Enterprise management intelligent decision system integrated with big data analysis

Abstract

The application discloses an enterprise management intelligent decision system integrated with big data analysis, and relates to the technical field of enterprise intelligent management.The system comprises: an installation package receiving and analyzing module, which is used for connecting an enterprise management software system and receiving and analyzing an installation package when upgrading is triggered; an installation package portrait building module, which extracts a function label word vector and a key repair module based on NLP technology to form an upgrade package portrait; a data portrait building module, which extracts and analyzes multi-source historical operation data to form an enterprise data portrait containing module usage heat and fault records; and a quantitative matching module, which performs upgrade matching calculation according to the two types of portraits, outputs upgrade quantitative indexes, and compares the indexes with preset indexes to generate an upgrade decision result. Thus, the technical effects of optimizing upgrade adaptability and enhancing self-analysis and active decision capabilities are achieved.

Description

Technical Field

[0001] This invention relates to the field of enterprise intelligent management technology, and in particular to an enterprise management intelligent decision-making system integrating big data analysis. Background Technology

[0002] Currently, most enterprise management relies on digital software for operations. This software is regularly upgraded to introduce new features, fix bugs, or improve performance. However, enterprise software upgrades are typically pushed out uniformly by software vendors, with upgrade content primarily based on general needs or industry trends, rather than the specific needs of individual enterprises. This "one-size-fits-all" upgrade approach may lead enterprises to the following problems:

[0003] The upgraded features may not match the company's actual business needs, leading to wasted resources; the upgrade may introduce unnecessary complexity or performance degradation, especially when the upgrade content is incompatible with the existing system's operation; the company lacks effective tools to assess the necessity of the upgrade and can only passively accept the upgrade instead of making proactive decisions. Summary of the Invention

[0004] This invention provides an intelligent decision-making system for enterprise management that integrates big data analytics to solve the technical problems of poor software upgrade adaptability and lack of independent analysis and proactive decision-making capabilities in existing technologies, thereby achieving the technical effects of online monitoring, complete monitoring, and good early warning capabilities.

[0005] The intelligent decision-making system for enterprise management integrating big data analytics provided by this invention includes:

[0006] The installation package receiving and parsing module is used to connect to the enterprise management software system. When the enterprise management software system triggers an upgrade, it receives and parses the upgrade installation package.

[0007] The installation package profiling module is used to perform semantic tag recognition on the upgrade installation package based on NLP technology to construct an upgrade package profile, which includes a set of functional tag word vectors and a set of key repair modules for the upgrade package.

[0008] The data profiling module is used to extract multi-source historical operating data from multiple heterogeneous software systems of the target enterprise, analyze the multi-source historical operating data, and construct an enterprise data profile. The enterprise data profile includes a set of functional modules based on usage frequency and a set of faulty modules.

[0009] The quantitative matching module is used to perform quantitative calculations for upgrade matching based on the upgrade package profile and the enterprise data profile, output upgrade quantitative indicators, compare the upgrade quantitative indicators with preset quantitative indicators, and return the upgrade decision result.

[0010] In one feasible implementation, upgrade matching quantification calculation is performed based on the upgrade package profile and the enterprise data profile. The quantification matching module includes:

[0011] The overlap matching degree calculation unit is used to call the function tag word vector set of the upgrade package profile and the usage popularity set of the enterprise data profile to perform cosine similarity matching and output the function overlap matching degree.

[0012] The correlation matching degree calculation unit is used to call the key repair module set of the upgrade package profile and the fault module set of the enterprise data profile to perform fault correlation identification and output the fault correlation matching degree.

[0013] The quantitative indicator generation unit is used to calculate the functional overlap matching degree and the fault association matching degree, and output the upgrade quantitative indicator based on the upgrade installation package.

[0014] In one feasible implementation, upgrade matching quantification calculation is performed based on the upgrade package profile and the enterprise data profile. The quantification matching module further includes:

[0015] The data analysis unit is used to analyze the multi-source historical operation data and obtain a set of functional module time factors, wherein the functional module time factors do not represent the interval between the last upgrade and each functional module.

[0016] The weight calculation and matching degree update unit is used to calculate the weight of modules with overlapping functions based on the set of time factors of the functional modules and update the functional overlap matching degree; and to calculate the weight of modules with fault association based on the set of time factors of the functional modules and update the fault association matching degree.

[0017] The indicator update unit is used to recalculate the updated functional overlap matching degree and the updated fault association matching degree, and update the upgrade quantification indicator.

[0018] In one feasible implementation, the upgrade quantification index is compared with a preset quantification index, and an upgrade decision result is returned. The execution steps of the quantification matching module include:

[0019] If the upgrade quantification index is less than the preset quantification index, the upgrade installation package is cached.

[0020] If the upgrade quantification index is greater than or equal to the preset quantification index, the enterprise management software system is upgraded according to the upgrade installation package.

