Information management method and device based on smart business

Abstract

This invention relates to the field of data management technology and discloses a method and device for information management based on smart business. This method effectively achieves compatibility of heterogeneous data from multiple sources. Raw data is acquired through a pre-set standardized protocol and converted into a standardized data set using a format conversion algorithm. This addresses the challenge of diverse data sources and formats, laying the foundation for subsequent processing. A clustering algorithm groups and processes feature attribute data to meet heterogeneity threshold requirements, further regularizing the data and making it more usable.

Description

Technical Field

[0001] The present invention relates to the field of data management technology, and in particular to an information management method and device based on smart business. Background Art

[0002] As a core pillar of smart business development, information management is directly related to a company's decision-making efficiency and competitive advantage in complex market environments. Its importance is self-evident. With the increasing diversification of business scenarios, the integration and intelligent processing of multi-source data has become key to driving business innovation.

[0003] Existing technologies typically process information or data streams separately, treating each piece of data separately. However, due to the lack of unified standards at the data collection layer, compatibility issues between heterogeneous data frequently arise. The data processing layer lacks the ability to model dynamic relationships, making it difficult to capture deep information relevance. Furthermore, the data analysis layer and the presentation layer are poorly connected, making it difficult to quickly translate analytical conclusions into actionable business insights.

[0004] In summary, the existing technology is separated in the data processing and analysis links, resulting in low efficiency of information flow and failure to meet the requirements of real-time and accuracy. Summary of the Invention

[0005] The present invention provides an information management method and device based on smart business to achieve compatibility, data modeling and dynamic presentation of multi-source heterogeneous data.

[0006] In a first aspect, in order to solve the above technical problems, the present invention provides an information management method based on smart business, comprising:

[0007] Obtain multi-source raw data through a preset standardized protocol, map it into a unified structure through a format conversion algorithm, and obtain a standardized data set;

[0008] Extracting characteristic attribute data from the normalized data set, grouping the characteristic attribute data using a clustering algorithm, and using grouping results that meet a heterogeneity threshold as a classified data set;

[0009] Acquire time series features according to the classified data set, and perform dynamic change sampling using a sliding window technique based on the difficulty of acquiring the time series features to obtain dynamic data samples;

[0010] Construct a relationship network based on the dynamic data samples and model and mine implicit associations, optimize the information flow path and transmission based on the results, and obtain target transmission data;

[0011] Extracting key indicators based on the target transmission data, and obtaining a predicted value of short-term resource demand and a real-time load of the current system resource usage status, fusing the predicted value and the real-time load through a weighted average algorithm to obtain an adjusted resource demand assessment value;

[0012] Determine a weight distribution ratio according to the adjusted resource demand assessment value, and perform error prediction on the weight distribution ratio through a regression algorithm to obtain analysis result data;

[0013] Generate a multidimensional feature vector based on the analysis result data, perform dimensionality reduction processing on the data analysis connection requirements through the principal component analysis algorithm, and optimize feature extraction to adjust the display parameter range to obtain associated data;

[0014] Constructing a visualization model based on the dimensionality reduction analysis data, converting the multidimensional feature vector into a graphic element through data mapping, and adjusting the display parameters of the graphic element according to the intuitive presentation capability requirements to obtain visualization presentation parameters;

[0015] User feedback is obtained based on the visualization presentation parameters, and feedback data is obtained by optimizing the system-level collaborative requirements through an iterative update algorithm. Optimized system data is obtained by adjusting the system level based on the feedback data and preset improvement conditions.

[0016] In an optional embodiment, characteristic attribute data is extracted from the normalized data set, the characteristic attribute data is grouped using a clustering algorithm, and grouping results that meet a heterogeneity threshold are used as a classification data set, including:

[0017] Acquire structural features according to the classification data set, perform category distribution processing using statistical tools, and obtain a distribution data set;

[0018] Extracting key fields from the distributed data set, and if the key fields meet the consistency check, determining a consistent data set; if not, adjusting the field content through a field mapping tool to obtain an adjusted data set;

[0019] Adjusting the data set according to its category distribution, identifying outliers through a clustering algorithm, and obtaining an abnormally marked data set;

[0020] According to the labeling information of the abnormal labeled data set, the abnormal points are corrected by using a mean filling tool to obtain a corrected data set;

[0021] Extracting related fields based on the modified data set, constructing a query structure using an indexing tool, and obtaining a query data set;

[0022] According to the complete records of the query data set, adjusting the order according to the category distribution by a sorting tool to obtain a sorted data set;

[0023] According to the structural characteristics of the sorted data set, integrity verification is performed using a consistency verification tool to obtain a verification data set.

[0024] In an optional embodiment, a relationship network is constructed based on the dynamic data sample and implicit associations are mined by modeling, and information flow paths and transmission are optimized based on the results to obtain target transmission data, including:

[0025] Constructing a relational network based on the dynamic data samples, performing multi-layer data relationship modeling through a graph neural network algorithm, and mining implicit associations between nodes requiring deep data processing to obtain relation-enhanced data;

[0026] Obtaining an information flow path based on the relationship enhancement data, optimizing the data transmission sequence using a shortest path algorithm, and adjusting the path weight according to information flow efficiency requirements to obtain target transmission data;

[0027] In an optional embodiment, key indicators are extracted based on the target transmission data, and a predicted value of short-term resource demand and a real-time load of the current system resource usage status are obtained. The predicted value and the real-time load are integrated using a weighted average algorithm to obtain an adjusted resource demand assessment value, including:

[0028] Obtaining key indicators based on the target transmission data, combining the predicted value of short-term resource demand, and performing real-time load fusion through a weighted average method to obtain preliminary resource demand data;

[0029] Determining whether there is abnormal fluctuation based on the sudden increase trend of the preliminary resource demand data, and obtaining fluctuation detection data;

[0030] According to the downward trend in the fluctuation detection data, adjusting the resource allocation order by a sorting tool to obtain rearranged resource data;

[0031] According to the system status of the rearranged resource data, balanced load data is obtained by acquiring load balancing distribution;

[0032] Acquire data fusion characteristics according to the balanced load data, and if the sudden increase trend exceeds a preset threshold, adjust resource demand to obtain revised demand data;

[0033] Obtaining verification demand data by determining whether resource requirements satisfy system status based on the estimated value of the modified demand data;

[0034] According to the key indicators of the verification demand data, resource allocation adjustment is performed through a mean filling tool to obtain a resource demand assessment value.

[0035] In an optional embodiment, determining a weight distribution ratio according to the adjusted resource demand assessment value, and performing error prediction on the weight distribution ratio through a regression algorithm to obtain analysis result data, including: determining the weight distribution ratio according to the adjusted resource demand assessment value, and predicting the real-time requirement and the accuracy requirement through a regression analysis algorithm to obtain preliminary prediction data;

[0036] If the prediction error in the preliminary prediction data is lower than a preset threshold, the analysis result data is determined; if it is higher than the preset threshold, the model parameters are adjusted to perform a re-prediction to obtain the revised prediction data;

[0037] According to the real-time requirement of the revised forecast data, trend analysis data is obtained by analyzing the changing trend of the allocation ratio;

[0038] According to the accuracy requirements of the trend analysis data, the weight distribution is adjusted through the mean filling tool to obtain balanced distribution data;

[0039] Based on the analysis results of the balanced allocation data, the fluctuation of resource demand is detected to obtain fluctuation detection data;

[0040] According to the change trend of the fluctuation detection data, the estimated value of the resource demand is adjusted to obtain the final demand data;

[0041] Key indicators are extracted according to the final demand data, and verification result data is obtained as analysis result data by judging whether the real-time requirement meets the system status.

