A data asset quality evaluation system, method, electronic device and storage medium

Abstract

The application relates to a data asset quality evaluation system, which adopts key means such as a multi-dimensional evaluation model, objective quantitative indicators and an automatic processing flow. A multi-dimensional data asset quality stereoscopic multi-dimensional model is constructed, and data asset quality is evaluated from multiple dimensions in combination with a standard rule library and an evaluation data model based on statistics, machine learning and weight calculation. In the evaluation process, the evaluation requirements are first set, data is collected and preprocessed, that is, standardized conversion is performed, then multi-model comprehensive evaluation calculation is performed, and finally the results are visualized. The technical scheme effectively solves the problem that data asset quality evaluation lacks effective methods and systems, resulting in low efficiency and poor result accuracy, realizes comprehensive, objective and efficient evaluation of data asset quality, can more accurately reflect the data asset quality status, and meets the large-scale evaluation requirements of data assets in value application such as entry into accounts and tables and circulation and utilization.

Description

Technical Field

[0001] This application relates to the field of data processing, and in particular to a data asset quality assessment system, method, computer device, and computer-readable storage medium. Background Technology

[0002] Currently, data, alongside traditional factors such as land, labor, capital, and technology, has become the fifth major factor of production in the digital economy era. Data assets, as an emerging asset type in the process of digital transformation of the economy and society, are increasingly becoming an important strategic resource for promoting the construction of Digital China and accelerating the development of the digital economy. However, the quality of data assets is becoming increasingly prominent, seriously affecting their value and application. For example, the process of data asset accounting, circulation, utilization, and pledge financing faces challenges in data asset valuation. Therefore, it is urgent to study data asset quality assessment methods and develop a data asset quality assessment system to effectively evaluate data asset quality.

[0003] Existing data asset quality assessment systems and methods suffer from the following main problems: First, they lack a single assessment dimension: a comprehensive assessment across multiple dimensions is lacking. Second, the assessment methods are highly subjective: they rely heavily on human experience and judgment, lacking objective and quantifiable assessment indicators and methods. Third, the assessment is inefficient: the assessment process is cumbersome and complex, making it difficult to meet the needs of large-scale data asset quality assessment.

[0004] Currently, no effective solutions have been proposed to address the problems existing in current methods for assessing the quality of technical data assets. Summary of the Invention

[0005] This application provides a data asset quality assessment system, method, computer device, and computer-readable storage medium to at least address the problems of low assessment efficiency and poor accuracy in related technologies.

[0006] In a first aspect, embodiments of this application provide a data asset quality assessment system. The system is deployed on a computer device and implemented by executing a computer program. The system includes: a configuration module, a data collection module, and a quality assessment module, wherein:

[0007] The configuration module is used to configure the assessment objectives and data scope related to data asset quality assessment based on user interaction information.

[0008] The data collection module is used to acquire historical data related to data asset assessment, and to acquire the data to be assessed from the historical data according to the data range;

[0009] The quality assessment module is used to assess the data asset quality of the data to be assessed based on the assessment objectives and through a pre-constructed multi-dimensional quality assessment model, and obtain the assessment results. The multi-dimensional quality assessment model is based on multiple interrelated three-dimensional dimensions and combines a standard rule base and a multi-dimensional calculation model to conduct a comprehensive quality assessment of the data to be assessed.

[0010] In some embodiments, the multi-dimensional quality assessment model includes: a three-dimensional multi-dimensional model, a standard rule base, and a quality assessment model.

[0011] In some embodiments, the three-dimensional multidimensional model stores multiple quality assessment dimensions that are interconnected during the process of forming data assets from raw data resources, and each quality assessment dimension includes multiple evaluation indicators to quantify the quality characteristics under that quality assessment dimension.

[0012] The standard rule base includes: a standard sub-base, used to define the scoring criteria for each evaluation indicator under the quality assessment dimension;

[0013] The evaluation index sub-library is used to define the calculation methods, threshold ranges, and corresponding raw data resources for each evaluation index;

[0014] An evaluation sample library is used to store historical evaluation cases and parameter configuration information under typical historical scenarios, which can be used to match current evaluation needs.

[0015] In some embodiments, the quality assessment dimensions include, but are not limited to: data asset quality infrastructure dimension, whose assessment indicators include, but are not limited to: quality standards, quality testing, and quality certification; data supply quality dimension, whose assessment indicators include, but are not limited to: supply methods and supply mechanisms; data product quality dimension, whose assessment indicators include, but are not limited to: accuracy, completeness, simplicity, applicability, consistency, and timeliness; data industry quality dimension, whose assessment indicators include, but are not limited to: the level of industry data resource collection and aggregation, the level of industry big data application, and the benefits of industry big data application; data ecosystem service quality dimension, whose assessment indicators include, but are not limited to: the characteristics of application scenarios in various industries; data brand dimension, whose assessment indicators include, but are not limited to: brand awareness, brand satisfaction, brand loyalty, user search, and social media interaction; past transactions or events dimension, whose assessment indicators include, but are not limited to: transaction date, transaction amount, description of transaction assets, bank statements, and transaction vouchers; effective control dimension, whose assessment indicators include, but are not limited to: access control and compliance management; expected value dimension, whose assessment indicators include, but are not limited to: expected future cash flows and expected profits; and asset measurement dimension, whose assessment indicators include, but are not limited to: cost measurement and fair value.