[0021] In one feasible implementation, the quantitative matching module, which returns the upgrade decision result, also includes:

[0022] An enterprise management cloud center establishment unit is used to establish an enterprise management cloud center, which connects multiple enterprise management software systems, wherein each enterprise management software system corresponds to one enterprise user.

[0023] The multi-terminal upgrade decision and execution unit is used to obtain multiple upgrade decision results returned by the multi-terminal enterprise management software system according to the enterprise management cloud center, and upgrade the multi-terminal enterprise management software system with the multiple upgrade decision results.

[0024] In one feasible implementation, the execution steps of the installation package profiling module include obtaining a set of functional tag word vectors:

[0025] The upgrade installation package is processed using NLP technology to output the upgrade installation text.

[0026] Define functional keyword samples and obtain the context word vector of each functional keyword in the upgrade installation text.

[0027] The context word vectors are aggregated to obtain functional labels representing each keyword, and the functional label word vector set is output.

[0028] In one feasible implementation, the key repair module set is obtained, and the execution steps of the installation package profiling module include:

[0029] Define a sample of repair action keywords, identify the relevant modules for each repair action keyword in the upgrade installation text, and output a set of key repair modules.

[0030] In one feasible implementation, the data profiling module obtains a usage popularity set, and its execution steps include:

[0031] Analyze the multi-source historical operation data, which includes call logs, operation frequency, exception records, and response latency.

[0032] The call frequency, continuous call cycle, and number of users operating each functional module are obtained.

[0033] The call frequency, continuous call cycle, and number of users of each functional module are normalized to obtain the usage popularity set.

[0034] In one feasible implementation, the data profiling module obtains a set of fault modules, and the execution steps of the data profiling module include:

[0035] By analyzing the multi-source historical operating data, the frequency, level, and scope of failure of each functional module are obtained.

[0036] The frequency, severity, and impact range of each functional module are normalized to obtain a set of fault modules.

[0037] This invention discloses an intelligent decision-making system for enterprise management integrating big data analytics, comprising: an installation package receiving and parsing module for connecting to an enterprise management software system, receiving and parsing an upgrade installation package when the system triggers an upgrade; an installation package profiling module for semantic tag recognition of the upgrade installation package based on NLP technology, constructing an upgrade package profiling including a set of functional tag word vectors and a set of key repair modules; a data profiling module for extracting multi-source historical operating data from multiple heterogeneous software systems of the target enterprise, and constructing an enterprise data profiling through analysis, including a set of usage popularity and a set of faulty modules for each functional module; and a quantitative matching module for performing quantitative calculations of upgrade matching based on the upgrade package profiling and the enterprise data profiling, generating upgrade quantitative indicators, comparing these indicators with preset quantitative indicators, and finally outputting upgrade decision results. This intelligent decision-making system for enterprise management integrating big data analytics solves the technical problems of poor software upgrade adaptability and the lack of independent analysis and proactive decision-making capabilities for enterprises, achieving the technical effects of optimizing upgrade adaptability and enhancing independent analysis and proactive decision-making capabilities. Attached Figure Description

[0038] Figure 1 This is a schematic diagram of the structure of the enterprise management intelligent decision-making system integrating big data analysis, as described in this invention.

[0039] Figure 2 This is a flowchart illustrating the process of obtaining functional tag word vector sets in the enterprise management intelligent decision-making system that integrates big data analysis, as described in this invention.

[0040] Figure labeling: 11 Installation package receiving and parsing module, 12 Installation package profile building module, 13 Data profile building module, 14 Quantitative matching module. Detailed Implementation

[0041] The above technical solutions will now be described in detail with reference to the accompanying drawings and specific embodiments to provide a better understanding of them. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. It should be understood that the present invention is not limited to the exemplary embodiments used only to explain the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention. Furthermore, it should be noted that, for ease of description, only the parts related to the present invention are shown in the drawings, not all of them.

[0042] Example, Figure 1This is a schematic diagram of the structure of the enterprise management intelligent decision-making system integrating big data analysis according to the present invention, wherein the enterprise management intelligent decision-making system integrating big data analysis includes:

[0043] The installation package receiving and parsing module 11 is used to connect to the enterprise management software system and receive and parse the upgrade installation package when the enterprise management software system triggers an upgrade.

[0044] Specifically, the installation package receiving and parsing module 11 is used to connect with the enterprise management software system and complete the receiving and parsing of the upgrade installation package before the system upgrade.

[0045] Specifically, the module establishes a communication channel with the enterprise management software system through a network interface (such as API or WebSocket) to ensure the stability and real-time performance of data transmission. When the enterprise management software system triggers an upgrade, the module obtains the binary file of the upgrade installation package from the server through a preset communication protocol (such as HTTP or FTP). Then, the obtained installation package is structured to extract key information (such as function description, repair modules, version number, etc.).

