Intelligent matching and transaction recommendation method and system for scientific and technological achievements based on knowledge graph

Abstract

The present invention provides a method and system for intelligent matching and transaction recommendation of scientific and technological achievements based on knowledge graphs, which relates to the field of science and technology, including extracting technology gene sequences based on scientific and technological achievement data sets, generating technology state vectors, constructing technology association strength matrices and technology iteration path diagrams, and calculating technology potential energy distribution fields to form an initial knowledge graph. The fluctuation of the knowledge graph state feature set is monitored by an intelligent sensor, and the knowledge graph is dynamically optimized according to the fluctuation. The user's technical requirements are converted into query states, and the state similarity is calculated with the knowledge graph to identify the optimal combination of technical units, predict its state evolution trajectory, evaluate regional adaptability, and generate a technical implementation plan. By dynamically optimizing the knowledge graph and accurately matching user needs, the present invention can effectively improve the efficiency and accuracy of scientific and technological achievement matching and transaction recommendations, and promote the transformation of scientific and technological achievements.

Description

Technical Field

[0001] The present invention relates to science and technology, and in particular to a method and system for intelligent matching and transaction recommendation of scientific and technological achievements based on knowledge graphs. Background Art

[0002] The transformation of scientific and technological achievements is a key component of scientific and technological innovation and an important path to achieving the integrated development of science, technology, and the economy. Traditional methods for matching and trading scientific and technological achievements rely primarily on manual searches and offline matchmaking, which are inefficient and costly, making it difficult to meet the growing demand for the transformation of scientific and technological achievements. In recent years, with the rapid development of technologies such as big data and artificial intelligence, methods for intelligently matching and recommending scientific and technological achievements based on knowledge graphs have gradually emerged, providing new ideas and methods for the transformation of scientific and technological achievements.

[0003] Insufficient mining of technical relevance: ** Most existing methods match scientific and technological achievements based on keyword matching or simple semantic similarity calculations, which makes it difficult to deeply explore the deep relevance between scientific and technological achievements, resulting in low matching accuracy and high missed matching rate.

[0004] Insufficient prediction of technological evolution trends: **Existing methods lack the ability to predict technological development trends, making it difficult to effectively evaluate the future development potential and market value of scientific and technological achievements, leading to blind trading decisions.

[0005] Insufficient consideration of the regional adaptability of technology implementation plans: **Existing methods rarely consider regional differences in technology implementation, such as policy environment, industrial foundation, talent resources and other factors, resulting in low implementation and feasibility of technology implementation plans. Summary of the Invention

[0006] The embodiments of the present invention provide a method and system for intelligent matching and transaction recommendation of scientific and technological achievements based on knowledge graphs, which can solve the problems in the existing technology.

[0007] According to a first aspect of the embodiments of the present invention,

[0008] Provides intelligent matching and transaction recommendation methods for scientific and technological achievements based on knowledge graphs, including:

[0009] Acquire a scientific and technological achievement data set, extract a technology gene sequence from the scientific and technological achievement data set through a deep feature extractor; input the technology gene sequence into a high-dimensional coding network to generate a technology state vector; construct a technology association intensity matrix based on the technology state vector, wherein the technology association intensity matrix records the degree of coupling between technology units; calculate the technology evolution coefficient according to the coupling degree and generate a technology iteration path map; use the technology iteration path map to quantitatively calculate the innovation potential energy value of each technology unit to form a technology potential energy distribution field; integrate the technology association intensity matrix, the technology iteration path map, and the technology potential energy distribution field into an initial knowledge graph, and output a state feature set of the initial knowledge graph;

[0010] Deploy intelligent sensors to monitor the fluctuation of the state feature set in real time; when the monitoring result of the state fluctuation set exceeds a preset dynamic threshold, trigger a graph optimization signal; according to the graph optimization signal, call a coupling calculation unit to recalculate the technology association strength matrix; based on the updated technology association strength matrix, use a dynamic optimization algorithm to optimize the technology iteration path map; use the optimized technology iteration path map to recalculate the technology potential energy distribution field; integrate the updated technology association strength matrix, the technology iteration path map, and the technology potential energy distribution field into a new knowledge graph state, and output an updated state feature set;

[0011] The user's technical requirements are converted into query states, and the state similarity is calculated with the updated state feature set to obtain the state similarity; the optimal technology unit combination is identified based on the state similarity; the state evolution trajectory of the optimal technology unit combination is predicted using the technology iteration path diagram; the regional adaptability of the technology implementation is evaluated in combination with the technology potential energy distribution field, a technology implementation plan is generated, and the user's feedback information on the technology implementation plan is recorded; the feedback information is converted into a state correction factor, which is used to adjust the dynamic threshold to achieve dynamic optimization of the new knowledge graph state.

[0012] Calculate the technology evolution coefficient based on the coupling degree to generate a technology iteration path map; use the technology iteration path map to quantitatively calculate the innovation potential value of each technology unit to form a technology potential energy distribution field; integrate the technology association intensity matrix, the technology iteration path map, and the technology potential energy distribution field into an initial knowledge graph, and output the state feature set of the initial knowledge graph, including:

[0013] Inputting the coupling degree into a technology evolution prediction model, the technology evolution prediction model calculates the evolution coefficient of the technology unit based on the dominance ratio, correlation strength ratio, and technology maturity difference between the technology units; combining the evolution coefficient with technology development constraints, the technology development constraints including timing constraints of technology development, technology generation span constraints, and technology branch convergence constraints; generating a technology iteration path map based on the evolution coefficient and the technology development constraints, wherein the edge weight of the technology iteration path map is determined by the evolution coefficient ratio of the source technology unit to the target technology unit;

[0014] Calculate the innovation potential energy value of the technology unit using the technology iteration path map, map the innovation potential energy value to a geographic space coordinate system, and construct a technology potential energy distribution field based on spatial aggregation effect, regional correlation, and regional technology foundation;

[0015] The technology association intensity matrix, the technology iteration path map and the technology potential energy distribution field are tensor-fused to construct an initial knowledge graph with multi-dimensional associations; the initial knowledge graph includes a topological structure layer, a technology evolution layer and a potential energy distribution layer of technology units;

[0016] Feature extraction is performed on the initial knowledge graph. The association features of the topological structure layer are extracted based on the graph neural network. The temporal features of the technology evolution layer are extracted based on the recursive neural network. The spatial features of the potential energy distribution layer are extracted based on the convolutional neural network. The association features, the temporal features and the spatial features are fused to form a state feature set of the knowledge graph.

[0017] Deploying intelligent sensors to monitor fluctuations of the state feature set in real time; triggering a graph optimization signal when a monitoring result of the state fluctuation set exceeds a preset dynamic threshold; calling a coupling calculation unit to recalculate the technology association strength matrix based on the graph optimization signal; and optimizing the technology iteration path map using a dynamic optimization algorithm based on the updated technology association strength matrix, including:

[0018] Deploy intelligent sensors to monitor fluctuations in the state feature set, wherein the intelligent sensors collect technology node feature data, technology association feature data, and graph global feature data, calculate feature deviation based on the technology node feature data, calculate coupling change rate based on the technology association feature data, and calculate situation evolution rate based on the graph global feature data;

[0019] Inputting the feature deviation, the coupling change rate and the situation evolution rate into a feature fusion model, the feature fusion model uses a weighted combination method to generate a state feature fluctuation index, and the state feature fluctuation index represents the real-time situation of the knowledge graph;

[0020] Constructing a dynamic threshold calculation model, wherein the dynamic threshold calculation model establishes a baseline threshold based on the historical data distribution of the state characteristic fluctuation index, and dynamically adjusts the baseline threshold according to the technology development cycle function and environmental impact factors to obtain an adaptive dynamic threshold;

[0021] Comparing the state feature fluctuation index with the adaptive dynamic threshold in real time, and when the state feature fluctuation index exceeds the adaptive dynamic threshold, calculating the impact of the fluctuation region, and generating a graph optimization signal including an optimization region, an optimization depth, and an optimization direction based on the impact;

[0022] The graph optimization signal triggers the coupling calculation unit to perform an update operation, wherein the coupling calculation unit obtains the latest technical data, recalculates the technical correlation strength of the corresponding area based on the latest technical data and the optimized area, and generates an updated technical correlation strength matrix;

[0023] The updated technology association intensity matrix is input into the path optimization model. The path optimization model identifies the path nodes that need to be adjusted based on the optimization depth, constructs path evolution constraints according to the optimization direction, and uses a dynamic programming algorithm to obtain the optimal technology iteration path diagram based on the path nodes and the path evolution constraints.

[0024] The graph optimization signal triggers the coupling calculation unit to perform an update operation. The coupling calculation unit obtains the latest technical data, recalculates the technical correlation strength of the corresponding area based on the latest technical data and the optimized area, and generates an updated technical correlation strength matrix, which includes:

[0025] The graph optimization signal triggers the coupling calculation unit to perform an update operation. The coupling calculation unit uses a locality-sensitive hashing algorithm to divide the influence area into multiple sub-areas, calculates the local density and relative distance of technology nodes in the multiple sub-areas, identifies the technology nodes, and constructs an influence propagation network with the technology nodes as the center.

[0026] The coupling calculation unit obtains the latest technology data from a multi-source technology database, performs semantic analysis on the latest technology data using a deep learning model, extracts technology feature vectors containing technology function attributes, application scenario characteristics, and innovation point descriptions, and establishes a temporal feature extraction module to capture dynamic trend information of technology development; the coupling calculation unit calculates the technology correlation strength within the impact area based on the technology feature vectors and the dynamic trend information;

[0027] The coupling calculation unit recalculates the technical correlation strength of the impact area using a multi-level coupling calculation method, and generates an updated technical correlation strength matrix based on the calculation results of the multi-level coupling calculation method.

