A scientific and technological innovation big data evaluation method and system for industrial brain

Abstract

The present invention provides a scientific and technological innovation big data evaluation method and system for industrial brain, which relates to the technical field of big data processing, receives scientific and technological innovation big data streams from different data sources; uses a hash algorithm to assign a unique identifier to the scientific and technological innovation big data stream, performs encryption processing, and obtains a data identifier; analyzes the collaborative network strength between scientific and technological innovation subjects and the knowledge flow efficiency in the data identifier according to a graph theory algorithm, and generates a quantitative index of innovation network effect; uses a Bayesian network algorithm to process the uncertainty information of scientific and technological innovation achievement data, and performs probabilistic inference to obtain a market value and market risk assessment report; inputs the quantitative index of innovation network effect and the market value and market risk assessment report into a reinforcement learning algorithm, simulates different scientific and technological innovation path strategies, selects the optimal scientific and technological innovation path strategy, generates a development trend forecast report and optimization suggestions, realizes a comprehensive analysis and evaluation of scientific and technological innovation big data streams, and improves the accuracy and reliability of scientific and technological innovation evaluation.

Description

Technical Field

[0001] The embodiments of the present invention relate to the field of big data processing technology, and in particular to a scientific and technological innovation big data evaluation method and system for industrial brain. Background Art

[0002] With the rapid development of science and technology, technological innovation has become a key driving force for promoting social and economic progress. However, in the face of massive amounts of big data on technological innovation, how to effectively mine and utilize the potential value in this data has become an urgent problem to be solved.

[0003] At present, existing solutions mainly rely on the experience and judgment of experts and simple statistical analysis methods, including descriptive statistical analysis, regression analysis, etc. Existing evaluation methods only focus on the performance of a single scientific and technological innovation project and cannot fully reflect the overall impact of scientific and technological innovation activities; in the face of massive scientific and technological innovation data from different data sources, existing methods cannot effectively process and utilize these data; in addition, existing methods often ignore the impact of uncertain factors, resulting in inaccurate evaluation results. Summary of the invention

[0004] The embodiments of the present invention provide a scientific and technological innovation big data evaluation method and system for industrial brain, which are used to solve the problems in the prior art that the overall impact of scientific and technological innovation activities cannot be fully reflected, massive data cannot be effectively processed and utilized, and the evaluation results are not accurate enough.

[0005] In a first aspect, an embodiment of the present invention provides a scientific innovation big data evaluation method for an industrial brain, comprising: Receive scientific and technological innovation big data streams from different data sources; Using a hash algorithm to assign a unique identifier to the scientific and technological innovation big data stream, and encrypting the scientific and technological innovation big data stream according to the unique identifier to obtain a data identifier; Analyze the intensity of the collaborative network between the scientific and technological innovation entities and the efficiency of knowledge flow in the data identification according to the graph theory algorithm to generate a quantitative index of the innovation network effect. The graph theory algorithm includes social network analysis, community detection algorithm, web page ranking algorithm and random walk algorithm. The intensity of the collaborative network between the scientific and technological innovation entities represents the degree of mutual cooperation between different scientific and technological innovation entities. Using the Bayesian network algorithm to process the uncertainty information of the scientific and technological innovation data in the scientific and technological innovation big data stream, and performing probabilistic inference processing on the scientific and technological innovation data according to the processing results to obtain a market value and market risk assessment report; The innovation network effect quantitative index and the market value and market risk assessment report are input into the reinforcement learning algorithm, so that the reinforcement learning algorithm simulates the selection of the optimal scientific and technological innovation path strategy under different scientific and technological innovation path strategies, and generates a long-term development trend forecast report and optimization suggestions for the optimal scientific and technological innovation path strategy.

[0006] Optionally, the use of a Bayesian network algorithm to process uncertainty information of the scientific and technological innovation achievement data in the scientific and technological innovation big data stream, and performing probabilistic inference processing on the scientific and technological innovation achievement data according to the processing result to obtain a market value and market risk assessment report, includes: Identify uncertainty information in scientific and technological innovation data, and initialize the prior probability distribution in the Bayesian network model based on the uncertainty information; Based on the prior probability distribution, a Bayesian network model is constructed, and the scientific and technological innovation achievement data is input into the Bayesian network model as evidence; Using the Bayesian update rule, updating the node probability distribution in the Bayesian network model according to the evidence to obtain a posterior probability distribution; The posterior probability distribution is used as a processing result, and based on the processing result, the market value and market risk of the scientific and technological achievement data are probabilistically inferred to generate a market value and market risk assessment report.

[0007] Optionally, the probabilistic inference of the market value and market risk of the scientific and technological achievement data based on the processing result to generate a market value and market risk assessment report includes: According to the processing results, the market value of the scientific and technological innovation achievement data is sampled multiple times using a Monte Carlo simulation algorithm to obtain distribution characteristics of the market value, wherein the distribution characteristics include the expected value and variance of the market value; Analyzing the distribution characteristics using statistical methods to obtain a probability distribution of market value; Based on the probability distribution, the market risk of the scientific and technological innovation achievement data is evaluated by applying the risk value model to obtain the market risk level; In combination with the probability distribution and the market risk level, a multi-criteria decision analysis method is adopted to comprehensively evaluate the market performance of the scientific and technological achievements data and generate a market value and market risk assessment report.

[0008] Optionally, the probability distribution and the market risk level are combined to adopt a multi-criteria decision analysis method to comprehensively evaluate the market performance of the scientific and technological achievement data and generate a market value and market risk assessment report, including: Using the probability distribution and the market risk level as evaluation criteria in a multi-criteria decision analysis; The evaluation criteria are hierarchically modeled by using the hierarchical analysis method to obtain the hierarchical modeling results, and the priority weight of each criterion is obtained by the expert scoring method to obtain the importance information of the evaluation criteria; The fuzziness of the evaluation criteria is processed by using a fuzzy comprehensive evaluation method, and the membership function of each evaluation criterion is determined according to the fuzziness; Based on the hierarchical modeling result, the evaluation criteria importance information and the membership function, a multi-criteria decision analysis model is constructed, and a solution is obtained based on the multi-criteria decision analysis model; Based on the solution results, the market performance of the scientific and technological innovation achievement data is comprehensively evaluated to generate a market value and market risk assessment report.

[0009] Optionally, the innovation network effect quantitative index and the market value and market risk assessment report are input into a reinforcement learning algorithm, so that the reinforcement learning algorithm simulates the selection of the optimal scientific and technological innovation path strategy under different scientific and technological innovation path strategies, and generates a long-term development trend forecast report and optimization suggestions for the optimal scientific and technological innovation path strategy, including: Initializing the state space in the reinforcement learning algorithm according to the innovation network effect quantitative index and the market value and market risk assessment report, wherein the reinforcement learning algorithm includes a Q learning algorithm; Using the Q learning algorithm, construct the reward function under different scientific and technological innovation path strategies, associate the reward function with the state space, simulate different scientific and technological innovation path strategies, and update the Q value corresponding to different scientific and technological innovation path strategies through a trial and error mechanism; According to the updated Q values ​​corresponding to different science and technology innovation path strategies, the optimal science and technology innovation path strategy corresponding to the largest Q value is selected; Generate a long-term development trend forecast report for the optimal scientific and technological innovation path strategy, and based on the long-term development trend forecast report and combined with the optimal scientific and technological innovation path strategy, generate optimization suggestions for the optimal scientific and technological innovation path strategy.