[0046] For example, the upgrade installation package may include update files for software functional modules, configuration parameters, database structure change instructions, script files, etc. The installation package receiving and parsing module 11 performs parsing operations including: verifying the integrity of the installation package (e.g., using algorithms such as MD5 and SHA); decompressing the installation package contents and identifying the module types and version information contained therein; parsing the upgrade script or configuration file and extracting the key parameters required for the upgrade; and passing the parsing results to subsequent modules for generating an upgrade package profile. For instance, suppose an enterprise management software system triggers an upgrade, and the upgrade installation package is a ZIP file containing a functional description file (JSON format) and a list of repair modules (XML format). The ZIP file is received via HTTP protocol, decompressed, and the metadata in the JSON and XML files is extracted for parsing. The parsing results show that the upgrade package contains 3 new functional modules and 2 repair modules. This information will be passed to subsequent modules for generating an upgrade package profile and performing matching calculations.

[0047] The installation package receiving and parsing module 11 serves as the entry point in the entire system, ensuring that subsequent modules can perform analysis and make decisions based on accurate installation package information.

[0048] The installation package profile building module 12 is used to perform semantic tag recognition on the upgrade installation package based on NLP technology to build an upgrade package profile, which includes a set of functional tag word vectors and a set of key repair modules of the upgrade package.

[0049] Specifically, the installation package image construction module 12 is used to perform semantic analysis on the content of the installation package based on natural language processing (NLP) technology after receiving and parsing the upgraded installation package, so as to construct a function image of the upgrade package to assist subsequent upgrade evaluation, risk prediction, and module-level dependency management.

[0050] Specifically, the function label word vector set is a set of vectors formed by converting function keywords into word vectors (such as through the Word2Vec model) to quantify the semantic information of function modules. The key repair module set is a set of modules related to repair actions identified in the obtained installation package, such as "permission control module", "data synchronization module", "report engine module", etc., for subsequent matching calculations.

[0051] In some implementation manners, such as Figure 2 shown, to obtain the function label word vector set, the execution steps of the installation package image construction module 12 include:

[0052] S100: Perform text processing on the upgraded installation package based on NLP technology, and output the upgraded installation text; S200: Define function keyword samples, and obtain the context word vectors of each function keyword in the upgraded installation text; S300: Aggregate the context word vectors to obtain function labels representing each keyword, and output the function label word vector set.

[0053] Specifically, the function keyword samples are a predefined set of keyword collections used to identify the function module break points in the installation package, that is, the keywords corresponding to the function change points. These keywords can be defined by domain experts or extracted from historical data through machine learning models.

[0054] Specifically, the function label is a semantic label assigned to each function keyword based on the context semantic aggregation result, such as "performance enhancement class", "compatibility update class", "security repair class", etc.

[0055] Specifically, first, based on natural language processing (NLP) technology, including semantic recognition models such as BERT, Word2Vec, FastText, etc., perform preprocessing operations on the text information in the upgraded installation package to obtain the upgraded installation text. Among them, the upgraded installation text may include change logs, upgrade instruction documents, script comments, configuration instructions, etc. The text processing process includes character cleaning, language detection, word segmentation (such as splitting "customer management module" into "customer", "management", "module"),词性标注 (positional tagging), stop word filtering (such as "of", "is", "and", etc.), and stemming (such as unifying "management" and "manager" as "management").

[0056] It should be noted that there is a misspelling in the original text. "词性标注" should be "positional tagging" in English. I have corrected it in the translation.Specifically, the process then loads predefined functional keyword samples (such as "customer management", "order processing", "data analysis" etc.) and uses them as a matching benchmark to locate the predefined functional keywords in the upgrade installation text. Based on the context window mechanism (such as 5 words before and after), the context content of these keywords when they appear in the text is extracted. Then, through a pre-trained word vector model (such as Word2Vec, GloVe, BERT, etc.), each word in the text and its context information are converted into a fixed-length vector representation (i.e., context word vector) to capture the semantic features of the words.

[0057] Furthermore, the context word vectors corresponding to each functional keyword are aggregated (e.g., average pooling, weighted aggregation, attention mechanisms, etc.) to generate a vector representing the semantic features of that keyword, which serves as the functional label vector for that functional keyword. Then, the functional label vectors of all functional keywords are aggregated to form the functional label word vector set of the upgrade installation package, which is output as part of the upgrade package profile for subsequent analysis and use.

[0058] In the above process, by aggregating contextual word vectors, the module can capture the semantic features of functional keywords, avoid semantic bias caused by isolated keywords, and thus improve semantic matching accuracy. At the same time, by dynamically updating functional keyword samples, it can adapt to changes in functional descriptions in different fields and enterprises, reduce manual intervention and improve the intelligence of the system.

[0059] In some implementations, the steps of obtaining the key repair module set, and the execution steps of the installation package profiling module 12, include:

[0060] Define a sample of repair action keywords, identify the relevant modules for each repair action keyword in the upgrade installation text, and output a set of key repair modules.