[0028] The technology potential energy distribution field is recalculated using the optimized technology iteration path map; the updated technology association intensity matrix, the technology iteration path map, and the technology potential energy distribution field are integrated into a new knowledge graph state, and the updated state feature set is output, including:

[0029] Extracting the path node degree centrality, path betweenness centrality, and path cluster cohesion in the optimized technology iteration path graph to construct a technology diffusion matrix, which records the diffusion intensity and coverage of technology knowledge in the path network;

[0030] Recalculate the technology potential energy distribution field based on the path node degree centrality, the path betweenness centrality, the path cluster cohesion and the technology diffusion matrix. The technology potential energy distribution field includes basic potential energy, position potential energy and development potential energy. The basic potential energy is calculated based on the level of technological innovation, the position potential energy is calculated based on the network structure status, and the development potential energy is calculated based on the future evolution trend.

[0031] Performing data standardization on the updated technology correlation intensity matrix, converting the technology iteration path diagram into a standard network representation, converting the technology potential energy distribution field into a unified digital format, and constructing a multi-dimensional data integration framework;

[0032] A vertical mapping relationship and a horizontal connection relationship are established in the multidimensional data integration framework. The vertical mapping relationship determines the correspondence between data at different levels, and the horizontal connection relationship determines the interaction between data at the same level. Common features of multidimensional data are extracted based on the tensor decomposition method. The common features are input into the feature alignment module. The feature alignment module unifies the expression of different dimensions, uses an adaptive weight distribution algorithm to determine the importance of features in each dimension, and integrates the aligned features into a new knowledge graph state.

[0033] Based on the new knowledge graph state, static features and dynamic features are extracted, the static features include structural features, distribution features and attribute features, and the dynamic features include evolution features, flow features and interaction features; the static features and the dynamic features are input into the feature fusion device, and the feature fusion device uses a multi-layer perceptron to combine features, and realizes feature dimensionality reduction through the principal component analysis method to generate an updated state feature set.

[0034] Converting the user's technical requirements into a query state, performing state similarity calculation with the updated state feature set to obtain state similarity; identifying an optimal technology unit combination based on the state similarity; and predicting a state evolution trajectory of the optimal technology unit combination using the technology iteration path diagram, including:

[0035] Receive user technical requirements, use a deep semantic analysis model to extract technical field features, functional requirement features, and performance indicator features in the user technical requirements, and construct a feature representation matrix that quantitatively describes the multidimensional attributes of the requirements;

[0036] Based on the feature representation matrix, a query state vector is constructed through a feature mapping network, wherein the feature mapping network includes a static feature encoding layer and a dynamic feature encoding layer, wherein the static feature encoding layer processes technical attributes and functional attributes, and the dynamic feature encoding layer processes temporal features and evolutionary features;

[0037] Calculate the similarity between the query state vector and the updated state feature set, use the attention mechanism to assign adaptive weights to different feature dimensions, calculate the distance between state vectors through the deep metric learning model, and generate a state similarity matrix;

[0038] The state similarity matrix is used to construct a technology unit affinity graph, where the nodes of the technology unit affinity graph represent technology units and the edge weights represent state similarities. A community discovery algorithm is used to identify highly cohesive technology unit combinations, and the optimal technology unit combination is selected based on a combination scoring function.

[0039] Performing a historical evolution analysis on the optimal technology unit combination, and extracting an evolutionary feature sequence based on the technology iteration path diagram, wherein the evolutionary feature sequence includes changes in technology maturity, innovation activity, and association strength;

[0040] Inputting the evolution feature sequence into a time series prediction model, the time series prediction model adopts a long short-term memory network structure, captures long-term dependencies through a gating mechanism, and combines an attention mechanism to identify time series patterns to generate a technology evolution prediction sequence;

[0041] A multidimensional evolution path diagram is constructed based on the technology evolution prediction sequence. The nodes of the multidimensional evolution path diagram represent the technology status, and the edges represent the state transition probability. The path selection strategy is optimized through the Monte Carlo tree search algorithm to output the optimal evolution trajectory.

[0042] Evaluating the regional adaptability of technology implementation in combination with the technology potential energy distribution field, generating a technology implementation plan, and recording user feedback on the technology implementation plan; converting the feedback information into a state correction factor for adjusting the dynamic threshold to achieve dynamic optimization of the new knowledge graph state, including:

[0043] Assessing the regional suitability of technology implementation based on the technology potential energy distribution field, and generating a technology implementation plan based on the regional suitability, wherein the technology implementation plan includes a technology path planning and a resource allocation plan based on the regional suitability;

[0044] A multimodal feedback collection system is used to record user feedback on the implementation effect of the technical implementation plan, and the multimodal feedback collection system obtains a multidimensional feature feedback dataset, and the multidimensional feature feedback dataset includes text feedback data, voice feedback data, and behavioral feedback data; the multidimensional feature feedback dataset is input into a deep semantic understanding network, and the deep semantic understanding network extracts semantic features and sentiment features to generate a feedback feature vector;

[0045] Constructing a state correction factor based on the feedback feature vector, the state correction factor includes a technology implementation effect score and an optimization direction indication; inputting the state correction factor into a dynamic threshold adjustment module, and the dynamic threshold adjustment module updates the evolution rule trigger threshold of the knowledge graph state according to the technology implementation effect score and the optimization direction indication;

[0046] A graph structure optimization network is used to achieve dynamic optimization of the knowledge graph state. The graph structure optimization network determines the rule triggering conditions according to the dynamic threshold adjustment module and applies the rules that meet the triggering conditions to the structural update of the knowledge graph. A multi-objective optimization framework is constructed. The multi-objective optimization framework uses the state correction factor as an optimization constraint to guide the evolution direction of the knowledge graph state.

[0047] A new knowledge graph state is generated based on the optimization results of the multi-objective optimization framework, and the optimization effect of the new knowledge graph state is fed back to the multimodal feedback acquisition system to form a closed-loop mechanism for continuous optimization.

[0048] According to a second aspect of the embodiments of the present invention,

[0049] Provides an intelligent matching and transaction recommendation system for scientific and technological achievements based on knowledge graphs, including:

[0050] The first unit is used to obtain a scientific and technological achievement data set, extract a technology gene sequence from the scientific and technological achievement data set through a deep feature extractor; input the technology gene sequence into a high-dimensional coding network to generate a technology state vector; construct a technology association intensity matrix based on the technology state vector, and the technology association intensity matrix records the degree of coupling between technology units; calculate the technology evolution coefficient according to the coupling degree to generate a technology iteration path map; use the technology iteration path map to quantitatively calculate the innovation potential energy value of each technology unit to form a technology potential energy distribution field; integrate the technology association intensity matrix, the technology iteration path map, and the technology potential energy distribution field into an initial knowledge graph, and output the state feature set of the initial knowledge graph;

[0051] The second unit is used to deploy intelligent sensors to monitor the fluctuation of the state feature set in real time; when the monitoring result of the state fluctuation set exceeds a preset dynamic threshold, a graph optimization signal is triggered; according to the graph optimization signal, the coupling calculation unit is called to recalculate the technology association strength matrix; based on the updated technology association strength matrix, the dynamic optimization algorithm is used to optimize the technology iteration path map; using the optimized technology iteration path map, the technology potential energy distribution field is recalculated; the updated technology association strength matrix, the technology iteration path map, and the technology potential energy distribution field are integrated into a new knowledge graph state, and an updated state feature set is output;

[0052] The third unit is used to convert the user's technical requirements into a query state, calculate the state similarity with the updated state feature set, and obtain the state similarity; identify the optimal technology unit combination based on the state similarity; use the technology iteration path diagram to predict the state evolution trajectory of the optimal technology unit combination; combine the technology potential energy distribution field to evaluate the regional adaptability of the technology implementation, generate a technology implementation plan, and record the user's feedback information on the technology implementation plan; convert the feedback information into a state correction factor, which is used to adjust the dynamic threshold to achieve dynamic optimization of the new knowledge graph state.

[0053] According to a third aspect of the embodiments of the present invention,

[0054] An electronic device is provided, comprising:

[0055] processor;

[0056] a memory for storing processor-executable instructions;

[0057] The processor is configured to call the instructions stored in the memory to execute the aforementioned method.

[0058] According to a fourth aspect of the embodiments of the present invention,

[0059] A computer-readable storage medium is provided, on which computer program instructions are stored. When the computer program instructions are executed by a processor, the method described above is implemented.

[0060] The beneficial effects of this application are as follows:

[0061] 1. Improve the matching and transaction efficiency of scientific and technological achievements: Through technical means such as technology gene sequence, technology state vector, and technology association intensity matrix, a scientific and technological achievement knowledge graph is constructed, which can more accurately match user technology needs with scientific and technological achievements, reduce information asymmetry, and improve transaction efficiency.

[0062] 2. Accurately recommend technological achievements: Based on state similarity calculation, technology evolution trajectory prediction, and regional adaptability assessment, the system can recommend optimal technology unit combinations and technology implementation plans to users, avoiding blind technology selection and improving the success rate of technological achievement transformation.

