Enterprise operation data analysis system and method based on big data analysis

Abstract

The invention discloses an enterprise operation data analysis system and method based on big data analysis, enterprise operation data is collected through multiple channels, it is ensured that the data comprehensively covers all aspects of an enterprise, the timeliness and accuracy of the data are ensured through real-time or regular collection, the enterprise is assisted to grasp market dynamics and internal operation, and the enterprise operation efficiency is improved. The complementarity between data sets is evaluated, the data sets which can be mutually complemented are identified, the comprehensiveness and the accuracy of data analysis are improved, data resources are efficiently integrated by utilizing data complementarity indexes, redundancy is avoided, data value maximization is realized, and the relevance and the similarity of data items are revealed based on a similarity matrix constructed by data feature vectors; and a deep analysis module combines a complementarity index and a similarity matrix, applies a deep learning algorithm, automatically identifies key business indexes, predicts enterprise development trends, identifies potential risks, provides decision support, visually displays analysis results, and improves management layer decision efficiency.

Description

Technical Field

[0001] The present invention relates to the technical field of enterprise operation data analysis, and particularly to an enterprise operation data analysis system and method based on big data analysis. Background Art

[0002] With the rapid development of information technology, the scale and complexity of enterprise operation data have increased rapidly, which has brought unprecedented challenges to enterprise data analysis and decision-making. Although there are various enterprise operation data analysis systems on the current market, these systems still have a series of significant defects in processing large-scale and diverse data and providing in-depth and comprehensive analysis results.

[0003] First of all, the existing systems have obvious deficiencies in data processing capabilities. Facing the massive and diverse enterprise operation data, traditional data processing technologies and tools often seem powerless. The efficiency of data preprocessing, cleaning, and formatting is low, resulting in a long time-consuming construction of the data warehouse and it is difficult to ensure the accuracy and consistency of the data. This not only affects the timeliness of data analysis but also reduces the reliability of the analysis results.

[0004] Secondly, the existing systems have limitations in the depth of data analysis. Although some systems provide basic data mining and analysis functions, in identifying key business indicators, trend prediction, and potential risks, they often can only provide superficial and shallow analysis results and cannot deeply explore the potential value and patterns behind the data.

[0005] In addition, the existing systems also have obvious defects in the evaluation of data complementarity and similarity. The complementarity between different data sets and the similarity between data items are crucial for improving the accuracy and reliability of analysis results. However, the existing systems often lack effective technical means to evaluate these indicators, resulting in enterprises being difficult to give full play to the complementary advantages of data when integrating and using different data sources and also difficult to discover the potential connections and patterns between data. Summary of the Invention

[0006] In view of this, the present invention proposes an enterprise operation data analysis system and method based on big data analysis, which can effectively solve the defects existing in the prior art in terms of data processing capabilities, analysis depth, data complementarity, and similarity evaluation.

[0007] The technical solution of the present invention is realized as follows:

[0008] An enterprise operation data analysis system based on big data analysis, comprising:

[0009] A data acquisition module, configured to collect enterprise operation data from multiple enterprise data sources;

[0010] A data preprocessing module for cleaning and formatting the collected data and constructing a standardized data warehouse;

[0011] A data complementarity index calculation module for evaluating the complementarity degree between different data sets using a preset algorithm and generating a data complementarity index;

[0012] A similarity matrix construction module for constructing a similarity matrix between data items based on data feature vectors;

[0013] A deep analysis module for deeply mining enterprise operation data by combining the data complementarity index and the similarity matrix and using deep learning algorithms to identify key business indicators, trend predictions, and potential risks;

[0014] A visualization report generation module for visually displaying the analysis results.

[0015] As a further optional solution of the enterprise operation data analysis system based on big data analysis, the data complementarity index calculation module uses a preset algorithm to evaluate the complementarity degree between different data sets and generate a data complementarity index, specifically including:

[0016] Select key features that can reflect the characteristics of the data set according to the data characteristics and analysis requirements;

[0017] Use natural language processing technology to extract the features of the data set;

[0018] Calculate the information volume of each data set and its subsets;

[0019] Calculate the unique information volume of each data set and the common information volume between different data sets according to the results of feature extraction and information volume calculation;

[0020] Calculate the data complementarity index based on the unique information volume of each data set and the common information volume between different data sets.

