Data acquisition and analysis method and system for ERP (Enterprise Resource Planning) system
By obtaining the system operation log and transaction information of the ERP system, identifying active and suspicious nodes, combining feedback information and log information, we can determine whether there are abnormalities in the ERP system, and solve the problem of difficult to identify ERP system exceptions in the existing technology and improve system stability Sex and management efficiency.
Patent Information
- Application Number
- CN202411521237.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-10-29
- Publication Date
- 2024-11-26
- Estimated Expiration
- 2044-10-29
AI Technical Summary
It is difficult for the existing technology to effectively identify whether there are abnormalities in the ERP system, which affects the normal operation of the enterprise.
By obtaining the system operation log, transaction information and benchmark active information of the ERP system, determine active and suspicious system nodes, combine feedback information, transaction information and log information, and integrate the abnormality degree threshold to determine whether there are abnormalities in the ERP system.
It realizes timely identification of ERP system abnormalities, improves the stability and management efficiency of the system, helps technicians deal with abnormalities in a timely manner, and avoids affecting corporate operations.
Smart Images

Figure CN119026067A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of computer technology, and in particular to a data collection and analysis method and system for an ERP system. Background Art
[0002] In order to meet the needs of enterprise information and data management, ERP (Enterprise Resource Planning) systems are widely used in enterprises, greatly improving the efficiency of enterprises.
[0003] In related technologies, more and more companies will use ERP systems. Abnormalities in ERP systems will seriously affect the normal operation of companies. How to identify whether abnormalities in ERP systems have occurred has become a hot topic of research. Summary of the invention
[0004] The present application embodiment provides a data collection and analysis method and system for an ERP system, which can identify anomalies in the ERP system. The technical solution is as follows: In one aspect, a data collection and analysis method for an ERP system is provided, the method comprising: Acquire a system operation log of an ERP system on a target date, transaction information of a target transaction corresponding to the ERP system, and baseline activity information of the ERP system, wherein the ERP system includes a plurality of system nodes, and the baseline activity information is used to indicate baseline activity levels of the plurality of system nodes; Based on the system operation log and the baseline activity information, determining a plurality of active system nodes and a plurality of suspicious system nodes on the target date from the plurality of system nodes, the active system nodes being system nodes whose activity levels meet a first activity level condition, the suspicious system nodes being system nodes whose activity levels meet a second activity level condition, and the baseline activity level being determined based on the baseline activity information; Based on usage feedback information of the ERP system on the target date, the transaction information, the multiple active system nodes and the multiple suspicious system nodes, it is determined whether the ERP system has an abnormality on the target date, wherein the usage feedback information is feedback information submitted by an object using the ERP system on the target date.
[0005] In a possible implementation, determining the multiple active system nodes and the multiple suspicious system nodes on the target date from the multiple system nodes based on the system operation log and the baseline activity information includes: Determining the activity levels of the plurality of system nodes based on the system operation log; Determine a system node whose activity level is greater than or equal to an activity level threshold among the multiple system nodes as the active system node; Determining baseline activity levels of the plurality of system nodes based on the baseline activity information; A system node among the multiple system nodes, the deviation between the activity level and the benchmark activity level of which is greater than or equal to the deviation threshold, is determined as the suspicious system node.
[0006] In a possible implementation, the determining whether the ERP system has an abnormality on the target date based on the usage feedback information of the ERP system on the target date, the transaction information, the multiple active system nodes, and the multiple suspicious system nodes includes: Determining a first abnormality level of the ERP system based on the usage feedback information; determining a second abnormality degree of the ERP system based on the transaction information, the plurality of active system nodes, and the plurality of suspicious system nodes; determining a third abnormality degree of the ERP system based on first node logs of the plurality of active system nodes on the target date and second node logs of the plurality of suspicious system nodes on the target date, the first node log and the second node log belonging to the system operation log; Based on the first abnormality level, the second abnormality level, and the third abnormality level, it is determined whether the ERP system has an abnormality on the target date.
[0007] In a possible implementation, determining the first abnormality level of the ERP system based on the usage feedback information includes: Clustering the plurality of sub-feedback information in the usage feedback information to obtain a plurality of sub-feedback information groups, wherein a sub-feedback information group includes at least one sub-feedback information, and similarities between the sub-feedback information in the same sub-feedback information group meet a similarity condition; Performing feature extraction on at least one sub-feedback information in each sub-feedback information group to obtain group features corresponding to each of the sub-feedback information groups; Determining a first reference abnormality degree of each sub-feedback information group based on a group feature corresponding to each of the sub-feedback information groups; The first reference abnormality degree of each sub-feedback information group is merged by using the first weight of each sub-feedback information group to obtain the first abnormality degree, and the first weight is positively correlated with the amount of sub-feedback information in the corresponding sub-feedback information group.