[0063] 3. Continuously optimize the knowledge graph and recommendation effects: Utilize user feedback information to dynamically adjust the dynamic threshold of the knowledge graph to achieve continuous optimization of the knowledge graph, thereby continuously improving the accuracy and effectiveness of scientific and technological achievement matching and transaction recommendations. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] Figure 1 This is a flow chart of a method for intelligently matching scientific and technological achievements and recommending transactions based on a knowledge graph according to an embodiment of the present invention;

[0065] Figure 2 This is a structural diagram of the scientific and technological achievements intelligent matching and transaction recommendation system based on the knowledge graph in an embodiment of the present invention. DETAILED DESCRIPTION

[0066] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0067] The following specific embodiments are used to describe the technical solution of the present invention in detail. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described in detail in some embodiments.

[0068] Figure 1 This is a flow chart of a method for intelligent matching and transaction recommendation of scientific and technological achievements based on knowledge graphs according to an embodiment of the present invention. Figure 1 As shown, the method includes:

[0069] S11. Obtain a scientific and technological achievement dataset, extract a technology gene sequence from the scientific and technological achievement dataset through a deep feature extractor; input the technology gene sequence into a high-dimensional encoding network to generate a technology state vector; construct a technology association intensity matrix based on the technology state vector, wherein the technology association intensity matrix records the degree of coupling between technology units; calculate the technology evolution coefficient according to the coupling degree and generate a technology iteration path map; use the technology iteration path map to quantitatively calculate the innovation potential energy value of each technology unit to form a technology potential energy distribution field; integrate the technology association intensity matrix, the technology iteration path map, and the technology potential energy distribution field into an initial knowledge graph, and output a state feature set of the initial knowledge graph;

[0070] S12. Deploy intelligent sensors to monitor the fluctuation of the state feature set in real time; when the monitoring result of the state fluctuation set exceeds a preset dynamic threshold, trigger a graph optimization signal; based on the graph optimization signal, call the coupling calculation unit to recalculate the technology association strength matrix; based on the updated technology association strength matrix, use a dynamic optimization algorithm to optimize the technology iteration path map; use the optimized technology iteration path map to recalculate the technology potential energy distribution field; integrate the updated technology association strength matrix, the technology iteration path map, and the technology potential energy distribution field into a new knowledge graph state, and output an updated state feature set;

[0071] S13. Convert the user's technical requirements into a query state, calculate the state similarity with the updated state feature set, and obtain the state similarity; identify the optimal technology unit combination based on the state similarity; use the technology iteration path diagram to predict the state evolution trajectory of the optimal technology unit combination; combine the technology potential energy distribution field to evaluate the regional adaptability of the technology implementation, generate a technology implementation plan, and record the user's feedback information on the technology implementation plan; convert the feedback information into a state correction factor, which is used to adjust the dynamic threshold to achieve dynamic optimization of the new knowledge graph state.

[0072] In an optional embodiment, the technology evolution coefficient is calculated based on the coupling degree to generate a technology iteration path map; the innovation potential value of each technology unit is quantitatively calculated using the technology iteration path map to form a technology potential energy distribution field; the technology association intensity matrix, the technology iteration path map, and the technology potential energy distribution field are integrated into an initial knowledge graph, and the state feature set of the initial knowledge graph is output, including:

[0073] Inputting the coupling degree into a technology evolution prediction model, the technology evolution prediction model calculates the evolution coefficient of the technology unit based on the dominance ratio, correlation strength ratio, and technology maturity difference between the technology units; combining the evolution coefficient with technology development constraints, the technology development constraints including timing constraints of technology development, technology generation span constraints, and technology branch convergence constraints; generating a technology iteration path map based on the evolution coefficient and the technology development constraints, wherein the edge weight of the technology iteration path map is determined by the evolution coefficient ratio of the source technology unit to the target technology unit;

[0074] Calculate the innovation potential energy value of the technology unit using the technology iteration path map, map the innovation potential energy value to a geographic space coordinate system, and construct a technology potential energy distribution field based on spatial aggregation effect, regional correlation, and regional technology foundation;

[0075] The technology association intensity matrix, the technology iteration path map and the technology potential energy distribution field are tensor-fused to construct an initial knowledge graph with multi-dimensional associations; the initial knowledge graph includes a topological structure layer, a technology evolution layer and a potential energy distribution layer of technology units;

[0076] Feature extraction is performed on the initial knowledge graph. The association features of the topological structure layer are extracted based on the graph neural network. The temporal features of the technology evolution layer are extracted based on the recursive neural network. The spatial features of the potential energy distribution layer are extracted based on the convolutional neural network. The association features, the temporal features and the spatial features are fused to form a state feature set of the knowledge graph.

[0077] First, obtain data on the degree of coupling between technology units. For example, by analyzing patent text data, scientific literature data, or technology collaboration network data, we can determine the frequency of co-occurrence between different technology units or the strength of collaborative relationships, which can be used as a measure of the degree of coupling between technology units. Suppose there are technology units A, B, C, and D. Through data analysis, we find that the degrees of coupling between them are: AB: 0.8, AC: 0.5, AD: 0.2, BC: 0.6, BD: 0.3, and CD: 0.7, respectively. This data will form a technology correlation strength matrix.

[0078] Next, the technology evolution coefficient is calculated based on the degree of coupling and a technology iteration path diagram is generated. The calculation of the technology evolution coefficient requires considering the dominance ratio, association strength ratio, and technology maturity difference between technology units. For example, if technology unit A has a higher dominance, stronger association strength, and higher technology maturity than technology unit B, the evolution coefficient from A to B will be lower; vice versa. Based on the calculated evolution coefficient and technology development constraints (such as the timing constraints of technology development, the technology generation span constraints, and the technology branch convergence constraints), a technology iteration path diagram can be generated. The nodes in the diagram represent technology units, and the edges represent the evolution paths between technology units. The edge weights are determined by the ratio of the evolution coefficients of the source and target technology units. For example, if the evolution coefficient from A to B is 0.2 and the evolution coefficient from B to C is 0.5, the weight of the edge from A to B is 0.2 / 0.5 = 0.4.

[0079] Then, the technology iteration path map is used to calculate the innovation potential energy value of the technology unit and construct the technology potential energy distribution field. The innovation potential energy value of a technology unit can be calculated by analyzing factors such as its position, connection relationship, and evolution coefficient in the technology iteration path map. For example, a technology unit that is located at the center of the technology iteration path map, connected to multiple other technology units, and has a high evolution coefficient generally has a higher innovation potential energy value. The calculated innovation potential energy value is mapped to a geographic coordinate system, and combined with factors such as spatial agglomeration effects, regional correlations, and regional technological foundations, a technology potential energy distribution field can be constructed. For example, if a region gathers multiple technology units with high innovation potential energy values, the overall technology potential energy of the region will also be relatively high.

[0080] Next, the technology association intensity matrix, technology iteration path map, and technology potential energy distribution field are integrated to construct an initial knowledge graph with multi-dimensional associations. This knowledge graph contains the topological structure layer, technology evolution layer, and potential energy distribution layer of technology units, corresponding to the technology association intensity matrix, technology iteration path map, and technology potential energy distribution field, respectively. For example, the topological structure layer information of technology unit A in the knowledge graph includes its connection relationship and connection strength with other technology units, the technology evolution layer information includes its position and evolution coefficient in the technology iteration path map, and the potential energy distribution layer information includes its position in the geographic coordinate system and innovation potential energy value.

[0081] Finally, feature extraction is performed on the initial knowledge graph to form the state feature set of the knowledge graph. A graph neural network is used to extract correlation features at the topological structure layer, a recurrent neural network is used to extract temporal features at the technological evolution layer, and a convolutional neural network is used to extract spatial features at the potential energy distribution layer. These features are then fused to form the state feature set of the knowledge graph, which is used for subsequent technological prediction and analysis. For example, the extracted state feature set can be input into a machine learning model to predict future technological development trends or identify technological fields with high innovation potential.

[0082] The solution of this application can:

[0083] Improving the accuracy of technology forecasts: By constructing a multi-dimensional, interconnected knowledge graph, we can more comprehensively understand the relationships and evolutionary patterns among technology units, thereby improving the accuracy of technology forecasts. Identifying technology fields with high innovation potential: By analyzing the distribution of technology potential energy, we can identify technology units and regions with high innovation potential, providing decision support for technology R&D and investment. Promoting technological innovation and development: By constructing a technology iteration roadmap, we can better understand the direction and path of technological evolution, thereby promoting technological innovation and development.

[0084] In an optional embodiment, an intelligent sensor is deployed to monitor the fluctuation of the state feature set in real time; when the monitoring result of the state fluctuation set exceeds a preset dynamic threshold, a graph optimization signal is triggered; based on the graph optimization signal, a coupling calculation unit is called to recalculate the technology association strength matrix; based on the updated technology association strength matrix, a dynamic optimization algorithm is used to optimize the technology iteration path map, including:

[0085] Deploy intelligent sensors to monitor fluctuations in the state feature set, wherein the intelligent sensors collect technology node feature data, technology association feature data, and graph global feature data, calculate feature deviation based on the technology node feature data, calculate coupling change rate based on the technology association feature data, and calculate situation evolution rate based on the graph global feature data;

[0086] Inputting the feature deviation, the coupling change rate and the situation evolution rate into a feature fusion model, the feature fusion model uses a weighted combination method to generate a state feature fluctuation index, and the state feature fluctuation index represents the real-time situation of the knowledge graph;

[0087] Constructing a dynamic threshold calculation model, wherein the dynamic threshold calculation model establishes a baseline threshold based on the historical data distribution of the state characteristic fluctuation index, and dynamically adjusts the baseline threshold according to the technology development cycle function and environmental impact factors to obtain an adaptive dynamic threshold;

[0088] Comparing the state feature fluctuation index with the adaptive dynamic threshold in real time, and when the state feature fluctuation index exceeds the adaptive dynamic threshold, calculating the impact of the fluctuation region, and generating a graph optimization signal including an optimization region, an optimization depth, and an optimization direction based on the impact;

[0089] The graph optimization signal triggers the coupling calculation unit to perform an update operation, wherein the coupling calculation unit obtains the latest technical data, recalculates the technical correlation strength of the corresponding area based on the latest technical data and the optimized area, and generates an updated technical correlation strength matrix;

[0090] The updated technology association intensity matrix is input into the path optimization model. The path optimization model identifies the path nodes that need to be adjusted based on the optimization depth, constructs path evolution constraints according to the optimization direction, and uses a dynamic programming algorithm to obtain the optimal technology iteration path diagram based on the path nodes and the path evolution constraints.