[0021] As a further optional solution of the enterprise operation data analysis system based on big data analysis, calculating the unique information volume of each data set and the common information volume between different data sets according to the results of feature extraction and information volume calculation specifically includes:

[0022] Calculate the information entropy of each feature, as well as the joint entropy and conditional entropy between features;

[0023] Calculate the unique information volume of each data set by multiplying the number of common features between data sets by the average information entropy of all features;

[0024] Calculate the common information volume between different data sets by defining the mutual information between different data sets based on joint entropy and conditional entropy.

[0025] As a further optional solution of the enterprise operation data analysis system based on big data analysis, the similarity matrix construction module constructs a similarity matrix between data items based on data feature vectors, specifically including:

[0026] Extract the feature vectors of each data item from the data set;

[0027] Assign a weight factor to each feature vector;

[0028] Calculate the similarity between each pair of data items according to the similarity measurement formula;

[0029] Store the calculation results in a two-dimensional array to form a similarity matrix between data items.

[0030] As a further optional solution of the enterprise operation data analysis system based on big data analysis, the specific similarity measurement formula is:

[0031]

[0032] Where A and B are the feature vectors of two data items, ai and bi are the values of A and B on the i-th feature respectively, wi is the weight factor of the i-th feature, and n is the number of features.

[0033] As a further optional solution of the enterprise operation data analysis system based on big data analysis, the deep analysis module combines the data complementarity index and the similarity matrix, and uses deep learning algorithms to deeply mine the enterprise operation data, identify key business indicators, trend predictions and potential risks, specifically including:

[0034] Calculate the comprehensive influence index of each business according to the data complementarity index and the similarity matrix, and identify the key business that has the greatest impact on the overall business;

[0035] Perform time series prediction on key business indicators to obtain the trend within a preset time;

[0036] Analyze the change trend and abnormal points of business indicators according to the key business and the trend within the preset time, and identify potential risks;

[0037] Formulate corresponding countermeasures according to the identified risks.

[0038] An enterprise operation data analysis method based on big data analysis specifically includes:

[0039] Data collection step: Automatically collect enterprise operation data from multiple enterprise data sources;

[0040] Data preprocessing steps: Clean and format the collected data, and build a standardized data warehouse;

[0041] Data complementarity index calculation steps: Use a preset algorithm to evaluate the complementarity degree between different data sets and generate a data complementarity index;

[0042] Similarity matrix construction steps: Based on the data feature vectors, construct a similarity matrix between data items;

[0043] Deep analysis steps: Combine the data complementarity index and the similarity matrix, and use deep learning algorithms to deeply mine the enterprise operation data, identify key business indicators, conduct trend prediction and potential risk analysis;

[0044] Visualization report generation steps: Visualize the results of the deep analysis.

[0045] A computing device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps of the above-mentioned enterprise operation data analysis method based on big data analysis.

[0046] A computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it implements the steps of the above-mentioned enterprise operation data analysis method based on big data analysis.

[0047] The beneficial effects of the present invention are as follows: By collecting data from multiple enterprise data sources, it can ensure that the obtained enterprise operation data is comprehensive, covering all aspects of the enterprise. Collecting data in real time or regularly ensures the timeliness and accuracy of the data, which helps the enterprise to obtain market dynamics and internal operation conditions in a timely manner. By evaluating the complementarity degree between different data sets, it is possible to identify which data sets can complement each other, thereby improving the accuracy and comprehensiveness of data analysis. Using the data complementarity index, data resources can be utilized and integrated more effectively, avoiding data redundancy and waste, and maximizing the data value. The similarity matrix constructed based on data feature vectors can reveal the correlation and similarity between data items, which helps to discover potential business rules and trends. The similarity matrix provides important input information for the deep analysis module, which helps to more deeply mine the business value in the data. Through deep learning algorithms, it is possible to automatically identify the key business indicators that affect the enterprise operation and provide decision-making support for the management. Combining the data complementarity index and the similarity matrix, the deep analysis module can predict the future development trend of the enterprise and identify potential business risks, helping the enterprise to take countermeasures in advance. Displaying the analysis results in a visual form enables the management to more intuitively understand the data and analysis results, improving the decision-making efficiency. Brief Description of the Drawings

[0048] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0049] Figure 1 It is a schematic diagram of the composition of an enterprise operation data analysis system based on big data analysis of the present invention;

