A data asset fair value intelligent evaluation method and system fusing multi-dimensional features
By integrating multi-dimensional features into a data asset fair value intelligent assessment system, the problem of insufficient data asset assessment quality in existing technologies has been solved, enabling dynamic and scientific data asset assessment and optimizing asset allocation and business decisions.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- ZHONGDE GAOLU CONSULTING (YUNNAN) CO LTD
- Filing Date
- 2026-03-11
- Publication Date
- 2026-06-19
AI Technical Summary
Existing technologies cannot process data assets according to their source, which affects the quality of the assessment and makes it difficult to screen key features in the value assessment.
The system employs an intelligent fair value assessment system for data assets that integrates multi-dimensional features. It includes modules for data acquisition, preprocessing, feature vector extraction, and model evaluation. The system crawls multi-dimensional data, cleans, transforms, and standardizes it, uses the PCA algorithm to screen key features, combines it with the linear regression algorithm to conduct fair value assessment, and displays the results through visualization.
It enables dynamic and scientific evaluation of data assets, overcomes the lag of static evaluation, optimizes asset allocation and business decisions, and improves the quality and accuracy of evaluation.
Smart Images
Figure CN122243549A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of fair value assessment technology, specifically to an intelligent assessment method and system for the fair value of data assets that integrates multi-dimensional features. Background Technology
[0002] Fair value assessment of data assets refers to the process of estimating the price of a data asset in an orderly transaction under specific market conditions when it is sold in an arm's length deal. This price reflects the current market information of market participants on the measurement date and can more comprehensively reflect the true value of a company. As data is a core asset in the digital economy era, its value is often underestimated in traditional financial statements. Fair value assessment can improve the accuracy of financial reports, enhance the confidence of investors and stakeholders, and help optimize resource allocation and improve management efficiency. By sorting out and assessing data assets, companies can clearly understand the value of their data resources, thereby supporting more scientific decision-making and exploring new business opportunities.
[0003] Existing methods for assessing the fair value of data assets cannot be tailored to their source, which affects the quality of the assessment and makes it difficult to select key features during the valuation process. Therefore, these methods do not meet current needs. To address this, we propose an intelligent assessment method and system for the fair value of data assets that integrates multi-dimensional features. Summary of the Invention
[0004] The purpose of this invention is to provide an intelligent assessment method and system for the fair value of data assets that integrates multi-dimensional features, in order to solve the problems mentioned in the background art, such as the inability to process data assets according to their source, which affects the assessment quality, and the inconvenience of screening key features in the value assessment.
[0005] To achieve the above objectives, the present invention provides the following technical solution: an intelligent assessment system for the fair value of data assets integrating multi-dimensional features, comprising a data acquisition module, a data preprocessing module, a feature vector extraction module, a model evaluation module, and an application display module:
[0006] The data acquisition module is used to capture raw data from multi-dimensional data sources related to data assets and input them in a unified manner;
[0007] The data preprocessing module is used to clean, transform, and standardize the input raw data.
[0008] The feature vector extraction module is used to extract multi-dimensional features from preprocessed data and construct feature vectors;
[0009] The model evaluation module is used to evaluate fair value based on feature vectors using a value evaluation model. The model evaluation module consists of a value evaluation model and an evaluation database. The value evaluation model compares the feature vectors with the built-in data in the evaluation database.
[0010] The application display module is used to show users the evaluation results and provide an interactive analysis window.
[0011] Preferably, the data acquisition module includes a data connector, a data crawler, an API interface caller, and a database connector. The data crawler is used to capture internal data and external information. The internal data consists of financial and business data from the enterprise's internal systems, and the external information consists of market conditions and industry report data.
[0012] Preferably, the data preprocessing module includes a data cleaning subunit, a data transformation subunit, and a data standardization subunit. The data cleaning subunit is used to identify and process missing values, outliers, and duplicate data in the original data. The data transformation subunit is used to convert unstructured data in the original data into structured data. The unstructured data includes text, images, and audio. The data standardization subunit is used to standardize the format and units of the original data.
[0013] Preferably, the feature vector extraction module includes a feature extraction subunit, a feature selection subunit, and a feature fusion subunit. The feature extraction subunit is used to extract key features from the original data. The key features include data scale, data quality, data scarcity, and data application scenario. The feature selection subunit is used to screen out the most valuable multi-dimensional features for evaluation using the principal component analysis (PCA) algorithm. The feature fusion subunit is used to fuse multi-dimensional features to form a comprehensive feature vector. The multi-dimensional features include technical features, market features, and legal features.