[0091] First, deploy intelligent sensors to monitor fluctuations in the technology landscape feature set in real time. These sensors collect data from multiple sources, including but not limited to patent databases, scientific literature databases, industry reports, and news information. The collected data covers three aspects: technology node feature data, such as technology maturity, market share, and development trends; technology association feature data, such as the cooperative, competitive, and interdependent relationships between technologies; and global graph feature data, such as the overall level of technology development, the distribution of hot topics, and the emergence of emerging technologies.

[0092] The intelligent sensor then extracts and calculates features based on the collected data. For example, the feature deviation is calculated based on the feature data of the technology node to measure the degree of deviation between the development status of each technology node and its expected status. Assuming that the expected maturity of a technology node is 0.8 and the actual maturity is 0.6, its feature deviation can be calculated as 0.2. The coupling change rate is calculated based on the technology association feature data to measure the speed of change of the association relationship between technologies. For example, if the cooperative relationship between two technologies is significantly strengthened over a period of time, their coupling change rate will increase accordingly. The situation evolution rate is calculated based on the global feature data of the graph to measure the speed of change of the overall technology situation. For example, if the speed of technological development in a certain field suddenly accelerates, the situation evolution rate in that field will increase accordingly.

[0093] Next, the calculated feature deviation, coupling change rate, and situation evolution rate are input into the feature fusion model. This model uses a weighted combination approach to fuse these three indicators into a single state feature fluctuation index, which is used to characterize the real-time state of the knowledge graph. For example, based on the experience of domain experts or historical data analysis, the three indicators can be assigned different weights and summed to obtain the state feature fluctuation index. Assuming that the weights of feature deviation, coupling change rate, and situation evolution rate are 0.4, 0.3, and 0.3, respectively, and their values are 0.2, 0.5, and 0.4, respectively, the state feature fluctuation index is 0.2*0.4+0.5*0.3+0.4*0.3=0.35.

[0094] To determine whether the situation has changed to the point where path optimization is necessary, a dynamic threshold calculation model is constructed. This model first establishes a baseline threshold based on the historical data distribution of the state characteristic fluctuation indicator. For example, the mean and standard deviation of the state characteristic fluctuation indicator over a period of time can be calculated, and the mean plus twice the standard deviation can be used as the baseline threshold. The baseline threshold is then dynamically adjusted based on the technology development cycle function and environmental impact factors to obtain an adaptive dynamic threshold. For example, during periods of rapid technological development, the threshold can be appropriately lowered to more sensitively capture situational changes; during periods of stable technological development, the threshold can be appropriately raised to avoid frequent path adjustments. Assuming the baseline threshold is 0.4, the current period is rapid technological development, and the environmental impact factor is 0.9, the adaptive dynamic threshold can be calculated as 0.4 * 0.9 = 0.36.

[0095] The state feature fluctuation index is compared with the adaptive dynamic threshold in real time. When the state feature fluctuation index exceeds the adaptive dynamic threshold, the impact of the area where the fluctuation occurs is calculated. For example, the number of technical nodes involved in the fluctuation area, their importance, and the correlation between them can be analyzed to evaluate the impact. Based on the impact, a graph optimization signal containing the optimization area, optimization depth, and optimization direction is generated. For example, if the impact is large, the optimization depth may need to be deeper, and the optimization direction may need to be adjusted more significantly. Assuming that the state feature fluctuation index is 0.37, which exceeds the adaptive dynamic threshold of 0.36, and the fluctuation is mainly concentrated in the field of artificial intelligence, the impact level is assessed to be high. Then the generated graph optimization signal may be: the optimization area is the field of artificial intelligence, the optimization depth is high, and the optimization direction is to strengthen research in the direction of deep learning technology.

[0096] The graph optimization signal triggers the coupling calculation unit to perform an update operation. The coupling calculation unit obtains the latest technical data and recalculates the technical correlation strength of the corresponding area based on the latest technical data and the optimized area, generating an updated technical correlation strength matrix. For example, if the technical correlation strength in the field of artificial intelligence needs to be updated, the coupling calculation unit will collect the latest data on the relationships between various technologies in the field of artificial intelligence and recalculate the correlation strength between them.

[0097] Finally, the updated technology correlation strength matrix is input into the path optimization model. The path optimization model identifies path nodes requiring adjustment based on the optimization depth and establishes path evolution constraints based on the optimization direction. For example, if the optimization depth is high, more path nodes may require adjustment; if the optimization direction is to strengthen deep learning technology, technology nodes related to deep learning should be prioritized. Based on the path nodes and path evolution constraints, a dynamic programming algorithm is used to obtain the optimal technology iteration path diagram. For example, a dynamic programming algorithm can be used to find the optimal path from the current technology state to the target technology state while satisfying the constraints.

[0098] The solution of this application can:

[0099] Improve technology R&D efficiency: By monitoring technology trends in real time and dynamically adjusting technology iteration paths, we can avoid wasting resources in the wrong direction, thereby improving the efficiency of technology R&D. Improve the success rate of technology R&D: By adaptively adjusting paths based on changes in technology trends, we can better adapt to technology development trends and market demands, thereby improving the success rate of technology R&D. Enhance the flexibility of technology path planning: The combination of dynamic thresholds and path optimization algorithms makes technology path planning more flexible and better able to cope with the complex and changing technology development environment.

[0100] In an optional embodiment, the graph optimization signal triggers the coupling calculation unit to perform an update operation, the coupling calculation unit obtains the latest technical data, and recalculates the technical correlation strength of the corresponding area based on the latest technical data and the optimized area, and generates an updated technical correlation strength matrix including:

[0101] The graph optimization signal triggers the coupling calculation unit to perform an update operation. The coupling calculation unit uses a locality-sensitive hashing algorithm to divide the influence area into multiple sub-areas, calculates the local density and relative distance of technology nodes in the multiple sub-areas, identifies the technology nodes, and constructs an influence propagation network with the technology nodes as the center.

[0102] The coupling calculation unit obtains the latest technology data from a multi-source technology database, performs semantic analysis on the latest technology data using a deep learning model, extracts technology feature vectors containing technology function attributes, application scenario characteristics, and innovation point descriptions, and establishes a temporal feature extraction module to capture dynamic trend information of technology development; the coupling calculation unit calculates the technology correlation strength within the impact area based on the technology feature vectors and the dynamic trend information;

[0103] The coupling calculation unit recalculates the technical correlation strength of the impact area using a multi-level coupling calculation method, and generates an updated technical correlation strength matrix based on the calculation results of the multi-level coupling calculation method.

[0104] First, the system receives a graph optimization signal, triggering the coupled computing unit to perform an update operation. For example, the system can generate a graph optimization signal periodically (e.g., daily or weekly), or trigger an update operation based on a user request or a specific event.

[0105] When the coupled computing unit receives a graph optimization signal, it first determines the impact area that needs to be updated. This impact area can be the entire technology graph or a specific portion of the graph, such as a specific technology field or technology cluster. For example, if rapid technological development in artificial intelligence is detected, AI can be designated as the impact area.

[0106] Next, the coupled computing unit uses a locality-sensitive hashing algorithm to divide the impact area into multiple sub-areas. For example, the impact area can be divided into different sub-areas based on the application scenario or functional attributes of the technology. For example, if the impact area is in the field of artificial intelligence, it can be divided into sub-areas such as computer vision, natural language processing, and machine learning.

[0107] The coupled computation unit then calculates the local density and relative distance of technology nodes in each subregion. Local density refers to the number of related technology nodes surrounding a given technology node. Relative distance refers to the semantic similarity or functional relevance between two technology nodes. By calculating local density and relative distance, representative technology nodes can be identified in each subregion. For example, in the computer vision subregion, "convolutional neural network" may be a representative technology node.

[0108] With these representative technology nodes at the center, coupled computing units are used to construct an influence propagation network. This network describes the mutual influence and correlation between different technology nodes. For example, "convolutional neural network" may be associated with other technology nodes such as "image recognition" and "object detection."

[0109] At the same time, the coupled computing unit obtains the latest technical data from multiple source technical databases (e.g., patent databases, scientific literature databases, news information websites, etc.) These data may include technical papers, patent specifications, news reports, etc.

[0110] After acquiring the latest technical data, the coupled computing unit uses deep learning models (e.g., BERT, GPT, etc.) to perform semantic analysis on this data, extracting a technical feature vector that contains technical functional attributes, application scenario characteristics, and descriptions of innovations. For example, for "autonomous driving" technology, the extracted technical feature vector may include features such as "vehicle control," "environmental perception," and "path planning." Assuming the latest technical data includes a paper on "autonomous driving technology based on deep learning," technical features such as "deep learning," "vehicle control," and "environmental perception" can be extracted.

[0111] Furthermore, the coupled computing unit also establishes a time series feature extraction module to capture dynamic trends in technological development. For example, it can count indicators such as the frequency of appearance and number of citations of a technology over a period of time to reflect its development trend. For example, if the number of mentions of "deep learning" technology has increased significantly over the past year, it can be considered a technological development trend.