[0042] In an optional embodiment, a multidimensional feature vector is generated based on the analysis result data, a principal component analysis algorithm is used to reduce the dimension of the data analysis connection requirements, and feature extraction is optimized to adjust the display parameter range to obtain associated data, including:

[0043] Generate a multidimensional feature vector based on the analysis result data, perform dimensionality reduction processing on the data analysis connection requirements through the principal component analysis algorithm, and obtain dimensionality reduction analysis data;

[0044] According to the optimized feature extraction process of the dimensionality reduction analysis data, the display parameter range is adjusted by the data change trend determination threshold to obtain the associated data of resource fluctuation and visualization structure, including:

[0045] Adjust the feature extraction process according to the dimensionality reduction analysis data, and obtain trend determination data by judging the sudden increase trend of the preset threshold;

[0046] Determine the downward trend of the data according to the trend, and obtain parameter optimization data by adjusting the display parameter range;

[0047] According to the resource fluctuation of the parameter optimization data, distribution detection data is obtained by detecting the distribution of the visualization structure;

[0048] According to the associated data of the distribution detection data, the core dimension of data optimization is extracted to obtain dimension extraction data;

[0049] If the dimension extraction data meets the judgment threshold, the trend judgment is adjusted through feature extraction to obtain adjustment analysis data;

[0050] According to the visualization structure of the adjustment analysis data, optimizing the allocation ratio of resource fluctuations to obtain ratio optimization data;

[0051] According to the parameter range of the ratio optimization data, the change detection data is obtained by judging the change of the sudden increase trend.

[0052] In an optional embodiment, a visualization model is constructed based on the dimensionality reduction analysis data, the multidimensional feature vector is converted into a graphic element through data mapping, and the display parameters of the graphic element are adjusted according to the presentation capability requirements to obtain visualization presentation parameters, including:

[0053] Obtaining a feature vector by parsing the dimensionality reduction analysis data;

[0054] According to the vector features, vector conversion is performed by using data mapping technology to obtain graphic elements;

[0055] Adjust display parameters according to presentation capabilities and determine the display format of graphic elements;

[0056] Build visualization models based on mapping technology to obtain presentation data;

[0057] If the presentation data meets a preset threshold, the distribution of the graphic elements is optimized by adjusting parameters to obtain adjusted presentation data;

[0058] Based on the adjusted presentation data, the improvement of presentation ability is judged by detecting the change trend of the feature vector;

[0059] The visualization model is updated according to the change trend to obtain optimized presentation data.

[0060] In an optional embodiment, user feedback is obtained based on the visualization presentation parameters, feedback data is obtained by optimizing the system-level collaboration requirements through an iterative update algorithm, and optimized system data is obtained by adjusting the system level based on the feedback data and preset improvement conditions, including:

[0061] Obtain user feedback based on user interaction data and obtain the distribution characteristics of feedback content through recording tools;

[0062] The feedback triggering situation is judged based on the distribution characteristics. If the preset improvement condition is triggered, the system level parameters are adjusted through an iterative algorithm to obtain the adjusted collaborative demand data;

[0063] Based on the adjusted collaborative demand data, the update direction of the processing parameters is obtained through the analysis process;

[0064] According to the update direction, the optimized system-level data is obtained by adjusting the processing parameters;

[0065] Based on the optimized system-level data, determine whether feedback triggers have decreased by detecting the changing trend of interaction data;

[0066] If the trend of change shows a decrease in triggers, the presentation format is updated through the analysis process to obtain optimized data;

[0067] Based on the optimization data, the matching degree of the collaborative requirements is tested by a verification tool to determine the optimized system data output by the final system.

[0068] In a second aspect, the present invention further provides an information management device based on smart business, comprising:

[0069] The data acquisition module is used to obtain multi-source raw data through a preset standardized protocol, and map it into a unified structure through a format conversion algorithm to obtain a standardized data set;

[0070] A data grouping module is used to extract characteristic attribute data from the normalized data set, group the characteristic attribute data using a clustering algorithm, and use the grouping results that meet the heterogeneity threshold as the classification data set;

[0071] A dynamic analysis module is used to obtain time series features based on the classified data set, and to perform dynamic change sampling using a sliding window technique based on the difficulty of collecting the time series features to obtain dynamic data samples;

[0072] A modeling and optimization module is used to construct a relationship network based on the dynamic data samples and to model and mine implicit associations, and to optimize the information flow path and transmission based on the results to obtain target transmission data;

[0073] A resource assessment module is configured to extract key indicators based on the target transmission data, obtain a predicted value of short-term resource demand and a real-time load of the current system resource usage status, and fuse the predicted value and the real-time load using a weighted average algorithm to obtain an adjusted resource demand assessment value;

[0074] An evaluation and prediction module is used to determine a weight distribution ratio according to the adjusted resource demand evaluation value, and to perform error prediction on the weight distribution ratio through a regression algorithm to obtain analysis result data;

[0075] A vector association module is used to generate a multidimensional feature vector based on the analysis result data, perform dimensionality reduction processing on the data analysis connection requirements through the principal component analysis algorithm, and optimize feature extraction to adjust the display parameter range to obtain associated data;

[0076] A parameter modeling module is used to construct a visualization model based on the dimensionality reduction analysis data, convert the multidimensional feature vector into a graphic element through data mapping, and adjust the display parameters of the graphic element according to the intuitive presentation capability requirements to obtain visualization presentation parameters;

[0077] A feedback optimization module is used to obtain user feedback based on the visualization presentation parameters, optimize the system level collaboration requirements through an iterative update algorithm to obtain feedback data, and compare the feedback data with preset improvement conditions to obtain optimized system data by adjusting the system level.

[0078] In a third aspect, the present invention also provides an electronic device comprising a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein when the processor executes the computer program, it implements any one of the above-mentioned information management methods based on smart business.

[0079] In a fourth aspect, the present invention also provides a computer-readable storage medium, which includes a stored computer program, wherein when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute any one of the above-mentioned information management methods based on smart business.

[0080] Compared with the prior art, the present invention has the following beneficial effects:

[0081] (1) The method of the present invention obtains multi-source raw data through a preset standardized protocol and maps it into a unified structure through a format conversion algorithm, and uses a clustering algorithm to group the characteristic attribute data. While improving the consistency and efficiency of data processing, it can screen out grouping results that meet the heterogeneity threshold as a classified data set, which helps to effectively classify the data, facilitates subsequent analysis, and improves the targetedness of data processing.

[0082] (2) The method of the present invention performs dynamic data sampling through sliding window technology, which can adapt to the dynamic variability of data. At the same time, by modeling dynamic data, it realizes the mining of implicit associations, optimizes the information flow path and transmission, obtains the target transmission data, improves the efficiency of information flow, makes data transmission more accurate, and meets the accuracy requirements.

[0083] (3) The present method uses visualization models and parameter adjustments to display data based on intuitive presentation needs. Finally, the system is optimized based on user feedback, and the system hierarchy is adjusted through iterative updates to make the system more aligned with actual business needs and improve the efficiency and quality of smart business operations. BRIEF DESCRIPTION OF THE DRAWINGS

[0084] Figure 1 This is a flowchart of an information management method based on smart business provided by the first embodiment of the present invention;

[0085] Figure 2 This is a structural diagram of an information management device based on smart business provided by the first embodiment of the present invention. DETAILED DESCRIPTION

[0086] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0087] Reference Figure 1 The first embodiment of the present invention provides an information management method based on smart business, comprising the following steps:

[0088] S101, obtaining multi-source raw data through a preset standardized protocol, mapping it into a unified structure through a format conversion algorithm, and obtaining a standardized data set;

[0089] S102, extracting characteristic attribute data from the normalized data set, grouping the characteristic attribute data using a clustering algorithm, and using grouping results that meet a heterogeneity threshold as a classified data set;

[0090] S103, acquiring time series features according to the classified data set, and performing dynamic change sampling using a sliding window technique according to the difficulty of acquiring the time series features to obtain dynamic data samples;

[0091] S104, constructing a relationship network based on the dynamic data sample and modeling and mining implicit associations, optimizing the information flow path and transmission based on the results, and obtaining target transmission data;

[0092] S105, extracting key indicators based on the target transmission data, and obtaining a predicted value of short-term resource demand and a real-time load of the current system resource usage status, fusing the predicted value and the real-time load using a weighted average algorithm to obtain an adjusted resource demand assessment value;

[0093] S106, determining a weight distribution ratio according to the adjusted resource demand assessment value, and performing error prediction on the weight distribution ratio using a regression algorithm to obtain analysis result data;

[0094] S107, generating a multidimensional feature vector based on the analysis result data, performing dimensionality reduction processing on the data analysis connection requirements through a principal component analysis algorithm, and optimizing feature extraction to adjust the display parameter range to obtain associated data;

[0095] S108, constructing a visualization model based on the dimensionality reduction analysis data, converting the multidimensional feature vector into a graphic element through data mapping, and adjusting the display parameters of the graphic element according to the intuitive presentation capability requirements to obtain visualization presentation parameters;

[0096] S109, obtaining user feedback based on the visualization presentation parameters, optimizing the system-level collaborative requirements through an iterative update algorithm to obtain feedback data, and comparing the feedback data with preset improvement conditions to obtain optimized system data by adjusting the system level.