[0061] Specifically, firstly, a set of repair action keyword samples is defined to represent common repair behaviors or actions in the upgrade installation package. These can be constructed through manual compilation, historical patch analysis, fault ticket mining, etc. For example, the repair action keyword samples include, but are not limited to: "repair", "correct", "adjust", "optimize", "patch", "resolve", "avoid", "tune", "correct", "eliminate", etc., which are used to locate semantic fragments related to defect repair, security hardening, performance optimization, etc. in the upgrade installation text, respectively.

[0062] Specifically, the upgrade installation text is scanned to identify sentences or paragraphs containing keywords related to the repair action. Entity recognition and dependency parsing are then used to further analyze the technical entities mentioned in the context, such as system modules, components, class names, function names, and configuration items. For example, if the text contains "repair the performance problem of the order processing module", then "repair" is associated with "order processing module".

[0063] Optionally, the extracted results can be lexical normalization (such as name deduplication and module name normalization mapping) to form a denoised module set (i.e., a key repair module set). This can be achieved by: using Named Entity Recognition (NER) technology to identify module names; matching based on existing module dictionaries or module mapping relationships; and performing reverse parsing by combining code comments or path information (e.g., "The token refresh logic in the com.company.auth permission control module has been fixed").

[0064] Optionally, the identified repair actions and their associated modules are structured and organized to form a set of key repair modules involved in the upgrade installation package, which is output as part of the upgrade package profile. The module set may include information such as module name, repair type, scope of impact, and repair location, thereby providing high-precision data support for subsequent matching calculations.

[0065] The data profiling module 13 is used to extract multi-source historical operation data from multiple heterogeneous software systems of the target enterprise, analyze the multi-source historical operation data, and construct an enterprise data profile. The enterprise data profile includes a set of functional modules based on usage frequency and a set of faulty modules.

[0066] Specifically, multi-source historical operational data refers to various types of data generated during the operation of an enterprise software system, including call logs, operation frequency, exception records, response latency, etc. Multi-source historical operational data corresponds to heterogeneous software systems (such as software systems applicable to different enterprises, scenarios, users, etc.). In other words, it is necessary to obtain multi-source historical operational data from all software systems (multiple heterogeneous software systems) involved in the software to be updated in order to determine the potential impact of the software update. This is because different heterogeneous software systems have their own requirements and operating environments, which leads to the possibility that software updates may have different impacts on different software systems (an upgrade of a single system may affect the normal operation of other systems).

[0067] Specifically, enterprise data profiling is a knowledge abstraction result based on the analysis of the aforementioned multi-source historical operational data, which is oriented towards the use and health status of software systems. For example, by performing semantic modeling, cluster analysis, and labeling on historical data, a comprehensive profile information that can characterize the actual use scenarios and module quality status of enterprise software can be constructed.

[0068] Specifically, the usage frequency set refers to the set of functional modules whose usage frequency and activity are quantified by analyzing historical operation data from multiple sources; the fault module set refers to the set of functional modules that frequently experience faults or performance problems by analyzing historical operation data from multiple sources.

[0069] The aforementioned data profiling module 13, through comprehensive analysis of multi-source historical operational data of the enterprise, can generate accurate usage popularity sets and fault module sets, providing reliable data support for subsequent matching calculations.

[0070] In some implementations, the data profiling module 13 obtains an applicable heat set, and the execution steps include:

[0071] The multi-source historical operation data is analyzed, including call logs, operation frequency, exception records, and response latency; the call frequency, continuous call cycle, and number of users operating each functional module are obtained; the call frequency, continuous call cycle, and number of users operating each functional module are normalized to obtain a usage popularity set.

[0072] Specifically, firstly, by interfacing with the software system deployed by the target enterprise, historical data of the system during actual operation is obtained, including: call logs, which record the call time, number of calls, and call source of each functional module or interface; operation frequency, which counts the number of times users operate on functional modules in the front-end interface or system operation; exception records, which include errors, exceptions, failure retries and other events generated during module operation; and response latency, which records the response time, latency fluctuations and other performance indicators of the module when it is called.

[0073] Furthermore, based on the aforementioned multi-source historical operational data, the following core usage metrics for each functional module are extracted: the number of times the module is called per unit time (call frequency), the time span during which a module is continuously accessed over a period of time (i.e., continuous call cycle), and the number of independent users accessing or operating the module within the statistical period (number of operating users). Then, to facilitate unified quantification and comparison among different metrics, the above three metrics are normalized. Methods such as linear normalization, Z-score standardization, and quantile normalization can be used to convert each metric value into a standardized value within the [0, 1] interval to eliminate differences in the dimensions and orders of magnitude of different metrics. The normalized call frequency, continuous call cycle, and number of operating users are then integrated to construct a usage heat vector for each functional module. This usage heat vector is used to quantitatively represent the usage heat of each functional module, and multiple usage heat vectors are aggregated to form the system's usage heat set.