[0016] In some embodiments, the multidimensional quality assessment model includes:

[0017] The first model is used to mine and describe the data characteristics of the data to be evaluated based on statistical indicators, and to identify quality anomalies through hypothesis testing. The statistical indicators include: mean, variance, and frequency distribution.

[0018] The second model is used to train a classification model, and / or a regression model, and / or a clustering model based on labeled samples, to estimate the quality of data assets and detect quality problems in the data to be evaluated.

[0019] The third model, based on the output of the second model of the first model, uses the analytic hierarchy process (AHP) and / or entropy method and / or principal component analysis to determine the principal weights of each relevant dimension, as well as the index weights of each evaluation index under the relevant dimensions.

[0020] In some embodiments, the multidimensional quality assessment model is used for:

[0021] Based on the data to be evaluated and the three-dimensional multi-dimensional model, the data to be evaluated is standardized using the index standardization method to obtain standardized index values.

[0022] The data to be evaluated is processed by the first model and the second model respectively to obtain data features and anomaly information;

[0023] The third model is used to calculate the sovereign weight and index weight corresponding to each standardized index value, and the correlation coefficient matrix between dimensions is quantified by the correlation coefficient matrix, and the interaction weight matrix is ​​calculated.

[0024] And by using a preset formula, based on the standardized index value, the sovereign weight, the index weight, the correlation coefficient matrix, and the interaction weight matrix, the single-dimensional score and the comprehensive dimension score are calculated respectively. In the calculation process, the initialization result is obtained first, and then the dimensions are calculated iteratively until the preset convergence condition is met to obtain the single-dimensional score and the comprehensive dimension score.

[0025] In some embodiments, the system further includes a display module, wherein the display module is used to generate visual information based on the evaluation results;

[0026] The visualized information will be displayed in multiple dimensions.

[0027] Secondly, embodiments of this application also provide a data asset quality assessment method, which is implemented on a computer device and by executing a computer program, including:

[0028] The configuration module allows users to configure assessment objectives and data scope related to data asset quality assessment based on user interaction information.

[0029] The data collection module acquires historical data related to data asset valuation, and retrieves the data to be valued from the historical data according to the data range.

[0030] Through the quality assessment module, based on the assessment objectives, the data asset quality of the data to be assessed is assessed using a pre-constructed multi-dimensional quality assessment model to obtain the assessment results. The multi-dimensional quality assessment model is based on multiple interrelated three-dimensional dimensions, combined with a standard rule base and a multi-dimensional calculation model, to conduct a comprehensive quality assessment of the data to be assessed.

[0031] Thirdly, embodiments of this application provide a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the method described in the first aspect above.

[0032] Fourthly, embodiments of this application provide a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method described in the first aspect above.

[0033] Compared to related technologies, the data asset quality assessment system and method provided in this application address the problems of single assessment dimensions, strong subjectivity, and low efficiency in existing technologies by employing key means such as multi-dimensional assessment models, objective quantitative indicators, and automated processing flows. A three-dimensional, multi-faceted model of data asset quality is constructed, combining a standard rule base with an assessment data model based on statistics, machine learning, and weight calculation to assess data asset quality from multiple dimensions. In the assessment process, assessment requirements are first defined, data is collected, organized, and preprocessed (standardized transformation), followed by comprehensive multi-model assessment calculations, and finally, the results are visualized. This technical solution effectively solves the problem of low efficiency and poor accuracy in data asset quality assessment due to the lack of effective methods and systems, achieving a comprehensive, objective, and efficient assessment of data asset quality. It can more accurately reflect the status of data asset quality and meet the large-scale assessment needs of data assets in value-added applications such as accounting, circulation, and utilization. Attached Figure Description

[0034] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings:

[0035] Figure 1This is a structural block diagram of a data evaluation system according to an embodiment of this application;

[0036] Figure 2 This is an architecture diagram of the quality assessment module according to an embodiment of this application;

[0037] Figure 3 This is a flowchart of a data asset quality assessment method according to an embodiment of this application;

[0038] Figure 4 This is a schematic diagram of the internal structure of an electronic device according to an embodiment of this application. Detailed Implementation

[0039] To make the objectives, technical solutions, and advantages of this application clearer, the application is described and illustrated below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application. All other embodiments obtained by those skilled in the art based on the embodiments provided in this application without inventive effort are within the scope of protection of this application.

[0040] Obviously, the accompanying drawings described below are merely some examples or embodiments of this application. Those skilled in the art can apply this application to other similar scenarios based on these drawings without any inventive effort. Furthermore, it is understood that although the efforts made in this development process may be complex and lengthy, for those skilled in the art related to the content disclosed in this application, any changes to design, manufacturing, or production based on the technical content disclosed in this application are merely conventional technical means and should not be construed as insufficient disclosure of the content of this application.

[0041] In this application, the reference to "embodiment" means that a specific feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places in the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment that is mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described in this application may be combined with other embodiments without conflict.

[0042] Unless otherwise defined, the technical or scientific terms used in this application shall have the ordinary meaning understood by one of ordinary skill in the art to which this application pertains. The terms “a,” “an,” “an,” “the,” and similar words used in this application do not indicate quantity limitation and may indicate singular or plural. The terms “comprising,” “including,” “having,” and any variations thereof used in this application are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or device that includes a series of steps or modules (units) is not limited to the listed steps or units, but may also include steps or units not listed, or may include other steps or units inherent to these processes, methods, products, or devices. The terms “connected,” “linked,” “coupled,” and similar words used in this application are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. “Multiple” used in this application refers to two or more. “And / or” describes the relationship between related objects, indicating that three relationships may exist; for example, “A and / or B” can represent: A alone, A and B simultaneously, and B alone. The character " / " generally indicates that the preceding and following objects are in an "or" relationship. The terms "first," "second," and "third" used in this application are merely to distinguish similar objects and do not represent a specific ordering of the objects.