[0097] In step S101, raw data of multi-source data is obtained through a preset standardized protocol, and the raw data from different sources are mapped to a unified structure through a format conversion algorithm to obtain a normalized data set;

[0098] It is worth noting that the original data set is extracted from multi-source data through a preset protocol, and the data content is obtained using a standard interface to obtain the initial data set. For the heterogeneous data in the initial data set, a format conversion algorithm is applied to process the data, and the data is mapped to a preset unified structure to obtain a first processed data set. If the first processed data set still contains unaligned fields, the data format is adjusted through a field matching tool to obtain a second processed data set. According to the structural characteristics of the second processed data set, a clustering algorithm is used to group the data to obtain a grouped data set. The category distribution in the grouped data set is used to determine whether there are abnormal data points. If there are abnormal data points, the mean filling method is used to correct them to obtain an optimized data set. The key fields in the optimized data set are obtained, and a consistency check is performed on the key fields to obtain a normalized data set. Using the complete records in the normalized data set, a query structure is constructed through a data indexing tool to obtain the final data set.

[0099] Specifically, extracting original data sets from multi-source data through preset protocols is actually pulling data scattered in different systems in a unified manner.

[0100] For example, suppose an enterprise's data comes from a sales system, inventory system, and customer management system. Through APIs or database query languages, sales records, inventory levels, and customer information can be retrieved according to agreed-upon field standards to form an initial data set. This initial data set might include sales data in JSON format, inventory tables in CSV format, and customer information in XML format. These data structures vary, and field names and types are not uniform. A format conversion algorithm is applied to the heterogeneous data in this initial data set, aiming to map the different formats into a unified structure.

[0101] For example, you can design a standard template containing fields such as "Time," "Product ID," "Quantity," and "Source." Convert the "date" field in sales data to "Time" and the "item_code" field in inventory data to "Product ID." Using a field mapping table and conversion rules, you can generate a first processed dataset. This first processed dataset may still have issues, such as the "Phone" field in customer data being missing from sales data, or misaligned fields still existing. If the first processed dataset contains misaligned fields, adjust the format using the field matching tool.

[0102] For example, fuzzy matching techniques can be used to identify "phone" and "tel" as the same field, thereby filling in missing values or removing redundant items to form a second processed data set. This approach improves data consistency and lays the foundation for subsequent analysis.

[0103] Specifically, a matching tool can be used to scan all records to find fields with similar semantics but different names, and then unified and adjusted to reduce redundancy. Based on the structural characteristics of the second processed data set, a clustering algorithm is used to group the data.

[0104] For example, using K-means clustering to categorize sales records into high-frequency and low-frequency groups based on the "Quantity" and "Time" fields creates a grouped data set. This grouping allows for clearer data distribution patterns, helping to identify potential patterns.

[0105] Preferably, if the data volume is large, you can first reduce the dimension through principal component analysis and then perform clustering to improve efficiency. By analyzing the category distribution of the grouped data set, you can determine abnormal data points.

[0106] For example, a record in a high-frequency trading group shows a "Quantity" of 10,000, while the average for similar data is only 50, which could be considered an anomaly. Using the mean-filling method, this data is corrected to the group average of 50, forming an optimized data set. This approach can smooth out abnormal fluctuations and improve data reliability.

[0107] It is understandable that if there are too many outliers, they can be eliminated instead of filled in based on business rules. Obtain key fields in the optimized data set, such as "product ID," "time," and "quantity," and perform consistency checks.

[0108] For example, check whether "time" conforms to the date format and whether "quantity" is a positive integer. If non-compliant records are found, they can be marked or corrected to obtain a normalized data set.

[0109] In one embodiment, the field format can be verified by regular expressions to ensure data integrity. The query structure is constructed using complete records in the normalized data set.

[0110] For example, by leveraging database indexing technology, a joint index based on "product ID" and "time" can be created to generate the final data set. This structure significantly improves query efficiency, enabling rapid response to business needs, especially in large data scenarios.

[0111] In one possible implementation, a distributed index may be introduced to further optimize performance.

[0112] For example, the entire process, from multi-source extraction to the final collection, can gradually resolve heterogeneity, anomalies, and consistency issues. The final data collection has a unified format and efficient query capabilities. The method of the present invention can significantly improve decision support efficiency in enterprise data integration while reducing data processing costs.

[0113] In step S102, characteristic attribute data is extracted from the normalized data set, the characteristic attribute data is grouped using a clustering algorithm, and the grouping results that meet the heterogeneity threshold are used as the classification data set, including:

[0114] S1021, obtaining structural features based on the classification data set, performing category distribution processing using statistical tools, and obtaining a distribution data set;

[0115] S1022, extracting key fields from the distributed data set. If the key fields meet the consistency check, determining a consistent data set; if not, adjusting the field content using a field mapping tool to obtain an adjusted data set.

[0116] S1023, adjusting the data set according to its category distribution, identifying outliers using a clustering algorithm, and obtaining an abnormally labeled data set;

[0117] S1024, correcting outliers using a mean filling tool based on the labeling information of the abnormal labeled data set to obtain a corrected data set;

[0118] S1025, extracting related fields based on the modified data set, constructing a query structure using an indexing tool, and obtaining a query data set;

[0119] S1026, adjusting the order of the complete records of the query data set according to the category distribution using a sorting tool to obtain a sorted data set;

[0120] S1027 , performing integrity verification using a consistency verification tool based on the structural characteristics of the sorted data set to obtain a verification data set.

[0121] Specifically, when obtaining structural features through classified data sets, statistical tools can be used to analyze category distribution.

[0122] For example, a company's sales data set contains fields such as "product category", "sales region", and "transaction volume". The statistical tool can calculate the transaction volume ratio of each product category to form a distribution data set.

[0123] For example, if a data set shows that a certain product's transaction volume in a certain region accounts for 60%, we can preliminarily determine its distribution characteristics. This method allows us to quickly grasp the overall patterns of the data. When extracting key fields from a distributed data set, let's assume that "transaction volume" and "sales region" are selected as key fields.

[0124] In one possible implementation, these fields need to be checked for consistency, such as whether "Transaction Volume" is always positive and contains no null values. If the consistency check is satisfied, a consistent data set is directly formed. If the "Transaction Volume" of some records is found to be negative, adjustments need to be made.

[0125] Specifically, you can use field mapping tools to map negative records to zero or infer reasonable values based on business rules, ultimately obtaining an adjusted data set. Clustering algorithms can be used to identify outliers based on the category distribution of the adjusted data set.

[0126] For example, based on the "transaction volume" and "region" fields, the DBSCAN algorithm is used to group the data. If the transaction volume of a record far exceeds the average value of the same region, it is marked as an outlier, forming an outlier marked data set.

[0127] It should be noted that this labeling depends on the setting of clustering parameters. Adjusting the parameters can change the sensitivity of anomaly identification. Based on the anomaly labeled data set, the outliers are corrected using the mean filling tool.

[0128] In one embodiment, if the “trading volume” of an outlier is 5000 and the mean value of the same group is 200, it is corrected to 200 to obtain a corrected data set.

[0129] It is understandable that this correction method can smooth out data fluctuations and facilitate the stability of subsequent analysis. From the corrected data set, related fields such as "product category" and "transaction volume" are extracted, and the query structure is constructed using indexing tools.

[0130] For example, create an index by "Product Category" to generate a query dataset. This structure can speed up searches by category, especially when the data volume is large. By querying the complete set of records in the dataset, the sorting tool can adjust the order based on the category distribution.

