A data multidimensional evaluation method and device and electronic equipment

Abstract

The application discloses a data multidimensional evaluation method and device and electronic equipment. The method comprises the following steps: collecting at least one piece of data; inputting the at least one piece of data into a pre-trained multidimensional fusion model to fuse strategic value, scientific value and market value in the at least one piece of data, and outputting a comprehensive value index on a strategic dimension, a scientific dimension and a market dimension, wherein the fusion model is trained by the strategic value, the scientific value and the market value of training data and the comprehensive value index; and generating a multidimensional evaluation result of the at least one piece of data according to the comprehensive value index. The method can more intelligently, dynamically and credibly represent and evaluate data value, and realizes intelligent prediction and scientific decision of target forward-looking technology application value.

Description

Technical Field

[0001] This application relates to the field of computer technology, and in particular to a method for multi-dimensional data evaluation, an electronic device, and a device for multi-dimensional data evaluation. Background Technology

[0002] Data value assessment involves multiple sources of information, including technology patents, technical papers, policy environment, market data, and industry chain data. With the rapid advancement of digital technologies such as big data, knowledge graphs, and artificial intelligence models, the role of science and technology intelligence value assessment as a science and technology management method and tool is becoming increasingly prominent in assisting and supporting national science and technology policy formulation, high-value technology foresight, industry chain risk warning, and socio-economic monitoring.

[0003] Currently, relevant technologies mainly utilize knowledge graphs to visualize and intuitively present target scientific and technological achievements, and also obtain data value assessment results through neural network model training. However, these methods suffer from problems such as limited assessment techniques, insufficient assessment mechanisms, and poor predictive capabilities in application value assessment. Therefore, there is an urgent need to construct an intelligent, interpretable, and dynamic multi-dimensional intelligent fusion assessment system for data to better support scientific decision-making and technological innovation. Summary of the Invention

[0004] This application aims to address, at least to some extent, one of the technical problems in related technologies. To this end, the first objective of this application is to propose a multi-dimensional data evaluation method. This method involves collecting at least one data point and inputting it into a pre-trained multi-dimensional fusion model to integrate the strategic, scientific, and market values ​​of the at least one data point. The model then outputs a comprehensive value index across the strategic, scientific, and market dimensions. The fusion model is trained using the strategic, scientific, and market values ​​of the training data and the comprehensive value index. Based on the comprehensive value index, a multi-dimensional evaluation result for at least one data point is generated. This allows for a more intelligent, dynamic, and reliable representation and evaluation of data value, enabling intelligent prediction and scientific decision-making regarding the application value of forward-looking technologies.

[0005] The second objective of this application is to propose a multi-dimensional data evaluation method.

[0006] The third objective of this application is to propose an electronic device.

[0007] The fourth objective of this application is to provide a computer-readable storage medium.

[0008] The fifth objective of this application is to propose a multi-dimensional data evaluation device.

[0009] The sixth objective of this application is to propose a multi-dimensional data evaluation device.

[0010] To achieve the above objectives, the first aspect of this application proposes a multi-dimensional data evaluation method applied in the model usage stage. The method includes the following steps: collecting at least one data point; inputting the at least one data point into a pre-trained multi-dimensional fusion model to fuse the strategic value, scientific value, and market value of the at least one data point, and outputting a comprehensive value index in the strategic, scientific, and market dimensions, wherein the fusion model is trained using the strategic value, scientific value, and market value of the training data and the comprehensive value index; and generating a multi-dimensional evaluation result for the at least one data point based on the comprehensive value index.

[0011] According to the data multi-dimensional evaluation method of this application embodiment, at least one data point is collected and input into a pre-trained multi-dimensional fusion model to integrate the strategic value, scientific value, and market value of the at least one data point, and output a comprehensive value index in the strategic, scientific, and market dimensions. The fusion model is trained by the strategic value, scientific value, market value, and comprehensive value index of the training data, and generates a multi-dimensional evaluation result of at least one data point based on the comprehensive value index. Thus, this method can more intelligently, dynamically, and reliably represent and evaluate the value of data, and realize intelligent prediction and scientific decision-making of the application value of forward-looking technologies.

[0012] To achieve the above objectives, a second aspect of this application proposes a multi-dimensional data evaluation method applied in the model building stage. The method includes the following steps: collecting a training sample set, wherein the training sample set includes multiple training data points; calculating the strategic value, scientific value, and market value of each training data point to generate a comprehensive value index across the strategic, scientific, and market dimensions; and training a model using the strategic value, scientific value, and market value of each training data point and the comprehensive value index to construct a multi-dimensional fusion model for outputting the comprehensive value index of data across the strategic, scientific, and market dimensions.

[0013] According to the data multi-dimensional evaluation method of this application embodiment, a training sample set is collected, wherein the training sample set includes multiple training data, the strategic value, scientific value, and market value of each training data are calculated to generate a comprehensive value index in the strategic, scientific, and market dimensions, and the model is trained using the strategic value, scientific value, market value, and comprehensive value index of each training data to construct a multi-dimensional fusion model for outputting the comprehensive value index of data in the strategic, scientific, and market dimensions. Thus, this method can more intelligently, dynamically, and reliably represent and evaluate the value of data, and realize intelligent prediction and scientific decision-making of the application value of forward-looking technologies.

[0014] In addition, the data multi-dimensional evaluation method according to the above embodiments of this application may also have the following additional technical features:

[0015] According to one embodiment of this application, calculating the strategic value, scientific value, and market value of each piece of training data includes: determining a reference database based on the training sample set; quantifying the reference database based on a combination of big data analysis, knowledge graphs, and artificial intelligence models to determine target indicators; and determining the strategic value, scientific value, and market value based on the target indicators.

[0016] According to one embodiment of this application, the target indicators include technological strategic alignment, policy support intensity, industrial chain synergy, and patent barrier index. The step of determining the strategic value based on the target indicators includes: determining the strategic value by weighting the strategic alignment with a first preset coefficient, the policy support intensity with a second preset coefficient, the industrial chain synergy with a third preset coefficient, and the patent barrier index with a fourth preset coefficient, wherein the weighted product is a power-weighted product.

[0017] According to one embodiment of this application, the target indicators include technology maturity, innovation breakthrough index, R&D feasibility, and technology diffusion degree. The step of determining the scientific value based on the target indicators includes: determining the scientific value by summing the technology maturity, the innovation breakthrough index, R&D feasibility, and technology diffusion degree with their respective weighting coefficients.