[0112] Based on the extracted technology feature vectors and dynamic trend information, the coupling calculation unit uses a multi-level coupling calculation method to recalculate the technology correlation strength within the impact area. This multi-level coupling calculation method can consider factors at different levels, such as semantic similarity, functional relevance, and development trends between technology nodes.

[0113] Finally, the coupling calculation unit generates an updated technology correlation strength matrix based on the results of the multi-level coupling calculation method. This matrix reflects the correlation strength between different technology nodes. For example, if the correlation strength between "deep learning" and "autonomous driving" is high, it indicates that there is a close connection between the two technologies.

[0114] The solution of this application can:

[0115] High degree of automation: The present invention can automatically obtain the latest technical data from multi-source technical databases and use deep learning models to perform semantic analysis, thereby realizing automatic updates of technical correlation strengths, greatly reducing the cost and workload of manual intervention. High accuracy: The present invention uses deep learning models to perform semantic analysis on technical data, which can more accurately capture the correlation between technologies, thereby improving the accuracy of technical correlation strength calculations. Strong dynamism: The present invention can capture dynamic trend information of technological development and integrate it into the calculation of technological correlation strength, so that the technology map can timely reflect the latest developments in technological development.

[0116] In an optional embodiment, the optimized technology iteration path map is used to recalculate the technology potential energy distribution field; the updated technology association intensity matrix, the technology iteration path map, and the technology potential energy distribution field are integrated into a new knowledge graph state, and the updated state feature set is output, including:

[0117] Extracting the path node degree centrality, path betweenness centrality, and path cluster cohesion in the optimized technology iteration path graph to construct a technology diffusion matrix, which records the diffusion intensity and coverage of technology knowledge in the path network;

[0118] Recalculate the technology potential energy distribution field based on the path node degree centrality, the path betweenness centrality, the path cluster cohesion and the technology diffusion matrix. The technology potential energy distribution field includes basic potential energy, position potential energy and development potential energy. The basic potential energy is calculated based on the level of technological innovation, the position potential energy is calculated based on the network structure status, and the development potential energy is calculated based on the future evolution trend.

[0119] Performing data standardization on the updated technology correlation intensity matrix, converting the technology iteration path diagram into a standard network representation, converting the technology potential energy distribution field into a unified digital format, and constructing a multi-dimensional data integration framework;

[0120] A vertical mapping relationship and a horizontal connection relationship are established in the multidimensional data integration framework. The vertical mapping relationship determines the correspondence between data at different levels, and the horizontal connection relationship determines the interaction between data at the same level. Common features of multidimensional data are extracted based on the tensor decomposition method. The common features are input into the feature alignment module. The feature alignment module unifies the expression of different dimensions, uses an adaptive weight distribution algorithm to determine the importance of features in each dimension, and integrates the aligned features into a new knowledge graph state.

[0121] Based on the new knowledge graph state, static features and dynamic features are extracted, the static features include structural features, distribution features and attribute features, and the dynamic features include evolution features, flow features and interaction features; the static features and the dynamic features are input into the feature fusion device, and the feature fusion device uses a multi-layer perceptron to combine features, and realizes feature dimensionality reduction through the principal component analysis method to generate an updated state feature set.

[0122] First, obtain a technology iteration path map. This describes the evolutionary relationships and development sequence between different technologies. For example, a technology iteration path map can be constructed based on patent data, literature data, industry reports, and other data. A node represents a technology, and an edge represents the evolutionary relationship between technologies. For a simplified example, consider three technologies, A, B, and C. If A evolves to B, and B evolves to C, then the path is A->B->C.

[0123] Next, calculate the technology association strength matrix. This matrix reflects the degree of association between different technologies. For example, technology association strength can be calculated based on technology co-occurrence, citation relationships, semantic similarity, and other factors. Example: Assume that the association strength between A and B is 0.8, the association strength between B and C is 0.7, and the association strength between A and C is 0.5. The matrix can be represented as [[1, 0.8, 0.5], [0.8, 1, 0.7], [0.5, 0.7, 1]].

[0124] Next, calculate the technology potential distribution field. The technology potential distribution field includes basic potential, positional potential, and development potential. Basic potential is calculated based on the level of technological innovation, such as the number of patents and citations. Positional potential is calculated based on the position of the network structure, such as the degree centrality and betweenness centrality of the node. Development potential is calculated based on future evolutionary trends, such as the speed of technological development and market demand. Example: Technology A has a basic potential of 10, a positional potential of 5, and a development potential of 8; Technology B has a basic potential of 8, a positional potential of 7, and a development potential of 9; Technology C has a basic potential of 6, a positional potential of 9, and a development potential of 7.

[0125] Recalculate the technology potential energy distribution field based on the optimized technology iteration path map. For example, consider technology development trends and market demand changes and adjust the development potential energy of technology C to 10.

[0126] The extraction technique iterates the path graph's node degree centrality, path betweenness centrality, and path cluster cohesion. Node degree centrality indicates the number of edges connecting a node, betweenness centrality indicates the number of times a node appears on the shortest path in the network, and cluster cohesion indicates the closeness of the groups to which nodes belong. Example: In the path A->B->C, B has the highest betweenness centrality.

[0127] Construct a technology diffusion matrix. The technology diffusion matrix records the diffusion intensity and coverage of technological knowledge in a path network. For example, the technology diffusion matrix can be calculated based on path length, node centrality, and other factors. Example: Assuming that the influence of A on B is 0.8, and the influence of B on C is 0.7, the diffusion matrix can be expressed as [[1, 0.8, 0.56], [0, 1, 0.7], [0, 0, 1]] (0.56 = 0.8 * 0.7).

[0128] Recalculate the technology potential distribution field based on the above indicators. For example, multiply the position potential and development potential by the degree centrality of the corresponding node to obtain the new position potential and development potential.

[0129] Normalize the updated technology correlation strength matrix, for example, scaling all values to between 0 and 1. Convert the technology iteration path diagram to a standard network representation, such as an adjacency matrix. Convert the technology potential energy distribution field to a unified digital format.

[0130] Build a multidimensional data integration framework. Establish vertical mapping relationships and horizontal linkages within the framework. For example, consider the technology linkage intensity matrix, technology iteration path map, and technology potential energy distribution field as different dimensions of the framework. Vertical mapping relationships determine how data at different levels correspond, such as linking different dimensions of data for the same technology. Horizontal linkage relationships determine how data at the same level interacts, such as determining the strength of linkages between different technologies.

[0131] Based on tensor decomposition, the common features of multidimensional data are extracted. These common features are input into the feature alignment module to unify the representation of different dimensions. An adaptive weight assignment algorithm is used to determine the importance of features in each dimension. The aligned features are then integrated into a new knowledge graph state.

[0132] Based on the new knowledge graph state, static and dynamic features are extracted. Static features include structural features (such as network density), distribution features (such as potential energy distribution), and attribute features (such as technology categories). Dynamic features include evolutionary features (such as the speed of technology development), flow features (such as the speed of technology diffusion), and interaction features (such as technological cooperation relationships).

[0133] The static and dynamic features are fed into a feature fusion machine, such as a multi-layer perceptron, for feature combination. Principal component analysis is used to reduce the feature dimensionality and generate an updated state feature set.

[0134] The solution of this application can:

[0135] Improve the accuracy of technology forecasts: The present invention integrates multi-dimensional data to construct a more comprehensive knowledge graph, which can more accurately capture the laws of technology development, thereby improving the accuracy of technology forecasts. Enhance the interpretability of technology forecasts: The present invention can reveal the inherent driving factors and evolutionary mechanisms of technology development and enhance the interpretability of technology forecasts by analyzing technology iteration paths and potential energy distribution fields. Support technology planning and decision-making: The technology forecast results provided by the present invention can provide a scientific basis for technology planning, investment decisions, etc., and help enterprises and governments better grasp the direction of technology development.

[0136] In an optional embodiment, converting the user's technical requirements into a query state, calculating state similarity with the updated state feature set to obtain state similarity; identifying an optimal technology unit combination based on the state similarity; and predicting a state evolution trajectory of the optimal technology unit combination using the technology iteration path diagram includes:

[0137] Receive user technical requirements, use a deep semantic analysis model to extract technical field features, functional requirement features, and performance indicator features in the user technical requirements, and construct a feature representation matrix that quantitatively describes the multidimensional attributes of the requirements;

[0138] Based on the feature representation matrix, a query state vector is constructed through a feature mapping network, wherein the feature mapping network includes a static feature encoding layer and a dynamic feature encoding layer, wherein the static feature encoding layer processes technical attributes and functional attributes, and the dynamic feature encoding layer processes temporal features and evolutionary features;

[0139] Calculate the similarity between the query state vector and the updated state feature set, use the attention mechanism to assign adaptive weights to different feature dimensions, calculate the distance between state vectors through the deep metric learning model, and generate a state similarity matrix;

[0140] The state similarity matrix is used to construct a technology unit affinity graph, where the nodes of the technology unit affinity graph represent technology units and the edge weights represent state similarities. A community discovery algorithm is used to identify highly cohesive technology unit combinations, and the optimal technology unit combination is selected based on a combination scoring function.

[0141] Performing a historical evolution analysis on the optimal technology unit combination, and extracting an evolutionary feature sequence based on the technology iteration path diagram, wherein the evolutionary feature sequence includes changes in technology maturity, innovation activity, and association strength;

[0142] Inputting the evolution feature sequence into a time series prediction model, the time series prediction model adopts a long short-term memory network structure, captures long-term dependencies through a gating mechanism, and combines an attention mechanism to identify time series patterns to generate a technology evolution prediction sequence;

[0143] A multidimensional evolution path diagram is constructed based on the technology evolution prediction sequence. The nodes of the multidimensional evolution path diagram represent the technology status, and the edges represent the state transition probability. The path selection strategy is optimized through the Monte Carlo tree search algorithm to output the optimal evolution trajectory.