[0074] For example, assuming a statistical period of 24 hours, we extract indicators and normalize them to construct a usage popularity vector, forming a usage popularity set:

[0075] Table 1 Core Usage Metrics of Functional Modules

[0076] Functional Module ID Call frequency (times / hour) Continuous call cycle (minutes) Number of users (persons) M001 3 180 2 M002 2 300 2 M003 1 60 1

[0077] Table 2. Normalized results of core usage metrics for functional modules

[0078] Functional Module ID Call frequency (normalized) Continuous call cycle (normalized) Number of operating users (normalized) M001 1.0 (180-60) / (300-60)=0.5 (2-1) / (2-1)=1.0 M002 (2-1) / (3-1)=0.5 (300-60) / (300-60)=1.0 (2-1) / (2-1)=1.0 M003 0.0 0.0 0.0

[0079] The normalized indicators are weighted and fused (with equal weights) to generate a module popularity vector:

[0080] Table 3. Use heat sets represented by use heat vectors

[0081] Functional Module ID Using heat vector M001 [1.0,0.5,1.0] M002 [0.5,1.0,1.0] M003 [0.0,0.0,0.0]

[0082] Optionally, in other implementations, the process of constructing the popularity set can be further optimized, including: introducing time decay weights to give higher weight to recent usage behavior; using clustering algorithms to classify module popularity patterns (such as high frequency, low frequency, and fluctuating type); combining user profiles to achieve module popularity difference analysis under different user groups; and introducing negative indicators such as anomaly rate and response latency to correct module popularity scores and improve stability assessment capabilities.

[0083] Through the above steps, the data profiling module 13 can achieve the following:

[0084] It accurately depicts the usage intensity and activity of each functional module; provides a quantitative ranking of module popularity, offering data support for subsequent test priority ranking, upgrade impact assessment, and resource scheduling; and facilitates comparative analysis of module popularity across time periods, versions, and enterprises.

[0085] In some implementations, the data profiling module 13 obtains a set of fault modules, and its execution steps include:

[0086] By analyzing the multi-source historical operating data, the fault frequency, fault level, and fault impact range of each functional module are obtained; the fault frequency, fault level, and fault impact range of each functional module are normalized to obtain the fault module set.

[0087] Specifically, to obtain the fault module set of each functional module in the target enterprise's software system, firstly, based on multi-source historical operational data, the frequency, severity, and impact range of faults for different functional modules are analyzed and extracted. This involves counting the number of times each module experiences a fault within a set time window, obtaining a severity-based classification (e.g., Level 1 fault: system unavailable, service interrupted; Level 2 fault: functional abnormality but system available; Level 3 fault: performance degradation or intermittent errors), and assessing the number of users, system components, or business process nodes affected by the fault. The severity-based classification mentioned above is the result of a classification under preset rules for the target scenario.

[0088] Furthermore, to achieve unified measurement among different indicators, the three indicators mentioned above are normalized, specifically including: linear normalization of fault frequency; numerical mapping of fault levels (e.g., level 1 mapped to 1.0, level 3 mapped to 0.3); and proportionalization of the impact range based on the number of users or modules. The normalized indicator values ​​will be uniformly converted to the [0, 1] interval using data fusion methods (e.g., weighted summation), and then subjected to frequency filtering according to preset selection rules to form a fault module set, facilitating subsequent fusion and sorting. Optionally, frequency filtering includes filtering based on relative percentages, such as selecting the top 50% of modules in the serialized fusion results as fault modules.

[0089] The quantitative matching module 14 is used to perform quantitative calculations for upgrade matching according to the upgrade package profile and the enterprise data profile, output upgrade quantitative indicators, compare the upgrade quantitative indicators with preset quantitative indicators, and return the upgrade decision result.

[0090] Specifically, upgrade matching quantification calculation refers to the process of comparing the upgrade package profile and the enterprise data profile using mathematical models or algorithms, and calculating the corresponding quantitative indicators. The obtained upgrade quantitative indicators are used to measure the degree of matching between the upgrade package and the enterprise's needs. Preset quantification refers to predefined thresholds used to determine whether the upgrade quantitative indicators meet the upgrade standards.

[0091] Specifically, the upgrade quantitative indicators are compared with preset quantitative indicators. If the quantitative indicator is greater than or equal to the preset value, an upgrade decision result of "upgrade recommended" can be returned; otherwise, an upgrade decision result of "upgrade not recommended" can be returned.

[0092] The quantitative matching module 14 can accurately assess the matching degree between the upgrade package and the enterprise's needs through quantitative calculation, reducing manual intervention and avoiding errors caused by subjective judgment.

[0093] In some embodiments, upgrade matching quantitative calculation is performed based on the upgrade package profile and the enterprise data profile. The quantitative matching module 14 includes:

[0094] The overlapping matching degree calculation unit is used to perform cosine similarity matching between the function tag word vector set of the upgrade package profile and the usage popularity set of the enterprise data profile, and output the function overlapping matching degree; the association matching degree calculation unit is used to perform fault association identification between the key repair module set of the upgrade package profile and the fault module set of the enterprise data profile, and output the fault association matching degree; the quantitative index generation unit is used to calculate the function overlapping matching degree and the fault association matching degree, and output the upgrade quantitative index based on the upgrade installation package.