[0050] Figure 2 It is a schematic diagram of the process of an enterprise operation data analysis method based on big data analysis of the present invention;

[0051] Figure 3 It is a schematic diagram of the composition of a computing device of the present invention. Detailed Embodiments

[0052] The following will clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, rather than all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0053] Refer to Figures 1 to 3 , an enterprise operation data analysis system based on big data analysis, including:

[0054] A data collection module, used to collect enterprise operation data from multiple enterprise data sources;

[0055] A data preprocessing module, used to clean, format the collected data, and build a standardized data warehouse;

[0056] A data complementarity index calculation module, used to evaluate the complementarity degree between different data sets using a preset algorithm, and generate a data complementarity index;

[0057] A similarity matrix construction module, used to construct a similarity matrix between data items based on data feature vectors;

[0058] A deep analysis module, used to combine the data complementarity index and the similarity matrix, and use deep learning algorithms to deeply mine enterprise operation data, identify key business indicators, trend predictions, and potential risks;

[0059] A visualization report generation module, used to visually display the analysis results.

[0060] In this embodiment, by collecting data from multiple enterprise data sources, it is possible to ensure the comprehensiveness of the obtained enterprise operation data, covering all aspects of the enterprise. Data is collected in real-time or regularly, ensuring the timeliness and accuracy of the data, which helps the enterprise to obtain market dynamics and internal operation conditions in a timely manner. By evaluating the complementarity degree between different data sets, it is possible to identify which data sets can complement each other, thereby improving the accuracy and comprehensiveness of data analysis. Using the data complementarity index, data resources can be utilized and integrated more effectively, avoiding data redundancy and waste, and maximizing the data value. The similarity matrix constructed based on the data feature vectors can reveal the correlation and similarity between data items, helping to discover potential business rules and trends. The similarity matrix provides important input information for the in-depth analysis module, contributing to a more in-depth exploration of the business value in the data. Through deep learning algorithms, it is possible to automatically identify the key business indicators that affect enterprise operation, providing decision-making support for the management. Combining the data complementarity index and the similarity matrix, the in-depth analysis module can predict the future development trend of the enterprise and identify potential business risks, helping the enterprise to take countermeasures in advance. The analysis results are presented in a visual form, enabling the management to more intuitively understand the data and the analysis results, and improving the decision-making efficiency.

[0061] Preferably, the data complementarity index calculation module uses a preset algorithm to evaluate the complementarity degree between different data sets and generates a data complementarity index, specifically including:

[0062] According to the data characteristics and analysis requirements, select the key features that can reflect the characteristics of the data set;

[0063] Use natural language processing technology to extract the features of the data set;

[0064] Calculate the information content of each data set and its subsets;

[0065] According to the results of feature extraction and information content calculation, calculate the unique information content of each data set and the common information content between different data sets;

[0066] Based on the unique information content of each data set and the common information content between different data sets, calculate the data complementarity index.

[0067] In this embodiment, according to the data characteristics and analysis requirements, key features that can reflect the characteristics of the dataset are selected. This step ensures the pertinence and effectiveness of subsequent analysis, avoiding redundancy and unnecessary calculations. Natural language processing technology is used to extract the features of the dataset. Natural language processing technology can process and understand human language data and extract useful information from it. In the calculation of the data complementarity index, it can help identify and understand the text information in the dataset and extract key features and indicators. For example, through techniques such as named entity recognition and dependency syntactic analysis, key information such as proper nouns and semantic relationships in the dataset can be accurately extracted, providing an accurate basis for subsequent complementarity calculations. Calculate the information content of each dataset and its subsets. Information content is an important indicator to measure the richness of information contained in a dataset. By calculating the information content, the information content and redundancy contained in each dataset and its subsets can be understood. According to the results of feature extraction and information content calculation, further calculate the unique information content of each dataset and the common information content between different datasets. The unique information content reflects the unique part of the dataset that is not contained in other datasets, while the common information content reveals the overlapping part of the information between different datasets. Based on the unique information content of each dataset and the common information content between different datasets, calculate the data complementarity index. This index can objectively reflect the degree of complementarity between different datasets and provides an important basis for data integration and analysis. By calculating the data complementarity index, it can be identified which datasets have a high degree of complementarity, so that these datasets can be used more effectively in data analysis, improving the utilization efficiency and accuracy of data. The technical solution of the data complementarity index calculation module can optimize the data analysis process, making the data analysis more efficient and accurate. Through precise feature extraction, accurate measurement of information content, and reliable calculation of the data complementarity index, more accurate and valuable data analysis results can be provided for enterprises.