[0010] Optionally, the Q learning algorithm is used to construct a reward function under different scientific and technological innovation path strategies, and the reward function is associated with the state space, and different scientific and technological innovation path strategies are simulated, and the Q values ​​corresponding to different scientific and technological innovation path strategies are updated through a trial and error mechanism, including: Define all states in the state space and define reward functions for actions under each technological innovation path strategy; Initialize the Q value of each state and action combination in the state space by using the Q learning algorithm, and provide feedback on the action execution results under different scientific and technological innovation path strategies based on the reward function to update the Q value, and determine whether the updated Q value reflects the benefits after taking the action; If not, repeat the steps of providing feedback on the results of action execution under different scientific and technological innovation path strategies until the predetermined convergence state is reached, the iterative update is completed, and the updated Q value is obtained. The predetermined convergence state is that the updated Q value reflects the benefit after taking the action, wherein the updated Q value in each iterative step is used to reflect the benefit after taking the action.

[0011] Optionally, the intensity of the collaborative network between scientific and technological innovation entities and the efficiency of knowledge flow in the data identifier are analyzed according to a graph theory algorithm to generate a quantitative index of the innovation network effect. The graph theory algorithm includes social network analysis, community detection algorithm, web page ranking algorithm and random walk algorithm. The intensity of the collaborative network between scientific and technological innovation entities characterizes the degree of mutual collaboration between different scientific and technological innovation entities, including: Using the degree centrality, closeness centrality and betweenness centrality algorithms in social network analysis, the importance and influence of scientific and technological innovation entities in the network are calculated, and quantitative indicators of the importance and influence of scientific and technological innovation entities in the network are obtained; A community detection algorithm is used to process the quantitative indicators of the importance and influence of the scientific and technological innovation subject in the network, to identify the close cooperation groups and the boundaries of the close cooperation groups among the scientific and technological innovation subjects, and to obtain the information of the close cooperation groups; Based on the information of the close cooperation group, the importance of the scientific and technological innovation subject in the information dissemination is analyzed by using the web page ranking algorithm to determine the subject information that plays a key role in the information diffusion process; Applying a random walk algorithm to process the propagation path of the subject information, evaluating the knowledge flow efficiency, calculating the probability of information reaching from one node to another, and obtaining the knowledge flow efficiency evaluation result; Based on the quantitative indicators of the importance and influence of the scientific and technological innovation entities in the network, the information of closely cooperating groups, the subject information and the evaluation results of knowledge flow efficiency, quantitative indicators of innovation network effects are generated.

[0012] In a second aspect, an embodiment of the present invention provides a scientific and technological innovation big data evaluation system for industrial brain, including: The receiving module is used to receive the scientific and technological innovation big data stream from different data sources; An encryption module, used to assign a unique identifier to the scientific and technological innovation big data stream using a hash algorithm, and encrypt the scientific and technological innovation big data stream according to the unique identifier to obtain a data identifier; An analysis module, for analyzing the intensity of the collaborative network between the scientific and technological innovation entities and the efficiency of knowledge flow in the data identifier according to a graph theory algorithm, and generating a quantitative index of the innovation network effect. The graph theory algorithm includes a social network analysis, a community detection algorithm, a web page ranking algorithm, and a random walk algorithm. The intensity of the collaborative network between the scientific and technological innovation entities represents the degree of mutual collaboration between different scientific and technological innovation entities. A processing module, used to process the uncertainty information of the scientific and technological innovation data in the scientific and technological innovation big data stream using a Bayesian network algorithm, and perform probabilistic inference processing on the scientific and technological innovation data according to the processing results to obtain a market value and market risk assessment report; A generation module is used to input the innovation network effect quantitative indicators and the market value and market risk assessment report into a reinforcement learning algorithm, so that the reinforcement learning algorithm simulates the selection of the optimal scientific and technological innovation path strategy under different scientific and technological innovation path strategies, and generates a long-term development trend forecast report and optimization suggestions for the optimal scientific and technological innovation path strategy.

[0013] In a third aspect, an embodiment of the present invention provides a computing device, comprising a processor and a memory, wherein the memory stores a computer program, and the processor is configured to run the computer program to execute any of the scientific and technological innovation big data evaluation methods for industrial brain as described in the first aspect.

[0014] In a fourth aspect, an embodiment of the present invention provides a computer storage medium having computer program instructions stored thereon, which, when executed by a processor, implements the scientific and technological innovation big data evaluation method for the industrial brain as described in any one of the first aspects.

[0015] In an embodiment of the present invention, a scientific and technological innovation big data stream is received from different data sources; a unique identifier is assigned to the scientific and technological innovation big data stream using a hash algorithm, and the scientific and technological innovation big data stream is encrypted according to the unique identifier to obtain a data identifier; the collaborative network strength between scientific and technological innovation entities and the knowledge flow efficiency in the data identifier are analyzed according to a graph theory algorithm to generate a quantitative index of innovation network effect, wherein the graph theory algorithm includes social network analysis, community detection algorithm, web page ranking algorithm and random walk algorithm, and the collaborative network strength between scientific and technological innovation entities characterizes the degree of mutual cooperation between different scientific and technological innovation entities; the uncertainty information of the scientific and technological innovation achievement data in the scientific and technological innovation big data stream is processed using a Bayesian network algorithm, and the scientific and technological innovation achievement data is probabilistically inferred according to the processing result to obtain a market value and market risk assessment report; the innovation network effect quantitative index and the market value and market risk assessment report are input into a reinforcement learning algorithm, so that the reinforcement learning algorithm simulates the selection of the optimal scientific and technological innovation path strategy under different scientific and technological innovation path strategies, and generates a long-term development trend forecast report and optimization suggestions for the optimal scientific and technological innovation path strategy. The technical solution provided by the present invention realizes the all-round analysis and evaluation of the big data stream of scientific and technological innovation, improves the accuracy and reliability of scientific and technological innovation evaluation, and also provides strong support for the selection of scientific and technological innovation paths, and promotes the effective management and sustainable development of scientific and technological innovation activities. Among them, through the multi-criteria decision analysis method, the market performance of scientific and technological innovation data is comprehensively evaluated, covering all aspects of market value and market risk; the hierarchical analysis method is used to determine the priority weights of each evaluation criterion to ensure the scientificity and rationality of the evaluation results; the fuzzy comprehensive evaluation method is used to deal with the fuzziness in the evaluation criteria to improve the accuracy and reliability of the evaluation; a detailed market value and market risk evaluation report is generated to provide a comprehensive scientific basis for the management and decision-making of scientific and technological innovation activities.

[0016] These and other aspects of the present invention will become more apparent from the following description of the embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0018] Figure 1 A flowchart of a scientific and technological innovation big data evaluation method for industrial brain provided by an embodiment of the present invention; Figure 2 A schematic diagram of the structure of a scientific and technological innovation big data evaluation system for industrial brain provided by an embodiment of the present invention; Figure 3 A schematic diagram of the structure of a computing device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0019] In order to enable those skilled in the art to better understand the solutions of the present invention, 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.

[0020] In some of the processes described in the specification and claims of the present invention and the above-mentioned figures, multiple operations that appear in a specific order are included, but it should be clearly understood that these operations may not be executed in the order in which they appear in this article or executed in parallel. The serial numbers of the operations, such as 101, 102, etc., are only used to distinguish different operations, and the serial numbers themselves do not represent any execution order. In addition, these processes may include more or fewer operations, and these operations may be executed in sequence or in parallel. It should be noted that the descriptions of "first", "second", etc. in this article are used to distinguish different messages, devices, modules, etc., do not represent the order of precedence, and do not limit the "first" and "second" to be different types.

[0021] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions 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 those skilled in the art without creative work are within the scope of protection of the present invention.