[0095] Specifically, by performing quantitative calculations to match upgrade package profiles with enterprise data profiles, intelligent upgrade decisions can be made within the enterprise environment. Overlap matching degree refers to the similarity between the functional tag word vector set of the upgrade package and the usage popularity set of the enterprise data, typically calculated using cosine similarity. Fault association matching degree refers to the degree of matching between the set of key repair modules in the upgrade package and the set of fault modules in the enterprise data, which can be calculated through set intersection or similarity.

[0096] Specifically, the function tag word vector set (from the upgrade package profile) and the usage popularity set (from the enterprise data profile) are used as inputs. The similarity between the two sets (i.e., the function overlap matching degree) is calculated by cosine similarity to measure the degree of overlap between the upgrade package functions and the enterprise's frequently used modules. The numerical range of the function overlap matching degree is [0, 1], and the larger the value, the higher the matching degree.

[0097] Meanwhile, taking the set of key repair modules (from the upgrade package profile) and the set of fault modules (from the enterprise data profile) as input, the module association is identified through methods such as module ID matching, function tag similarity, or call relationship graph analysis. The output result is used as the fault association matching degree to reflect the potential value of the upgrade package in improving the stability of the enterprise's current system.

[0098] For example, the fault association matching degree of two sets can be calculated by using the intersection of the sets. The formula for the fault association matching degree is:

[0099]

[0100] Furthermore, the two matching results are fused and calculated to output upgrade priority or value quantification scores. For example, the functional overlap matching degree and fault association matching degree are weighted and summed to generate upgrade quantification indicators, which are used to guide whether to upgrade, upgrade priority ranking, or upgrade resource allocation.

[0101] The above-mentioned units enable the following: intelligently assessing the suitability and value of upgrade packages in the enterprise environment based on actual business usage and failure risks; avoiding ineffective or low-value upgrades and improving the efficiency of upgrade resource utilization; supporting personalized upgrade recommendations and differentiated operation and maintenance strategy formulation; and improving system stability and functional coverage.

[0102] In some embodiments, the upgrade matching quantitative calculation is performed based on the upgrade package profile and the enterprise data profile. The quantitative matching module 14 further includes:

[0103] The data analysis unit is used to analyze the multi-source historical operation data to obtain a set of functional module time factors, wherein the functional module time factor represents the interval between each functional module and the last upgrade; the weight calculation and matching degree update unit is used to calculate the weight of modules with overlapping functions based on the set of functional module time factors and update the functional overlap matching degree, and to calculate the weight of modules with fault association based on the set of functional module time factors and update the fault association matching degree; the indicator update unit is used to recalculate the updated functional overlap matching degree and the updated fault association matching degree and update the upgrade quantitative indicator.

[0104] Specifically, based on the quantitative calculation of upgrade matching according to the upgrade package profile and the enterprise data profile, a set of functional module time factors is further introduced. This set of functional module time factors is used to introduce information on the upgrade time interval dimension, thereby adjusting the score of systems that have not been upgraded for a long time, so as to improve the rationality and timeliness of upgrade recommendations.

[0105] Specifically, firstly, time-related information for functional modules is extracted from multi-source historical operational data of the enterprise to construct a set of time factors for each functional module. This includes extracting the last upgrade time of each functional module in the current enterprise system (e.g., through system maintenance logs) and comparing it with the current time to calculate the upgrade interval of each module as the functional module's time factor. The larger the time factor, the longer the module has not been upgraded, potentially indicating security risks, technical debt, or compatibility issues, and thus a greater likelihood of it being updated. Conversely, if a module's updates are manually canceled or rolled back several times consecutively, it means that the current version of the module is likely the most suitable / stable / compliant with the target enterprise's processes, and in this case, the time factor may be extremely high.

[0106] Furthermore, the matching degree is adjusted based on the time factor, including:

[0107] Adjusting the functional overlap matching degree involves weighting the modules participating in the functional overlap matching by incorporating their time factors. For example, time-weighted averaging or weighted summation can be performed on the cosine similarity results; where, the larger the time factor of a functional module, the higher its weight in the matching degree calculation, reflecting its urgency for upgrading.

[0108] The adjustment of fault association matching involves weighting the modules participating in fault association matching in conjunction with their time factors. Specifically, if a module has both fault risk and has not been upgraded for a long time, its priority will be significantly increased; conversely, if a module skips updates several times in a row (corresponding to a very high functional module time factor), its priority can be reduced to the lowest level or set to not update automatically (manual review is required before updating).

[0109] Furthermore, the indicator update unit recalculates and outputs the upgrade quantitative indicators based on the updated matching results. This includes calling the updated functional overlap matching degree and fault association matching degree, recalculating the upgrade score according to the preset weighting function, and outputting the final upgrade quantitative indicators to support upgrade decision ranking, automated recommendation, or manual approval.

[0110] By introducing a time factor, the above-mentioned units and corresponding processes can achieve the following technical effects: dynamically identify modules that have not been upgraded for a long time, improve system security and technology update efficiency; and avoid ignoring the upgrade of old modules due to "low functional overlap" or "current no fault".