[0014] Preferably, the value assessment model has a built-in assessment index function and a linear regression algorithm. The assessment index function consists of mean squared error (MSE), mean absolute error (MAE), and coefficient of determination (R²).
[0015] MSE = (1 / n)*Σ(yᵢ-ŷᵢ)²;
[0016] MAE=(1 / n)*Σ|yᵢ-ŷᵢ|;
[0017] R² = 1 - (SS) e / SS t );
[0018] SS e =Σ(yᵢ-ŷᵢ);
[0019] SS t=Σ(yᵢ-ȳ)²;
[0020] Where yᵢ is the true value, ŷᵢ is the predicted value, n is the sample size, and SS e Sum of squared residuals, SS t Let be the total sum of squares, and ȳ be the mean of the true values;
[0021] The evaluation index function is used to quantify the deviation between the predicted value and the actual value.
[0022] Preferably, the application display module includes a web front-end interface, a visualization component, and a report generator, wherein the visualization component is a touch screen.
[0023] A method for intelligently assessing the fair value of data assets by integrating multi-dimensional features includes the following steps:
[0024] S1: Data Acquisition. The data crawler is connected to internal and external scenarios through data connectors, API interface callers and database connectors. The data crawler is used to capture financial and business data from the enterprise's internal systems as well as market conditions and industry report data from external scenarios. The raw data obtained is then uniformly input.
[0025] S2: Raw data processing, which preprocesses the input raw data by cleaning, transforming and standardizing it. Specifically, the data cleaning subunit identifies and processes missing values, outliers and duplicate data in the raw data, and the data transformation subunit converts the unstructured data of the raw data into structured data. Then, the data standardization subunit unifies the format and units of the raw data.
[0026] S3: Feature extraction and vector construction. Extract multi-dimensional features from preprocessed data and construct feature vectors. The feature extraction subunit extracts key features from the original data. Then, the feature selection subunit uses the principal component analysis (PCA) algorithm to select the most valuable multi-dimensional features for evaluation. Finally, the feature fusion subunit fuses the multi-dimensional features to form a comprehensive feature vector.
[0027] S4: Feature-based valuation, which assesses fair value based on feature vectors using a valuation model, and compares the valuation model with the built-in data in the valuation database using feature vectors;
[0028] S5: Evaluation display and interaction, intuitively displaying evaluation results through charts, dashboards and other forms, allowing users to adjust parameters, and automatically generating detailed evaluation reports in multiple formats.
[0029] Preferably, the linear regression algorithm is used to fit the linear relationship between the feature vector and the fair value, and the linear regression algorithm is y = β0 + β1x1 + ... + β k xk +ε;
[0030] Where y is the dependent variable, x1...x k Let β be the independent variable, β0 be the intercept term, and β1...β be the independent variable. k ε is the regression coefficient, and ε is the error term.
[0031] Compared with the prior art, the beneficial effects of the present invention are:
[0032] This invention comprehensively evaluates the input value, business value, market value, and external value of data assets, avoiding the valuation bias caused by the single dimension of traditional methods. By adjusting the parameters of the value assessment model, it reflects the changes in data value with the market and application scenarios in a timely manner, overcoming the lag of static assessment. Through high-quality data analysis and visualization results, it provides enterprises with a scientific basis to optimize asset allocation and business decisions. Attached Figure Description
[0033] Figure 1 This is a schematic diagram of the intelligent valuation system for fair value of data assets according to the present invention;
[0034] Figure 2 This is a flowchart of the intelligent valuation method for data assets according to the present invention. Detailed Implementation
[0035] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.
[0036] Please see Figure 1 The present invention provides an embodiment of a data asset fair value intelligent assessment system that integrates multi-dimensional features, comprising a data acquisition module, a data preprocessing module, a feature vector extraction module, a model evaluation module, and an application display module;
[0037] The data acquisition module is used to capture raw data from multi-dimensional data sources related to data assets and input them uniformly. The data acquisition module includes a data connector, a data crawler, an API interface caller, and a database connector. The data crawler is used to capture internal data and external information. The internal data consists of financial and business data from the enterprise's internal systems, while the external information consists of market conditions and industry report data.
[0038] The data preprocessing module is used to clean, transform, and standardize the input raw data. The data preprocessing module includes a data cleaning subunit, a data transformation subunit, and a data standardization subunit. The data cleaning subunit is used to identify and process missing values, outliers, and duplicate data in the raw data. The data transformation subunit is used to convert the unstructured data of the raw data into structured data. The unstructured data includes text, images, and audio. The data standardization subunit is used to standardize the format and units of the raw data.