[0144] First, the system receives a technical requirement text submitted by the user. For example, a user's requirement is to "develop a voice recognition system for smart homes, with a recognition accuracy of over 95% and a response speed of less than 1 second."

[0145] Next, the system uses a deep semantic analysis model to extract features from the user's technical requirements. For example, for the requirement "Develop a voice recognition system for smart homes, requiring recognition accuracy of at least 95% and a response speed of less than 1 second," the system extracts the technical domain features "voice recognition" and "smart home," the functional requirement features "voice recognition" and "control of smart home devices," and the performance indicator features "recognition accuracy > 95%" and "response speed < 1 second." The system then constructs these features into a feature representation matrix. Assuming this matrix contains four dimensions: technical domain, functional requirement, recognition accuracy, and response speed, the feature representation matrix for this requirement can be expressed as: [voice recognition, smart home, voice recognition, control of smart home devices, 0.95, 1], where 0.95 and 1 represent the numerical values of recognition accuracy and response speed, respectively.

[0146] The system then inputs the feature representation matrix into the feature mapping network to generate a query state vector. The feature mapping network consists of a static feature encoding layer and a dynamic feature encoding layer. The static feature encoding layer processes technical and functional attributes, such as encoding "speech recognition" and "smart home" into specific vector representations. The dynamic feature encoding layer processes temporal and evolutionary features. Since current requirements lack clear temporal and evolutionary features, the dynamic feature encoding layer can be supplemented based on current technological trends, such as the latest advances in speech recognition technology and the future development direction of the smart home market. Suppose the static feature encoding layer encodes "speech recognition" as [0.1, 0.2, 0.7], "smart home" as [0.8, 0.1, 0.1], "speech recognition" as [0.1, 0.2, 0.7], and "controlling smart home devices" as [0.5, 0.3, 0.2]. The dynamic feature encoding layer adds two additional dimensions based on current technological trends, such as market growth rate and technology maturity, encoding them as [0.9, 0.8]. The final query state vector generated is [0.1, 0.2, 0.7, 0.8, 0.1, 0.1, 0.1, 0.2, 0.7, 0.5, 0.3, 0.2, 0.9, 0.8].

[0147] The system calculates the similarity between the generated query state vector and the updated state feature set. Assuming the state feature set contains state vectors for multiple technology units, the system uses an attention mechanism to assign adaptive weights to different feature dimensions. A deep metric learning model is then used to calculate the distance between the query state vector and the state vector of each technology unit, ultimately generating a state similarity matrix. For example, assuming there are two technology units A and B in the state feature set, with state vectors of [0.1, 0.3, 0.6, ...] and [0.2, 0.2, 0.6, ...], respectively, the calculated state similarity matrix is [[1, 0.8], [0.8, 1]], where the diagonal elements are 1, indicating a self-similarity of 1, and the remaining elements represent the similarities between different technology units.

[0148] The system uses the state similarity matrix to construct an affinity graph for technology units. The nodes in the graph represent technology units, and the edge weights represent state similarities. For example, based on the aforementioned state similarity matrix, an affinity graph can be constructed containing two nodes, A and B, with an edge weight of 0.8 between nodes A and B. The system then uses a community discovery algorithm to identify highly cohesive combinations of technology units. In this example, since there are only two technology units, they form a combination. The system then selects the optimal combination of technology units based on a combination scoring function. Assuming that the scoring function takes into account factors such as the performance, cost, and reliability of the technology units, the combination of A and B is ultimately selected as the optimal combination of technology units.

[0149] The system analyzes the historical evolution of the optimal technology unit combination and extracts a sequence of evolutionary characteristics based on the technology iteration path diagram, such as changes in technology maturity, innovation activity, and correlation strength. Suppose the historical evolution data for technology units A and B shows that their technology maturity is constantly improving, their innovation activity remains stable, and their correlation strength is gradually increasing.

[0150] The system inputs the evolutionary feature sequence into a time series prediction model, such as a long short-term memory network, to predict the trend of technological evolution. Assume that the prediction results show that the technological maturity of technology units A and B will continue to improve, the innovation activity will increase, and the correlation strength will further increase.

[0151] Finally, the system constructs a multidimensional evolutionary path diagram based on the technology evolution prediction sequence and optimizes the path selection strategy using a Monte Carlo tree search algorithm, outputting the optimal evolutionary trajectory. For example, the prediction results may show that technology units A and B will merge into a new technology unit C, or they will develop into more advanced technology units A' and B' respectively.

[0152] The solution of this application can:

[0153] Improving the accuracy of technology forecasts: Through deep semantic analysis and deep metric learning, we more accurately understand user needs and identify matching technology units, thereby improving the accuracy of technology forecasts. Optimizing technology portfolio solutions: Through state similarity calculation and community discovery algorithms, we identify highly cohesive technology unit combinations and screen the optimal combination based on a combination scoring function to optimize technology portfolio solutions. Predicting technology evolution trajectories: Through historical evolution analysis and time series prediction models, we predict the future development trends of technology units and construct a multi-dimensional evolution path map to provide reference for technology planning and decision-making.

[0154] In an optional embodiment, the regional adaptability of the technology implementation is evaluated in combination with the technology potential energy distribution field, a technology implementation plan is generated, and user feedback on the technology implementation plan is recorded; the feedback information is converted into a state correction factor for adjusting the dynamic threshold to achieve dynamic optimization of the new knowledge graph state, including:

[0155] Assessing the regional suitability of technology implementation based on the technology potential energy distribution field, and generating a technology implementation plan based on the regional suitability, wherein the technology implementation plan includes a technology path planning and a resource allocation plan based on the regional suitability;

[0156] A multimodal feedback collection system is used to record user feedback on the implementation effect of the technical implementation plan, and the multimodal feedback collection system obtains a multidimensional feature feedback dataset, and the multidimensional feature feedback dataset includes text feedback data, voice feedback data, and behavioral feedback data; the multidimensional feature feedback dataset is input into a deep semantic understanding network, and the deep semantic understanding network extracts semantic features and sentiment features to generate a feedback feature vector;

[0157] Constructing a state correction factor based on the feedback feature vector, the state correction factor includes a technology implementation effect score and an optimization direction indication; inputting the state correction factor into a dynamic threshold adjustment module, and the dynamic threshold adjustment module updates the evolution rule trigger threshold of the knowledge graph state according to the technology implementation effect score and the optimization direction indication;

[0158] A graph structure optimization network is used to achieve dynamic optimization of the knowledge graph state. The graph structure optimization network determines the rule triggering conditions according to the dynamic threshold adjustment module and applies the rules that meet the triggering conditions to the structural update of the knowledge graph. A multi-objective optimization framework is constructed. The multi-objective optimization framework uses the state correction factor as an optimization constraint to guide the evolution direction of the knowledge graph state.

[0159] A new knowledge graph state is generated based on the optimization results of the multi-objective optimization framework, and the optimization effect of the new knowledge graph state is fed back to the multimodal feedback acquisition system to form a closed-loop mechanism for continuous optimization.

[0160] First, evaluate the regional adaptability of technology implementation. For example, assuming that a knowledge graph about "smart home" is to be constructed, it is necessary to evaluate the applicability of different technologies (e.g., natural language processing, knowledge representation learning, etc.) in different areas (e.g., smart home device control, home environment monitoring, etc.). The adaptability score of each technology in each area can be obtained through expert scoring, data analysis, and other methods to form a technology potential energy distribution field. For example, the adaptability score of natural language processing technology in the smart home device control area is 0.9, and the adaptability score in the home environment monitoring area is 0.7.

[0161] A technical implementation plan is generated based on the regional suitability. This includes technical path planning and resource allocation plans. For example, based on the aforementioned suitability scores, natural language processing technology is prioritized in the smart home device control area, and corresponding computing and human resources are allocated.

[0162] Next, a multimodal feedback collection system is used to record user feedback. This system can collect text feedback (e.g., user comments when using smart home apps), voice feedback (e.g., instructions when controlling devices through voice assistants), and behavioral feedback (e.g., user habits when using smart home devices). For example, if a user feedbacks "I hope to be able to control the color of the lights by voice," this feedback will be recorded.

[0163] The multi-dimensional feature feedback dataset is fed into a deep semantic understanding network. This network extracts semantic and sentiment features and generates a feedback feature vector. For example, the semantic features of the user feedback above can be represented as "voice control," "light," and "color," while the sentiment feature can be represented as "expectation."

[0164] A state correction factor is constructed based on the feedback feature vector. This factor includes a technical implementation effectiveness score and optimization direction. For example, based on the feedback feature vector, it can be determined that users have a strong demand for voice control of lighting color, resulting in a high technical implementation effectiveness score and a direction for optimization to enhance voice control.

[0165] The state correction factor is input into the dynamic threshold adjustment module. This module updates the trigger threshold of the knowledge graph state evolution rules based on the technical implementation effect score and optimization direction. For example, if the technical implementation effect score is high, the trigger threshold of the corresponding rule is lowered, making it more likely to be triggered, thereby accelerating the evolution of the knowledge graph.

[0166] A graph structure optimization network is used to dynamically optimize the state of the knowledge graph. This network determines the triggering conditions for rules based on a dynamic threshold adjustment module and applies the rules that meet the triggering conditions to update the structure of the knowledge graph. For example, if the triggering conditions for the "add new entity" rule are met, "light color" is added to the knowledge graph.