[0068] It should be noted that based on the unique information content of each dataset and the common information content between different datasets, the data complementarity index is calculated. The specific calculation formula is as follows:

[0069]

[0070] Among them, I(X) represents the information content of dataset X, which can be measured using information entropy;

[0071] A ∩ BC represents the part that exists in dataset A but does not exist in dataset B;

[0072] B ∩ A C represents the part that exists in dataset B but does not exist in dataset A;

[0073] I(A ∩ B) represents the information content jointly owned by dataset A and dataset B.

[0074] Preferably, according to the results of feature extraction and information quantity calculation, calculate the unique information quantity of each data set and the common information quantity between different data sets, specifically including:

[0075] Calculate the information entropy of each feature, as well as the joint entropy and conditional entropy between features;

[0076] Multiply the number of common features between data sets by the average information entropy of all features to calculate the unique information quantity of each data set;

[0077] Define the mutual information between different data sets based on joint entropy and conditional entropy, and calculate the common information quantity between different data sets.

[0078] In this embodiment, the information entropy is an index to measure the amount of information contained in a single feature in a data set. By calculating the information entropy of each feature, the importance of each feature in the data set can be understood, providing a basis for subsequent analysis. The calculation of information entropy helps to identify redundant features in the data set, that is, those features with lower information entropy and little contribution to the analysis results, which can be eliminated in subsequent analysis to improve the analysis efficiency; the joint entropy is used to measure the amount of information jointly contained by multiple features. By calculating the joint entropy, the correlation between different features can be understood, providing a basis for subsequent feature selection and combination. The conditional entropy is used to measure the amount of new information contained by other features when a certain feature is known. The calculation of conditional entropy helps to identify the dependence relationship between features, thereby optimizing the feature combination and improving the accuracy of the analysis results; by multiplying the number of common features between data sets by the average information entropy of all features to calculate the unique information quantity, the uniqueness of each data set at the feature level can be accurately measured. The calculation result of the unique information quantity provides an important basis for the selection of data sets. In the process of data integration and analysis, data sets with higher unique information quantity can be preferentially selected to improve the comprehensiveness and accuracy of the analysis results; by defining the mutual information between different data sets based on joint entropy and conditional entropy, the common information quantity between different data sets can be calculated, which helps to reveal the correlation between different data sets. The calculation result of the common information quantity helps to optimize the data integration strategy. In the process of data integration, data sets with higher common information quantity can be preferentially considered for merging or correlation analysis to improve the efficiency and accuracy of data integration; through the calculation of information entropy, joint entropy and conditional entropy, and the evaluation of unique information quantity and common information quantity, the data processing process is optimized, which helps to reduce redundancy and errors in the data processing process and improve the efficiency and accuracy of data processing.

[0079] It should be noted that the information entropy calculation:

[0080] For numerical features, they can be discretized into categorical features or the calculation method of continuous entropy can be used;

[0081] For categorical features, calculate the information entropy of each feature, specifically:

[0082]

[0083] where x is the feature, and P(x i ) is the probability that the feature value xi appears;

[0084] Joint entropy and conditional entropy:

[0085] Calculate the joint entropy of two or more features, specifically:

[0086]

[0087] Calculate the conditional entropy, specifically:

[0088] H(Y / X) = H(X,Y) - H(X).

[0089] Preferably, the similarity matrix construction module constructs a similarity matrix between data items based on data feature vectors, specifically including:

[0090] Extract the feature vectors of each data item from the dataset;

[0091] Assign a weight factor to each feature vector;

[0092] Calculate the similarity between each pair of data items according to the similarity measurement formula;

[0093] Store the calculation results in a two-dimensional array to form a similarity matrix between data items.

[0094] Preferably, the similarity measurement formula is specifically:

[0095]

[0096] where A and B are the feature vectors of two data items, ai and bi are the values of A and B on the i-th feature respectively, wi is the weight factor of the i-th feature, and n is the number of features.