[0022] Figure 1 A flowchart of a scientific and technological innovation big data evaluation method for an industrial brain is provided for an embodiment of the present invention, such as Figure 1 As shown, the method includes: In the context of the rapid development of scientific and technological innovation, how to effectively evaluate and manage massive scientific and technological innovation big data from different data sources has become an urgent problem to be solved. Traditional evaluation methods are difficult to meet the complex needs of modern scientific and technological innovation activities. Based on this, the present invention provides a scientific and technological innovation big data evaluation method for the industry brain, such as Figure 1 ,include: Step 101: receiving scientific and technological innovation big data streams from different data sources; In this step, the scientific and technological innovation big data flow refers to the data set related to scientific and technological innovation activities collected from different data sources, such as scientific research institutions, corporate R&D departments, government databases, etc. These data include scientific research papers, patent documents, project reports, market research reports, etc.

[0023] For example, data related to scientific and technological innovation activities are collected from multiple data sources such as scientific research institutions, corporate R&D departments, government databases, etc., including scientific research papers, patent documents, project reports, market research reports, etc.; these data sources cover various dimensions of scientific and technological innovation activities to ensure the comprehensiveness and diversity of the data; the data are aggregated into a central database through API interfaces, file transfers, etc.

[0024] Step 102: using a hash algorithm to assign a unique identifier to the scientific and technological innovation big data stream, and encrypting the scientific and technological innovation big data stream according to the unique identifier to obtain a data identifier; In this step, the hash algorithm refers to an encryption algorithm that converts data of any length into a string of fixed length. It is unique and irreversible and is often used for data integrity verification and encryption processing. The unique identifier refers to a unique code assigned to each data item, which is used to identify and distinguish different data items to ensure data traceability.

[0025] This step uses a hash algorithm, such as SHA-256, on the collected scientific and technological innovation big data streams to assign a unique identifier to each data stream, and uses the identifier to encrypt the data to ensure data security and traceability; the hash algorithm ensures that even if the original data changes, its hash value will change accordingly, thereby ensuring data integrity; the unique identifier gives each data stream a unique mark, which facilitates subsequent data management and analysis.

[0026] Step 103: Analyze the intensity of the collaborative network between the scientific and technological innovation entities and the efficiency of knowledge flow in the data identifier according to a graph theory algorithm to generate a quantitative index of the innovation network effect. The graph theory algorithm includes social network analysis, community detection algorithm, web page ranking algorithm and random walk algorithm. The intensity of the collaborative network between the scientific and technological innovation entities represents the degree of mutual collaboration between different scientific and technological innovation entities. This step uses algorithms such as degree centrality, closeness centrality, and betweenness centrality in social network analysis to calculate the importance and influence of scientific and technological innovation entities in the network; uses a community detection algorithm to identify close cooperation groups and their boundaries among scientific and technological innovation entities; analyzes the importance of scientific and technological innovation entities in information dissemination based on a web page ranking algorithm; and applies a random walk algorithm to evaluate the efficiency of knowledge flow. Through these graph theory algorithms, we can comprehensively evaluate the collaborative relationship and knowledge flow efficiency among scientific and technological innovation entities, generate quantitative indicators of innovation network effects, and reflect the degree of mutual cooperation between different scientific and technological innovation entities and their impact on innovation activities.

[0027] Step 104: using the Bayesian network algorithm to process the uncertainty information of the scientific and technological innovation achievement data in the scientific and technological innovation big data stream, and performing probability inference processing on the scientific and technological innovation achievement data according to the processing result to obtain a market value and market risk assessment report; In this step; Bayesian network algorithm refers to a graphical model based on probability theory, which is used to process uncertain information and perform probabilistic inference; reinforcement learning algorithm refers to a type of artificial intelligence algorithm that learns optimal strategies through a trial and error mechanism. Common algorithms include Q learning algorithm, etc.

[0028] This step initializes the prior probability distribution in the Bayesian network model according to the uncertainty information in the scientific and technological innovation data; based on the prior probability distribution, a Bayesian network model is constructed, and through model analysis, the market value and market risk of the scientific and technological innovation data are probabilistically inferred; the Bayesian network algorithm can process uncertain information and evaluate market value and market risk through probabilistic inference, providing a scientific basis for the commercialization of scientific and technological innovation results.

[0029] Step 105: Input the innovation network effect quantitative index and the market value and market risk assessment report into a reinforcement learning algorithm, so that the reinforcement learning algorithm simulates the selection of the optimal scientific and technological innovation path strategy under different scientific and technological innovation path strategies, and generates a long-term development trend forecast report and optimization suggestions for the optimal scientific and technological innovation path strategy; In this step, the quantitative indicators of innovation network effects and market value and market risk assessment reports are input into reinforcement learning algorithms, such as the Q-learning algorithm, to obtain the maximum Q value through model analysis to generate long-term development trend forecast reports and optimization suggestions for the optimal scientific and technological innovation path strategy; the reinforcement learning algorithm simulates different scientific and technological innovation path strategies, selects the optimal path, and generates long-term development trend forecast reports and optimization suggestions to provide scientific support for scientific and technological innovation decisions.

[0030] The embodiments of the present invention can achieve the following beneficial effects through the above steps: The data is encrypted through the hash algorithm to ensure the security and traceability of the data and prevent the data from being tampered with or illegally accessed during transmission and storage; Through graph theory algorithms, we comprehensively evaluate the collaborative relationship and knowledge flow efficiency among scientific and technological innovation entities, generate quantitative indicators of innovation network effects, and accurately reflect the degree of mutual collaboration among different scientific and technological innovation entities and their impact on innovation activities. The Bayesian network algorithm is used to process the uncertainty information in the scientific and technological innovation results data, and the probability inference is processed to obtain the market value and market risk assessment report, which provides a scientific basis for the market value assessment of scientific and technological innovation results and can effectively assess market risks; Through reinforcement learning algorithms, different scientific and technological innovation path strategies are simulated, the optimal path is selected, and long-term development trend forecast reports and optimization suggestions are generated, providing scientific guidance for scientific and technological innovation decisions and improving the accuracy and foresight of scientific and technological innovation path selection.

[0031] In the process of scientific and technological innovation, it is crucial to accurately evaluate the market value of scientific and technological achievements and their potential risks. However, since scientific and technological innovation achievements themselves are often accompanied by a high degree of uncertainty, traditional evaluation methods are difficult to fully capture these uncertainties. Based on this, the present invention provides a specific embodiment, in which step 104 uses a Bayesian network algorithm to process the uncertainty information of the scientific and technological achievement data in the scientific and technological innovation big data stream, and performs probabilistic inference processing on the scientific and technological achievement data based on the processing results to obtain a market value and market risk assessment report, which specifically includes the following steps: Step 410: Identify uncertainty information in the scientific and technological achievement data, and initialize the prior probability distribution in the Bayesian network model according to the uncertainty information; In this step, uncertainty information refers to some unknown or ambiguous factors existing in the data of scientific and technological achievements. These factors may lead to uncertainty in the evaluation results and need to be processed through statistical methods or models; the Bayesian network model refers to a graphical model based on probability theory, which is used to process uncertain information and perform probabilistic inference. The Bayesian network consists of a set of nodes and directed edges between them. The nodes represent random variables and the edges represent the dependencies between variables; the prior probability distribution refers to the assumption about the probability distribution of unknown parameters in Bayesian statistics before observing any data.