[0018] According to one embodiment of this application, the target indicators include market size index, application scenario breadth, competition intensity, and investment enthusiasm. The step of determining the market value based on the target indicators includes: taking the hyperbolic tangent of the weighted sum of the market size index with a fifth preset coefficient, the application scenario breadth with a sixth preset coefficient, the competition intensity with a seventh preset coefficient, and the investment enthusiasm with an eighth preset coefficient as the strategic value.

[0019] According to one embodiment of this application, a multi-dimensional fusion model for outputting a comprehensive value index of data across strategic, scientific, and market dimensions is constructed using the following formula:

[0020]

[0021] Where V(t) represents the comprehensive value index at time t, β represents the balance factor, γ represents the adjustment factor, λ represents the time decay coefficient, SV(t) represents the strategic value at time t, SJ(t) represents the scientific value at time t, and MF(τ) represents the market value at time τ. This represents the weighted integral of the market value over the past n years.

[0022] According to one embodiment of this application, determining a reference database based on the training sample set includes: performing data cleaning and labeling on the training sample set; classifying the processed training sample set to obtain multiple types of sub-databases; performing structuring and standardization on each type of sub-database to obtain a corresponding sub-reference database; and merging multiple sub-reference databases to obtain the reference database.

[0023] To achieve the above objectives, an electronic device is proposed in the third aspect of this application, including a memory, a processor, and a program stored in the memory and executable on the processor. When the processor executes the program, it implements the above-described multi-dimensional data evaluation method.

[0024] The electronic device according to the embodiments of this application, by executing the above-described multi-dimensional data evaluation method, can more intelligently, dynamically, and reliably characterize and evaluate the value of data, and realize intelligent prediction and scientific decision-making of the application value of forward-looking technologies.

[0025] To achieve the above objectives, a fourth aspect of this application provides a computer-readable storage medium having a program stored thereon that, when executed by a processor, implements the above-described multi-dimensional data evaluation method.

[0026] The computer-readable storage medium according to the embodiments of this application, by implementing the above-described multi-dimensional data evaluation method during execution, can more intelligently, dynamically, and reliably characterize and evaluate the value of data, and realize intelligent prediction and scientific decision-making of the application value of forward-looking technologies.

[0027] To achieve the above objectives, a fifth aspect of this application proposes a multi-dimensional data evaluation device applied in the model usage stage. The device includes: an acquisition module for collecting at least one data point; a simulation module for inputting the at least one data point into a pre-trained multi-dimensional fusion model to fuse the strategic value, scientific value, and market value of the at least one data point and output a comprehensive value index in the strategic, scientific, and market dimensions, wherein the fusion model is trained from the strategic value, scientific value, and market value of the training data and the comprehensive value index; the simulation module is further used to generate a multi-dimensional evaluation result of the at least one data point based on the comprehensive value index.

[0028] According to the data multi-dimensional evaluation device of this application embodiment, the acquisition module is used to collect at least one data point, and the simulation module is used to input the at least one data point into a pre-trained multi-dimensional fusion model to fuse the strategic value, scientific value, and market value of the at least one data point, and output a comprehensive value index in the strategic, scientific, and market dimensions. The fusion model is trained using the strategic value, scientific value, and market value of the training data and the comprehensive value index. The simulation module is also used to generate a multi-dimensional evaluation result for the at least one data point based on the comprehensive value index. Therefore, this device can more intelligently, dynamically, and reliably characterize and evaluate the value of data, enabling intelligent prediction and scientific decision-making regarding the application value of forward-looking technologies.

[0029] To achieve the above objectives, a sixth aspect of this application proposes a multi-dimensional data evaluation device applied in the model building stage. The device includes: an acquisition module for collecting a training sample set, wherein the training sample set includes multiple training data points; a calculation module for calculating the strategic value, scientific value, and market value of each training data point to generate a comprehensive value index across the strategic, scientific, and market dimensions; and a training module for training a model using the strategic value, scientific value, and market value of each training data point and the comprehensive value index to construct a multi-dimensional fusion model for outputting the comprehensive value index of data across the strategic, scientific, and market dimensions.

[0030] According to the data multi-dimensional evaluation device of this application embodiment, the acquisition module is used to collect a training sample set, wherein the training sample set includes multiple training data points. The calculation module is used to calculate the strategic value, scientific value, and market value of each training data point to generate a comprehensive value index in the strategic, scientific, and market dimensions. The training module is used to train a model using the strategic value, scientific value, market value, and comprehensive value index of each training data point to construct a multi-dimensional fusion model for outputting the comprehensive value index of data in the strategic, scientific, and market dimensions. Therefore, this device can more intelligently, dynamically, and reliably characterize and evaluate the value of data, realizing intelligent prediction and scientific decision-making regarding the application value of forward-looking technologies.

[0031] The beneficial effects of this application are:

[0032] (1) This application inputs at least one piece of data into a pre-trained multi-dimensional fusion model to integrate the strategic value, scientific value and market value of at least one piece of data, and outputs a comprehensive value index in the strategic dimension, scientific dimension and market dimension. The fusion model is trained by the strategic value, scientific value and market value and comprehensive value index of the training data, and generates a multi-dimensional evaluation result of at least one piece of data based on the comprehensive value index. It can more intelligently, dynamically and reliably represent and evaluate the value of the data, and realize intelligent prediction and scientific decision-making of the application value of the target forward-looking technology.

[0033] Additional aspects and advantages of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this application. Attached Figure Description

[0034] Figure 1 This is a flowchart of a data multi-dimensional evaluation method according to an embodiment of this application;

[0035] Figure 2 This is a flowchart of a multi-dimensional data evaluation method according to another embodiment of this application;

[0036] Figure 3 Here is a flowchart of a multi-dimensional data evaluation method based on a specific example of this application;

[0037] Figure 4 This is a block diagram of an electronic device according to an embodiment of this application;

[0038] Figure 5 This is a block diagram of a data multi-dimensional evaluation device according to an embodiment of this application;

[0039] Figure 6 This is a block diagram of a data multidimensional evaluation device according to another embodiment of this application. Detailed Implementation

[0040] The embodiments of this application are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this application, and should not be construed as limiting this application.

[0041] The following description, with reference to the accompanying drawings, outlines a data multidimensional evaluation method, electronic device, and data multidimensional evaluation apparatus according to embodiments of this application.

[0042] Figure 1 This is a flowchart of a data multi-dimensional evaluation method according to an embodiment of this application.

[0043] like Figure 1 As shown, the data multi-dimensional evaluation method of this application embodiment may include the following steps:

[0044] S1, collect at least one data point.