[0097] In this embodiment, through feature vector extraction and weight factor assignment, this technical solution can more accurately represent the main attributes and importance of data items, thereby improving the accuracy of data representation. By using a self-created similarity metric formula, it can accurately calculate the similarity between data items, avoiding errors and biases in similarity calculations, which helps improve the accuracy of subsequent data analysis and mining results. The similarity matrix can intuitively display the similarity degree between data items, providing strong support for data analysis and mining. Through the similarity matrix, potential associations and patterns between data items can be discovered. This technical solution uses a two-dimensional array to store the similarity matrix, which has high computational efficiency and scalability. As the data volume increases, the size of the two-dimensional array can be adjusted to adapt to the new data scale and maintain the stability of computational performance.

[0098] Preferably, the in-depth analysis module combines the data complementarity index and the similarity matrix, and uses deep learning algorithms to deeply mine the enterprise operation data to identify key business indicators, trend prediction, and potential risks. Specifically, it includes:

[0099] Calculate the comprehensive influence index of each business based on the data complementarity index and the similarity matrix, and identify the key business that has the greatest impact on the overall business;

[0100] Perform time series prediction on key business indicators to obtain trends within a preset time;

[0101] Analyze the change trends and abnormal points of business indicators based on the key business and trends within the preset time, and identify potential risks;

[0102] Formulate corresponding countermeasures according to the identified risks.

[0103] In this embodiment, based on the data complementarity index and the similarity matrix, this module can calculate the comprehensive influence index of each business. The key to this step lies in reasonably setting the construction method of the data complementarity index and the similarity matrix to ensure the accuracy and reliability of the calculation results; through the comprehensive influence index, the module can identify the key business that has the greatest impact on the overall business. This helps the enterprise concentrate resources, optimize the business structure, and enhance overall competitiveness; perform time series prediction on key business indicators to obtain trends within a preset time. This step depends on the selection and training of deep learning algorithms, as well as the accuracy and integrity of time series data; based on the key business and trends within the preset time, the module can analyze the change trends and abnormal points of business indicators, thereby identifying potential risks. This helps the enterprise take measures in advance to prevent risks and ensure the stable development of the business; according to the identified risks, the module can formulate corresponding countermeasures. These countermeasures should be targeted and operable, and can effectively solve the problems and challenges faced by the enterprise.

[0104] It should be noted that according to the data complementarity index and similarity matrix, the comprehensive influence index of each business is calculated, and the specific formula is:

[0105] CII(Bi) = w1·P(Bi) + w2·C(Bi) + w3·R(Bi);

[0106] Among them, P(Bi): The performance index of business Bi, which can be sales, market share, profit, etc. This index measures the performance of the business itself;

[0107] C(B i ): The complementarity score of business Bi with other businesses. This score can be obtained by calculating the average or weighted average of the complementarity indexes of business Bi with all other businesses. The complementarity index can be designed based on factors such as the similarity of business feature vectors, the overlap of market demands, and the degree of resource sharing;

[0108] R(B i ): The relative position of business B i in the entire business portfolio, which can be calculated through the similarity matrix. This index measures the degree of association of business Bi with other businesses and its importance in the network. The PageRank algorithm or a similar graph network analysis method is used to calculate the relative position score of each business;

[0109] w 1 , w 2 , w 3 : Weight factors used to adjust the contributions of the performance index, complementarity score, and relative position score to the comprehensive influence index. These weights can be determined according to business requirements, data characteristics, and analysis objectives.

[0110] An enterprise operation data analysis method based on big data analysis specifically includes:

[0111] Data collection step: Automatically collect enterprise operation data from multiple enterprise data sources;

[0112] Data preprocessing step: Clean and format the collected data and build a standardized data warehouse;

[0113] Data complementarity index calculation step: Use a preset algorithm to evaluate the complementarity degree between different data sets and generate a data complementarity index;

[0114] Similarity matrix construction step: Based on data feature vectors, construct a similarity matrix between data items;

[0115] Deep analysis step: Combining the data complementarity index and the similarity matrix, using deep learning algorithms to deeply mine the enterprise operation data, identify key business indicators, conduct trend prediction and potential risk analysis;

[0116] Visualization report generation step: Visualize the results of the deep analysis.

[0117] A computing device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the above-mentioned enterprise operation data analysis method based on big data analysis are implemented.