[0039] The feature vector extraction module is used to extract multi-dimensional features from preprocessed data and construct feature vectors. The feature vector extraction module includes a feature extraction subunit, a feature selection subunit, and a feature fusion subunit. The feature extraction subunit is used to extract key features from the original data. Key features include data scale, data quality, data scarcity, and data application scenario. The feature selection subunit is used to screen out the most valuable multi-dimensional features for evaluation using the principal component analysis (PCA) algorithm. The feature fusion subunit is used to fuse multi-dimensional features and form a comprehensive feature vector. Multi-dimensional features include technical features, market features, and legal features.
[0040] The application display module is used to show users the evaluation results and provide an interactive analysis window. The application display module includes a web front-end interface, visualization components, and a report generator. The visualization components are touch screens.
[0041] The model evaluation module is used to evaluate fair value based on feature vectors using a value evaluation model. The model evaluation module consists of a value evaluation model and an evaluation database. The value evaluation model compares the feature vectors with the built-in data in the evaluation database. The value evaluation model has built-in evaluation index functions and linear regression algorithms. The evaluation index functions consist of mean squared error (MSE), mean absolute error (MAE), and coefficient of determination (R²).
[0042] MSE = (1 / n)*Σ(yᵢ-ŷᵢ)²;
[0043] MAE=(1 / n)*Σ|yᵢ-ŷᵢ|;
[0044] R² = 1 - (SS) e / SS t );
[0045] SS e =Σ(yᵢ-ŷᵢ);
[0046] SS t =Σ(yᵢ-ȳ)²;
[0047] Where yᵢ is the true value, ŷᵢ is the predicted value, n is the sample size, and SS e Sum of squared residuals, SS tLet be the total sum of squares, and ȳ be the mean of the true values;
[0048] The linear regression algorithm is y = β0 + β1x1 + ... + β k x k +ε;
[0049] Where y is the dependent variable, x1...x k Let β be the independent variable, β0 be the intercept term, and β1...β be the independent variable. k ε is the regression coefficient, and ε is the error term. The linear regression algorithm is used to fit the linear relationship between the feature vector and the fair value, and the evaluation index function is used to quantify the deviation between the predicted value and the true value.
[0050] Please see Figure 2 A method for intelligently assessing the fair value of data assets by integrating multi-dimensional features includes the following steps:
[0051] S1: Data Acquisition. The data crawler is connected to internal and external scenarios through data connectors, API interface callers and database connectors. The data crawler is used to capture financial and business data from the enterprise's internal systems as well as market conditions and industry report data from external scenarios. The raw data obtained is then uniformly input.
[0052] S2: Raw data processing, which preprocesses the input raw data by cleaning, transforming and standardizing it. Specifically, the data cleaning subunit identifies and processes missing values, outliers and duplicate data in the raw data, and the data transformation subunit converts the unstructured data of the raw data into structured data. Then, the data standardization subunit unifies the format and units of the raw data.
[0053] S3: Feature extraction and vector construction. Extract multi-dimensional features from preprocessed data and construct feature vectors. The feature extraction subunit extracts key features from the original data. Then, the feature selection subunit uses the principal component analysis (PCA) algorithm to select the most valuable multi-dimensional features for evaluation. Finally, the feature fusion subunit fuses the multi-dimensional features to form a comprehensive feature vector.
[0054] S4: Feature-based valuation, which assesses fair value based on feature vectors using a valuation model, and compares the valuation model with the built-in data in the valuation database using feature vectors;
[0055] S5: Evaluation display and interaction, intuitively displaying evaluation results through charts, dashboards and other forms, allowing users to adjust parameters, and automatically generating detailed evaluation reports in multiple formats.
[0056] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered in all respects as exemplary and non-limiting, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the present invention. No reference numerals in the claims should be construed as limiting the scope of the claims.
Claims
1. A data asset fair value intelligent assessment system integrating multi-dimensional features, characterized in that, It includes a data acquisition module, a data preprocessing module, a feature vector extraction module, a model evaluation module, and an application demonstration module. The data acquisition module is used to capture raw data from multi-dimensional data sources related to data assets and input them in a unified manner; The data preprocessing module is used to clean, transform, and standardize the input raw data. The feature vector extraction module is used to extract multi-dimensional features from preprocessed data and construct feature vectors; The model evaluation module is used to evaluate fair value based on feature vectors using a value evaluation model. The model evaluation module consists of a value evaluation model and an evaluation database. The value evaluation model compares the feature vectors with the built-in data in the evaluation database. The application display module is used to show users the evaluation results and provide an interactive analysis window.