[0032] This step identifies the key uncertainties that affect the market value and risk of scientific and technological achievements through data analysis and expert opinions; for each uncertain factor, a reasonable prior probability distribution is set as the initial probability estimate based on existing knowledge and assumptions. For example, various uncertainties are identified from the data of scientific and technological achievements, such as market acceptance, technology maturity, competitor reactions, etc.; based on the identified uncertainty information, a prior probability distribution is assigned to each node in the Bayesian network model, such as normal distribution, Poisson distribution, etc.

[0033] Step 411: constructing a Bayesian network model based on the prior probability distribution, and inputting the scientific and technological innovation achievement data as evidence into the Bayesian network model; In this step, the posterior probability distribution refers to the new probability distribution obtained by updating the prior probability distribution through Bayes' theorem after observing the data.

[0034] In this step, when constructing a Bayesian network model, the connection between nodes is determined based on the causal or logical relationship between uncertain information; the actual observed data, i.e., evidence, is input into the model as the basis for updating the prior probability distribution.

[0035] Step 412: using the Bayesian update rule to update the node probability distribution in the Bayesian network model according to the evidence to obtain a posterior probability distribution; In this step, the Bayesian update rule refers to an application of Bayes' theorem to update the known probability distribution based on new evidence.

[0036] This step uses the Bayesian formula to update the probability distribution of each node using the input evidence and calculates the posterior probability distribution; the posterior probability distribution reflects the new probability estimates of each uncertainty factor after considering the actual observed data.

[0037] Step 413: taking the posterior probability distribution as a processing result, and based on the processing result, probabilistically inferring the market value and market risk of the scientific and technological achievement data, and generating a market value and market risk assessment report; The posterior probability distribution of this step provides the latest probability estimates of various uncertainty factors after taking into account the actual observed data; by analyzing the posterior probability distribution, more accurate probabilistic inferences can be made about the market value and market risk of scientific and technological achievements; the final evaluation report will describe in detail the expected levels of market value and risk and their uncertainty range.

[0038] The embodiments of the present invention can achieve the following beneficial effects through the above steps: Using the Bayesian network model to process the uncertainty information in the scientific and technological innovation data can more accurately reflect the market value and market risk and provide a scientific basis; Using the Bayesian update rule, the prior probability distribution is updated according to the actual observation data to obtain the posterior probability distribution, which provides a more reliable probability inference result for the market value and market risk assessment of scientific and technological achievements. The market value and market risk assessment report generated based on the posterior probability distribution can provide decision makers with detailed market analysis and risk warnings, thereby improving the scientific nature and reliability of decision-making.

[0039] In order to accurately evaluate the market value and potential risks of scientific and technological achievements, the present invention provides a specific embodiment. The step 413 performs probabilistic inference on the market value and market risk of the scientific and technological achievement data based on the processing results to generate a market value and market risk assessment report, which specifically includes the following steps: Step 421: According to the processing result, the market value of the scientific and technological achievement data is sampled multiple times using a Monte Carlo simulation algorithm to obtain distribution characteristics of the market value, wherein the distribution characteristics include the expected value and variance of the market value; In this step, the Monte Carlo simulation algorithm refers to a numerical calculation method that solves mathematical, physical and engineering problems by random sampling or observing historical data, and is often used to simulate the uncertain behavior of the system; distribution characteristics refer to the characteristics that describe the statistical properties of random variables, including expected value, variance, etc.

[0040] For example, based on the posterior probability distribution obtained by the Bayesian network model, random variables of market value are set, such as market acceptance, technology maturity, etc.; through the Monte Carlo simulation algorithm, a large number of random samplings are performed on these random variables to simulate different possible scenarios of market value; the sampling results are collected and analyzed, and the expected value and variance of the market value are calculated to obtain the distribution characteristics of the market value.

[0041] Step 422: Analyze the distribution characteristics using statistical methods to obtain a probability distribution of market value; In this step, statistical methods refer to a range of methods used for data analysis, including descriptive statistics and inferential statistics, which are used to summarize data characteristics and infer population characteristics.

[0042] This step uses statistical methods, such as descriptive statistical analysis, to analyze the distribution characteristics of market value, and fits the probability distribution of market value based on the analysis results. The analysis includes calculating statistics such as mean and variance. Based on these statistics, fit the probability distribution of market value, such as normal distribution, Poisson distribution, etc., to better describe the randomness of market value.

[0043] Step 423: Based on the probability distribution, the market risk of the scientific and technological innovation achievement data is evaluated by applying a risk value model to obtain a market risk level; In this step, the Value at Risk (VaR) model refers to a method used to assess the maximum loss that a financial asset or investment portfolio may suffer within a certain period of time. The Value at Risk model usually takes into account market fluctuations and other uncertainties.

[0044] This step sets the risk value model, such as the parameters of the VaR model, such as confidence level, time window, etc., according to the probability distribution of market value; the market risk level is calculated through the risk value model, that is, the maximum extent to which the market value may decline under a certain confidence level, thereby assessing the market risk.

[0045] Step 423: Combining the probability distribution and the market risk level, using a multi-criteria decision analysis method, comprehensively evaluating the market performance of the scientific and technological achievement data, and generating a market value and market risk assessment report; In this step, Multi-Criteria Decision Making (MCDM) refers to a method that considers multiple evaluation criteria simultaneously in the decision-making process, aiming to find the optimal or satisfactory solution.

[0046] This step defines the evaluation criteria in the multi-criteria decision analysis based on the probability distribution of market value and the market risk level, such as the expected value, variance and market risk level of market value; uses the analytic hierarchy process (AHP) or other multi-criteria decision analysis methods to assign weights to these evaluation criteria and determine the importance of each criterion; combines the weights and evaluation results of each criterion to comprehensively evaluate the market performance of scientific and technological innovation results data and generate a market value and market risk evaluation report.

[0047] The embodiments of the present invention can achieve the following beneficial effects through the above steps: The distribution characteristics of market value are obtained through Monte Carlo simulation algorithm, and these characteristics are analyzed by statistical methods to obtain the probability distribution of market value, so as to more accurately evaluate market value; Quantify market risk levels using risk value models, provide scientific risk assessment methods, and help decision makers better understand potential risks; Through multi-criteria decision analysis methods, market value and market risk are comprehensively evaluated, and detailed evaluation reports are generated to provide comprehensive support for decision-making.

[0048] In order to more accurately evaluate the market performance of scientific and technological achievements data, based on this, the present invention provides a specific embodiment, wherein the step 423, combining the probability distribution and the market risk level, adopts a multi-criteria decision analysis method to comprehensively evaluate the market performance of the scientific and technological achievements data, and generates a market value and market risk assessment report, which specifically includes the following steps: Step 431: using the probability distribution and the market risk level as evaluation criteria in multi-criteria decision analysis; In this step of multi-criteria decision analysis, the evaluation criteria must first be defined. In this embodiment, the evaluation criteria include the expected value, variance, and market risk level of the market value; these criteria reflect different aspects of market performance and can comprehensively evaluate the market potential and risks of scientific and technological achievements.

[0049] Step 432: Use the hierarchical analysis method to perform hierarchical modeling on the evaluation criteria to obtain a hierarchical modeling result, and obtain the priority weight of each criterion through the expert scoring method to obtain the importance information of the evaluation criteria; In this step, the Analytic Hierarchy Process (AHP) refers to a multi-criteria decision analysis method that determines the priority weight of each criterion by constructing a hierarchical model and an expert scoring method; the expert scoring method refers to a method for obtaining the priority weight of each criterion, and the weight is determined by experts scoring the importance of each criterion.