[0043] Currently, data, alongside traditional factors such as land, labor, capital, and technology, has become the fifth major factor of production in the digital economy era. Data assets, as an emerging asset type in the process of digital transformation of the economy and society, are increasingly becoming an important strategic resource for promoting the construction of Digital China and accelerating the development of the digital economy. However, the quality of data assets is becoming increasingly prominent, seriously affecting their value and application. For example, in the process of enabling applications such as data asset accounting, circulation, utilization, and pledge financing, there is a lack of methods and systems for data asset quality assessment, leading to difficulties in data asset valuation. Therefore, it is urgent to study methods for data asset quality assessment and develop a data asset quality assessment system to effectively evaluate data asset quality.

[0044] Existing data asset quality assessment systems and methods suffer from the following main problems: First, they lack a single assessment dimension: a comprehensive assessment across multiple dimensions is lacking. Second, the assessment methods are highly subjective: they rely heavily on human experience and judgment, lacking objective and quantifiable assessment indicators and methods. Third, the assessment is inefficient: the assessment process is cumbersome and complex, making it difficult to meet the needs of large-scale data asset quality assessment.

[0045] Specifically, Chinese patent with publication number CN1 16701890A provides a data asset quality assessment system. This patent provides a system consisting of a data dictionary metadata model, a metadata model, a data asset quality assessment model, an assessment task management module, an assessment task scheduling module, a quality monitoring and alarm module, and an assessment result storage module.

[0046] Chinese patent CN113034023 discloses a market-based asset valuation system and method. This patent uses a calculation module, an input module, a template library, a transaction case library, and an auxiliary parameter library to value the assets of the company to be valued. However, none of the aforementioned prior art solutions address the issue of constructing a standardized calculation system for data asset quality assessment, which suffers from low efficiency and poor generalization ability.

[0047] In view of this, embodiments of this application provide a data asset quality assessment system. This system has comprehensive assessment dimensions, covering all key stages from the formation to the application of data assets, and can comprehensively consider the quality status of data assets. It uses scientific and objective assessment indicators and quantitative methods to minimize the interference of human subjective factors on the assessment results, ensuring the fairness and accuracy of the assessment results. At the same time, the highly automated process design greatly improves the assessment efficiency, enabling the rapid processing of large-scale data asset quality assessment tasks, meeting the urgent needs for data asset quality assessment in various scenarios, and truly realizing a comprehensive, objective, and efficient assessment of data asset quality.

[0048] Figure 1 This is a structural block diagram of a data evaluation system according to an embodiment of this application, such as... Figure 1 As shown, the system is deployed on a computer device and implemented by executing a computer program. The system includes: a configuration module 10, a data collection module 20, a quality assessment module 30, and a display module 40, wherein:

[0049] Configuration module 10 is used to configure the assessment objectives and data scope related to data asset quality assessment based on user interaction information;

[0050] Within the overall data asset quality assessment system, configuration module 10 undertakes the preliminary preparation work. Its core function is to configure assessment objectives and data scope closely related to data asset quality assessment based on user interaction information.

[0051] This module precisely configures the evaluation objectives and data scope based on user interaction information. Optionally, users can provide interactive information through system interface input, document upload, etc. For example, if a user wants to evaluate the quality of the company's sales data assets in a precision marketing scenario, as well as the sales data range for a specific quarter, the configuration module 10 will receive the information, parse and process it, and transform the user's needs into system-executable evaluation objectives, such as determining whether the accuracy, completeness, and timeliness of the evaluation data meet the requirements of precision marketing.

[0052] The data collection module 20 is used to acquire historical data related to data asset assessment and to obtain the data to be assessed from the historical data according to the data scope;

[0053] Specifically, within the data asset quality assessment system architecture, the data collection module 20 performs specific data processing tasks. This module first collects historical data related to data asset assessment from diverse data sources, such as local databases, distributed file systems, or external data interfaces. This historical data contains information generated throughout the entire process of developing data resources into data assets.

[0054] Furthermore, based on the data range set by configuration module 10, a data filtering algorithm is used to accurately extract the data to be evaluated from the collected historical data set. Optionally, the data filtering algorithm performs matching and filtering based on set filtering conditions, such as time intervals, business categories to which the data belongs, etc.

[0055] In addition, after data collection and screening, the data collection module 20 performs preprocessing operations such as cleaning and transformation on the acquired data. During the data cleaning stage, deduplication algorithms are used to remove duplicate records, and data validation rules are used to identify and correct data with formatting and logical errors. Furthermore, during the data transformation process, standardization algorithms are used to normalize numerical data, giving it a unified dimension; for non-numerical data, encoding conversion is performed to transform it into a format that is easy for subsequent evaluation models to process, thereby improving data asset quality and providing a reliable data foundation for subsequent data asset quality assessment.

[0056] The quality assessment module 30 is used to assess the quality of the data to be assessed based on the assessment objectives and through a pre-built multi-dimensional quality assessment model to obtain the assessment results. The multi-dimensional quality assessment model is based on multiple interrelated three-dimensional dimensions and combines a standard rule base with a multi-dimensional calculation model to conduct a comprehensive quality assessment of the data to be assessed.