[0045] S2, input at least one data point into a pre-trained multi-dimensional fusion model to integrate the strategic value, scientific value and market value of at least one data point, and output a comprehensive value index in the strategic dimension, scientific dimension and market dimension. The fusion model is trained by the strategic value, scientific value and market value and comprehensive value index of the training data.

[0046] S3 generates a multi-dimensional evaluation result based on at least one data point, according to the comprehensive value index.

[0047] Specifically, during the model usage phase, at least one data point is collected first. This data can come from various sources, depending on the evaluation target and field (e.g., the vehicle field). For example, various types of data (including but not limited to scientific and technological intelligence) can come from government websites, industry reports, academic databases, patent databases (such as the website of the State Intellectual Property Office), etc. Alternatively, it can come from paid market research reports, data provided by industry analysis agencies, or from the company's own R&D reports, market research results, technical documents, etc., or from public opinion data on industry news websites and social media platforms (such as Weibo and Twitter).

[0048] When collecting data, readily available web crawling tools can be developed or used to collect data from public websites. For example, policy documents can be crawled from government websites, and technical papers from academic databases. Alternatively, APIs (Application Programming Interfaces) provided by data providers can be used to acquire data in batches. For example, patent information can be obtained through the API of a patent database. For data that cannot be obtained automatically, it can be collected manually, such as extracting key information from internal documents.

[0049] In addition, the collected data can be preprocessed to ensure its quality and usability, such as removing duplicate data records, converting the data into a uniform format, such as unifying the date format to "YYYY-MM-DD", correcting errors or outliers in the data, handling missing data, and classifying and labeling the data, such as labeling the data type (technology patents, market reports, etc.) and industry (new energy, artificial intelligence, etc.).

[0050] After acquiring the data, at least one data point can be input into a pre-trained multi-dimensional fusion model to integrate the strategic, scientific, and market values ​​of that data point, outputting a comprehensive value index across the strategic, scientific, and market dimensions. The fusion model is trained using the strategic, scientific, and market values ​​of the training data, along with the comprehensive value index. In other words, the model's main function is to integrate the strategic, scientific, and market values ​​of the input data and output a comprehensive value index. For example, the input layer of the model can receive pre-processed data, including basic information (such as title, author, and publication date) and key indicators (such as technology maturity and market size). The feature extraction layer can extract key features of the data using algorithms (such as knowledge graphs and big data analysis). The value assessment layer can calculate the strategic, scientific, and market values ​​of the data separately. The fusion layer integrates the values ​​of the three dimensions to calculate the comprehensive value index. Finally, the output layer outputs the comprehensive value index, representing the data's comprehensive value across the strategic, scientific, and market dimensions.

[0051] After obtaining the comprehensive value index, a multi-dimensional evaluation result for at least one data point can be generated based on the index. For example, the evaluation result may include an indication of the data's comprehensive value across three dimensions: strategy, science, and market. It may also include classifying the data into different evaluation levels based on the range of the comprehensive value index, such as high value, medium value, and low value. Furthermore, it may include generating a detailed evaluation report, such as basic data information, a detailed analysis of the value of each dimension, and a comprehensive evaluation conclusion. Thus, by analyzing the value of each dimension, potential risks within the data can be identified. For example, if a technology has a low market value, further market research or adjustments to the technological direction may be necessary. This can also help companies or institutions concentrate their limited resources on the most promising areas, improving resource utilization efficiency.

[0052] Furthermore, to facilitate understanding and application, the evaluation results can be displayed using visualization tools. For example, bar charts can be used to show the value distribution of data across the three dimensions of strategy, science, and market; radar charts can visually demonstrate the overall performance of data across these three dimensions; and heatmaps can show the intensity of data value across different dimensions.

[0053] Therefore, the above steps can address the issues of limited quantitative indicators and weak evaluation mechanisms in data assessment. Simultaneously, it proposes multi-dimensional quantitative indicators encompassing strategy, science, and market, and optimizes the data assessment process through a fusion model, achieving a comprehensive, dynamic, and scientific evaluation of data. This process not only provides a comprehensive value index of data across the strategic, scientific, and market dimensions but also offers decision-makers detailed evaluation reports and visualizations, thereby supporting scientific decision-making and optimal resource allocation.

[0054] In summary, the data multi-dimensional evaluation method according to the embodiments of this application collects at least one data point and inputs it into a pre-trained multi-dimensional fusion model to integrate the strategic value, scientific value, and market value of the at least one data point, and outputs a comprehensive value index in the strategic, scientific, and market dimensions. The fusion model is trained by the strategic value, scientific value, market value, and comprehensive value index of the training data, and generates a multi-dimensional evaluation result of at least one data point based on the comprehensive value index. Therefore, this method can more intelligently, dynamically, and reliably represent and evaluate the value of data, and realize intelligent prediction and scientific decision-making on the application value of forward-looking technologies.

[0055] Corresponding to the above embodiments, this application also proposes a multi-dimensional data evaluation method.

[0056] like Figure 2 As shown, the data multi-dimensional evaluation method of this application embodiment may include the following steps:

[0057] S11, Collect the training sample set, which includes multiple training data points.

[0058] S12 calculates the strategic value, scientific value, and market value of each training data point to generate a comprehensive value index across the strategic, scientific, and market dimensions.

[0059] S13 utilizes the strategic value, scientific value, market value, and comprehensive value index of each training data point to train the model, thereby constructing a multi-dimensional fusion model for outputting the comprehensive value index of data in the strategic, scientific, and market dimensions.

[0060] Specifically, in the model building phase, the first step is to collect a training sample set. The training sample set refers to the dataset used for model training. It includes multiple training data points, such as vehicle-related technical patents, technical papers, market reports, and policy documents. Each training data point should include the following: basic information (e.g., data source, publication date, author, field of study); specific content (e.g., technical description, market analysis, policy provisions); and annotation information (e.g., the type of data (e.g., technical patents, papers), the industry (e.g., new energy, artificial intelligence), and possible tags (e.g., technology maturity, market potential). When collecting the training sample set, publicly available data can be collected from government websites, academic databases, and industry report websites through web scraping, or professional databases can be obtained for a fee, or offline expert interviews and practical surveys can be conducted.

[0061] After acquiring multiple training data points, the strategic, scientific, and market values ​​of each data point can be calculated to generate a comprehensive value index across these three dimensions. For example, typical algorithms such as big data, knowledge graphs, and artificial intelligence models can be used to quantify the training data, obtaining quantified values ​​for strategic, scientific, and market values, thus generating a comprehensive value index across these dimensions. The strategic, scientific, and market values ​​of each training data point, along with the comprehensive value index, can then be used to train a model to construct a multi-dimensional fusion model for outputting the comprehensive value index of data across these three dimensions.