[0050] This step first constructs a hierarchical model, dividing the evaluation criteria into the overall goal, criterion level, and program level; obtains the expert's scoring results on the importance of each criterion, and determines the priority weight by comparing the relative importance of different criteria; based on the expert's scoring results, integrates the importance information of each criterion to provide a basis for subsequent comprehensive evaluation.

[0051] Step 433: using a fuzzy comprehensive evaluation method to process the fuzziness of the evaluation criteria, and determining the membership function of each evaluation criterion according to the fuzziness; In this step, fuzzy comprehensive evaluation (FCE) refers to a method for dealing with the fuzziness of evaluation criteria, which converts fuzzy evaluation into numerical representation through membership function; membership function refers to the function used to describe the degree of membership of elements in a fuzzy set in fuzzy comprehensive evaluation.

[0052] In actual evaluation, some criteria may be ambiguous. For example, the market risk level is difficult to quantify accurately. The fuzzy comprehensive evaluation method is used to transform the fuzzy evaluation into the form of membership function to better describe the fuzziness.

[0053] Step 434: constructing a multi-criteria decision analysis model based on the hierarchical modeling result, the evaluation criteria importance information and the membership function, and solving the multi-criteria decision analysis model to obtain a solution result; This step combines the importance information obtained by the hierarchical analysis method with the membership function obtained by the fuzzy comprehensive evaluation method to construct a complete multi-criteria decision analysis model; the model is used to solve the market performance of scientific and technological achievements data to obtain the final solution.

[0054] Step 434: Based on the solution result, comprehensively evaluate the market performance of the scientific and technological achievement data to generate a market value and market risk evaluation report; This step comprehensively evaluates the performance of scientific and technological achievements data in terms of market value and market risk based on the solution results; based on the comprehensive evaluation results, a detailed market value and market risk evaluation report is generated to provide a scientific basis for decision-making.

[0055] More specifically, the present invention also provides a comprehensive evaluation function to comprehensively evaluate the market performance of the scientific and technological achievement data. The specific calculation formula is as follows: ; in, is a comprehensive evaluation function; is the membership function of each evaluation criterion; Representing the probability distribution of market value Expected value; is the market risk level; It is The priority weight of each evaluation criterion; is the total number of evaluation criteria; It is The membership function of the evaluation criteria represents the probability distribution of the given market value. and market risk level The fuzzy evaluation results under This step effectively combines multiple evaluation criteria by providing a comprehensive evaluation function, and provides a comprehensive and scientific market value and market risk evaluation result through weight allocation and fuzzy processing.

[0056] The calculation formula of the fuzzy evaluation result is as follows: ; in, is the probability distribution of market value Expected value; is the market risk level; It is a threshold used to adjust the size of the membership value and can be set according to specific circumstances; The purpose of designing the membership function in this step is to quantify the ambiguity in the evaluation criteria, so as to better reflect the ambiguity of market value and market risk in the evaluation criteria and make the evaluation results more accurate and reliable; The calculation formula for market risk level is as follows: ; ; in, is the market risk level; Indicates the time interval Confidence level The maximum possible loss under Represents the current value of the portfolio; is the confidence level corresponding to the standard normal distribution quantile of ; is the standard deviation of the portfolio return; It is a time interval, measured in days, weeks, or months; The market risk level in this step is a key indicator for evaluating the market performance of scientific and technological achievements data. It reflects the possible impact of market fluctuations on the market value of scientific and technological achievements within a certain time frame. The purpose of designing the market risk level is to quantify market risks and provide decision makers with a clear understanding of potential risks.

[0057] The embodiments of the present invention can achieve the following beneficial effects through the above steps: Through multi-criteria decision analysis methods, the market performance of scientific and technological innovation data is comprehensively evaluated, covering all aspects of market value and market risk; Use the analytic hierarchy process to determine the priority weight of each evaluation criterion to ensure the scientificity and rationality of the evaluation results; The fuzzy comprehensive evaluation method is used to deal with the fuzziness in the evaluation criteria and improve the accuracy and reliability of the evaluation; Generate detailed market value and market risk assessment reports to provide a comprehensive scientific basis for the management and decision-making of scientific and technological innovation activities.

[0058] In order to provide a comprehensive and scientific basis for decision-making, based on this, the present invention provides a specific embodiment, in which the step 105 inputs the innovation network effect quantitative index and the market value and market risk assessment report into a reinforcement learning algorithm, so that the reinforcement learning algorithm simulates the selection of the optimal scientific and technological innovation path strategy under different scientific and technological innovation path strategies, and generates a long-term development trend forecast report and optimization suggestions for the optimal scientific and technological innovation path strategy, which specifically includes the following steps: Step 501: Initializing the state space in the reinforcement learning algorithm according to the innovation network effect quantitative index and the market value and market risk assessment report, wherein the reinforcement learning algorithm includes a Q learning algorithm; In this step, the state space refers to the set of all possible states in the reinforcement learning algorithm; the Q-learning algorithm refers to a model-free reinforcement learning algorithm that learns what actions to take in a given state to obtain the maximum cumulative reward by updating the Q value.

[0059] This step takes the quantitative indicators of innovation network effects and the market value and market risk assessment report as input; initializes the state space of the reinforcement learning algorithm and defines all possible states; in the state space, defines possible actions for each state, that is, the scientific and technological innovation path strategy.

[0060] Step 502: construct reward functions under different technological innovation path strategies using a Q learning algorithm, associate the reward functions with the state space, simulate different technological innovation path strategies, and update the Q values ​​corresponding to different technological innovation path strategies through a trial and error mechanism; In this step, the reward function refers to the function used to define the immediate reward obtained after taking a certain action in reinforcement learning; the trial and error mechanism refers to the process of learning the optimal strategy through continuous trial and error correction.

[0061] This step constructs the reward function under different scientific and technological innovation path strategies, defines the immediate reward obtained after adopting a certain scientific and technological innovation path strategy; associates the reward function with the state space so that each action in each state has a corresponding Q value; during the simulation process, the agent selects actions based on the current state and obtains immediate rewards based on the reward function; through the trial and error mechanism, the agent continuously updates the Q value and learns what actions to take in each state to obtain the maximum cumulative reward.

[0062] Step 503: According to the updated Q values ​​corresponding to different scientific and technological innovation path strategies, the optimal scientific and technological innovation path strategy corresponding to the largest Q value is selected; This step calculates the updated Q values ​​corresponding to different scientific and technological innovation path strategies. After multiple iterations and updates, the Q values ​​corresponding to different scientific and technological innovation path strategies are obtained; the scientific and technological innovation path strategy with the largest Q value is selected as the optimal strategy.

[0063] Step 504: Generate a long-term development trend forecast report of the optimal scientific and technological innovation path strategy, and generate optimization suggestions for the optimal scientific and technological innovation path strategy based on the long-term development trend forecast report and the optimal scientific and technological innovation path strategy; This step uses the optimal science and technology innovation path strategy to predict long-term development trends and generate a long-term development trend forecast report; based on the forecast report, combined with the optimal science and technology innovation path strategy, generate optimization suggestions to provide a reference for science and technology innovation decisions.

[0064] The embodiments of the present invention can achieve the following beneficial effects through the above steps: Through the Q-learning algorithm, different scientific and technological innovation path strategies are simulated and the optimal path is selected, which improves the scientificity and reliability of scientific and technological innovation decision-making; Generate a long-term development trend forecast report on the optimal scientific and technological innovation path strategy, providing a basis for the long-term planning of scientific and technological innovation activities; Based on the long-term development trend forecast report and combined with the optimal scientific and technological innovation path strategy, optimization suggestions are generated to provide comprehensive support for the selection of scientific and technological innovation paths.