[0057] It should be noted that the core of the data asset quality assessment system includes a three-dimensional, multi-dimensional model of data asset quality, a standard rule base, and a data asset quality assessment data model.

[0058] Figure 2This is an architecture diagram of the quality assessment module according to an embodiment of this application, such as... Figure 2 As shown, the multi-dimensional data asset quality model has 10 dimensions and numerous evaluation indicators, comprehensively covering all aspects related to data asset quality. The standard rule base provides evaluation criteria, indicators, and examples, offering specifications and references for the evaluation. Based on the evaluation objectives and referring to the standard rule base, the data asset quality assessment model determines the data asset quality dimensions, evaluation methods, and indicator parameters. Using statistical, machine learning, and weight calculation models, combined with the dimensions and indicators of the multi-dimensional model, it performs calculations to output data asset quality indicator values, achieving a comprehensive evaluation. All modules collaborate to complete the data asset quality assessment.

[0059] Specifically, the three-dimensional, multi-dimensional model covers 10 dimensions. These quality assessment dimensions include: data asset quality infrastructure dimension, with assessment indicators including quality standards, quality testing, and quality certification; data supply quality dimension, with assessment indicators including supply methods and mechanisms; data product quality dimension, with assessment indicators including accuracy, completeness, simplicity, applicability, consistency, and timeliness; data industry quality dimension, with assessment indicators including the level of industry data resource collection and aggregation, the level of industry big data application, and the benefits of industry big data application; data ecosystem service quality dimension, with assessment indicators including the characteristics of application scenarios in various industries; data brand dimension, with assessment indicators including brand awareness, brand satisfaction, brand loyalty, user search, and social media interaction; historical transaction or event dimension, with assessment indicators including transaction date, transaction amount, description of transaction assets, bank statements, and transaction vouchers; effective control dimension, with assessment indicators including access control and compliance management; expected value dimension, with assessment indicators including expected future cash flows and expected profits; and asset measurement dimension, with assessment indicators including cost measurement and fair value.

[0060] From the perspectives of data asset quality infrastructure, we focus on the basic environment and conditions for data generation; from the perspective of data supply quality, we consider the stability and timeliness of data supply; from the perspective of data product quality, we focus on the quality characteristics of the data product itself; from the perspective of data ecosystem service quality, we focus on the service effectiveness of data in the ecosystem; from the perspective of data industry quality, we assess the impact and value of data on the industry to which it belongs; from the perspective of data brand, we measure the brand influence of data; from the perspective of past transactions or events, we assess the quality of data assets based on past transactions; from the perspective of effective control, we judge the management and control of data assets; from the perspective of expected value, we predict the future value potential of data assets; and from the perspective of asset measurement, we determine the quantitative measurement standards for data assets.

[0061] Compared with existing assessment models and evaluation indicators in the market, the data asset quality assessment system in this application has significant advantages. Existing assessment methods are often relatively singular, mostly focusing on only a specific angle, making it difficult to comprehensively reflect the true state of data asset quality. In contrast, the multi-dimensional data asset quality model constructed in this application considers the entire process from developing data resources into data products and forming data assets. It includes multiple interconnected and interacting quality dimensions, such as the data quality infrastructure dimension and the data supply quality dimension, totaling 10 dimensions. Each dimension is equipped with an independent and scientific indicator system, deeply analyzing data asset quality from different levels.

[0062] For example, in the data product quality dimension, its indicator system covers key indicators such as data accuracy, completeness, and consistency, which can accurately measure the quality of data products; the data ecosystem service quality dimension sets indicators around the service effectiveness of data in the ecosystem. This multi-dimensional evaluation framework breaks through the limitations of traditional evaluation, with each dimension complementing and coordinating with the others, providing a more comprehensive, in-depth, and accurate evaluation result for data asset quality, and powerfully promoting the efficient application of data assets in various fields.

[0063] In this embodiment, the three-dimensional multi-dimensional model of data asset quality consists of 10 interrelated quality dimension planes D1, D2, ..., D... 10 Composition, each dimension Di contains n i One evaluation index {d i1 d i2 , ..., d ini}

[0064] It is understandable that the role of this multidimensional model is to define the "dimensional scope" and "indicator system" of the assessment, solve the problems of "from which angles to assess" and "what indicators to use for quantification"; establish 10 core dimensions (such as data product quality and expected value dimensions) and specific indicators under each dimension (such as "data integrity" and "compliance" indicators), forming the basic framework of the assessment.

[0065] Furthermore, the standard rule base includes an evaluation standard sub-base, an evaluation indicator sub-base, and an evaluation sample base.

[0066] The evaluation standards sub-library stores the norms and guidelines for evaluating the quality of various data assets. These standards serve as the basis for judging the quality of data assets, clearly defining the quality levels that data should achieve in different dimensions, such as the compliance rate of data accuracy and the specific requirements for data integrity, providing a unified scale and standard framework for evaluation work.

[0067] The evaluation indicator sub-library defines specific indicators for measuring data asset quality, corresponding to the dimensions in the multi-dimensional data asset quality model. Each indicator has its own specific calculation method and meaning, providing a quantitative assessment of data asset quality from different perspectives.

[0068] The evaluation sample library collects a large number of representative data asset quality evaluation cases. These cases cover data assets of different industries and types, demonstrating how to use evaluation standards and indicators in actual evaluation processes. They provide practical references for evaluators, enabling them to draw on past experience and conduct data asset quality evaluations more accurately and efficiently.