[0167] A multi-objective optimization framework is constructed, using state correction factors as optimization constraints to guide the evolution of the knowledge graph state. For example, the optimization direction indicated by user feedback can be used as one of the optimization objectives to ensure that the evolution direction of the knowledge graph is consistent with user needs.

[0168] Finally, a new knowledge graph state is generated based on the optimization results of the multi-objective optimization framework. The optimization results of the new knowledge graph state are fed back to the multimodal feedback collection system, forming a closed-loop mechanism for continuous optimization. For example, the updated knowledge graph can be applied to a smart home system, and user feedback on the new features can be collected to further optimize the knowledge graph.

[0169] The solution of this application can:

[0170] Improving the quality of the knowledge graph: By driving the dynamic evolution of the knowledge graph through user feedback, we can continuously improve the content and structure of the knowledge graph, making it more accurate, complete, and more aligned with user needs. Enhancing the user experience: By continuously optimizing the knowledge graph, we can improve the performance and user experience of applications such as smart homes, for example, by enabling more precise voice control and smarter scenario recommendations. Personalizing services: By analyzing user feedback, we can understand users' individual needs and provide more personalized services, such as customizing smart home scenarios based on their lifestyle habits.

[0171] Figure 2 This is a schematic diagram of the structure of the scientific and technological achievements intelligent matching and transaction recommendation system based on the knowledge graph according to an embodiment of the present invention. Figure 2 As shown, the system includes:

[0172] The first unit is used to obtain a scientific and technological achievement data set, extract a technology gene sequence from the scientific and technological achievement data set through a deep feature extractor; input the technology gene sequence into a high-dimensional coding network to generate a technology state vector; construct a technology association intensity matrix based on the technology state vector, and the technology association intensity matrix records the degree of coupling between technology units; calculate the technology evolution coefficient according to the coupling degree to generate a technology iteration path map; use the technology iteration path map to quantitatively calculate the innovation potential energy value of each technology unit to form a technology potential energy distribution field; integrate the technology association intensity matrix, the technology iteration path map, and the technology potential energy distribution field into an initial knowledge graph, and output the state feature set of the initial knowledge graph;

[0173] The second unit is used to deploy intelligent sensors to monitor the fluctuation of the state feature set in real time; when the monitoring result of the state fluctuation set exceeds a preset dynamic threshold, a graph optimization signal is triggered; according to the graph optimization signal, the coupling calculation unit is called to recalculate the technology association strength matrix; based on the updated technology association strength matrix, the dynamic optimization algorithm is used to optimize the technology iteration path map; using the optimized technology iteration path map, the technology potential energy distribution field is recalculated; the updated technology association strength matrix, the technology iteration path map, and the technology potential energy distribution field are integrated into a new knowledge graph state, and an updated state feature set is output;

[0174] The third unit is used to convert the user's technical requirements into a query state, calculate the state similarity with the updated state feature set, and obtain the state similarity; identify the optimal technology unit combination based on the state similarity; use the technology iteration path diagram to predict the state evolution trajectory of the optimal technology unit combination; combine the technology potential energy distribution field to evaluate the regional adaptability of the technology implementation, generate a technology implementation plan, and record the user's feedback information on the technology implementation plan; convert the feedback information into a state correction factor, which is used to adjust the dynamic threshold to achieve dynamic optimization of the new knowledge graph state.

[0175] According to a third aspect of the embodiments of the present invention,

[0176] An electronic device is provided, comprising:

[0177] processor;

[0178] a memory for storing processor-executable instructions;

[0179] The processor is configured to call the instructions stored in the memory to execute the aforementioned method.

[0180] According to a fourth aspect of the embodiments of the present invention,

[0181] A computer-readable storage medium is provided, on which computer program instructions are stored. When the computer program instructions are executed by a processor, the method described above is implemented.

[0182] The present invention may be a method, an apparatus, a system and / or a computer program product. The computer program product may include a computer-readable storage medium carrying computer-readable program instructions for executing various aspects of the present invention.

[0183] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the above embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. The method for intelligent matching and transaction recommendation of scientific and technological achievements based on knowledge graph is characterized by: include: Acquire a scientific and technological achievement dataset, and extract a technical gene sequence from the scientific and technological achievement dataset using a deep feature extractor; Inputting the technology gene sequence into a high-dimensional coding network to generate a technology state vector; constructing a technology association strength matrix based on the technology state vector, wherein the technology association strength matrix records the degree of coupling between technology units; Calculating the technology evolution coefficient based on the coupling degree to generate a technology iteration path map; using the technology iteration path map to quantitatively calculate the innovation potential energy value of each technology unit to form a technology potential energy distribution field; Integrate the technology association intensity matrix, the technology iteration path diagram, and the technology potential energy distribution field into an initial knowledge graph, and output a state feature set of the initial knowledge graph; Deploying intelligent sensors to monitor fluctuations of the state feature set in real time; When the monitoring result of the state fluctuation set exceeds the preset dynamic threshold, the graph optimization signal is triggered; Calling a coupling calculation unit to recalculate the technology correlation strength matrix according to the graph optimization signal; Based on the updated technology association intensity matrix, a dynamic optimization algorithm is used to optimize the technology iteration path map; and the technology potential energy distribution field is recalculated using the optimized technology iteration path map. Integrate the updated technology association intensity matrix, the technology iteration path map, and the technology potential energy distribution field into a new knowledge graph state, and output an updated state feature set; The user's technical requirements are converted into query states, and the state similarity is calculated with the updated state feature set to obtain the state similarity; the optimal technology unit combination is identified based on the state similarity; the state evolution trajectory of the optimal technology unit combination is predicted using the technology iteration path diagram; the regional adaptability of the technology implementation is evaluated in combination with the technology potential energy distribution field, a technology implementation plan is generated, and the user's feedback information on the technology implementation plan is recorded; the feedback information is converted into a state correction factor, which is used to adjust the dynamic threshold to achieve dynamic optimization of the new knowledge graph state.

2. The method according to claim 1, characterized in that Calculating the technology evolution coefficient based on the coupling degree to generate a technology iteration path map; using the technology iteration path map to quantitatively calculate the innovation potential energy value of each technology unit to form a technology potential energy distribution field; Integrating the technology association intensity matrix, the technology iteration path map, and the technology potential energy distribution field into an initial knowledge graph, and outputting a state feature set of the initial knowledge graph includes: Inputting the coupling degree into a technology evolution prediction model, the technology evolution prediction model calculates the evolution coefficient of the technology unit based on the dominance ratio, correlation strength ratio, and technology maturity difference between the technology units; combining the evolution coefficient with technology development constraints, the technology development constraints including timing constraints of technology development, technology generation span constraints, and technology branch convergence constraints; generating a technology iteration path map based on the evolution coefficient and the technology development constraints, wherein the edge weight of the technology iteration path map is determined by the evolution coefficient ratio of the source technology unit to the target technology unit; Calculate the innovation potential energy value of the technology unit using the technology iteration path map, map the innovation potential energy value to a geographic space coordinate system, and construct a technology potential energy distribution field based on spatial aggregation effect, regional correlation, and regional technology foundation; The technology association intensity matrix, the technology iteration path map and the technology potential energy distribution field are tensor-fused to construct an initial knowledge graph with multi-dimensional associations; the initial knowledge graph includes a topological structure layer, a technology evolution layer and a potential energy distribution layer of technology units; Feature extraction is performed on the initial knowledge graph. The association features of the topological structure layer are extracted based on the graph neural network. The temporal features of the technology evolution layer are extracted based on the recursive neural network. The spatial features of the potential energy distribution layer are extracted based on the convolutional neural network. The association features, the temporal features and the spatial features are fused to form a state feature set of the knowledge graph.

3. The method according to claim 1, characterized in that Deploy intelligent sensors to monitor the fluctuation of the state feature set in real time; when the monitoring result of the state fluctuation set exceeds a preset dynamic threshold, trigger a graph optimization signal; Calling a coupling calculation unit to recalculate the technology correlation strength matrix according to the graph optimization signal; Optimizing the technology iteration path map using a dynamic optimization algorithm based on the updated technology association intensity matrix includes: Deploy intelligent sensors to monitor fluctuations in the state feature set, wherein the intelligent sensors collect technology node feature data, technology association feature data, and graph global feature data, calculate feature deviation based on the technology node feature data, calculate coupling change rate based on the technology association feature data, and calculate situation evolution rate based on the graph global feature data; Inputting the feature deviation, the coupling change rate and the situation evolution rate into a feature fusion model, the feature fusion model uses a weighted combination method to generate a state feature fluctuation index, and the state feature fluctuation index represents the real-time situation of the knowledge graph; Constructing a dynamic threshold calculation model, wherein the dynamic threshold calculation model establishes a baseline threshold based on the historical data distribution of the state characteristic fluctuation index, and dynamically adjusts the baseline threshold according to the technology development cycle function and environmental impact factors to obtain an adaptive dynamic threshold; Comparing the state feature fluctuation index with the adaptive dynamic threshold in real time, and when the state feature fluctuation index exceeds the adaptive dynamic threshold, calculating the impact of the fluctuation region, and generating a graph optimization signal including an optimization region, an optimization depth, and an optimization direction based on the impact; The graph optimization signal triggers the coupling calculation unit to perform an update operation, wherein the coupling calculation unit obtains the latest technical data, recalculates the technical correlation strength of the corresponding area based on the latest technical data and the optimized area, and generates an updated technical correlation strength matrix; The updated technology association intensity matrix is input into the path optimization model. The path optimization model identifies the path nodes that need to be adjusted based on the optimization depth, constructs path evolution constraints according to the optimization direction, and uses a dynamic programming algorithm to obtain the optimal technology iteration path diagram based on the path nodes and the path evolution constraints.