[0065] To update the Q value to determine the scientific and technological innovation path strategy. Based on this, the present invention provides a specific embodiment, in which step 502 uses the Q learning algorithm to construct a reward function under different scientific and technological innovation path strategies, associates the reward function with the state space, simulates different scientific and technological innovation path strategies, and updates the Q values ​​corresponding to different scientific and technological innovation path strategies through a trial and error mechanism, specifically including the following steps: Step 511: define all states in the state space and define a reward function for each action under each technological innovation path strategy; In this step, the scientific and technological innovation path strategy, in the present invention, refers to different scientific and technological innovation path options, such as investing in R&D funds, selecting partners, etc.

[0066] This step defines all states in the state space based on the quantitative indicators of innovation network effects and the market value and market risk assessment report. The state space contains all possible market conditions and network effect indicators, such as market acceptance, technology maturity, partnerships, etc. For each technological innovation path strategy, such as increasing R&D investment, expanding international markets, etc., a corresponding reward function is defined to measure the immediate benefits after adopting the strategy.

[0067] Step 512: Initialize the Q value of each state and action combination in the state space using the Q learning algorithm, and provide feedback on the action execution results under different technological innovation path strategies based on the reward function to update the Q value, and determine whether the updated Q value reflects the benefit after taking the action; In this step, the Q value of each state and action combination in the state space is initially set to a default value, such as zero; based on the reward function, the action execution results under each technological innovation path strategy are fed back, that is, the Q value is updated according to the actual benefits; and it is checked whether the updated Q value reflects the actual benefits after taking the action.

[0068] Step 513: If not, the step of providing feedback on the results of the actions executed under different technological innovation path strategies is repeated until a predetermined convergence state is reached, the iterative update is completed, and an updated Q value is obtained. The predetermined convergence state is that the updated Q value reflects the benefit after the action is taken, wherein the updated Q value in each iterative step is used to reflect the benefit after the action is taken; In this step, the convergence state refers to the state in reinforcement learning when the algorithm reaches a stable state and no longer changes significantly. At this time, the algorithm believes that the optimal strategy has been found. In this embodiment, the convergence state is the updated Q value that reflects the benefit after taking the action.

[0069] If after one iteration it is found that the Q value does not accurately reflect the benefits after taking an action, the feedback step will continue to be repeated; until the updated Q value reflects the benefits after taking the action, at which time it is considered that convergence has been reached; after completing the iterative update, the final Q value is obtained, which reflects the expected benefits after adopting each technological innovation path strategy.

[0070] The embodiments of the present invention can achieve the following beneficial effects through the above steps: The Q value is updated through the Q learning algorithm, and the optimal scientific and technological innovation path strategy is selected, which improves the scientificity and reliability of scientific and technological innovation decision-making; Use reward functions to accurately evaluate the immediate benefits of different scientific and technological innovation path strategies and provide a basis for decision-making; Continuously adjust strategies through trial and error mechanism to ensure that strategies can be continuously optimized in a dynamically changing market environment.

[0071] In order to comprehensively evaluate the network effects in scientific and technological innovation activities and provide a scientific basis for evaluation. Based on this, the present invention provides a specific embodiment, in which the step 103 analyzes the intensity of the collaborative network between the scientific and technological innovation subjects and the efficiency of knowledge flow in the data identifier according to a graph theory algorithm to generate a quantitative index of the innovation network effect. The graph theory algorithm includes social network analysis, community detection algorithm, web page ranking algorithm and random walk algorithm. The intensity of the collaborative network between scientific and technological innovation subjects represents the degree of mutual cooperation between different scientific and technological innovation subjects, and specifically includes the following steps: Step 301: using degree centrality, closeness centrality and betweenness centrality algorithms in social network analysis, calculate the importance and influence of the scientific and technological innovation subject in the network, and obtain quantitative indicators of the importance and influence of the scientific and technological innovation subject in the network; In this step, Social Network Analysis (SNA) refers to the method of studying individuals and their relationships in social networks, that is, the relationship between nodes and edges, which can be used to evaluate the centrality, betweenness and other properties of nodes in the network; Degree Centrality refers to the number of direct connections between a node and other nodes in the network. Nodes with high degree centrality usually have more direct neighbors in the network; Closeness Centrality refers to the average shortest path length from a node to all other nodes in the network. Nodes with high closeness centrality are more likely to reach other nodes in the network; Betweenness Centrality refers to the frequency of measuring the shortest path between nodes in the network. Nodes with high betweenness centrality play a key intermediary role in the flow of network information.

[0072] This step constructs a network diagram by collecting cooperation information between scientific and technological innovation entities, such as jointly publishing papers and jointly applying for patents. The nodes in the diagram represent scientific and technological innovation entities, and the edges represent the cooperative relationships between them. The degree centrality algorithm is used to calculate the number of direct connections of each node. Nodes with high degree centrality have more direct neighbors in the network, indicating that they have a higher direct influence in the network. The closeness centrality algorithm is used to calculate the average shortest path length from each node to all other nodes. Nodes with high closeness centrality are more likely to reach other nodes in the network, indicating that they have a higher central position in the network. The betweenness centrality algorithm is used to calculate the frequency of each node acting as the shortest path between other nodes in the network. Nodes with high betweenness centrality play a key intermediary role in the flow of network information. The results of these three centralities are integrated together to form a quantitative indicator of the importance and influence of scientific and technological innovation entities in the network.

[0073] Step 302: using a community detection algorithm to process the quantitative indicators of the importance and influence of the scientific and technological innovation subject in the network, identifying the close cooperation groups among the scientific and technological innovation subjects and the boundaries of the close cooperation groups, and obtaining the close cooperation group information; In this step, the community detection algorithm is used to identify closely connected subgroups in the network. Common algorithms include the Louvain method and the Infomap algorithm.

[0074] This step uses a community detection algorithm to process the constructed network graph to find closely connected subgroups in the network; the subgroups identified by the algorithm reflect the close cooperation relationship between scientific and technological innovation entities; determine the boundaries of these subgroups, that is, which nodes belong to the same closely cooperating group, which helps to understand the cooperation pattern in the network; integrate the information of the identified closely cooperating groups to obtain a clear division of the cooperative groups.

[0075] Step 303: Based on the close cooperation group information, the importance of the scientific and technological innovation subject in the information dissemination is analyzed using a web page ranking algorithm to determine the subject information that plays a key role in the information diffusion process; In this step, PageRank refers to the algorithm originally used for search engine rankings and can also be used to evaluate the importance of nodes in the network.

[0076] This step uses the web page ranking algorithm to score each node in the network graph. Nodes with high scores have a higher influence in the dissemination of network information. Based on the scoring results, it is determined which nodes play a key role in the dissemination of network information. These nodes may be nodes with high degree centrality, closeness centrality, and betweenness centrality. The information of these key nodes is integrated to obtain the main information that plays a key role in the process of information diffusion.

[0077] Step 304: Apply a random walk algorithm to process the propagation path of the subject information, evaluate the knowledge flow efficiency, calculate the probability of information reaching from one node to another, and obtain a knowledge flow efficiency evaluation result; In this step, the random walk algorithm evaluates the efficiency of information transmission between nodes by simulating the random walk process in the network.

[0078] This step uses the random walk algorithm to simulate the information propagation path in the network and evaluate the efficiency of information flow in the network; calculate the probability of information reaching from one node to another, and the path with high probability indicates that the information propagation efficiency is higher; integrate these calculation results to obtain the evaluation result of the knowledge flow efficiency in the entire network.