[0062] For example, the strategic value, scientific value, market value, and comprehensive value index of each training data point calculated above can be used as training data. This data can be divided into training and validation sets for model training and validation. Model selection can be based on machine learning models such as linear regression, decision trees, and random forests; deep learning models such as neural networks (including convolutional neural networks and recurrent neural networks); or hybrid models combining multiple algorithms (such as big data analytics, knowledge graphs, and machine learning). Key features of the data, such as technical keywords, market data indicators, and policy relevance, should be extracted and normalized to ensure the stability of model training. The training set can be used to train the model, and the validation set can be used to adjust model parameters (such as learning rate and regularization coefficient). Cross-validation and other methods can be used to evaluate model performance, avoiding overfitting or underfitting.

[0063] The trained model can be applied to new data, for example, to calculate its comprehensive value index across strategic, scientific, and market dimensions in real time. Decision support: It provides scientific basis for science and technology policy formulation, technology investment decisions, and supply chain risk warnings. Furthermore, model parameters can be regularly updated or the model retrained based on changes in the market and technological environment to ensure its adaptability and accuracy.

[0064] Therefore, an efficient, scientific, and dynamically adaptable multi-dimensional fusion model can be constructed to comprehensively assess the strategic, scientific, and market value of data, providing strong support for scientific decision-making.

[0065] According to one embodiment of this application, calculating the strategic value, scientific value, and market value of each piece of training data includes: determining a reference database based on a training sample set; quantifying the reference database using indicators based on at least one of big data analysis, knowledge graphs, and artificial intelligence models to determine target indicators; and determining the strategic value, scientific value, and market value based on the target indicators.

[0066] Furthermore, according to one embodiment of this application, determining a reference database based on a training sample set includes: cleaning and labeling the training sample set; classifying the processed training sample set to obtain multiple types of sub-databases; structuring and standardizing each type of sub-database to obtain a corresponding sub-reference database; and merging multiple sub-reference databases to obtain a reference database.

[0067] Specifically, when calculating the strategic, scientific, and market value of each piece of training data, a reference database can first be determined using the training sample set. That is, when determining the reference database based on the training sample set, steps such as data cleaning and labeling can be performed on the training sample set to construct a reference database that meets the requirements of hybrid computing. Data cleaning removes duplicate data, handles missing values, standardizes data formats, corrects data errors, and handles outliers. For example, for text data, meaningless symbols and spaces are removed; for numerical data, missing values ​​are filled or records with excessive missing values ​​are deleted. Data labeling involves classifying and labeling data as needed, such as labeling scientific data as technical patents, technical papers, policy documents, etc., or labeling it with information such as its industry and discipline. Labeling helps to better classify and process data subsequently. In short, data cleaning and labeling are standardization processes that remove duplicates from raw data, handle missing values, standardize data formats, correct data errors, and handle outliers, which can significantly improve the quality of raw data.

[0068] Furthermore, given the large volume of multi-source information, this application's implementation case requires classifying and filtering the training sample set to obtain multiple types of sub-databases. That is, based on annotation information or data characteristics, the processed training sample set is divided into multiple sub-databases. For example, it can be divided into technical patent sub-databases, technical paper sub-databases, policy document sub-databases, etc., according to data type, or into new energy vehicle sub-databases, artificial intelligence sub-databases, etc. Each type of sub-database undergoes structuring and standardization processing to obtain a corresponding sub-reference database. Multiple sub-reference databases are then merged to obtain a reference database. Structuring processing converts unstructured or semi-structured data into structured data. For example, for technical papers, information such as title, author, abstract, keywords, and publication date are extracted and stored in structured tables; for technical patents, information such as patent number, invention name, applicant, application date, and patent abstract are extracted. Standardization processing unifies the data format and unit of measurement. For example, the date format is standardized to "YYYY-MM-DD", and the currency unit is standardized to RMB or USD, etc. Finally, the various sub-reference databases that have undergone structured and standardized processing are merged into a complete reference database for subsequent unified analysis and processing.

[0069] For example, market data on new energy vehicles can be categorized and filtered according to dimensions such as product type (pure electric, range-extended, hybrid), market region (North China, South China; domestic, international, etc.), and consumer group (consumers of different ages, interests, and family attributes) to form sub-databases of different categories or clusters. This facilitates centralized analysis and comparison of similar data. Alternatively, data can be categorized according to subject matter, such as technical data (LiDAR, smart cockpit, chassis, etc.), market data (market penetration rate, investment status, etc.), and policy data (policy documents published by the National Development and Reform Commission, the Ministry of Industry and Information Technology, the Ministry of Science and Technology, etc., as well as provincial and municipal policy support). After structuring and standardizing various sub-databases, they can be merged into a unified reference database, improving the processing efficiency for quantifying indicators for assessing the value of scientific and technological data.

[0070] Next, the reference database can be quantified using at least one of the following methods: big data analysis, knowledge graphs, and artificial intelligence models, to determine the target indicators. In other words, in the embodiments of this application, multiple combinations of typical algorithms such as big data analysis, knowledge graphs, and artificial intelligence models can be used in series, parallel, or simultaneously to quantify the reference database and determine the target indicators. For example, the k-means algorithm can be used to cluster the data in the reference database to discover potential patterns and similarities. For example, technology patents can be clustered according to technical fields or application scenarios to better analyze their strategic and market value, and association rules can be mined using algorithms such as Apriori to find the relationships between different data. For example, a correlation can be found between the publication of certain technical papers and the application of specific technology patents, thus providing a basis for assessing scientific value. Constructing a knowledge graph can transform entities (such as technical concepts, enterprises, research institutions, etc.) in the reference database and the relationships between them (such as cooperation relationships, citation relationships, etc.) into a knowledge graph. For example, information such as inventors, applicants, and other cited patents in technology patents can be constructed into a knowledge graph for analyzing the influence of the technology and the degree of synergy in the industrial chain. Knowledge graphs can be used for reasoning and querying to obtain relevant information about target data. For example, a knowledge graph can be used to query the upstream and downstream supply chain relationships of a technology, providing data support for the supply chain synergy indicator in strategic value. Artificial intelligence models, such as linear regression, decision trees, and random forests, can be used to train data in a reference database to predict target indicators. For example, a linear regression model can be trained based on historical data to predict the technology-strategy fit indicator in the strategic value of technology patents.