[0131] Preferably, sorting by "transaction volume" from high to low can intuitively reflect which categories perform best, forming a sorted data set. This adjustment helps prioritize core data. Based on the structural characteristics of the sorted data set, consistency checking tools can verify its integrity.

[0132] For example, check whether the "trading volume" field still has empty values or abnormal formats. If problems are found, mark and correct them, and finally obtain the verification data set.

[0133] In one embodiment, a visualization tool can be used to assist in verification and intuitively demonstrate data consistency. This approach can effectively improve the credibility of data and provide reliable support for business decisions.

[0134] Step S103, obtaining time series features based on the classified data set, and performing dynamic change sampling using a sliding window technique based on the difficulty of collecting the time series features to obtain dynamic data samples, including:

[0135] S1031, obtaining dynamic change data from the time series using a windowing technique to obtain a dynamic data sample;

[0136] S1032, using statistical tools to calculate sequence distribution based on the changing trend of the dynamic data sample to obtain a distribution data sample;

[0137] S1033, extracting time features from the distribution data samples, identifying dynamic change patterns through a clustering algorithm, and obtaining pattern data samples;

[0138] S1034, constructing a time series query structure using an indexing tool based on the sequence distribution in the pattern data sample to obtain a query data sample;

[0139] S1035: If the change trend in the query data sample meets the preset threshold, a consistent data sample is determined; if not, the dynamic sample is adjusted by a mean filling tool to obtain an adjusted data sample.

[0140] S1036, by adjusting the time features in the data sample, using a sorting tool to adjust the order according to the sequence distribution, to obtain a sorted data sample.

[0141] S1037, verifying the integrity through a consistency verification tool according to the dynamic change pattern in the sorted data sample to obtain a verification data sample.

[0142] Specifically, sliding window technology is often used to extract dynamically changing data from time series.

[0143] For example, a company's sales data includes daily transaction volume, spanning 30 days. Using sliding window technology, we can set a window size of 7 days, sliding from day 1 to day 24, to gradually obtain dynamic weekly data samples. This method can capture short-term fluctuations in transaction volume.

[0144] For example, if the transaction volume within a window gradually increases from 100 to 150, it can be preliminarily determined that there is an upward trend. For the changing trend in dynamic data samples, statistical tools can calculate the sequence distribution.

[0145] Specifically, the average transaction volume and standard deviation in each window can be counted to obtain a distribution data sample.

[0146] In one possible implementation, the average transaction volume in a window is 120, and the standard deviation is 15, reflecting that the data is relatively concentrated.

[0147] It should be noted that this distribution analysis helps determine data stability. When extracting temporal features from distributed data samples, the window start time and rate of change can be selected as key points. Clustering algorithms such as K-Means can be used to identify dynamic change patterns.

[0148] For example, multiple windows are divided into three categories: "stable", "rising" and "falling", and pattern data samples are obtained.

[0149] In one embodiment, a pattern showing a three-window increase in transaction volume from 100 to 130 is classified as an "upward" pattern. This classification facilitates understanding the patterns of change. Based on the sequence distribution in the pattern data sample, the indexing tool can construct a time series query structure.

[0150] Preferably, an index is created by timestamp to generate query data samples.

[0151] For example, querying all "rising" windows within a specific timeframe can quickly locate relevant records. If the trend of a queried data sample meets a preset threshold, such as an increase rate greater than 10%, it is identified as a consistent data sample. Conversely, if the increase rate in a window is only 5%, this can be adjusted using the mean padding tool.

[0152] For example, an adjusted data sample is obtained by padding the abnormally low value from 80 to the window mean of 100. By adjusting the time characteristics in the data sample, the sorting tool can adjust the order according to the sequence distribution.

[0153] For example, sort by transaction volume change rate from high to low to generate a sorted data sample.

[0154] In one embodiment, a window change rate of 20% is ranked first, preferentially reflecting a significant trend.

[0155] It is understandable that this sorting can highlight key change points. Based on the dynamic change patterns in the sorted data samples, the consistency check tool verifies the integrity.

[0156] For example, check whether all window data are complete and without missing, and finally obtain the verification data sample.

[0157] For example, if the transaction volume in a window is empty, it can be marked and filled with the mean value of 120. This method ensures data reliability and provides support for subsequent analysis.

[0158] Step S104, constructing a relationship network based on the dynamic data sample and modeling and mining implicit associations, optimizing the information flow path and transmission based on the results, and obtaining target transmission data, including:

[0159] S1041, constructing a relationship network based on the dynamic data sample, performing multi-layer data relationship modeling using a graph neural network algorithm, and mining implicit associations between nodes requiring data processing depth to obtain relationship-enhanced data;

[0160] S1042, obtaining an information flow path based on the relationship enhancement data, optimizing the data transmission order using a shortest path algorithm, and adjusting the path weight according to information flow efficiency requirements to obtain target transmission data;

[0161] According to step S1041, a relational network is constructed through dynamic data, and a graph neural network algorithm is used to mine multi-layer implicit associations to obtain enhanced relational data. The association strength between nodes is obtained from the enhanced relational data, and the distribution characteristics are calculated using statistical tools to obtain distribution characteristic data. According to the changes in the implicit associations in the distribution characteristic data, a clustering algorithm is used to identify node groups to obtain group division data. Through the relational network characteristics in the group division data, the hierarchical structure between nodes is obtained to obtain hierarchical structure data. If the association strength in the hierarchical structure data meets the preset threshold, the consistency data is determined. If not, the distribution characteristics are adjusted through the mean filling tool to obtain adjusted distribution data. According to the node group characteristics in the adjusted distribution data, the hierarchical order is adjusted using the sorting tool to obtain sorted hierarchical data. Through the multi-layer association characteristics in the sorted hierarchical data, an enhanced relational network structure is obtained to obtain verification network data.

[0162] According to step S1042, the information flow path is obtained through the relationship enhancement data, and the transmission order is adjusted using the shortest path algorithm to obtain preliminary transmission data. The path weight is extracted from the preliminary transmission data, and the weight distribution is adjusted using the mean filling tool to obtain balanced weight data. According to the information flow characteristics in the balanced weight data, the path calculation result is calculated to obtain the optimized path data. If the flow efficiency in the optimized path data meets the preset threshold, the consistency data is determined. If it does not meet the threshold, the transmission order is adjusted through the sorting tool to obtain the rearranged transmission data. By rearranging the path weight distribution in the transmission data, the optimized transmission structure is obtained to obtain the verified transmission data. According to the data calculation characteristics in the verified transmission data, it is judged whether the efficiency requirements are met to obtain the final transmission data.

[0163] Specifically, when obtaining the information flow path through relationship enhancement data, it can be regarded as a transaction delivery process in the enterprise sales network.

[0164] For example, assume a sales chain includes customer A, product X, and customer B, and the information flow path is from A purchasing X and then passing it to B. The principle of the shortest path algorithm to adjust the transmission order is to identify the optimal connection between nodes.

[0165] For example, the weight of the direct transaction path from customer A to product X is 10, while the weight of the path through intermediary C is 15. The shortest path algorithm will give priority to the direct path with a weight of 10 and generate preliminary transmission data.

[0166] Specifically, preliminary transmission data may show that the total path weight of AXB is 20. After extracting the path weights from the preliminary transmission data, the mean filling tool can adjust the weight distribution in a way that balances outliers.

[0167] In one possible implementation, if AX has a weight of 10 and XB has a weight of 30, the mean fill tool will adjust the lower weight of 10 up to a value closer to the mean of 20, resulting in a balanced weighted data. This adjustment can smooth out fluctuations in the flow of information.

[0168] For example, balanced weight data may show that the weights of AX and XB are both 20, reflecting more stable transmission characteristics. When calculating path results based on the information flow characteristics in balanced weight data, it can be understood as analyzing transmission efficiency.

[0169] Preferably, if the total weight of path AXB is 40 and the preset efficiency threshold is 50, then optimization is required. In the embodiment where the sorting tool adjusts the transmission order, the priority of XB can be increased, the intermediate delay can be reduced, and rearranged transmission data can be generated.

[0170] For example, after rearrangement, the path becomes XBA, and the total weight is reduced to 35, but the efficiency is improved. When obtaining an optimized transmission structure by rearranging the path weight distribution in the transmission data, the generation of verification transmission data depends on the weight balance.