[0069] In this embodiment, the data asset quality assessment data model specifically includes setting data asset quality dimensions according to the assessment objectives, matching assessment samples to select assessment method standards, determining assessment indicator parameters, and calculating and outputting data asset quality indicator values.

[0070] The first model is a statistical model, which uses statistical indicators such as mean, variance, and frequency distribution to describe the data characteristics of the data to be evaluated, and uses hypothesis testing to determine quality anomalies.

[0071] Among them, statistical indicators such as mean, variance, and frequency distribution can characterize data from different perspectives. The mean reflects the central tendency of the data; if the data asset is a company's sales data, the mean can show the average sales amount and provide insight into sales levels. Variance measures the dispersion of the data; a large variance means that the sales data fluctuates greatly and has poor stability. Frequency distribution presents the frequency of occurrence of various data types, helping to discover patterns in the data. Hypothesis testing is used to determine whether the quality of the data asset meets expectations. Taking data accuracy as an example, an acceptable error range is set as the null hypothesis. By testing the actual data, it is determined whether to reject the null hypothesis. If rejected, it indicates that the data may have accuracy issues.

[0072] Furthermore, the second model is a machine learning model used to train a classification model, and / or a regression model, and / or a clustering model based on labeled samples to detect quality issues in the data to be evaluated;

[0073] Specifically, classification models trained based on labeled samples can categorize data asset quality into different levels, such as high, medium, and low, for easy and intuitive judgment. Regression models are used to predict a quality-related numerical value of a data asset, such as predicting the accuracy score of the data. Clustering models can group similar data assets into one category, uncovering potential data structures. Through these models, abnormal patterns in the data can be detected. For example, if clustering reveals that the characteristics of a certain type of data asset differ significantly from other categories, it may indicate a quality problem; classification and regression models can also identify data points that do not conform to normal patterns.

[0074] Furthermore, the third model is a weight calculation model, which uses the results of the second model output by the first model to determine the main weights of each relevant dimension and the index weights of each evaluation index under the relevant dimension by adopting / using the analytic hierarchy process, entropy method, and principal component analysis.

[0075] Among these methods, the analytic hierarchy process (AHP), entropy method, and principal component analysis (PCA) can determine the weights of each dimension and indicator. The degree of influence of each dimension and indicator on the quality of data assets varies depending on the evaluation scenario. For example, when data assets are used for strategic decision-making, the expected value dimension may have a higher weight; when data assets are used for daily operational monitoring, the data timeliness dimension may have a more important weight. The third model combines the outputs of the first and second models, comprehensively considering various factors to determine the weights, and then calculates the evaluation results, making the evaluation results more scientific and reasonable.

[0076] It is understandable that this data asset quality assessment data model is used to design the "computational logic" and "multi-model fusion strategy" of the assessment, and to solve the problem of "how to calculate results through data".

[0077] The calculation formulas include formulas for single-dimensional quality assessment and comprehensive multi-dimensional quality assessment, as detailed below:

[0078] The formula for calculating single-dimensional quality assessment is as follows:

[0079]

[0080] Where σ(dij) is the standardized value of indicator dij; Wij is the weight of the indicator within the dimension, ∑wij=1; ρik is the correlation coefficient between dimensions; λi is the correlation correction factor, used to control the intensity of cross-dimensional influence.

[0081] The comprehensive multi-dimensional quality assessment calculation formula is as follows:

[0082]

[0083] Where Wi is the dimensional sovereign weight, ∑Wi=1; Wij is the dimensional interaction weight; ρij is the correlation coefficient between dimensions (-1≤ρ≤1).

[0084] In one specific embodiment, the data processing procedure of the multi-dimensional quality assessment model includes the following steps:

[0085] 1. Data preprocessing and indicator standardization:

[0086] Based on the data to be evaluated and the three-dimensional multi-dimensional model, the data to be evaluated is standardized using the index standardization method to obtain standardized index values.

[0087] Optionally, data preprocessing can be performed using the following formula:

[0088]

[0089] Where, dij = Xij / Yij, Xij represents the number of data set elements that meet the quality standard requirements in the ij-th dimension indicator; Yij represents the number of data set elements evaluated in the ij-th dimension indicator; this calculation method is used to determine the value of the original data dij to measure the degree to which the data meets the quality standard under this dimension indicator, and to provide basic data for subsequent standardization calculations.

[0090] The significance of this formula lies in its ability to map data to a specific interval, transforming raw data with different dimensions and value ranges into data of a unified scale, eliminating the impact of differences in dimensions and value ranges on the evaluation results, and making the data of each indicator comparable.

[0091] 2. Multi-model evaluation calculation;

[0092] Statistical models utilize statistical indicators such as mean, variance, and frequency distribution to describe the characteristics and anomalies of data asset quality, and employ statistical hypothesis testing methods to determine whether data asset quality meets expectations or exhibits significant differences. Machine learning models use labeled data asset quality samples to train classification or regression models to estimate data asset quality, and leverage algorithms such as clustering and anomaly detection to identify abnormal patterns and potential quality problems in the data assets. Weight calculation models employ methods such as the analytic hierarchy process (AHP) and entropy to calculate weights, determining the weights Wij within each dimension and the dimension's sovereign weight Wi. Correlation coefficient matrices quantify dimensional relationships, and interaction weight matrices are calculated to construct the correlation relationships between quality dimensions.