[0079] Step 305: Generate a quantitative index of innovation network effect based on the quantitative index of the importance and influence of the scientific and technological innovation subject in the network, the information of the close cooperation group, the subject information and the evaluation result of the knowledge flow efficiency; This step integrates the quantitative indicators of the importance and influence of scientific and technological innovation entities in the network, information on closely cooperating groups, information on key entities, and evaluation results of knowledge flow efficiency; based on the integrated information, quantitative indicators of innovation network effects are generated, which comprehensively reflect the importance, influence, and knowledge flow efficiency of scientific and technological innovation entities in the network.

[0080] The embodiments of the present invention can achieve the following beneficial effects through the above steps: Through various algorithms in social network analysis, the importance and influence of scientific and technological innovation entities in the network, as well as the efficiency of knowledge flow, are comprehensively evaluated to generate quantitative indicators of innovation network effects; Use web page ranking algorithms and community detection algorithms to identify entities that play a key role in information dissemination and provide key support for scientific and technological innovation activities; The random walk algorithm is used to evaluate the efficiency of knowledge flow, which helps to understand the propagation path and efficiency of information in the network and provides a basis for optimizing information flow.

[0081] Figure 2 A schematic diagram of the structure of a scientific and technological innovation big data evaluation system for an industry brain is provided for an embodiment of the present invention. Figure 2 As shown, the system includes: A receiving module 21 is used to receive scientific and technological innovation big data streams from different data sources; The encryption module 22 is used to assign a unique identifier to the scientific and technological innovation big data stream using a hash algorithm, and encrypt the scientific and technological innovation big data stream according to the unique identifier to obtain a data identifier; An analysis module 23 is used to analyze the intensity of the collaborative network between the scientific and technological innovation entities and the efficiency of knowledge flow in the data identifier according to a graph theory algorithm to generate a quantitative index of the innovation network effect. The graph theory algorithm includes social network analysis, community detection algorithm, web page ranking algorithm and random walk algorithm. The intensity of the collaborative network between the scientific and technological innovation entities represents the degree of mutual collaboration between different scientific and technological innovation entities. The processing module 24 is used to process the uncertainty information of the scientific and technological innovation data in the scientific and technological innovation big data stream using the Bayesian network algorithm, and perform probability inference processing on the scientific and technological innovation data according to the processing results to obtain a market value and market risk assessment report; The generation module 25 is used to input the innovation network effect quantitative indicators and the market value and market risk assessment report into the reinforcement learning algorithm, so that the reinforcement learning algorithm simulates the selection of the optimal scientific and technological innovation path strategy under different scientific and technological innovation path strategies, and generates a long-term development trend forecast report and optimization suggestions for the optimal scientific and technological innovation path strategy.

[0082] Figure 2 The scientific and technological innovation big data evaluation system for industrial brain can be implemented Figure 1 The implementation principle and technical effect of the scientific and technological innovation big data evaluation method for industrial brain described in the embodiment shown will not be repeated. The specific way in which each module and unit performs operations in the scientific and technological innovation big data evaluation system for industrial brain in the above embodiment has been described in detail in the embodiment of the method, and will not be elaborated here.

[0083] In one possible design, Figure 2 The scientific and technological innovation big data evaluation method for the industrial brain of the embodiment shown can be implemented as a computing device, such as Figure 3 As shown, the computing device may include a storage component 31 and a processing component 32; The storage component 31 stores one or more computer instructions, wherein the one or more computer instructions are called and executed by the processing component 32 .

[0084] The processing component 32 is used to: receive scientific and technological innovation big data streams from different data sources; use a hash algorithm to assign a unique identifier to the scientific and technological innovation big data stream, and encrypt the scientific and technological innovation big data stream according to the unique identifier to obtain a data identifier; analyze the collaborative network strength between scientific and technological innovation entities and the knowledge flow efficiency in the data identifier according to a graph theory algorithm to generate a quantitative index of innovation network effect, wherein the graph theory algorithm includes social network analysis, community detection algorithm, web page ranking algorithm and random walk algorithm, and the collaborative network strength between scientific and technological innovation entities represents the degree of mutual cooperation between different scientific and technological innovation entities; use a Bayesian network algorithm to process the uncertainty information of the scientific and technological innovation achievement data in the scientific and technological innovation big data stream, and perform probabilistic inference processing on the scientific and technological innovation achievement data according to the processing results to obtain a market value and market risk assessment report; input the innovation network effect quantitative index and the market value and market risk assessment report into a reinforcement learning algorithm, so that the reinforcement learning algorithm simulates the selection of the optimal scientific and technological innovation path strategy under different scientific and technological innovation path strategies, and generates a long-term development trend forecast report and optimization suggestions for the optimal scientific and technological innovation path strategy.

[0085] The processing component 32 may include one or more processors to execute computer instructions to complete all or part of the steps in the above method. Of course, the processing component may also be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors or other electronic components to perform the above method.

[0086] The storage component 31 is configured to store various types of data to support operations at the terminal. The storage component can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk.

[0087] Of course, the computing device may also include other components, such as input / output interfaces, display components, communication components, etc.

[0088] The input / output interface provides an interface between the processing component and the peripheral interface module, which may be an output device, an input device, etc.

[0089] The communication component is configured to facilitate, among other things, wired or wireless communications between the computing device and other devices.

[0090] Among them, the computing device can be a physical device or an elastic computing host provided by a cloud computing platform, etc. In this case, the computing device can refer to a cloud server, and the above-mentioned processing components, storage components, etc. can be basic server resources rented or purchased from the cloud computing platform.

[0091] The embodiment of the present invention further provides a computer storage medium storing a computer program, which can achieve the above-mentioned Figure 1 The illustrated embodiment is a scientific and technological innovation big data evaluation method for industrial brain.

[0092] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0093] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0094] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0095] 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 aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A scientific and technological innovation big data evaluation method for industrial brain, characterized by: include: Receive scientific and technological big data streams from different data sources; Using a hash algorithm to assign a unique identifier to the scientific and technological innovation big data stream, and encrypting the scientific and technological innovation big data stream according to the unique identifier to obtain a data identifier; Analyze the intensity of the collaborative network between the scientific and technological innovation entities and the efficiency of knowledge flow in the data identification according to the graph theory algorithm to generate a quantitative index of the innovation network effect. The graph theory algorithm includes social network analysis, community detection algorithm, web page ranking algorithm and random walk algorithm. The intensity of the collaborative network between the scientific and technological innovation entities represents the degree of mutual cooperation between different scientific and technological innovation entities. Using the Bayesian network algorithm to process the uncertainty information of the scientific and technological innovation data in the scientific and technological innovation big data stream, and performing probabilistic inference processing on the scientific and technological innovation data according to the processing results to obtain a market value and market risk assessment report; The innovation network effect quantitative index and the market value and market risk assessment report are input into the reinforcement learning algorithm, so that the reinforcement learning algorithm simulates the selection of the optimal scientific and technological innovation path strategy under different scientific and technological innovation path strategies, and generates a long-term development trend forecast report and optimization suggestions for the optimal scientific and technological innovation path strategy.

2. The method according to claim 1, characterized in that The Bayesian network algorithm is used to process the uncertainty information of the scientific and technological innovation data in the scientific and technological innovation big data stream, and the scientific and technological innovation data is subjected to probabilistic inference processing according to the processing results to obtain a market value and market risk assessment report, including: Identify uncertainty information in scientific and technological innovation data, and initialize the prior probability distribution in the Bayesian network model based on the uncertainty information; Based on the prior probability distribution, a Bayesian network model is constructed, and the scientific and technological innovation achievement data is input into the Bayesian network model as evidence; Using the Bayesian update rule, updating the node probability distribution in the Bayesian network model according to the evidence to obtain a posterior probability distribution; The posterior probability distribution is used as a processing result, and based on the processing result, the market value and market risk of the scientific and technological achievement data are probabilistically inferred to generate a market value and market risk assessment report.