[0171] Specifically, if the weight of XB is 15 and BA is 20, optimizing the transmission structure may reveal that B is the core node.

[0172] In one embodiment, verification of the transmission data calculation characteristics shows that the turnover time is reduced from 5 days to 3 days, and the efficiency requirement is met.

[0173] It should be noted that the final determination of the transmitted data is based on the efficiency threshold.

[0174] For example, if the threshold is 3 days, the consistency data is confirmed if the verification data meets the requirements. This method can help companies optimize their sales paths.

[0175] For example, client B, as a core node, can be allocated resources first to improve overall circulation efficiency.

[0176] Understandably, this optimization can also reveal potential bottlenecks. For example, a high weight of the XB path may indicate the need for logistics improvements, thereby supporting sales decisions.

[0177] According to step S105, key indicators are extracted based on the target transmission data, and a predicted value of short-term resource demand and a real-time load of the current system resource usage status are obtained. The predicted value and the real-time load are integrated using a weighted average algorithm to obtain an adjusted resource demand assessment value, including:

[0178] Obtaining key indicators based on the target transmission data, combining the predicted value of short-term resource demand, and performing real-time load fusion through a weighted average method to obtain preliminary resource demand data;

[0179] Determining whether there is abnormal fluctuation based on the sudden increase trend of the preliminary resource demand data, and obtaining fluctuation detection data;

[0180] According to the downward trend in the fluctuation detection data, adjusting the resource allocation order by a sorting tool to obtain rearranged resource data;

[0181] According to the system status of the rearranged resource data, balanced load data is obtained by acquiring load balancing distribution;

[0182] Acquire data fusion characteristics according to the balanced load data, and if the sudden increase trend exceeds a preset threshold, adjust resource demand to obtain revised demand data;

[0183] Obtaining verification demand data by determining whether resource requirements satisfy system status based on the estimated value of the modified demand data;

[0184] According to the key indicators of the verification demand data, resource allocation adjustment is performed through a mean filling tool to obtain a resource demand assessment value.

[0185] Step S106, determining a weight distribution ratio based on the adjusted resource demand assessment value, and performing error prediction on the weight distribution ratio using a regression algorithm to obtain analysis result data, including:

[0186] S1061: Determine a weight distribution ratio based on the adjusted resource demand assessment value, and predict the real-time requirements and accuracy requirements using a regression analysis algorithm to obtain preliminary prediction data;

[0187] S1062, if the prediction error in the preliminary prediction data is lower than a preset threshold, determining analysis result data; if it is higher than the preset threshold, adjusting the model parameters to perform re-prediction to obtain revised prediction data;

[0188] S1063, according to the real-time requirement of the revised forecast data, obtaining trend analysis data by analyzing the changing trend of the allocation ratio;

[0189] S1064, adjusting the weight distribution using a mean filling tool according to the accuracy requirement of the trend analysis data to obtain balanced distribution data;

[0190] S1065 , based on the analysis result of the balanced allocation data, detecting the fluctuation of resource demand to obtain fluctuation detection data;

[0191] S1066, adjusting the estimated value of resource demand according to the change trend of the fluctuation detection data to obtain final demand data;

[0192] S1067, extracting key indicators based on the final demand data, and determining whether the real-time requirements meet the system status to obtain verification result data as analysis result data.

[0193] Specifically, in resource demand forecasting, the weight distribution ratio is determined by the adjusted resource demand estimate. For example, the weight of the real-time requirement is set to 6, and the weight of the accuracy requirement is set to 4. A multivariate linear regression analysis algorithm is used to construct a regression model with historical resource usage data as the training set. Assuming that the current system's disk IO utilization rate is 45% and the network delay is 30ms, the regression model predicts that the disk IO utilization rate may rise to 60% and the network delay may increase to 40ms in the next half hour. By calculating the error between the predicted value and the actual value, if the error is lower than the preset threshold of 5%, the predicted result is directly used; if the error is higher than the threshold, the regression model parameters are adjusted, such as increasing the time window of historical data or adjusting the feature weights, and re-forecasting.

[0194] For example, after adjustments, the predicted disk I / O utilization rate was 58%, and the network latency was 38ms, with an error reduced to 4%, meeting requirements. Ultimately, the analysis results are combined with system resource scheduling strategies to dynamically adjust resource allocation, such as proactively increasing disk I / O bandwidth or optimizing network routing based on the predicted results, to ensure stable system performance. This regression-based prediction method effectively balances real-time performance with accuracy, providing a reliable basis for resource scheduling.

[0195] Step S107: Generate a multidimensional feature vector based on the analysis result data, perform dimensionality reduction processing on the data analysis connection requirements through the principal component analysis algorithm, and optimize feature extraction to adjust the display parameter range to obtain related data, including:

[0196] S1071, generating a multidimensional feature vector based on the analysis result data, performing dimensionality reduction processing on the data analysis connection requirements using a principal component analysis algorithm, and obtaining dimensionality reduction analysis data;

[0197] By analyzing the result data, a multidimensional feature vector is constructed. The principal component analysis algorithm is used to reduce the dimensionality of the connection requirements to obtain the reduced dimensionality analysis data. The mean filling tool is used to adjust the information dimensions of the core information in the reduced dimensionality analysis data to obtain balanced adjustment data. Based on the conversion requirements in the balanced adjustment data, the changing trends of business insights are calculated to obtain trend analysis data. If the connection requirements in the trend analysis data meet the preset threshold, the allocation ratio is determined to obtain proportional allocation data. The retained dimensions in the proportional allocation data are used to detect fluctuations in the data construction to obtain fluctuation detection data. Based on the changing trends in the fluctuation detection data, the distribution structure of the multidimensional features is adjusted to obtain optimized distribution data. Based on the information dimensions in the optimized distribution data, key indicators are extracted to determine whether the business insights meet the system status and obtain verification result data.

[0198] Specifically, in a commercial data analysis scenario, we first extract original features from user behavior logs, including 20 initial dimensions such as daily visit frequency (e.g., user A visits an average of 12 times), page dwell time (e.g., product page average of 95 seconds), and conversion rate (e.g., 2%), to form the original feature matrix. Using the principal component analysis algorithm, we set the variance contribution rate threshold to 85%, and calculated the eigenvalues and eigenvectors using the covariance matrix. We found that the cumulative contribution rate of the first five principal components reached 87%, with the first principal component (PC1) having the highest loads for visit frequency (62) and dwell time (58), reflecting user activity. The second principal component (PC2) highlights conversion rate (71) and average order value (65), representing consumption quality. The 20-dimensional data is projected into a 5-dimensional space. For example, the original feature vector of user B [15, 120, 5% ... ] is reduced to [34, -87, 0 ... ] after linear transformation. The K-means clustering algorithm (k=4, Euclidean distance) was used to cluster the reduced-dimensional data into user groups. Cluster 1 (high activity, low conversion) accounted for 32%, while Cluster 2 (low activity, high conversion) accounted for 18%. The clustering effectiveness was verified by a silhouette coefficient of 68. The reduced-dimensional data and cluster labels were ultimately output to the recommendation system. For example, high-priced items could be prioritized for users in Cluster 2, achieving a balance between business needs and computational efficiency.

[0199] S1072, according to the optimized feature extraction process of the dimensionality reduction analysis data, adjusting the display parameter range by using the data change trend determination threshold to obtain the associated data of resource fluctuation and visualization structure, including:

[0200] S10721, adjusting the feature extraction process according to the dimensionality reduction analysis data, and obtaining trend determination data by judging a sudden increase trend of a preset threshold;

[0201] S10722, determining a downward trend of the data based on the trend, and obtaining parameter optimization data by adjusting a display parameter range;

[0202] S10723, detecting the distribution of the visualization structure according to the resource fluctuation of the parameter optimization data to obtain distribution detection data;

[0203] S10724, extracting the core dimension of data optimization based on the associated data of the distribution detection data to obtain dimension extraction data;

[0204] S10725, if the dimension extraction data meets the determination threshold, adjusting the trend determination through feature extraction to obtain adjustment analysis data;

[0205] S10726, optimizing the allocation ratio of resource fluctuations according to the visualization structure of the adjustment analysis data to obtain ratio optimization data;

[0206] S10727: Obtain change detection data by determining a change in the sudden increase trend based on a parameter range of the proportional optimization data.