[0008] In a possible implementation, determining the second abnormality level of the ERP system based on the transaction information, the multiple active system nodes, and the multiple suspicious system nodes includes: Based on the transaction information, determining a plurality of reference active system nodes, wherein the reference active system nodes are theoretically active system nodes when processing the target transaction; determining a second reference abnormality degree of the ERP system based on the plurality of reference active system nodes and the plurality of active system nodes; determining a third reference abnormality level of the ERP system based on deviations between the activity levels of the plurality of suspicious system nodes and a baseline activity level; The second reference abnormality level and the third reference abnormality level are merged to obtain a second abnormality level of the ERP system.
[0009] In a possible implementation, determining the third abnormality degree of the ERP system based on the first node logs of the multiple active system nodes on the target date and the second node logs of the multiple suspicious system nodes on the target date includes: Determine, based on the first node logs of the plurality of active system nodes on the target date, a first interaction abnormality degree and a first sending and receiving abnormality degree of each of the active system nodes; Determine, based on the second node logs of the plurality of suspicious system nodes on the target date, a second interaction abnormality degree and a second sending and receiving abnormality degree of each of the suspicious system nodes; Based on the first interaction abnormality degree and the first transceiver abnormality degree of each of the active system nodes, and the second interaction abnormality degree and the second transceiver abnormality degree of each of the suspicious system nodes, a third abnormality degree of the ERP system is determined.
[0010] In a possible implementation manner, determining the first interaction abnormality degree and the first sending and receiving abnormality degree of each of the active system nodes based on the first node logs of the multiple active system nodes on the target date includes: For any active system node among the multiple active system nodes, obtaining a data interaction log and a data transceiving log of the active system node from a first node log of the active system node; determining a first interaction abnormality degree of the active system node based on a node type of the active system node and the data interaction log; determining a first transceiving abnormality degree of the active system node based on a node type of the active system node and the data transceiving log; The determining, based on the second node logs of the plurality of suspicious system nodes on the target date, the second interaction abnormality degree and the second sending and receiving abnormality degree of each of the suspicious system nodes comprises: For any suspicious system node among the multiple suspicious system nodes, obtain the data interaction log and the data sending and receiving log of the suspicious system node from the second node log of the suspicious system node; determine the second interaction abnormality degree of the suspicious system node based on the node type of the suspicious system node and the data sending and receiving log; determine the second sending and receiving abnormality degree of the suspicious system node based on the node type of the suspicious system node and the data sending and receiving log.
[0011] In a possible implementation, determining whether the ERP system has an abnormality on the target date based on the first abnormality degree, the second abnormality degree, and the third abnormality degree includes: Merging the first abnormality degree, the second abnormality degree, and the third abnormality degree to obtain a target abnormality degree; When the target abnormality level is greater than or equal to the abnormality level threshold, determining that the ERP system has an abnormality on the target date; When the target abnormality level is less than the abnormality level threshold, it is determined that there is no abnormality in the ERP system on the target date.
[0012] In a possible implementation manner, after determining whether the ERP system has an abnormality on the target date based on the usage feedback information of the ERP system on the target date, the transaction information, the multiple active system nodes, and the multiple suspicious system nodes, the method further includes: In the case that the ERP system has an abnormality on the target date, determining at least one abnormal system node among the multiple system nodes based on first node logs of the multiple active system nodes on the target date and second node logs of the multiple suspicious system nodes on the target date; Based on the at least one abnormal system node, a cause of the abnormality of the ERP system is determined.
[0013] In one aspect, a data collection and analysis system for an ERP system is provided, the system comprising: an acquisition module, configured to acquire a system operation log of an ERP system on a target date, transaction information of a target transaction corresponding to the ERP system, and baseline activity information of the ERP system, wherein the ERP system includes a plurality of system nodes, the baseline activity information is used to indicate a baseline activity level of the plurality of system nodes, and the target transaction is a transaction processed using the ERP system on the target date; a node determination module, configured to determine, based on the system operation log and the baseline activity information, a plurality of active system nodes and a plurality of suspicious system nodes on the target date from the plurality of system nodes, wherein the active system nodes are system nodes whose activity levels meet a first activity level condition, and the suspicious system nodes are system nodes whose activity levels meet a second activity level condition, and the baseline activity level is determined based on the baseline activity information; an abnormality identification module, for determining whether the ERP system has an abnormality on the target date based on usage feedback information of the ERP system on the target date, the transaction information, the multiple active system nodes, and the multiple suspicious system nodes, wherein the usage feedback information is feedback information submitted by an object using the ERP system on the target date.