2. The intelligent assessment system for the fair value of data assets integrating multi-dimensional features according to claim 1, characterized in that: The data acquisition module includes a data connector, a data crawler, an API interface caller, and a database connector. The data crawler is used to capture internal data and external information. The internal data consists of financial and business data from the enterprise's internal systems, while the external information consists of market trends and industry report data.
3. The intelligent assessment system for the fair value of data assets integrating multi-dimensional features according to claim 2, characterized in that: The data preprocessing module includes a data cleaning subunit, a data transformation subunit, and a data standardization subunit. The data cleaning subunit is used to identify and process missing values, outliers, and duplicate data in the original data. The data transformation subunit is used to convert unstructured data in the original data into structured data. The unstructured data includes text, images, and audio. The data standardization subunit is used to standardize the format and units of the original data.
4. The intelligent assessment system for the fair value of data assets integrating multi-dimensional features according to claim 3, characterized in that: The feature vector extraction module includes a feature extraction subunit, a feature selection subunit, and a feature fusion subunit. The feature extraction subunit is used to extract key features from the raw data. The key features include data scale, data quality, data scarcity, and data application scenario. The feature selection subunit is used to screen out the most valuable multi-dimensional features for evaluation using the principal component analysis (PCA) algorithm. The feature fusion subunit is used to fuse multi-dimensional features to form a comprehensive feature vector. The multi-dimensional features include technical features, market features, and legal features.
5. The intelligent assessment system for the fair value of data assets integrating multi-dimensional features according to claim 4, characterized in that: The value assessment model has a built-in assessment index function and a linear regression algorithm. The assessment index function consists of mean squared error (MSE), mean absolute error (MAE), and coefficient of determination (R²). MSE = (1 / n)*Σ(yᵢ-ŷᵢ)²; MAE=(1 / n)*Σ|yᵢ-ŷᵢ|; R²=1-(SS e / SS t ); SS e =Σ(yᵢ-ŷᵢ); SS t =Σ(yᵢ-ȳ)²; Where yᵢ is the true value, ŷᵢ is the predicted value, n is the sample size, and SS e Sum of squared residuals, SS t Let be the total sum of squares, and ȳ be the mean of the true values; The evaluation index function is used to quantify the deviation between the predicted value and the actual value.
6. The intelligent assessment system for the fair value of data assets integrating multi-dimensional features according to claim 5, characterized in that: The application demonstration module includes a web front-end interface, a visualization component, and a report generator. The visualization component is a touch screen.
7. A method for intelligently assessing the fair value of data assets by integrating multi-dimensional features, and a system for intelligently assessing the fair value of data assets by integrating multi-dimensional features according to any one of claims 1-6, characterized in that, Includes the following steps: S1: Data Acquisition. The data crawler is connected to internal and external scenarios through data connectors, API interface callers and database connectors. The data crawler is used to capture financial and business data from the enterprise's internal systems as well as market conditions and industry report data from external scenarios. The raw data obtained is then uniformly input. S2: Raw data processing, which preprocesses the input raw data by cleaning, transforming and standardizing it. Specifically, the data cleaning subunit identifies and processes missing values, outliers and duplicate data in the raw data, and the data transformation subunit converts the unstructured data of the raw data into structured data. Then, the data standardization subunit unifies the format and units of the raw data. S3: Feature extraction and vector construction. Extract multi-dimensional features from preprocessed data and construct feature vectors. The feature extraction subunit extracts key features from the original data. Then, the feature selection subunit uses the principal component analysis (PCA) algorithm to select the most valuable multi-dimensional features for evaluation. Finally, the feature fusion subunit fuses the multi-dimensional features to form a comprehensive feature vector. S4: Feature-based valuation, which assesses fair value based on feature vectors using a valuation model, and compares the valuation model with the built-in data in the valuation database using feature vectors; S5: Evaluation display and interaction, intuitively displaying evaluation results through charts, dashboards and other forms, allowing users to adjust parameters, and automatically generating detailed evaluation reports in multiple formats.
8. The intelligent assessment method for the fair value of data assets integrating multi-dimensional features according to claim 7, characterized in that: The linear regression algorithm is used to fit the linear relationship between the feature vector and the fair value. The linear regression algorithm is y = β0 + β1x1 + ... + β k x k +ε; Where y is the dependent variable, x1...x k Let β be the independent variable, β0 be the intercept term, and β1...β be the independent variable. k ε is the regression coefficient, and ε is the error term.