[0093] Optionally, various methods can be used to calculate weights, such as the analytic hierarchy process (AHP), entropy method, principal component analysis (PCA), Delphi method, regression analysis, fuzzy comprehensive evaluation, and neural network method. Each of these methods has its own characteristics. For example, the AHP can decompose complex problems into multiple levels and determine the weight of each factor through pairwise comparisons; the entropy method determines weights based on the dispersion of the data—the greater the dispersion, the higher the weight; and the PCA method reduces dimensionality, transforming multiple related indicators into a few uncorrelated comprehensive indicators and determining their weights.

[0094] Furthermore, through the correlation coefficient matrix R 10×10 =[ρ ij Quantify dimensional correlations and calculate the interaction weight matrix.

[0095] It should be noted that the multi-dimensional model of data asset quality includes 10 dimensions, such as data asset quality infrastructure and data supply quality, which are interconnected. The correlation coefficient matrix can accurately measure the strength of the linear relationship between any two dimensions, and its value ranges from -1 to ρ to 1. The closer ρ is to 1, the stronger the positive correlation between the two dimensions.

[0096] Using formula Calculate the interaction weight matrix (Hadamard product). Here, W is the dimensional sovereign weight matrix, and WT is its transpose. The interaction weight matrix reflects the weight relationships between different dimensions, comprehensively considering the correlation between dimensions (represented by R) and the sovereign weight of each dimension (represented by W). When comprehensively assessing the quality of data assets, the interaction weight matrix affects the information transfer and comprehensive calculation between dimensions, thus more comprehensively and accurately reflecting the contribution of each dimension to the overall data asset quality assessment under interrelationship conditions.

[0097] 4. Iteratively calculate dimensional scores:

[0098] initialization Bin iteration correction: Until

[0099] The initialization step calculates the initial dimension score based on the standardized indicator values ​​σ(dij) and the indicator weights wi within the dimension. For example, in data asset quality assessment, for the data asset quality infrastructure dimension, the initial score of this dimension is calculated based on the standardized indicator values ​​of each indicator under this dimension, such as the standardized values ​​of data storage stability and data transmission reliability, combined with their respective weights. This initial score serves as the starting point for iterative calculations.

[0100] 5. Comprehensive multi-dimensional quality assessment calculation

[0101] Where D = [D1, D2, ..., D] 10 ] T

[0102] Among them, the comprehensive multi-dimensional quality assessment calculation is a key step in the comprehensive quantitative assessment of data asset quality. It integrates the results of the previous steps and calculates a comprehensive value through a specific formula to reflect the overall quality of the data assets.

[0103] In the formula, D is a vector composed of the scores of each dimension. These dimension scores are obtained through iterative calculation in 3.4.4, reflecting the quality assessment results of each dimension after multiple revisions. W represents the dimensional sovereign weight vector, which is determined in the weight allocation stage and reflects the relative importance of different dimensions in the comprehensive evaluation. Winter is the interaction weight matrix, which is constructed from the quality dimension correlation steps using the formula. The calculations show the weights reflecting the interactions between dimensions; R is the correlation coefficient matrix, used to quantify dimensional correlation, with element values ​​ranging from -1 to 1, measuring the strength of the linear relationship between any two dimensions; (symbols omitted). This represents the Hadamard product, which is the element-wise multiplication of two matrices.

[0104] During the calculation, first calculate This step comprehensively considers the main weights of each dimension and the adjustments made to the weights based on the relationships between dimensions. Then, a matrix operation is performed with DT and D, and finally, the square root of the result is taken to obtain the comprehensive multi-dimensional quality assessment value S. This calculation process integrates information from each dimension in the three-dimensional multi-dimensional model of data asset quality, including dimension scores, main weights of dimensions, and the relationships between dimensions, comprehensively reflecting the overall quality status of data assets. Through this comprehensive value, users can intuitively understand the overall level of data asset quality, providing strong quantitative evidence for the value-added applications of data assets, such as accounting, circulation, utilization, and pledge financing.

[0105] Furthermore, the formula Di(τ+1)=(1-λi)Di(τ)+λi∑k=iρikDk(τ) is used for iterative correction until |Di(τ+1)-Di(τ)|<∈. In this formula, Di(τ) represents the score of the i-th dimension in the τ-th iteration, λi is the correlation correction factor used to control the strength of cross-dimensional influence, ρik is the correlation coefficient between dimensions, reflecting the degree of correlation between the i-th and k-th dimensions, and Dk(τ) is the score of the k-th dimension in the τ-th iteration. In each iteration, the score of the current dimension is adjusted by comprehensively considering the previous score of the current dimension, the scores of other dimensions, and the correlation between dimensions. When the difference between the dimension scores obtained from two adjacent iterations is less than the set threshold ∈, the iteration is considered to have converged, and the dimension scores obtained at this time are relatively stable and accurate.

[0106] This embodiment of the multi-dimensional data asset quality assessment model mainly comprises three parts: a three-dimensional multi-dimensional data asset quality model, a standard rule base, and a data asset quality assessment data model. The three-dimensional multi-dimensional model covers 10 dimensions, assessing the entire process from data resource development to the formation of data assets, with multiple evaluation indicators for each dimension. The standard rule base provides assessment standards, indicators, and sample support. The assessment data model integrates statistics, machine learning, and weight calculation models. By setting dimensions, matching samples, determining parameters, and performing calculations, it derives the assessment results. The specific calculation steps involved ensure the scientific rigor and accuracy of the assessment, comprehensively and efficiently evaluating data asset quality.