[0071] Finally, strategic value, scientific value, and market value can be determined based on the target indicators. Strategic value and scientific value are two fundamental dimensions for evaluating the application value of the target's forward-looking technology. Strategic value reflects the alignment of the technology with national policies and corporate strategies, while scientific value reflects the maturity and innovativeness of the technology. Market value reflects the future development trend of the target's forward-looking technology application value. For example, the relationship between target indicators and strategic, scientific, and market values ​​can be determined in advance. Once the target indicators are determined, the corresponding relationships can be directly invoked to obtain the strategic, scientific, and market values.

[0072] It should be noted that big data analysis algorithms may include one or more of the following: partitioning-based algorithms such as k-means, k-medoids, and clarans; hierarchical algorithms such as Birch, Cure, and Chameleon; density-based algorithms such as DBSCAN, Optics, and Denclue; and grid-based algorithms such as Sting, Clique, and Wave-Cluster. Knowledge graph algorithms may include one or more of the following: knowledge representation and reasoning algorithms, entity recognition and extraction algorithms, various recommendation algorithms, graph database algorithms, time series analysis algorithms, label propagation algorithms, and maximum flow algorithms. Artificial intelligence model algorithms may include one or more of the following: linear regression, logistic regression, decision trees, support vector machines, random forests, k-nearest neighbors, k-means, ensemble learning, reinforcement learning, and neural network algorithms.

[0073] According to one embodiment of this application, the target indicators include technological strategic alignment, policy support intensity, industrial chain synergy, and patent barrier index. Determining strategic value based on these target indicators includes: determining strategic value through a weighted product of strategic alignment with a first preset coefficient, policy support intensity with a second preset coefficient, industrial chain synergy with a third preset coefficient, and the patent barrier index with a fourth preset coefficient, wherein the weighted product is a power-weighted product. The first, second, third, and fourth preset coefficients can be determined according to actual circumstances.

[0074] Specifically, the target indicators may include the degree of alignment between technology and strategy, the intensity of policy support, the degree of synergy of the industrial chain, and the patent barrier index. When determining the strategic value based on the target indicators, the strategic value can be determined by a weighted product (power-weighted product) of the degree of alignment with the first preset coefficient, the intensity of policy support with the second preset coefficient, the degree of synergy of the industrial chain with the third preset coefficient, and the patent barrier index with the fourth preset coefficient.

[0075] For example, through formula The strategic value is determined by defining SV(t) as the numerical value of the strategic value dimension indicator at time t, Si as the various indicators of the strategic value dimension (strategic fit, policy support intensity, industrial chain synergy, and patent barrier index), and Wi as the corresponding weights, which can be derived from historical data analysis, expert evaluation, and model validation. In the implementation case of this application, the strategic fit S1 reflects the degree of matching between the target technology and the company's existing technological roadmap, and can be determined through text analysis algorithms (such as TF-IDF (Term Frequency-Inverse)). Document Frequency (Word Frequency-Inverse Document Frequency), Word2Vec, etc., are used to extract the degree of matching between the target technology and the enterprise strategy. The weight value (first preset coefficient) can be adjusted between 0.25 and 0.35 based on national policy requirements and the enterprise's strategic priorities. This weight value can then be combined with the calculated technology-strategy fit. Policy support intensity S2 reflects the government's support for the target technology. This support intensity (0-1) can be analyzed using machine learning algorithms, such as sentiment analysis and keyword extraction, to quantify the policy's support for the target technology. Its weight value (second preset coefficient) can be adjusted based on changes in the policy environment. The weight value (s3) can be adjusted between 0.15 and 0.25 to calculate the strength of policy support based on the weighted values. The industry chain synergy degree (S3) can be calculated by constructing the connection relationships between upstream and downstream nodes in the industry chain using knowledge graph algorithms (such as graph databases and relation extraction). Its weight value (the third preset coefficient) can be adjusted between 0.2 and 0.3 based on the industry chain structure, thus calculating the industry chain synergy degree. The patent barrier index (S4) reflects the strength of intellectual property protection for the target technology. It can be calculated using patent analysis algorithms (such as patent similarity calculation and patent citation analysis), thus calculating the patent barrier index based on the weighted values. For example, it can be calculated by multiplying the size of the technology patent family by the citation rate. Its weight value (the fourth preset coefficient) can be adjusted between 0.2 and 0.3 based on the strength of intellectual property protection.

[0076] According to one embodiment of this application, the target indicators include technology maturity, innovation breakthrough index, R&D feasibility and technology diffusion degree. The scientific value is determined based on the target indicators, including: determining the scientific value by summing the technology maturity, innovation breakthrough index, R&D feasibility and technology diffusion degree with their respective weight coefficients.

[0077] Specifically, the target indicators may also include technology maturity, innovation breakthrough index, R&D feasibility and technology diffusion. When determining the scientific value based on the target indicators, the scientific value can be determined by summing the technology maturity, innovation breakthrough index, R&D feasibility and technology diffusion with their respective weight coefficients.

[0078] For example, through formula Determine the scientific value. Here, SJ(t) is the value of the scientific judgment dimension index (scientific value) at time t, Jj are the various indicators of the scientific judgment dimension (technology maturity, innovation breakthrough index, R&D feasibility and technology diffusion), and Wj are the corresponding weights, which can be given empirical values ​​based on historical data analysis, expert evaluation, model validation, etc. In this application implementation case, the technology maturity J1 reflects the development stage of the target technology and is a core indicator for scientific judgment. It can be obtained by quantifying (0-1) according to the Gartner curve, and its weight value can be adjusted between 0.3 and 0.4 depending on the technology development stage. The innovation breakthrough index J2 reflects the innovativeness and disruptiveness of the target technology and can be calculated by multiplying the growth rate of papers in the past three years by the frequency of disruptive keywords. Its weight value can be adjusted between 0.25 and 0.35 depending on the technological innovation. The R&D feasibility J3 reflects the feasibility and resource requirements of the target technology in the R&D process and can be calculated based on the expert evaluation matrix (equipment, talent, or funding availability). Its weight value can be adjusted between 0.15 and 0.25 depending on the R&D resource requirements. The technology diffusion degree J4 reflects the degree of dissemination and application of the target technology in the industry and can be calculated by the diffusion path entropy value from the knowledge graph. Its weight value can be adjusted between 0.1 and 0.2 depending on the degree of technology diffusion.

[0079] According to one embodiment of this application, the target indicators include market size index, application scenario breadth, competition intensity, and investment enthusiasm. Determining market value based on these target indicators includes: using the hyperbolic tangent of the weighted sum of the market size index with a fifth preset coefficient, the application scenario breadth with a sixth preset coefficient, the competition intensity with a seventh preset coefficient, and the investment enthusiasm with an eighth preset coefficient as the strategic value. The fifth, sixth, seventh, and eighth preset coefficients can be determined according to actual circumstances.