[0111] In some implementations, the upgrade quantification indicators are compared with preset quantification indicators, and an upgrade decision result is returned. The execution steps of the quantification matching module 14 include:

[0112] If the upgrade quantification index is less than the preset quantification index, the upgrade installation package is cached; if the upgrade quantification index is greater than or equal to the preset quantification index, the enterprise management software system is upgraded according to the upgrade installation package.

[0113] Specifically, upgrade quantification metrics are numerical values ​​calculated to represent the degree of match between the upgrade package and the enterprise's needs. Preset quantification metrics are predefined thresholds used to determine whether the upgrade quantification metrics meet the upgrade standards.

[0114] Specifically, if the upgrade quantification index is less than the preset quantification index, the upgrade installation package will be temporarily stored in the system for future use (when the upgrade quantification index is greater than the preset quantification index); if the upgrade quantification index is greater than or equal to the preset quantification index, the current upgrade version can be considered to have a high degree of matching with the current system, and the enterprise management software system can be directly upgraded according to the upgrade installation package to ensure that the enterprise software system can be updated in a timely manner and improve its functions and performance.

[0115] Optionally, the upgrade threshold can be configured to support dynamic adjustment, thereby automatically changing based on peak business periods, budget cycles, or operational strategies.

[0116] Optionally, the upgrade decision results can be output to the approval process interface, supplemented by a hybrid strategy of "manual confirmation + automatic execution". At the same time, a multi-level threshold mechanism can be introduced to realize various strategies such as "forced upgrade", "suggested upgrade" and "delayed upgrade".

[0117] Optionally, a periodic review mechanism can be set up for cached upgrade packages to promptly clean up unnecessary upgrade packages or simplify upgrade packages, retaining only the incremental / changed parts, thus avoiding the waste of system resources caused by long-term backlog.

[0118] In some implementations, the quantitative matching module 14 also includes the following in return for the upgrade decision result:

[0119] An enterprise management cloud center establishment unit is used to establish an enterprise management cloud center, which connects to multiple enterprise management software systems, wherein each enterprise management software system corresponds to one enterprise user; a multi-terminal upgrade decision and execution unit is used to obtain multiple upgrade decision results returned by the multiple enterprise management software systems according to the enterprise management cloud center, and upgrade the multiple enterprise management software systems with the multiple upgrade decision results.

[0120] Specifically, firstly, a unified enterprise management cloud platform is built through the establishment of an enterprise management cloud center to centrally manage the upgrade tasks of multiple enterprise management software systems. For example, this includes: establishing an enterprise management cloud center and connecting multiple enterprise management software systems deployed on different terminals or in different enterprise environments, with each enterprise user corresponding to an independent instance of the enterprise management software system; then, through the unified data collection, indicator calculation, upgrade package distribution, and upgrade strategy control capabilities of the enterprise management cloud center, information such as the operating status, upgrade history, and functional modules of each enterprise system is centrally managed.

[0121] Furthermore, through multi-terminal upgrade decision-making and execution units, and based on the enterprise management cloud center, the upgrade decision-making and execution processes of multiple terminal systems are uniformly coordinated. Specifically, this includes: obtaining upgrade quantification indicators and upgrade decision results returned by multiple enterprise management software systems from the enterprise management cloud center; then, based on a unified threshold set by the cloud center or a custom threshold defined by the enterprise, and combined with the upgrade decision results of each system, executing the upgrade action; for example, if the upgrade quantification indicators of a certain enterprise system meet the upgrade conditions, the corresponding upgrade installation package is automatically pushed to that system and the upgrade is executed; if the conditions are not met, the upgrade package is cached in the system, awaiting subsequent review or administrator confirmation.

[0122] Through the above process, centralized upgrade management of multiple enterprises and multiple terminal systems can be achieved, meeting the multi-tenant upgrade needs of group enterprises or large service providers.

[0123] In summary, the enterprise management intelligent decision-making system integrating big data analysis provided by this invention has the following technical effects:

[0124] The installation package receiving and parsing module connects to the enterprise management software system, receiving and parsing the upgrade installation package when the system triggers an upgrade. The installation package profiling module uses NLP technology to perform semantic tag recognition on the upgrade installation package, constructing an upgrade package profile that includes a set of functional tag word vectors and a set of key repair modules. The data profiling module extracts multi-source historical operating data from multiple heterogeneous software systems of the target enterprise and constructs an enterprise data profile through analysis, including a set of usage popularity and faulty modules for each functional module. The quantitative matching module performs quantitative calculations for upgrade matching based on the upgrade package profile and the enterprise data profile, generates upgrade quantitative indicators, compares these indicators with preset quantitative indicators, and finally outputs the upgrade decision result, thereby achieving the technical effects of optimizing upgrade adaptability and enhancing autonomous analysis and proactive decision-making capabilities.

