching score formula, the matching degree between the park space characteristics and the production and R&D project requirements is evaluated.

2. The method for evaluating the matching degree between a production R&D project and a park according to claim 1, characterized in that: The step of obtaining target demand data of the production R&D project and constructing a production R&D project demand map according to target demand attribute information of the target demand data of the production R&D project includes: Obtain target demand data for production and R&D projects, standardize the target demand data, and obtain target demand attribute information; According to the defined production and research and development project demand types, identifying the production and research and development project demand type to which the target demand attribute information belongs; Create each production R&D project requirement type as a first-class level node; If there is a mutual influence relationship between any two types of production and R&D project requirements, a first-class first-class edge is created between the first-class first-class nodes corresponding to the two types of production and R&D project requirements, and the first-class first-class edge is directional and is determined according to the dependency relationship; Based on the first-class first-level nodes and the first-class first-level edges created, a production and R&D project demand graph is constructed.

3. The method for evaluating the matching degree between a production R&D project and a park according to claim 2, characterized in that: The construction of the production R&D project demand map also includes: Creating the target demand attribute information under the first-class first-class node as a first-class second-class node; A first-type secondary edge is constructed between the first-type primary node and the first-type secondary node according to the attribute relationship.

4. The method for evaluating the matching degree between a production R&D project and a park according to claim 3 is characterized in that: The obtaining of the park space feature data and constructing a park space feature map according to the park space feature attribute information of the park space feature data includes: Acquire the park space characteristic data, perform standardization on the park space characteristic data, and obtain the park space characteristic attribute information; According to the defined park space feature types, identifying the park space feature type to which the park space feature attribute information belongs; Each park space feature type is created as a second-class first-level node; If there is a mutual influence relationship between any two park space feature types, a second-type first-level edge is created between the second-type first-level nodes corresponding to the two park space feature types, and the second-type first-level edge is directional and is determined according to the dependency relationship; Based on the created second-type first-level nodes and the second-type first-level edges, a park spatial feature map is constructed.

5. The method for evaluating the matching degree between a production R&D project and a park according to claim 4, characterized in that: The construction of the park space feature map also includes: The park space feature attribute information under the second-type first-level node is created as a second-type second-level node; A second type of secondary edge is constructed between the second type of primary node and the second type of secondary node according to the attribute relationship.

6. The method for evaluating the matching degree between a production R&D project and a park according to claim 5, characterized in that: The adopting of the subgraph isomorphism algorithm to search for a subgraph isomorphic to the production and R&D project demand graph in the park space feature graph and obtaining the number of isomorphic subgraphs includes: The first-class nodes and the first-class edges contained in the production and R&D project demand graph are respectively encoded using a one-hot encoding technology to obtain a feature vector set of the first-class nodes and a feature vector set of the first-class edges, wherein the first-class nodes include the first-class primary nodes and the first-class secondary nodes, and the first-class edges include the first-class primary edges and the first-class secondary edges; The second-type nodes and the second-type edges contained in the park space feature map are respectively encoded using a one-hot encoding technology to obtain a feature vector set of the second-type nodes and a feature vector set of the second-type edges, wherein the second-type nodes include the second-type primary nodes and the second-type secondary nodes, and the second-type edges include the second-type primary edges and the second-type secondary edges; Based on the correspondence between the first-type nodes and the first-type edges, integrating the feature vector set of the first-type nodes and the feature vector set of the first-type edges to generate a project feature vector set; Based on the correspondence between the second-type nodes and the second-type edges, the feature vector set of the second-type nodes and the feature vector set of the second-type edges are integrated to generate a park feature vector set; Using a subgraph isomorphism algorithm to compare the project feature vector set and the park feature vector set, to obtain a matching result for each subgraph, wherein the matching result includes a successful match and a failed match; The number of successfully matched subgraphs is counted as the number of isomorphic subgraphs.

7. The method for evaluating the matching degree between a production R&D project and a park according to claim 6, characterized in that: According to the number of isomorphic subgraphs and the matching score formula, the matching degree between the park space characteristics and the production and R&D project requirements is evaluated, including: The matching score formula is used to output the evaluation value of the park space characteristics and the production and R&D project requirements, wherein the magnitude of the evaluation value represents the degree of matching between the park space characteristics and the production and R&D project requirements; The matching score formula is as follows: Where: It represents the overall matching score between the park's spatial characteristics and the production and R&D project requirements. Indicates the number of subgraphs that are successfully matched. Represents the i-th matching subgraph Matching subgraph with j The similarity between The calculation of is based on the similarity between node feature vectors and edge feature vectors.

8. The method for evaluating the matching degree between a production R&D project and a park according to claim 6, characterized in that: Before using the one-hot encoding technology to encode the production and R&D project demand map and the park space feature map, the method further includes: The first type of nodes included in the production and R&D project demand map and the second type of nodes included in the park space feature map are normalized.

9. The method for evaluating the matching degree between a production R&D project and a park according to claim 6, characterized in that: The subgraph isomorphism algorithm is used to compare the project feature vector set and the park feature vector set to obtain the matching result of each subgraph, including: Construct an empty set of matching subgraphs; Select any feature vector from the project feature vector set as a starting point, and traverse each feature vector in the park feature vector set. If the starting point successfully matches any of the park feature vectors, a successful match result is output, and a recursive search is continued for the next feature vector consisting of a first-class node and a first-class edge adjacent to the starting point. If the match fails, the path is marked, and the path represents the first-class nodes and first-class edges when the starting point fails to match any feature vector in the park feature vector, and backtracks to the last successfully matched feature vector, and continues to recursively search for other unmarked paths based on the last successfully matched feature vector; All successfully matched subgraphs are put into the matching subgraph set.

10. The method for evaluating the matching degree between a production R&D project and a park according to claim 6, characterized in that: Whether two feature vectors are matched successfully is determined based on the calculation result of the cosine similarity formula, which is: Where A represents an eigenvector, B represents another eigenvector, · represents the dot product, represents the modulus of the one eigenvector, represents the modulus of the other eigenvector.

11. A system for evaluating the matching degree between production and R&D projects and parks, characterized in that: include: A production R&D project demand map construction module is used to: obtain target demand data of the production R&D project, and construct a production R&D project demand map according to target demand attribute information of the target demand data of the production R&D project, wherein the production R&D project demand map is generated by multiple production R&D project demand types and the association relationship between the production R&D project demand types, and the association relationship is a mutual influence relationship and a dependency relationship; A park space feature map construction module is used to: obtain park space feature data, and construct a park space feature map according to park space feature attribute information of the park space feature data, wherein the park space feature map is generated by a plurality of park space feature types and associations between the park space feature types, wherein the associations are mutual influence relationships and dependency relationships; A subgraph statistics module is used to: use a subgraph isomorphism algorithm to search for subgraphs that are isomorphic to the production and R&D project demand graph in the park space feature graph, and obtain the number of isomorphic subgraphs; The evaluation and analysis module is used to evaluate the matching degree between the park space characteristics and the production and R&D project requirements according to the number of isomorphic subgraphs and the matching degree scoring formula.

Images