4. The method according to claim 3, characterized in that The graph optimization signal triggers the coupling calculation unit to perform an update operation. The coupling calculation unit obtains the latest technical data, recalculates the technical correlation strength of the corresponding area based on the latest technical data and the optimized area, and generates an updated technical correlation strength matrix, which includes: The graph optimization signal triggers the coupling calculation unit to perform an update operation. The coupling calculation unit uses a locality-sensitive hashing algorithm to divide the influence area into multiple sub-areas, calculates the local density and relative distance of technology nodes in the multiple sub-areas, identifies the technology nodes, and constructs an influence propagation network with the technology nodes as the center; The coupling calculation unit obtains the latest technology data from a multi-source technology database, performs semantic analysis on the latest technology data using a deep learning model, extracts technology feature vectors containing technology function attributes, application scenario characteristics, and innovation point descriptions, and establishes a temporal feature extraction module to capture dynamic trend information of technology development; the coupling calculation unit calculates the technology correlation strength within the impact area based on the technology feature vectors and the dynamic trend information; The coupling calculation unit recalculates the technical correlation strength of the impact area using a multi-level coupling calculation method, and generates an updated technical correlation strength matrix based on the calculation results of the multi-level coupling calculation method.

5. The method according to claim 1, wherein The technology potential energy distribution field is recalculated using the optimized technology iteration path map; the updated technology association intensity matrix, the technology iteration path map, and the technology potential energy distribution field are integrated into a new knowledge graph state, and the updated state feature set is output, including: Extracting the path node degree centrality, path betweenness centrality, and path cluster cohesion in the optimized technology iteration path graph to construct a technology diffusion matrix, which records the diffusion intensity and coverage of technology knowledge in the path network; Recalculate the technology potential energy distribution field based on the path node degree centrality, the path betweenness centrality, the path cluster cohesion and the technology diffusion matrix. The technology potential energy distribution field includes basic potential energy, position potential energy and development potential energy. The basic potential energy is calculated based on the level of technological innovation, the position potential energy is calculated based on the network structure status, and the development potential energy is calculated based on the future evolution trend. Performing data standardization on the updated technology correlation intensity matrix, converting the technology iteration path diagram into a standard network representation, converting the technology potential energy distribution field into a unified digital format, and constructing a multi-dimensional data integration framework; A vertical mapping relationship and a horizontal connection relationship are established in the multidimensional data integration framework. The vertical mapping relationship determines the correspondence between data at different levels, and the horizontal connection relationship determines the interaction between data at the same level. Common features of multidimensional data are extracted based on the tensor decomposition method. The common features are input into the feature alignment module. The feature alignment module unifies the expression of different dimensions, uses an adaptive weight distribution algorithm to determine the importance of features in each dimension, and integrates the aligned features into a new knowledge graph state. Based on the new knowledge graph state, static features and dynamic features are extracted, the static features include structural features, distribution features and attribute features, and the dynamic features include evolution features, flow features and interaction features; the static features and the dynamic features are input into the feature fusion device, and the feature fusion device uses a multi-layer perceptron to combine features, and realizes feature dimensionality reduction through the principal component analysis method to generate an updated state feature set.

6. The method according to claim 1, characterized in that Converting the user's technical requirements into a query state, performing state similarity calculation with the updated state feature set to obtain state similarity; and identifying the optimal technology unit combination based on the state similarity; Predicting the state evolution trajectory of the optimal technology unit combination using the technology iteration path diagram includes: Receive user technical requirements, use a deep semantic analysis model to extract technical field features, functional requirement features, and performance indicator features in the user technical requirements, and construct a feature representation matrix that quantitatively describes the multidimensional attributes of the requirements; Based on the feature representation matrix, a query state vector is constructed through a feature mapping network, wherein the feature mapping network includes a static feature encoding layer and a dynamic feature encoding layer, wherein the static feature encoding layer processes technical attributes and functional attributes, and the dynamic feature encoding layer processes temporal features and evolutionary features; Calculate the similarity between the query state vector and the updated state feature set, use the attention mechanism to assign adaptive weights to different feature dimensions, calculate the distance between state vectors through the deep metric learning model, and generate a state similarity matrix; The state similarity matrix is used to construct a technology unit affinity graph, where the nodes of the technology unit affinity graph represent technology units and the edge weights represent state similarities. A community discovery algorithm is used to identify highly cohesive technology unit combinations, and the optimal technology unit combination is selected based on a combination scoring function. Performing a historical evolution analysis on the optimal technology unit combination, and extracting an evolutionary feature sequence based on the technology iteration path diagram, wherein the evolutionary feature sequence includes changes in technology maturity, innovation activity, and association strength; Inputting the evolution feature sequence into a time series prediction model, the time series prediction model adopts a long short-term memory network structure, captures long-term dependencies through a gating mechanism, and combines an attention mechanism to identify time series patterns to generate a technology evolution prediction sequence; A multidimensional evolution path diagram is constructed based on the technology evolution prediction sequence. The nodes of the multidimensional evolution path diagram represent the technology status, and the edges represent the state transition probability. The path selection strategy is optimized through the Monte Carlo tree search algorithm to output the optimal evolution trajectory.

7. The method according to claim 1, characterized in that Evaluate the regional adaptability of technology implementation in combination with the technology potential energy distribution field, generate a technology implementation plan, and record user feedback on the technology implementation plan; Converting the feedback information into a state correction factor for adjusting the dynamic threshold to achieve dynamic optimization of the new knowledge graph state includes: Assessing the regional suitability of technology implementation based on the technology potential energy distribution field, and generating a technology implementation plan based on the regional suitability, wherein the technology implementation plan includes a technology path planning and a resource allocation plan based on the regional suitability; A multimodal feedback collection system is used to record user feedback on the implementation effect of the technical implementation plan, and the multimodal feedback collection system obtains a multidimensional feature feedback dataset, and the multidimensional feature feedback dataset includes text feedback data, voice feedback data, and behavioral feedback data; the multidimensional feature feedback dataset is input into a deep semantic understanding network, and the deep semantic understanding network extracts semantic features and sentiment features to generate a feedback feature vector; Constructing a state correction factor based on the feedback feature vector, the state correction factor includes a technology implementation effect score and an optimization direction indication; inputting the state correction factor into a dynamic threshold adjustment module, and the dynamic threshold adjustment module updates the evolution rule trigger threshold of the knowledge graph state according to the technology implementation effect score and the optimization direction indication; A graph structure optimization network is used to achieve dynamic optimization of the knowledge graph state. The graph structure optimization network determines the rule triggering conditions according to the dynamic threshold adjustment module and applies the rules that meet the triggering conditions to the structural update of the knowledge graph. A multi-objective optimization framework is constructed. The multi-objective optimization framework uses the state correction factor as an optimization constraint to guide the evolution direction of the knowledge graph state. A new knowledge graph state is generated based on the optimization results of the multi-objective optimization framework, and the optimization effect of the new knowledge graph state is fed back to the multimodal feedback acquisition system to form a closed-loop mechanism for continuous optimization.

8. A knowledge graph-based intelligent matching and transaction recommendation system for scientific and technological achievements, used to implement the method described in any one of claims 1 to 7, characterized in that: include: The first unit is used to obtain a scientific and technological achievement dataset and extract a technical gene sequence from the scientific and technological achievement dataset through a deep feature extractor; Inputting the technology gene sequence into a high-dimensional coding network to generate a technology state vector; constructing a technology association strength matrix based on the technology state vector, wherein the technology association strength matrix records the degree of coupling between technology units; Calculating the technology evolution coefficient based on the coupling degree to generate a technology iteration path map; using the technology iteration path map to quantitatively calculate the innovation potential energy value of each technology unit to form a technology potential energy distribution field; Integrate the technology association intensity matrix, the technology iteration path diagram, and the technology potential energy distribution field into an initial knowledge graph, and output a state feature set of the initial knowledge graph; The second unit is used to deploy an intelligent sensor to monitor the fluctuation of the state feature set in real time; when the monitoring result of the state fluctuation set exceeds a preset dynamic threshold, a graph optimization signal is triggered; Calling a coupling calculation unit to recalculate the technology correlation strength matrix according to the graph optimization signal; Based on the updated technology association intensity matrix, a dynamic optimization algorithm is used to optimize the technology iteration path map; and the technology potential energy distribution field is recalculated using the optimized technology iteration path map. Integrate the updated technology association intensity matrix, the technology iteration path map, and the technology potential energy distribution field into a new knowledge graph state, and output an updated state feature set; The third unit is used to convert the user's technical requirements into a query state, calculate the state similarity with the updated state feature set, and obtain the state similarity; identify the optimal technology unit combination based on the state similarity; use the technology iteration path diagram to predict the state evolution trajectory of the optimal technology unit combination; combine the technology potential energy distribution field to evaluate the regional adaptability of the technology implementation, generate a technology implementation plan, and record the user's feedback information on the technology implementation plan; convert the feedback information into a state correction factor, which is used to adjust the dynamic threshold to achieve dynamic optimization of the new knowledge graph state.

9. An electronic device, characterized in that: include: processor; a memory for storing processor-executable instructions; The processor is configured to call the instructions stored in the memory to execute the method according to any one of claims 1 to 7.

10. A computer-readable storage medium having computer program instructions stored thereon, characterized in that: When the computer program instructions are executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images