[0207] Specifically, in a resource monitoring scenario, by analyzing historical resource usage data, key features are extracted across 15 initial dimensions, including CPU utilization (e.g., Server A averages 75%), memory usage (e.g., averages 45GB), and network bandwidth (e.g., averages 120Mbps), to construct an original feature matrix. Using the Isolation Forest algorithm, with an anomaly score threshold set to 65, the feature values and anomaly scores are calculated to identify anomalies indicating sudden increases or decreases in resource usage. For example, Server B's CPU utilization suddenly spiked to 95% during a certain period, while its memory usage dropped to 30GB. Based on the distribution of these anomalies, the visualization parameters are adjusted to display CPU utilization from 0% to 100%, memory usage from 0GB to 64GB, and network bandwidth from 0Mbps to 200Mbps. By dynamically adjusting these display parameters and combining them with time series analysis algorithms (such as the ARIMA model), future resource usage trends can be predicted. For example, it can be predicted that Server C's CPU utilization will stabilize between 80% and 85% over the next 24 hours. Finally, the resource fluctuation data is associated with the visualization structure to generate a dynamic monitoring chart, which reflects the resource usage status in real time and provides data support for resource scheduling.

[0208] Step S108, constructing a visualization model based on the dimensionality reduction analysis data, converting the multidimensional feature vector into a graphic element through data mapping, and adjusting the display parameters of the graphic element according to the presentation capability requirements to obtain visualization presentation parameters, including:

[0209] S1081, obtaining a feature vector by parsing the dimensionality reduction analysis data;

[0210] S1082, performing vector conversion based on the vector features using data mapping technology to obtain graphic elements;

[0211] S1083, adjusting display parameters according to the presentation capability to determine the display format of the graphic elements;

[0212] S1084, constructing a visualization model based on mapping technology to obtain presentation data;

[0213] S1085, if the presentation data meets a preset threshold, optimizing the distribution of the graphic elements by adjusting parameters to obtain adjusted presentation data;

[0214] S1086, judging the improvement of the presentation capability by detecting the change trend of the feature vector based on the adjusted presentation data;

[0215] S1087: Update the visualization model according to the change trend to obtain optimized presentation data.

[0216] Specifically, in the financial transaction data analysis scenario, the original data contains 20 feature dimensions, including trading volume (e.g., average daily trading volume of stock X is 12 million shares), price volatility (e.g., standard deviation 15), and bid-ask spread (e.g., average 0.3 yuan). Principal component analysis (PCA) is used to reduce the dimensions to 5, retaining 95% of the variance contribution. The first principal component has a weight of 42, reflecting the dominance of price volatility. The t-SNE algorithm is used to map the reduced feature vectors to a two-dimensional space. The perplexity parameter is set to 30, the learning rate is 200, and 1000 iterations are performed. A scatter plot cluster distribution is generated, for example, with the center coordinates of a high-frequency trading cluster at (-2, 8) and the center coordinates of a low-frequency trading cluster at (5, -6). Based on the clustering results, the heat map color gradient is dynamically adjusted, categorizing the trading volume ranges as 0-5 million (blue), 5-15 million (green), and above 15 million (red), and the price volatility ranges as 0-1 (light), 1-3 (medium), and above 3 (dark). Incorporating the DBSCAN clustering algorithm, with a neighborhood radius of 5 and a minimum sample size of 10, we identify unusual trading patterns, such as an outlier where trading volume suddenly surges to 30 million and volatility reaches 4 during a specific period. By dynamically adjusting the candlestick chart's time granularity, switching from a 1-minute chart to a 5-minute chart, we smooth out noise. We also overlay a Bollinger Band indicator (with a period of 20 and a standard deviation of 2) to display price channels. A red alert is triggered when the price breaks through the upper band. Finally, we generate an interactive visualization panel that simultaneously displays dimensionality reduction clustering, heat maps, and candlestick analysis results, enabling the visualization of multi-dimensional trading characteristics.

[0217] Step S109, obtaining user feedback based on the visualization parameters, optimizing the system-level collaborative requirements through an iterative update algorithm to obtain feedback data, and comparing the feedback data with preset improvement conditions to obtain optimized system data by adjusting the system level, including:

[0218] S1091, obtaining user feedback based on the user interaction data, and obtaining distribution characteristics of the feedback content through a recording tool;

[0219] S1092, judging the feedback triggering situation based on the distribution characteristics, and if a preset improvement condition is triggered, adjusting the system level parameters through an iterative algorithm to obtain adjusted collaboration demand data;

[0220] S1093, obtaining an update direction of processing parameters through a parsing process based on the adjusted collaboration demand data;

[0221] S1094, obtaining optimized system-level data by adjusting processing parameters according to the update direction;

[0222] S1095, based on the optimized system-level data, determining whether feedback triggers are decreasing by detecting a change trend in the interaction data;

[0223] S1096, if the change trend shows a decrease in triggers, then the presentation format is updated through the analysis process to obtain optimized data;

[0224] S1097: Based on the optimization data, a verification tool is used to detect the matching degree of the collaborative requirements to determine the optimized system data output by the final system.

[0225] Specifically, during user interaction with the visualization system, the system captures user clicks on hot spots in the cluster distribution in real time. For example, a click focusing on a data point within a radius of 5 near coordinate (2, -8) triggers density-based local data recalculation. A Gaussian mixture model (GMM) is used to re-cluster the clicked area data, setting the initial number of clusters to 3, the covariance type to full covariance, the maximum number of iterations to 500, and the convergence threshold to 1e-4. This results in new mean vectors (1, -7), (3, -9), and (0, -6), with covariance matrices of [[2, 05], [05, 3]], [[3, 1], [1, 4]], and [[1, 02], [02, 2]], respectively. Based on the new clustering results, the system automatically adjusts the weights of the parallel coordinate axes, increasing the original features of order book thickness (original weight 15) and liquidity gap (original weight 08) to 22 and 12, respectively, while reducing the weight of position change rate from 2 to 1. Using a hidden Markov model (HMM) to analyze continuous user operation sequences, with a set number of states of 4, the number of observation symbols of 6, and the initial value of the transition probability matrix uniformly distributed, the forward-backward algorithm iteratively updated the user's preferred operation pattern of "first checking the volatility distribution, then checking for volume anomalies" with a probability of 78%. Based on this pattern, the system dynamically loads the LSTM prediction module, inputs a window length of 10, and 64 hidden layer units. It outputs the predicted value of the liquidity risk indicator for the next five time steps (for example, the predicted value at time t+1 is 45, with a confidence interval of [38, 52]), and overlays the predicted result onto the existing visualization as a semi-transparent ribbon layer. When a user stays on an asset category for more than 30 seconds, a feature importance analysis based on a gradient boosted tree (GBDT) is triggered. The tree depth is set to 6, the learning rate is 1, and the number of iterations is 200. The buy-sell pressure ratio (importance score 35) and the order book slope (importance score 28) in the current view are calculated as key influencing factors. The real-time change curves of these two indicators are then highlighted in the sidebar of the interface, and the sampling frequency is increased from the default 1 second / time to 2 seconds / time.

[0226] The method of the present invention firstly achieves compatibility of multi-source heterogeneous data, and through operations such as normalization, clustering and grouping, data from different sources can be effectively integrated and utilized. Secondly, data modeling is performed with the help of various algorithms, such as building a relationship network to mine implicit associations, regression analysis to predict errors, etc., to improve the depth and accuracy of data processing. Furthermore, data can be dynamically displayed, and through visualization models and parameter adjustments, data can be displayed according to intuitive presentation requirements. Finally, the system is optimized in combination with user feedback, and the system hierarchy is adjusted through iterative updates to make the system more in line with actual business needs and improve the efficiency and quality of smart business operations.