[0107] It should also be noted that this system also includes a display module 40, which is used to generate visual information based on the evaluation results and display the visual information in a multi-dimensional form.

[0108] Understandably, the visualization module is a crucial component. Based on the assessment results, it transforms the data asset quality status into visual information, such as charts and reports. Furthermore, the visualization module presents the results from multiple dimensions, including assessments across 10 dimensions such as data asset quality infrastructure and data supply quality, all displayed intuitively. This multi-dimensional visualization allows users to comprehensively and clearly understand the quality performance of data assets in different aspects, facilitating a quick grasp of the overall data asset quality status and providing strong support for applications such as data asset accounting, circulation, utilization, and pledge financing.

[0109] This system addresses the shortcomings of existing technologies, such as limited evaluation dimensions, strong subjectivity, and low efficiency. It employs key methods including multi-dimensional evaluation models, objective quantitative indicators, and automated processing. A multi-dimensional model of data asset quality is constructed, combining a standard rule base with an evaluation data model based on statistics, machine learning, and weight calculation to assess data asset quality from multiple dimensions. The evaluation process involves first defining evaluation requirements, collecting and organizing data, and preprocessing (standardization transformation), followed by comprehensive multi-model evaluation calculations, and finally visualizing the results. This technical solution effectively solves the problem of low efficiency and poor accuracy in data asset quality assessment due to the lack of effective methods and systems. It achieves comprehensive, objective, and efficient evaluation of data asset quality, more accurately reflecting the status of data assets and meeting the large-scale evaluation needs of data assets in value-added applications such as accounting, circulation, and utilization.

[0110] On the other hand, embodiments of this application also provide a data asset quality assessment method, which is implemented by running on a computer device and executing a computer program. Figure 3 This is a flowchart of a data asset quality assessment method according to an embodiment of this application, such as... Figure 3 As shown, the process includes the following steps:

[0111] S301, through the configuration module, configure the assessment objectives and data scope related to data asset quality assessment based on user interaction information;

[0112] S302, through the data collection module, acquire historical data related to data asset assessment, and obtain the data to be assessed from the historical data according to the data scope;

[0113] S303, through the quality assessment module, based on the assessment objectives, uses a pre-built multi-dimensional quality assessment model to assess the data asset quality of the data to be assessed and obtain the assessment results. The multi-dimensional quality assessment model is based on multiple interrelated three-dimensional dimensions, combined with a standard rule base and a multi-dimensional calculation model, to conduct a comprehensive quality assessment of the data to be assessed.

[0114] S304 generates visual information based on the evaluation results and displays the visual information in a multi-dimensional format.

[0115] Through steps S301 to S304 above, compared to existing technologies that suffer from single evaluation dimensions, strong subjectivity, and low efficiency, this solution employs key methods such as multi-dimensional evaluation models, objective quantitative indicators, and automated processing. It constructs a multi-dimensional, three-dimensional model of data asset quality, combining a standard rule base with an evaluation data model based on statistics, machine learning, and weight calculation to assess data asset quality from multiple dimensions. The evaluation process involves first defining evaluation requirements, collecting and organizing data, and preprocessing (standardization transformation), followed by comprehensive multi-model evaluation calculations, and finally visualizing the results. This technical solution effectively solves the problem of low efficiency and poor accuracy in data asset quality assessment due to the lack of effective methods and systems. It achieves comprehensive, objective, and efficient evaluation of data asset quality, more accurately reflecting the status of data assets and meeting the large-scale evaluation needs of data assets in value-added applications such as accounting, circulation, and utilization.

[0116] In one embodiment, Figure 4 This is a schematic diagram of the internal structure of an electronic device according to an embodiment of this application, such as... Figure 4 As shown, an electronic device is provided, which can be a server, and its internal structure diagram can be as follows. Figure 4As shown, the electronic device includes a processor, a network interface, internal memory, and non-volatile memory connected via an internal bus. The non-volatile memory stores an operating system, computer programs, and a database. The processor provides computing and control capabilities, the network interface communicates with external terminals via a network, the internal memory provides an environment for the operating system, the computer programs are executed by the processor to implement a data asset quality assessment method, and the database stores data.

[0117] Those skilled in the art will understand that Figure 4 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the electronic device to which the present application is applied. The specific electronic device may include more or fewer components than shown in the figure, or combine certain components, or have different component arrangements.

[0118] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. This computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and RAMbus dynamic RAM (RDRAM), etc.

[0119] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0120] The above embodiments merely illustrate several implementation methods of this application, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this patent application should be determined by the appended claims.

Claims

1. A data asset quality assessment system, characterized by, The system is deployed on a computer device and is implemented by a computer program, and comprises a configuration module, a data collection module and a quality evaluation module, wherein: The configuration module is configured to configure evaluation targets and data ranges related to data asset quality evaluation according to user interaction information; The data collection module is configured to obtain historical data related to data asset evaluation and obtain to-be-evaluated data from the historical data according to the data ranges; The quality evaluation module is configured to perform data asset quality evaluation on the to-be-evaluated data based on the evaluation targets by using a pre-constructed multi-dimensional quality evaluation model to obtain evaluation results, wherein the multi-dimensional quality evaluation model is based on multiple interrelated three-dimensional dimensions, combines a standard rule base and a multi-dimensional calculation model, and performs comprehensive quality evaluation on the to-be-evaluated data; The multi-dimensional quality evaluation model comprises: A first model configured to mine and describe data features of the to-be-evaluated data according to statistical indexes and judge quality abnormalities by hypothesis testing, wherein the statistical indexes comprise mean, variance and frequency distribution; A second model configured to train a classification model, a regression model and / or a clustering model based on labeled samples, estimate data asset quality, and detect quality problems of the to-be-evaluated data; A third model configured to determine main weights of each related dimension and index weights of each evaluation index under the related dimension by using an analytic hierarchy process, an entropy method and / or a principal component analysis method based on results output by the second model of the first model.