[0080] Specifically, the target indicators include market size index, application scenario breadth, competition intensity, and investment enthusiasm. When determining market value based on the target indicators, the hyperbolic tangent of the result of weighted summation of market size index with the fifth preset coefficient, application scenario breadth with the sixth preset coefficient, competition intensity with the seventh preset coefficient, and investment enthusiasm with the eighth preset coefficient can be used as strategic value.

[0081] For example, through formula Determine the strategic value. Here, MF(t) is the market value dimension indicator (strategic value) at time t, Mk represents various indicators within the market value dimension (market size index, application scenario breadth, competition intensity, and investment enthusiasm), and Wk represents the corresponding weights, which can be derived from historical data analysis, expert evaluation, and model validation. In this application's implementation case, the market size index M1 reflects the market potential and growth prospects of the target technology. It can be obtained by collecting and organizing market data from relevant industries using big data analytics, including historical sales, growth rates, and market share, and then using time series analysis (such as the ARIMA model) to predict future market size. For example, the compound annual growth rate (CAGR) over the next five years can be predicted, and its weight (the fifth preset coefficient) can be adjusted between 0.4 and 0.6 depending on the characteristics of the target industry and market; the breadth of application scenarios (M2) reflects the potential of the technology to be applied in different fields, and can be measured by the PageRank value of the application nodes in the knowledge graph, and its weight (the sixth preset coefficient) can be adjusted between 0.2 and 0.4 depending on the potential of the technology's application; the intensity of competition (M3) reflects the degree of competition in the target technology market, and can be measured by the Herfindahl-Hirschman Index (HHI) and the technology similarity matrix, and its weight (the seventh preset coefficient) can be adjusted between 0.1 and 0.2; the investment enthusiasm (M4) reflects the degree of attention the capital market pays to the target technology, and can be measured by the amount of venture capital events and the number of strategic alliances. For example, an investment knowledge graph can be constructed, including nodes of investment institutions, investment projects, and investment amounts, and the investment enthusiasm can be calculated by querying and analyzing the graph database, and its weight (the eighth preset coefficient) can be adjusted between -0.1 and 0 depending on the activity of the capital market (negative values ​​indicate the negative impact of overheated investment).

[0082] According to one embodiment of this application, a multi-dimensional fusion model for outputting a comprehensive value index of data across strategic, scientific, and market dimensions is constructed using the following formula:

[0083]

[0084] Where V(t) represents the comprehensive value index at time t, β represents the balancing factor, γ represents the adjustment factor, λ represents the time decay coefficient, SV(t) represents the strategic value at time t, SJ(t) represents the scientific value at time t, and MF(τ) represents the market value at time τ. It represents the weighted integral of the market value over the past n years.

[0085] Specifically, the comprehensive value index of data can be intelligently, reliably, and dynamically represented by multi-dimensional indicators calculated according to a specific fusion model. For example, the comprehensive value index can be determined by the above formula (1). Among them, the balance factor β reflects the current status of the development of forward-looking technologies based on the combined influence of strategic value and scientific value. It can be adjusted between 0.5 and 0.7 according to the specific industry and technology characteristics. The adjustment factor γ is used to balance the impact of the current status of the development of forward-looking technologies (strategic value and scientific value) and market forecast (market value). It can be adjusted between 0.6 and 0.8 according to the importance that enterprises attach to the current status and trends. The time decay coefficient λ is used to reflect the impact of historical market forecast data on the current assessment. The impact of historical data should gradually weaken over time. The exponential decay function e -λ(t-τ) To achieve this decay effect, the value can be adjusted between 0.4 and 0.6, depending on the frequency of market forecast data updates and the period of influence.

[0086] Therefore, a multi-dimensional fusion model can simultaneously consider three key dimensions: strategy, science, and market. This comprehensive evaluation approach can more accurately reflect the overall value of data, enabling a comprehensive, dynamic, and scientific assessment of data value, providing strong support for scientific decision-making, improving resource utilization efficiency, and effectively addressing various challenges in complex environments. It avoids bias; for example, a technology may be highly innovative scientifically (high scientific value), but if it doesn't match national policies or market needs (low strategic and market value), its practical application value may be limited. The multi-dimensional fusion model avoids overlooking potential risks or opportunities due to biased evaluation. Furthermore, the balance factor β, adjustment factor γ, and time decay coefficient λ in the model can be adjusted according to different industries, technological fields, or market environments. This flexibility allows the model to adapt to evaluation needs in different scenarios. By introducing the time decay coefficient λ, the model can better handle the impact of historical data on current evaluations. As time progresses, the importance of earlier data gradually decreases; this dynamic adjustment mechanism ensures that the evaluation results always reflect the latest market and technological trends.

[0087] The following is combined Figure 3 The method described in this application is used to describe the method.

[0088] As a specific example, the multi-dimensional data evaluation method of this application may include the following steps:

[0089] S101, Collect and train a sample set, wherein the training sample set includes multiple training data.

[0090] S102, perform data cleaning and labeling on the training sample set.

[0091] S103. Classify the processed training sample set to obtain multiple types of sub-databases. Perform structuring and standardization processing on each type of sub-database to obtain the corresponding sub-reference database. Merge multiple sub-reference databases to obtain the reference database.

[0092] S104, quantify the reference database based on at least one of big data analysis, knowledge graph and artificial intelligence model to determine the target indicators, and proceed to steps S105, S106 and S107 respectively.

[0093] S105, determine the strategic value based on the weighted product of strategic fit with the first preset coefficient, policy support intensity with the second preset coefficient, industrial chain synergy with the third preset coefficient, and patent barrier index with the fourth preset coefficient, wherein the weighted product is a power-weighted product, and proceed to step S108.

[0094] S106, the scientific value is determined by summing the weighted coefficients of technology maturity, innovation breakthrough index, R&D feasibility and technology diffusion, and then proceeds to step S108.

[0095] S107 uses the hyperbolic tangent of the result of weighted summation of market size index with the fifth preset coefficient, application scenario breadth with the sixth preset coefficient, competition intensity with the seventh preset coefficient, and investment enthusiasm with the eighth preset coefficient as the strategic value.

[0096] S108 calculates the strategic value, scientific value, and market value of each training data point to generate a comprehensive value index across the strategic, scientific, and market dimensions.

[0097] S109 utilizes the strategic value, scientific value, market value, and comprehensive value index of each training data point to train the model, thereby constructing a multi-dimensional fusion model for outputting the comprehensive value index of data in the strategic, scientific, and market dimensions.