[0227] The embodiment of the present invention also provides an information management device based on smart business, such as Figure 2 Shown, including:

[0228] The data acquisition module is used to obtain multi-source raw data through a preset standardized protocol, and map it into a unified structure through a format conversion algorithm to obtain a standardized data set;

[0229] A data grouping module is used to extract characteristic attribute data from the normalized data set, group the characteristic attribute data using a clustering algorithm, and use the grouping results that meet the heterogeneity threshold as the classification data set;

[0230] A dynamic analysis module is used to obtain time series features based on the classified data set, and to perform dynamic change sampling using a sliding window technique based on the difficulty of collecting the time series features to obtain dynamic data samples;

[0231] A modeling and optimization module is used to construct a relationship network based on the dynamic data samples and to model and mine implicit associations, and to optimize the information flow path and transmission based on the results to obtain target transmission data;

[0232] A resource assessment module is configured to extract key indicators based on the target transmission data, obtain a predicted value of short-term resource demand and a real-time load of the current system resource usage status, and fuse the predicted value and the real-time load using a weighted average algorithm to obtain an adjusted resource demand assessment value;

[0233] An evaluation and prediction module is used to determine a weight distribution ratio according to the adjusted resource demand evaluation value, and to perform error prediction on the weight distribution ratio through a regression algorithm to obtain analysis result data;

[0234] A vector association module is used to generate a multidimensional feature vector based on the analysis result data, perform dimensionality reduction processing on the data analysis connection requirements through the principal component analysis algorithm, and optimize feature extraction to adjust the display parameter range to obtain associated data;

[0235] A parameter modeling module is used to construct a visualization model based on the dimensionality reduction analysis data, convert the multidimensional feature vector into a graphic element through data mapping, and adjust the display parameters of the graphic element according to the intuitive presentation capability requirements to obtain visualization presentation parameters;

[0236] A feedback optimization module is used to obtain user feedback based on the visualization presentation parameters, optimize the system level collaboration requirements through an iterative update algorithm to obtain feedback data, and compare the feedback data with preset improvement conditions to obtain optimized system data by adjusting the system level.

[0237] It should be noted that the information management device based on smart business provided in an embodiment of the present invention is used to execute all the process steps of the information management method based on smart business in the above embodiment. The working principles and beneficial effects of the two correspond one to one, so they will not be repeated here.

[0238] An embodiment of the present invention further provides an electronic device. The electronic device includes: a processor, a memory, and a computer program stored in the memory and executable on the processor, such as a data acquisition program. When the processor executes the computer program, the steps in the above-mentioned embodiments of the information management method based on smart business are implemented, such as Figure 1 Alternatively, when the processor executes the computer program, the functions of the modules / units in the above-mentioned device embodiments are realized, such as the data acquisition module.

[0239] Exemplarily, the computer program may be divided into one or more modules / units, which are stored in the memory and executed by the processor to implement the present invention. The one or more modules / units may be a series of computer program instruction segments capable of implementing specific functions, and the instruction segments are used to describe the execution process of the computer program in the electronic device.

[0240] The electronic device may be a computing device such as a desktop computer, notebook, PDA, or smart tablet. The electronic device may include, but is not limited to, a processor and memory. Those skilled in the art will appreciate that the aforementioned components are merely examples of electronic devices and do not constitute a limitation of the electronic device. The electronic device may include more or fewer components than those described above, or a combination of certain components, or different components. For example, the electronic device may also include input / output devices, network access devices, buses, and the like.

[0241] The processor may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc. The processor is the control center of the electronic device, connecting various parts of the entire electronic device using various interfaces and lines.

[0242] The memory can be used to store the computer programs and / or modules. The processor implements the various functions of the electronic device by running or executing the computer programs and / or modules stored in the memory and accessing the data stored in the memory. The memory may primarily include a program storage area and a data storage area. The program storage area may store an operating system and at least one application required for a function (such as a sound playback function or an image playback function); the data storage area may store data generated based on the use of the mobile phone (such as audio data, a phone book, etc.). Furthermore, the memory may include high-speed random access memory and non-volatile memory, such as a hard disk, internal memory, a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, at least one disk storage device, a flash memory device, or other volatile solid-state storage device.

[0243] If the module / unit integrated into the electronic device is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present invention can implement all or part of the process steps in the above-mentioned method embodiments by using a computer program to instruct the relevant hardware. The computer program can be stored in a computer-readable storage medium. When executed by a processor, the computer program can implement the steps of each of the above-mentioned method embodiments. The computer program includes computer program code, which can be in source code form, object code form, executable file, or some intermediate form. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard drive, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal, and software distribution medium. It should be noted that the content of the computer-readable medium can be appropriately increased or decreased based on the requirements of legislation and patent practice in a jurisdiction. For example, in some jurisdictions, based on legislation and patent practice, computer-readable media does not include electric carrier signals and telecommunication signals.

[0244] It should be noted that the device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed across multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the present embodiment. In addition, in the drawings of the device embodiments provided by the present invention, the connection relationship between the modules indicates that there is a communication connection between them, which may be specifically implemented as one or more communication buses or signal lines. A person of ordinary skill in the art can understand and implement the present invention without inventive effort.

[0245] The specific embodiments described above further illustrate the objectives, technical solutions, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the scope of protection of the present invention. In particular, it should be noted that any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included within the scope of protection of the present invention for those skilled in the art.

Claims

1. An information management method based on smart business, characterized in that: include: Obtain multi-source raw data through a preset standardized protocol, map it into a unified structure through a format conversion algorithm, and obtain a standardized data set; Extracting characteristic attribute data from the normalized data set, grouping the characteristic attribute data using a clustering algorithm, and using grouping results that meet a heterogeneity threshold as a classified data set; Acquire time series features according to the classified data set, and perform dynamic change sampling using a sliding window technique based on the difficulty of acquiring the time series features to obtain dynamic data samples; Construct a relationship network based on the dynamic data samples and model and mine implicit associations, optimize the information flow path and transmission based on the results, and obtain target transmission data; Extracting key indicators based on the target transmission data, and obtaining a predicted value of short-term resource demand and a real-time load of the current system resource usage status, fusing the predicted value and the real-time load through a weighted average algorithm to obtain an adjusted resource demand assessment value; Determine a weight distribution ratio according to the adjusted resource demand assessment value, and perform error prediction on the weight distribution ratio through a regression algorithm to obtain analysis result data; Generate a multidimensional feature vector based on the analysis result data, perform dimensionality reduction processing on the data analysis connection requirements through the principal component analysis algorithm, and optimize feature extraction to adjust the display parameter range to obtain associated data; Constructing a visualization model based on the dimensionality reduction analysis data, converting the multidimensional feature vector into a graphic element through data mapping, and adjusting the display parameters of the graphic element according to the intuitive presentation capability requirements to obtain visualization presentation parameters; Obtain user feedback based on the visualization presentation parameters, optimize the system-level collaborative requirements through an iterative update algorithm to obtain feedback data, and compare the feedback data with preset improvement conditions to obtain optimized system data by adjusting the system level; Among them, building a relationship network based on the dynamic data samples and modeling and mining implicit associations, optimizing the information flow path and transmission based on the results, and obtaining target transmission data, including: Constructing a relational network based on the dynamic data samples, performing multi-layer data relationship modeling through a graph neural network algorithm, and mining implicit associations between nodes requiring deep data processing to obtain relation-enhanced data; The information flow path is obtained according to the relationship enhancement data, the data transmission order is optimized through the shortest path algorithm, and the path weight is adjusted according to the information flow efficiency requirement to obtain the target transmission data.

2. The information management method based on smart business according to claim 1, characterized in that: Extracting characteristic attribute data from the normalized data set, grouping the characteristic attribute data using a clustering algorithm, and using grouping results that meet a heterogeneity threshold as a classification data set, including: Acquire structural features according to the classification data set, perform category distribution processing using statistical tools, and obtain a distribution data set; Extracting key fields from the distributed data set, and if the key fields meet the consistency check, determining a consistent data set; if not, adjusting the field content through a field mapping tool to obtain an adjusted data set; Adjusting the data set according to its category distribution, identifying outliers through a clustering algorithm, and obtaining an abnormally marked data set; According to the labeling information of the abnormal labeled data set, the abnormal points are corrected by using a mean filling tool to obtain a corrected data set; Extracting related fields based on the modified data set, constructing a query structure using an indexing tool, and obtaining a query data set; According to the complete records of the query data set, adjusting the order according to the category distribution by a sorting tool to obtain a sorted data set; According to the structural characteristics of the sorted data set, integrity verification is performed using a consistency verification tool to obtain a verification data set.