[0125] It should be understood that the embodiments disclosed in this invention and the above description enable those skilled in the art to implement this invention. However, this invention is not limited to the embodiments mentioned above. It should be understood that those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this invention, and should all be included within the protection scope of this invention.

Claims

1. An intelligent decision-making system for enterprise management integrating big data analytics, characterized in that: include: The installation package receiving and parsing module is used to connect to the enterprise management software system. When the enterprise management software system triggers an upgrade, it receives and parses the upgrade installation package. The installation package profile building module is used to perform semantic tag recognition on the upgrade installation package based on NLP technology and build an upgrade package profile, which includes a set of functional tag word vectors and a set of key repair modules of the upgrade package. The data profiling module is used to extract multi-source historical operation data from multiple heterogeneous software systems of the target enterprise, analyze the multi-source historical operation data, and construct an enterprise data profile. The enterprise data profile includes a set of functional modules based on usage frequency and a set of faulty modules. The quantitative matching module is used to perform quantitative calculations for upgrade matching according to the upgrade package profile and the enterprise data profile, output upgrade quantitative indicators, compare the upgrade quantitative indicators with preset quantitative indicators, and return the upgrade decision result. The upgrade matching quantitative calculation is performed based on the upgrade package profile and the enterprise data profile. The quantitative matching module includes: The overlap matching degree calculation unit is used to call the function tag word vector set of the upgrade package profile and the usage popularity set of the enterprise data profile to perform cosine similarity matching, and output the function overlap matching degree. The correlation matching degree calculation unit is used to call the key repair module set of the upgrade package profile and the fault module set of the enterprise data profile to perform fault correlation identification and output the fault correlation matching degree. The quantitative indicator generation unit is used to calculate the functional overlap matching degree and the fault association matching degree, and output the upgrade quantitative indicator based on the upgrade installation package. The quantitative matching module further includes: performing upgrade matching quantitative calculations based on the upgrade package profile and the enterprise data profile. The data analysis unit is used to analyze the multi-source historical operation data and obtain a set of functional module time factors, wherein the functional module time factors do not represent the interval time between the last upgrade for each functional module. The weight calculation and matching degree update unit is used to calculate the weight of modules with overlapping functions based on the set of time factors of the functional modules and update the functional overlap matching degree; and to calculate the weight of modules with fault association based on the set of time factors of the functional modules and update the fault association matching degree. The indicator update unit is used to recalculate the updated functional overlap matching degree and the updated fault association matching degree, and update the upgrade quantification indicator.

2. The enterprise management intelligent decision-making system integrating big data analysis as described in claim 1, characterized in that, The upgrade quantitative indicators are compared with preset quantitative indicators, and an upgrade decision result is returned. The execution steps of the quantitative matching module include: If the upgrade quantification index is less than the preset quantification index, cache the upgrade installation package; If the upgrade quantification index is greater than or equal to the preset quantification index, the enterprise management software system is upgraded according to the upgrade installation package.

3. The enterprise management intelligent decision-making system integrating big data analysis as described in claim 1, characterized in that, The quantitative matching module also includes the following: (This is part of the upgrade decision process.) An enterprise management cloud center establishment unit is used to establish an enterprise management cloud center, which connects multiple enterprise management software systems, wherein each enterprise management software system corresponds to one enterprise user; The multi-terminal upgrade decision and execution unit is used to obtain multiple upgrade decision results returned by the multi-terminal enterprise management software system according to the enterprise management cloud center, and upgrade the multi-terminal enterprise management software system with the multiple upgrade decision results.

4. The enterprise management intelligent decision-making system integrating big data analysis as described in claim 1, characterized in that, The execution steps of the installation package profiling module include obtaining a set of functional tag word vectors: The upgrade installation package is processed using NLP technology to output the upgrade installation text. Define functional keyword samples and obtain the context word vector of each functional keyword in the upgrade installation text; The context word vectors are aggregated to obtain functional labels representing each keyword, and the functional label word vector set is output.

5. The enterprise management intelligent decision-making system integrating big data analysis as described in claim 4, characterized in that, The steps of the installation package profiling module to obtain the key repair module set include: Define a sample of repair action keywords, identify the relevant modules for each repair action keyword in the upgrade installation text, and output a set of key repair modules.

6. The enterprise management intelligent decision-making system integrating big data analysis as described in claim 1, characterized in that, The data profiling module, which obtains usage popularity sets, performs the following steps: Analyze the multi-source historical operation data, which includes call logs, operation frequency, exception records, and response latency; The call frequency, continuous call cycle, and number of users operating each functional module are obtained. The call frequency, continuous call cycle, and number of users of each functional module are normalized to obtain the usage popularity set.

7. The enterprise management intelligent decision-making system integrating big data analysis as described in claim 6, characterized in that, The data profiling module, which acquires a set of fault modules, includes the following execution steps: By analyzing the multi-source historical operating data, the fault frequency, fault level, and fault impact range of each functional module are obtained. The frequency, severity, and impact range of each functional module are normalized to obtain a set of fault modules.

Images