3. The method according to claim 2, characterized in that The probabilistic inference of the market value and market risk of the scientific and technological achievement data based on the processing results to generate a market value and market risk assessment report includes: According to the processing results, the market value of the scientific and technological innovation achievement data is sampled multiple times using a Monte Carlo simulation algorithm to obtain distribution characteristics of the market value, wherein the distribution characteristics include the expected value and variance of the market value; Analyzing the distribution characteristics using statistical methods to obtain a probability distribution of market value; Based on the probability distribution, the market risk of the scientific and technological innovation achievement data is evaluated by applying the risk value model to obtain the market risk level; In combination with the probability distribution and the market risk level, a multi-criteria decision analysis method is adopted to comprehensively evaluate the market performance of the scientific and technological achievements data and generate a market value and market risk assessment report.

4. The method according to claim 3, characterized in that The method combines the probability distribution and the market risk level, adopts a multi-criteria decision analysis method, comprehensively evaluates the market performance of the scientific and technological achievement data, and generates a market value and market risk assessment report, including: Using the probability distribution and the market risk level as evaluation criteria in a multi-criteria decision analysis; The evaluation criteria are hierarchically modeled by using the hierarchical analysis method to obtain the hierarchical modeling results, and the priority weight of each criterion is obtained by the expert scoring method to obtain the importance information of the evaluation criteria; The fuzziness of the evaluation criteria is processed by using a fuzzy comprehensive evaluation method, and the membership function of each evaluation criterion is determined according to the fuzziness; Based on the hierarchical modeling result, the evaluation criteria importance information and the membership function, a multi-criteria decision analysis model is constructed, and a solution is obtained based on the multi-criteria decision analysis model; Based on the solution results, the market performance of the scientific and technological innovation achievement data is comprehensively evaluated to generate a market value and market risk assessment report.

5. The method according to claim 1, characterized in that The said inputting the said innovation network effect quantitative index and the said market value and market risk assessment report into the reinforcement learning algorithm, so that the said reinforcement learning algorithm simulates the selection of the optimal scientific and technological innovation path strategy under different scientific and technological innovation path strategies, and generates the long-term development trend forecast report and optimization suggestions of the optimal scientific and technological innovation path strategy, including: Initializing the state space in the reinforcement learning algorithm according to the innovation network effect quantitative index and the market value and market risk assessment report, wherein the reinforcement learning algorithm includes a Q learning algorithm; Using the Q learning algorithm, construct the reward function under different scientific and technological innovation path strategies, associate the reward function with the state space, simulate different scientific and technological innovation path strategies, and update the Q value corresponding to different scientific and technological innovation path strategies through a trial and error mechanism; According to the updated Q values ​​corresponding to different science and technology innovation path strategies, the optimal science and technology innovation path strategy corresponding to the largest Q value is selected; Generate a long-term development trend forecast report for the optimal scientific and technological innovation path strategy, and based on the long-term development trend forecast report and combined with the optimal scientific and technological innovation path strategy, generate optimization suggestions for the optimal scientific and technological innovation path strategy.

6. The method according to claim 5, characterized in that The Q learning algorithm is used to construct a reward function under different scientific and technological innovation path strategies, and the reward function is associated with the state space, and different scientific and technological innovation path strategies are simulated, and the Q values ​​corresponding to different scientific and technological innovation path strategies are updated through a trial and error mechanism, including: Define all states in the state space and define reward functions for actions under each technological innovation path strategy; Initialize the Q value of each state and action combination in the state space by using the Q learning algorithm, and provide feedback on the action execution results under different scientific and technological innovation path strategies based on the reward function to update the Q value, and determine whether the updated Q value reflects the benefits after taking the action; If not, repeat the steps of providing feedback on the results of action execution under different scientific and technological innovation path strategies until the predetermined convergence state is reached, the iterative update is completed, and the updated Q value is obtained. The predetermined convergence state is that the updated Q value reflects the benefit after taking the action, wherein the updated Q value in each iterative step is used to reflect the benefit after taking the action.

7. The method according to claim 1, characterized in that The graph theory algorithm is used to analyze the intensity of the collaborative network between the scientific and technological innovation entities and the efficiency of knowledge flow in the data identification to generate a quantitative index of the innovation network effect. The graph theory algorithm includes social network analysis, community detection algorithm, web page ranking algorithm and random walk algorithm. The intensity of the collaborative network between scientific and technological innovation entities characterizes the degree of mutual cooperation between different scientific and technological innovation entities, including: Using the degree centrality, closeness centrality and betweenness centrality algorithms in social network analysis, the importance and influence of scientific and technological innovation entities in the network are calculated, and quantitative indicators of the importance and influence of scientific and technological innovation entities in the network are obtained; A community detection algorithm is used to process the quantitative indicators of the importance and influence of the scientific and technological innovation subject in the network, to identify the close cooperation groups and the boundaries of the close cooperation groups among the scientific and technological innovation subjects, and to obtain the information of the close cooperation groups; Based on the information of the close cooperation group, the importance of the scientific and technological innovation subject in the information dissemination is analyzed by using the web page ranking algorithm to determine the subject information that plays a key role in the information diffusion process; Applying a random walk algorithm to process the propagation path of the subject information, evaluating the knowledge flow efficiency, calculating the probability of information reaching from one node to another, and obtaining the knowledge flow efficiency evaluation result; Based on the quantitative indicators of the importance and influence of the scientific and technological innovation entities in the network, the information of closely cooperating groups, the subject information and the evaluation results of knowledge flow efficiency, quantitative indicators of innovation network effects are generated.

8. A big data processing system for implementing hybrid data analysis, characterized in that: include: The receiving module is used to receive the scientific and technological innovation big data stream from different data sources; An encryption module, used to assign a unique identifier to the scientific and technological innovation big data stream using a hash algorithm, and encrypt the scientific and technological innovation big data stream according to the unique identifier to obtain a data identifier; An analysis module, for analyzing the intensity of the collaborative network between the scientific and technological innovation entities and the efficiency of knowledge flow in the data identifier according to a graph theory algorithm, and generating a quantitative index of the innovation network effect. The graph theory algorithm includes a social network analysis, a community detection algorithm, a web page ranking algorithm, and a random walk algorithm. The intensity of the collaborative network between the scientific and technological innovation entities represents the degree of mutual collaboration between different scientific and technological innovation entities. A processing module, used to process the uncertainty information of the scientific and technological innovation data in the scientific and technological innovation big data stream using a Bayesian network algorithm, and perform probabilistic inference processing on the scientific and technological innovation data according to the processing results to obtain a market value and market risk assessment report; A generation module is used to input the innovation network effect quantitative indicators and the market value and market risk assessment report into a reinforcement learning algorithm, so that the reinforcement learning algorithm simulates the selection of the optimal scientific and technological innovation path strategy under different scientific and technological innovation path strategies, and generates a long-term development trend forecast report and optimization suggestions for the optimal scientific and technological innovation path strategy.

9. A computing device, characterized in that It comprises a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to implement a scientific and technological innovation big data evaluation method for industrial brain as described in any one of claims 1 to 7.

10. A computer storage medium, characterized in that: A computer program is stored, and when the computer program is executed by a computer, a scientific and technological innovation big data evaluation method for industrial brain as described in any one of claims 1 to 7 is implemented.

Images