[0098] S110 inputs at least one data point into a multi-dimensional fusion model to integrate the strategic value, scientific value, and market value of at least one data point, and outputs a comprehensive value index in the strategic, scientific, and market dimensions.

[0099] S111, generate a multi-dimensional evaluation result based on at least one data point according to the comprehensive value index.

[0100] In summary, the data multi-dimensional evaluation method according to the embodiments of this application collects a training sample set, which includes multiple training data points. The strategic value, scientific value, and market value of each training data point are calculated to generate a comprehensive value index across the strategic, scientific, and market dimensions. The strategic value, scientific value, market value, and comprehensive value index of each training data point are used to train a model to construct a multi-dimensional fusion model for outputting the comprehensive value index of the data across the strategic, scientific, and market dimensions. Therefore, this method can more intelligently, dynamically, and reliably characterize and evaluate data value, enabling intelligent prediction and scientific decision-making regarding the application value of forward-looking technologies.

[0101] Corresponding to the above embodiments, this application also proposes an electronic device.

[0102] like Figure 4 As shown, the electronic device 200 of this application embodiment may include: a memory 210, a processor 220, and a program stored on the memory 210 and executable on the processor 220. When the processor 220 executes the program, it implements the above-mentioned multi-dimensional data evaluation method.

[0103] The electronic device according to the embodiments of this application, by executing the above-described multi-dimensional data evaluation method, can more intelligently, dynamically, and reliably characterize and evaluate the value of data, and realize intelligent prediction and scientific decision-making of the application value of forward-looking technologies.

[0104] Corresponding to the above embodiments, this application also proposes a computer-readable storage medium.

[0105] The computer-readable storage medium of this application embodiment stores a program that, when executed by a processor, implements the above-described multi-dimensional data evaluation method.

[0106] The computer-readable storage medium according to the embodiments of this application, by executing the above-described multi-dimensional data evaluation method, can more intelligently, dynamically, and reliably characterize and evaluate the value of data, and realize intelligent prediction and scientific decision-making of the application value of forward-looking technologies.

[0107] Corresponding to the above embodiments, this application also proposes a data multi-dimensional evaluation device.

[0108] like Figure 5 As shown, the data multi-dimensional evaluation device 100 of this application embodiment includes: an acquisition module 110 and a simulation module 120.

[0109] The acquisition module 110 is used to collect at least one data point. The simulation module 120 is used to input the at least one data point into a pre-trained multi-dimensional fusion model to integrate the strategic value, scientific value, and market value of the at least one data point, and output a comprehensive value index in the strategic, scientific, and market dimensions. The fusion model is trained using the strategic value, scientific value, and market value of the training data and the comprehensive value index. The simulation module 120 is also used to generate a multi-dimensional evaluation result for the at least one data point based on the comprehensive value index.

[0110] According to the data multi-dimensional evaluation device of this application embodiment, the acquisition module is used to collect at least one data point, and the simulation module is used to input the at least one data point into a pre-trained multi-dimensional fusion model to fuse the strategic value, scientific value, and market value of the at least one data point, and output a comprehensive value index in the strategic, scientific, and market dimensions. The fusion model is trained using the strategic value, scientific value, and market value of the training data and the comprehensive value index. The simulation module is also used to generate a multi-dimensional evaluation result for the at least one data point based on the comprehensive value index. Therefore, this device can more intelligently, dynamically, and reliably characterize and evaluate the value of data, enabling intelligent prediction and scientific decision-making regarding the application value of forward-looking technologies.

[0111] Corresponding to the above embodiments, this application also proposes a data multi-dimensional evaluation device.

[0112] like Figure 6 As shown, the data multi-dimensional evaluation device 300 of this application embodiment includes: an acquisition module 310, a calculation module 320 and a training module 330.

[0113] The acquisition module 310 is used to collect a training sample set, which includes multiple training data points. The calculation module 320 is used to calculate the strategic value, scientific value, and market value of each training data point to generate a comprehensive value index across the strategic, scientific, and market dimensions. The training module 330 is used to train a model using the strategic value, scientific value, market value, and comprehensive value index of each training data point to construct a multi-dimensional fusion model for outputting the comprehensive value index of the data across the strategic, scientific, and market dimensions.

[0114] According to one embodiment of this application, the calculation module 320 calculates the strategic value, scientific value, and market value of each piece of training data, specifically for: determining a reference database based on the training sample set; quantifying the reference database based on at least one of big data analysis, knowledge graphs, and artificial intelligence models to determine target indicators; and determining the strategic value, scientific value, and market value based on the target indicators.

[0115] According to one embodiment of this application, the target indicators include technology strategy fit, policy support intensity, industrial chain synergy, and patent barrier index. The calculation module 320 determines the strategic value based on the target indicators, specifically by: determining the strategic value by weighting the strategic fit with a first preset coefficient, the policy support intensity with a second preset coefficient, the industrial chain synergy with a third preset coefficient, and the patent barrier index with a fourth preset coefficient, wherein the weighted product is a power-weighted product.

[0116] According to one embodiment of this application, the target indicators include technology maturity, innovation breakthrough index, R&D feasibility and technology diffusion degree. The calculation module 320 determines the scientific value based on the target indicators, specifically: the scientific value is determined by summing the technology maturity, innovation breakthrough index, R&D feasibility and technology diffusion degree with their respective weight coefficients.

[0117] According to one embodiment of this application, the target indicators include market size index, application scenario breadth, competition intensity, and investment enthusiasm. The calculation module 320 determines the market value based on the target indicators, specifically by using the hyperbolic tangent value of the weighted sum of the market size index with the fifth preset coefficient, the application scenario breadth with the sixth preset coefficient, the competition intensity with the seventh preset coefficient, and the investment enthusiasm with the eighth preset coefficient as the strategic value.

[0118] According to one embodiment of this application, the training module 330 constructs a multi-dimensional fusion model for outputting a comprehensive value index of data across strategic, scientific, and market dimensions using the following formula:

[0119]

[0120] Where V(t) represents the comprehensive value index at time t, β represents the balancing factor, γ represents the adjustment factor, SV(t) represents the strategic value at time t, λ represents the time decay coefficient, SJ(t) represents the scientific value at time t, and MF(τ) represents the market value at time τ. It represents the weighted integral of the market value over the past n years.

[0121] According to one embodiment of this application, the calculation module 320 determines a reference database based on the training sample set, specifically for: performing data cleaning and labeling on the training sample set; classifying the processed training sample set to obtain multiple types of sub-databases; performing structuring and standardization on each type of sub-database to obtain a corresponding sub-reference database; and merging multiple sub-reference databases to obtain a reference database.