2. The system of claim 1, wherein, The multi-dimensional quality evaluation model comprises a three-dimensional multi-dimensional model, a standard rule base and a quality evaluation model.

3. The system of claim 2, wherein, The three-dimensional multi-dimensional model stores multiple quality evaluation dimensions that are interrelated in a process in which original data resources form data assets, and each quality evaluation dimension comprises multiple evaluation indexes, which are used to quantify quality features under the quality evaluation dimension; The standard rule base comprises a standard sub-base configured to define scoring standards of each evaluation index under the quality evaluation dimension, an evaluation index sub-base configured to define calculation methods, threshold ranges and corresponding original data resources of each evaluation index, and an evaluation sample sub-base configured to store parameter configuration information in historical evaluation cases and historical typical scenarios, and used to match current evaluation requirements. ​ ​ 4. The system of claim 3, wherein, The quality evaluation dimensions include: a data asset quality infrastructure dimension, whose evaluation indexes include: quality standards, quality detection, and quality certification; a data supply quality dimension, whose evaluation indexes include: supply methods and supply mechanisms; a data product quality dimension, whose evaluation indexes include: accuracy, completeness, conciseness, applicability, consistency, and timeliness; a data industry quality dimension, whose evaluation indexes include: industry data resource collection and aggregation levels, industry big data application levels, and industry big data application benefits; a data ecological service quality dimension, whose evaluation indexes include: characteristics of various industry application scenarios; a data brand dimension, whose evaluation indexes include: brand awareness, brand satisfaction, brand loyalty, user search, and social media interaction; a historical transaction or matter dimension, whose evaluation indexes include: transaction dates, transaction amounts, transaction asset descriptions, bank flow, and transaction vouchers; an effective control dimension, whose evaluation dimensions include: access control and compliance control; an expected value dimension, whose evaluation indexes include: expected future cash flow, expected profit; an asset measurement dimension, whose evaluation indexes include: cost measurement and fair value.

5. The system of claim 3, wherein, The multi-dimensional quality evaluation model is used for: Based on the to-be-evaluated data and the three-dimensional multi-dimensional model, using an index standardization method, the to-be-evaluated data is standardized to obtain standardized index values; Through the first model and the second model, the to-be-evaluated data is processed to obtain data features and abnormal information; Through the third model, the main weight and the index weight corresponding to each standardized index value are calculated, the correlation coefficient matrix between dimensions is quantified through a correlation coefficient matrix, and an interaction weight matrix is calculated; And through a preset formula, the single-dimensional score and the comprehensive dimensional score are calculated according to the standardized index value, the main weight, the index weight, the correlation coefficient matrix, and the interaction weight matrix, wherein, in the calculation process, first, an initialization result is obtained, and then the dimensions are calculated through iteration until a preset convergence condition is met, to obtain the single-dimensional score and the comprehensive dimensional score, wherein: The single-dimensional quality evaluation calculation formula is as follows: where D is the single-dimension score, is the index standardized value; is the intra-dimension index weight; is the inter-dimension correlation coefficient; is the correlation correction factor, used to control the strength of cross-dimension influence. The comprehensive multi-dimensional quality evaluation calculation formula is as follows: wherein, is the comprehensive multi-dimensional evaluation score, is the dimension main weight; is the dimension interaction weight; is the correlation coefficient between dimensions.

6. The system of claim 2, wherein, The system further includes a display module, wherein the display module is used for generating visual information based on the evaluation result; And the visual information is displayed in a multi-dimensional form.

7. A method of data asset quality assessment, characterized by, The method runs on a computer device and is implemented by executing a computer program, including: Through a configuration module, evaluation targets and data ranges related to data asset quality evaluation are configured according to user interaction information; Through a data collection module, historical data related to data asset evaluation is obtained, and to-be-evaluated data is obtained from the historical data according to the data range; Through a quality evaluation module, based on the evaluation targets, a multi-dimensional quality evaluation model pre-constructed is used to perform data asset quality evaluation on the to-be-evaluated data, to obtain an evaluation result, wherein the multi-dimensional quality evaluation model is based on multiple interrelated three-dimensional dimensions, combines a standard rule library and a multi-dimensional calculation model, and performs comprehensive quality evaluation on the to-be-evaluated data; The multi-dimensional quality evaluation model comprises: A first model configured to mine and describe data features of the data to be evaluated according to statistical indexes, and determine quality abnormalities through hypothesis testing, wherein the statistical indexes comprise mean, variance, and frequency distribution; A second model configured to train a classification model, and / or a regression model, and / or a clustering model based on labeled samples, estimate data asset quality, and detect quality problems of the data to be evaluated; A third model configured to determine main weights of each relevant dimension and index weights of each evaluation index under the relevant dimension based on results output by the second model of the first model by using analytic hierarchy process, and / or entropy method, and / or principal component analysis.

8. A computer device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, The processor executes the computer program to implement the method in claim 7.

9. A computer readable storage medium having stored thereon a computer program, characterized in that, The program is executed by the processor to implement the method in claim 7. The program is executed by the processor to implement the method in claim 7.

Images