[0118] A computer-readable storage medium stores a computer program. When the computer program is executed by a processor, the steps of the above-mentioned enterprise operation data analysis method based on big data analysis are implemented.

[0119] The above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. An enterprise management data analysis system based on big data analysis, characterized in that: include: A data collection module is used to collect enterprise operation data from multiple enterprise data sources; Data preprocessing module, used to clean and format the collected data and build a standardized data warehouse; A data complementarity index calculation module is used to evaluate the degree of complementarity between different data sets using a preset algorithm and generate a data complementarity index; A similarity matrix building module is used to build a similarity matrix between data items based on data feature vectors; The deep analysis module is used to combine data complementarity indicators and similarity matrices, apply deep learning algorithms, conduct in-depth mining of enterprise operating data, and identify key business indicators, trend forecasts, and potential risks; The visual report generation module is used to visualize the analysis results.

2. The enterprise management data analysis system based on big data analysis according to claim 1 is characterized in that: The data complementarity index calculation module uses a preset algorithm to evaluate the degree of complementarity between different data sets and generate a data complementarity index, specifically including: According to the data characteristics and analysis requirements, select key features that can reflect the characteristics of the data set; Use natural language processing techniques to extract features from data sets; Calculate the information content of each data set and its subsets; Based on the results of feature extraction and information calculation, the unique information of each data set and the common information between different data sets are calculated; The data complementarity index is calculated based on the unique information of each data set and the common information between different data sets.

3. The enterprise management data analysis system based on big data analysis according to claim 2 is characterized in that: The method of calculating the unique information of each data set and the common information between different data sets based on the results of feature extraction and information calculation specifically includes: Calculate the information entropy of each feature, as well as the joint entropy and conditional entropy between features; The unique information content of each data set is calculated by multiplying the number of common features between data sets by the average information entropy of all features; The mutual information between different data sets is defined based on joint entropy and conditional entropy, and the common information between different data sets is calculated.

4. The enterprise management data analysis system based on big data analysis according to claim 3 is characterized in that: The similarity matrix construction module constructs a similarity matrix between data items based on the data feature vector, specifically including: Extract the feature vector for each data item from the dataset; Assign a weight factor to each eigenvector; According to the similarity measurement formula, the similarity between each pair of data items is calculated; The calculation results are stored in a two-dimensional array to form a similarity matrix between data items.

5. The enterprise operation data analysis system based on big data analysis according to claim 4 is characterized in that: The similarity measurement formula is specifically: Among them, A and B are the feature vectors of the two data items, ai and bi are the values ​​of A and B on the i-th feature respectively, wi is the weight factor of the i-th feature, and n is the number of features.

6. The enterprise operation data analysis system based on big data analysis according to claim 5 is characterized in that: The in-depth analysis module combines data complementarity indicators and similarity matrices, uses deep learning algorithms, and conducts in-depth mining of enterprise business data to identify key business indicators, trend forecasts, and potential risks, including: Based on the data complementarity index and similarity matrix, calculate the comprehensive influence index of each business and identify the key businesses that have the greatest impact on the overall business; Conduct time series forecasts on key business indicators to obtain trends within a preset time period; Analyze the changing trends and abnormal points of business indicators based on key businesses and trends within a preset time period to identify potential risks; Develop appropriate response strategies based on the identified risks.

7. A method for analyzing business operation data based on big data analysis, characterized in that: Specifically include: Data collection steps: Automatically collect enterprise operating data from multiple enterprise data sources; Data preprocessing steps: clean and format the collected data and build a standardized data warehouse; Data complementarity index calculation steps: using a preset algorithm to evaluate the degree of complementarity between different data sets and generate a data complementarity index; Similarity matrix construction steps: Based on the data feature vector, construct the similarity matrix between data items; In-depth analysis steps: Combine data complementarity indicators and similarity matrices, use deep learning algorithms to conduct in-depth mining of corporate operating data, identify key business indicators, conduct trend forecasts and potential risk analysis; Visual report generation steps: Visualize the results of in-depth analysis.

8. A computing device, characterized in that It includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the enterprise business data analysis method based on big data analysis as described in claim 7 are implemented.

9. A computer-readable storage medium, characterized in that: The storage medium stores a computer program, which, when executed by a processor, implements the steps of the business data analysis method based on big data analysis as described in claim 7.

Images