[0122] It should be noted that for details not disclosed in the data multi-dimensional evaluation device of this application embodiment, please refer to the details disclosed in the data multi-dimensional evaluation method of this application embodiment, which will not be repeated here.

[0123] According to the data multi-dimensional evaluation device of this application embodiment, the acquisition module is used to collect a training sample set, wherein the training sample set includes multiple training data points. The calculation module is used to calculate the strategic value, scientific value, and market value of each training data point to generate a comprehensive value index in the strategic, scientific, and market dimensions. The training module is used to train a model using the strategic value, scientific value, market value, and comprehensive value index of each training data point to construct a multi-dimensional fusion model for outputting the comprehensive value index of data in the strategic, scientific, and market dimensions. Therefore, this device can more intelligently, dynamically, and reliably characterize and evaluate the value of data, realizing intelligent prediction and scientific decision-making regarding the application value of forward-looking technologies.

[0124] It should be noted that the logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of computer-readable media include: an electrical connection having one or more wires (electronic device), a portable computer disk drive (magnetic device), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Alternatively, the computer-readable medium may be paper or other suitable media on which the program can be printed, since the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in a computer memory.

[0125] It should be understood that various parts of this application can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented using software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0126] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0127] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this application, "multiple" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0128] In this application, unless otherwise expressly specified and limited, the terms "installation," "connection," "joining," and "fixing," etc., should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral part; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components, unless otherwise expressly limited. Those skilled in the art can understand the specific meaning of the above terms in this application according to the specific circumstances.

[0129] Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of this application.

Claims

1. A multi-dimensional data evaluation method, characterized in that, When applied to the model usage phase, the method includes the following steps: Collect at least one data point; The at least one piece of data is input into a pre-trained multi-dimensional fusion model to integrate the strategic value, scientific value, and market value of the at least one piece of data, and output a comprehensive value index in the strategic, scientific, and market dimensions. The fusion model is trained by the strategic value, scientific value, market value, and comprehensive value index of the training data. A multi-dimensional evaluation result of at least one data point is generated based on the comprehensive value index.

2. A multi-dimensional data evaluation method, characterized in that, Applied to the model building phase, the method includes the following steps: Collect a training sample set, wherein the training sample set includes multiple training data; Calculate the strategic value, scientific value, and market value of each piece of training data to generate a comprehensive value index across the strategic, scientific, and market dimensions; The strategic, scientific, and market value of each piece of training data, along with the comprehensive value index, are used to train a model to construct a multi-dimensional fusion model for outputting the comprehensive value index of data in the strategic, scientific, and market dimensions.

3. The data multi-dimensional evaluation method according to claim 2, characterized in that, The calculation of the strategic, scientific, and market value of each piece of training data includes: A reference database is determined based on the training sample set; The reference database is quantified using at least one of big data analytics, knowledge graphs, and artificial intelligence models to determine the target metrics. The strategic value, scientific value, and market value are determined based on the aforementioned target indicators.

4. The data multi-dimensional evaluation method according to claim 3, characterized in that, The target indicators include technological strategic alignment, policy support intensity, industrial chain synergy, and patent barrier index. Determining the strategic value based on these target indicators includes: The strategic value is determined by a weighted product of the strategic fit with a first preset coefficient, the policy support intensity with a second preset coefficient, the industrial chain synergy with a third preset coefficient, and the patent barrier index with a fourth preset coefficient, wherein the weighted product is a power-weighted product.

5. The data multi-dimensional evaluation method according to claim 3, characterized in that, The target indicators include technology maturity, innovation breakthrough index, R&D feasibility, and technology diffusion. Determining the scientific value based on the target indicators includes: The scientific value is determined by summing the aforementioned technology maturity, innovation breakthrough index, R&D feasibility, and technology diffusion degree with their respective weighting coefficients.

6. The data multi-dimensional evaluation method according to claim 3, characterized in that, The target indicators include market size index, application scenario breadth, competition intensity, and investment enthusiasm. Determining the market value based on the target indicators includes: The hyperbolic tangent of the weighted sum of the market size index with the fifth preset coefficient, the application scenario breadth with the sixth preset coefficient, the competition intensity with the seventh preset coefficient, and the investment enthusiasm with the eighth preset coefficient is taken as the strategic value.

7. The data multi-dimensional evaluation method according to claim 2, characterized in that, The following formula is used to construct a multi-dimensional fusion model for outputting a comprehensive value index of data across strategic, scientific, and market dimensions: Where V(t) represents the comprehensive value index at time t, β represents the balance factor, γ represents the adjustment factor, λ represents the time decay coefficient, SV(t) represents the strategic value at time t, SJ(t) represents the scientific value at time t, and MF(τ) represents the market value at time τ. This represents the weighted integral of the market value over the past n years.

8. The data multi-dimensional evaluation method according to claim 3, wherein determining the reference database based on the training sample set includes: The training sample set is cleaned and labeled. The processed training sample set is classified to obtain multiple types of sub-databases; Each type of sub-database is structured and standardized to obtain the corresponding sub-reference database; The multiple sub-reference databases are merged to obtain the reference database.

9. An electronic device, characterized in that, include: A memory, a processor, and a program stored in the memory and executable on the processor, wherein when the processor executes the program, it implements the data multidimensional evaluation method according to any one of claims 1 or 2-8.

10. A computer-readable storage medium, characterized in that, It stores a program that, when executed by a processor, implements the data multidimensional evaluation method according to any one of claims 1 or 2-8.

11. A multi-dimensional data evaluation device, characterized in that, The device is used in the model usage phase and includes: The acquisition module is used to collect at least one piece of data. The simulation module is used to input the at least one piece of data into a pre-trained multi-dimensional fusion model to fuse the strategic value, scientific value and market value of the at least one piece of data, and output a comprehensive value index in the strategic dimension, scientific dimension and market dimension. The fusion model is trained by the strategic value, scientific value and market value and comprehensive value index of the training data. The simulation module is also used to generate a multi-dimensional evaluation result of the at least one data point based on the comprehensive value index.

12. A multi-dimensional data evaluation device, characterized in that, Applied to the model building phase, wherein the apparatus includes: The acquisition module is used to collect a training sample set, wherein the training sample set includes multiple training data. The calculation module is used to calculate the strategic value, scientific value, and market value of each piece of training data to generate a comprehensive value index in the strategic, scientific, and market dimensions. The training module is used to train a model using the strategic value, scientific value, and market value of each piece of training data and the comprehensive value index, in order to construct a multi-dimensional fusion model for outputting the comprehensive value index of the data in the strategic, scientific, and market dimensions.

Images