3. The information management method based on smart business according to claim 1, characterized in that: Extract key indicators based on the target transmission data, obtain a predicted value of short-term resource demand and a real-time load of the current system resource usage status, and fuse the predicted value and the real-time load using a weighted average algorithm to obtain an adjusted resource demand assessment value, including: Obtaining key indicators based on the target transmission data, combining the predicted value of short-term resource demand, and performing real-time load fusion through a weighted average method to obtain preliminary resource demand data; Determining whether there is abnormal fluctuation based on the sudden increase trend of the preliminary resource demand data, and obtaining fluctuation detection data; According to the downward trend in the fluctuation detection data, adjusting the resource allocation order by a sorting tool to obtain rearranged resource data; Obtaining balanced load data by acquiring load balancing distribution according to the system status of the rearranged resource data; Acquire data fusion characteristics according to the balanced load data, and if the sudden increase trend exceeds a preset threshold, adjust resource demand to obtain revised demand data; Obtaining verification demand data by determining whether resource requirements satisfy system status based on the estimated value of the modified demand data; According to the key indicators of the verification demand data, resource allocation adjustment is performed through a mean filling tool to obtain a resource demand assessment value.

4. The information management method based on smart business according to claim 3, characterized in that: Determine a weight distribution ratio based on the adjusted resource demand assessment value, and perform error prediction on the weight distribution ratio using a regression algorithm to obtain analysis result data, including: Determine the weight distribution ratio based on the adjusted resource demand assessment value, predict the real-time requirements and accuracy requirements through a regression analysis algorithm, and obtain preliminary prediction data; If the prediction error in the preliminary prediction data is lower than a preset threshold, the analysis result data is determined; if it is higher than the preset threshold, the model parameters are adjusted to perform a re-prediction to obtain the revised prediction data; According to the real-time requirement of the revised forecast data, trend analysis data is obtained by analyzing the changing trend of the allocation ratio; According to the accuracy requirements of the trend analysis data, the weight distribution is adjusted through the mean filling tool to obtain balanced distribution data; Based on the analysis results of the balanced allocation data, the fluctuation of resource demand is detected to obtain fluctuation detection data; According to the change trend of the fluctuation detection data, the estimated value of the resource demand is adjusted to obtain the final demand data; Key indicators are extracted according to the final demand data, and verification result data is obtained as analysis result data by judging whether the real-time requirement meets the system status.

5. The information management method based on smart business according to claim 1, characterized in that: Generate a multi-dimensional feature vector based on the analysis result data, perform dimensionality reduction processing on the data analysis connection requirements through the principal component analysis algorithm, and optimize feature extraction to adjust the display parameter range to obtain related data, including: Generate a multidimensional feature vector based on the analysis result data, perform dimensionality reduction processing on the data analysis connection requirements through the principal component analysis algorithm, and obtain dimensionality reduction analysis data; According to the optimized feature extraction process of the dimensionality reduction analysis data, the display parameter range is adjusted by the data change trend determination threshold to obtain the associated data of resource fluctuation and visualization structure, including: Adjust the feature extraction process according to the dimensionality reduction analysis data, and obtain trend determination data by judging the sudden increase trend of the preset threshold; Determine the downward trend of the data according to the trend, and obtain parameter optimization data by adjusting the display parameter range; According to the resource fluctuation of the parameter optimization data, distribution detection data is obtained by detecting the distribution of the visualization structure; According to the associated data of the distribution detection data, the core dimension of data optimization is extracted to obtain dimension extraction data; If the dimension extraction data meets the judgment threshold, the trend judgment is adjusted through feature extraction to obtain adjustment analysis data; According to the visualization structure of the adjustment analysis data, optimizing the allocation ratio of resource fluctuations to obtain ratio optimization data; According to the parameter range of the ratio optimization data, the change detection data is obtained by judging the change of the sudden increase trend.

6. The information management method based on smart business according to claim 1, characterized in that: A visualization model is constructed based on the dimensionality reduction analysis data, the multidimensional feature vector is converted into a graphic element through data mapping, and the display parameters of the graphic element are adjusted according to the presentation capability requirements to obtain visualization presentation parameters, including: Obtaining a feature vector by parsing the dimensionality reduction analysis data; According to the characteristic vector, vector conversion is performed by using data mapping technology to obtain graphic elements; Adjust display parameters according to presentation capabilities and determine the display format of graphic elements; Build visualization models based on mapping technology to obtain presentation data; If the presentation data meets a preset threshold, the distribution of the graphic elements is optimized by adjusting parameters to obtain adjusted presentation data; Based on the adjusted presentation data, the improvement of presentation ability is judged by detecting the change trend of the feature vector; The visualization model is updated according to the change trend to obtain optimized presentation data.

7. The information management method based on smart business according to claim 1, characterized in that: Obtaining user feedback based on the visualization presentation parameters, optimizing the system-level collaborative requirements through an iterative update algorithm to obtain feedback data, and comparing the feedback data with preset improvement conditions to obtain optimized system data by adjusting the system level, including: Obtain user feedback based on user interaction data and obtain the distribution characteristics of feedback content through recording tools; The feedback triggering situation is judged based on the distribution characteristics. If the preset improvement condition is triggered, the system level parameters are adjusted through an iterative algorithm to obtain the adjusted collaborative demand data; Based on the adjusted collaborative demand data, the update direction of the processing parameters is obtained through the analysis process; According to the update direction, the optimized system-level data is obtained by adjusting the processing parameters; Based on the optimized system-level data, determine whether feedback triggers have decreased by detecting the changing trend of interaction data; If the trend of change shows a decrease in triggers, the presentation format is updated through the analysis process to obtain optimized data; Based on the optimization data, the matching degree of the collaborative requirements is tested by a verification tool to determine the optimized system data output by the final system.

8. An information management device based on smart business, characterized in that: A method for implementing the information management method based on smart business according to any one of claims 1 to 7, comprising: The data acquisition module is used to obtain multi-source raw data through a preset standardized protocol, and map it into a unified structure through a format conversion algorithm to obtain a standardized data set; A data grouping module is used to extract characteristic attribute data from the normalized data set, group the characteristic attribute data using a clustering algorithm, and use the grouping results that meet the heterogeneity threshold as the classification data set; A dynamic analysis module is used to obtain time series features based on the classified data set, and to perform dynamic change sampling using a sliding window technique based on the difficulty of collecting the time series features to obtain dynamic data samples; A modeling and optimization module is used to construct a relationship network based on the dynamic data samples and to model and mine implicit associations, and to optimize the information flow path and transmission based on the results to obtain target transmission data; A resource assessment module is configured to extract key indicators based on the target transmission data, obtain a predicted value of short-term resource demand and a real-time load of the current system resource usage status, and fuse the predicted value and the real-time load using a weighted average algorithm to obtain an adjusted resource demand assessment value; An evaluation and prediction module is used to determine a weight distribution ratio according to the adjusted resource demand evaluation value, and to perform error prediction on the weight distribution ratio through a regression algorithm to obtain analysis result data; A vector association module is used to generate a multidimensional feature vector based on the analysis result data, perform dimensionality reduction processing on the data analysis connection requirements through the principal component analysis algorithm, and optimize feature extraction to adjust the display parameter range to obtain associated data; A parameter modeling module is used to construct a visualization model based on the dimensionality reduction analysis data, convert the multidimensional feature vector into a graphic element through data mapping, and adjust the display parameters of the graphic element according to the intuitive presentation capability requirements to obtain visualization presentation parameters; A feedback optimization module is used to obtain user feedback based on the visualization presentation parameters, optimize the system level collaboration requirements through an iterative update algorithm to obtain feedback data, and compare the feedback data with preset improvement conditions to obtain optimized system data by adjusting the system level.

9. A computer-readable storage medium, characterized in that The computer-readable storage medium includes a stored computer program, wherein when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute the information management method based on smart business as described in any one of claims 1 to 7.

Images