[0014] In one possible implementation, the node determination module is used to determine the activity levels of the multiple system nodes based on the system operation log; determine the system nodes among the multiple system nodes whose activity levels are greater than or equal to an activity level threshold as the active system nodes; determine the baseline activity levels of the multiple system nodes based on the baseline activity information; and determine the system nodes among the multiple system nodes whose deviations from the baseline activity levels are greater than or equal to the deviation threshold as the suspicious system nodes.
[0015] In a possible implementation, the abnormality identification module is used to determine a first abnormality degree of the ERP system based on the usage feedback information; determine a second abnormality degree of the ERP system based on the transaction information, the multiple active system nodes and the multiple suspicious system nodes; determine a third abnormality degree of the ERP system based on a first node log of the multiple active system nodes on the target date and a second node log of the multiple suspicious system nodes on the target date, the first node log and the second node log belonging to the system operation log; determine whether the ERP system has an abnormality on the target date based on the first abnormality degree, the second abnormality degree and the third abnormality degree.
[0016] In a possible implementation, the abnormality identification module is configured to cluster multiple sub-feedback information in the usage feedback information to obtain multiple sub-feedback information groups, where one sub-feedback information group includes at least one sub-feedback information, and the similarity between the sub-feedback information in the same sub-feedback information group meets the similarity condition; perform feature extraction on at least one sub-feedback information in each sub-feedback information group to obtain a group feature corresponding to each of the sub-feedback information groups; determine a first reference abnormality degree for each of the sub-feedback information groups based on the group feature corresponding to each of the sub-feedback information groups; and fuse the first reference abnormality degrees of each of the sub-feedback information groups using a first weight of each of the sub-feedback information groups to obtain the first abnormality degree, where the first weight is positively correlated with the number of sub-feedback information in the corresponding sub-feedback information group.
[0017] In a possible implementation, the anomaly identification module is used to determine a plurality of reference active system nodes based on the transaction information, wherein the reference active system nodes are theoretical active system nodes when processing the target transaction; determine a second reference abnormality level of the ERP system based on the plurality of reference active system nodes and the plurality of active system nodes; determine a third reference abnormality level of the ERP system based on a deviation between the activity levels of the plurality of suspicious system nodes and a baseline activity level; and fuse the second reference abnormality level and the third reference abnormality level to obtain a second abnormality level of the ERP system.
[0018] In a possible implementation, the anomaly identification module is used to determine a first interaction anomaly degree and a first reception and transmission anomaly degree of each of the multiple active system nodes based on the first node logs of the multiple active system nodes on the target date; determine a second interaction anomaly degree and a second reception and transmission anomaly degree of each of the suspicious system nodes based on the second node logs of the multiple suspicious system nodes on the target date; and determine a third anomaly degree of the ERP system based on the first interaction anomaly degree and the first reception and transmission anomaly degree of each of the active system nodes, and the second interaction anomaly degree and the second reception and transmission anomaly degree of each of the suspicious system nodes.
[0019] In a possible implementation, the abnormality identification module is used to obtain, for any active system node among the multiple active system nodes, a data interaction log and a data sending and receiving log of the active system node from the first node log of the active system node; determine, based on the node type of the active system node and the data interaction log, a first interaction abnormality degree of the active system node; determine, based on the node type of the active system node and the data sending and receiving log, a first sending and receiving abnormality degree of the active system node; for any suspicious system node among the multiple suspicious system nodes, obtain the data interaction log and the data sending and receiving log of the suspicious system node from the second node log of the suspicious system node; determine, based on the node type of the suspicious system node and the data interaction log, a second interaction abnormality degree of the suspicious system node; determine, based on the node type of the suspicious system node and the data sending and receiving log, a second sending and receiving abnormality degree of the suspicious system node.
[0020] In a possible implementation, the anomaly identification module is used to fuse the first anomaly degree, the second anomaly degree and the third anomaly degree to obtain a target anomaly degree; when the target anomaly degree is greater than or equal to an anomaly degree threshold, it is determined that the ERP system has an anomaly on the target date; when the target anomaly degree is less than the anomaly degree threshold, it is determined that the ERP system has no anomaly on the target date.
[0021] In a possible implementation, the anomaly identification module is further used to determine, when an anomaly exists in the ERP system on the target date, at least one abnormal system node among the multiple system nodes based on the first node logs of the multiple active system nodes on the target date and the second node logs of the multiple suspicious system nodes on the target date; and determine the cause of the anomaly of the ERP system based on the at least one abnormal system node.
[0022] On the one hand, a computer device is provided, which includes one or more processors and one or more memories, wherein at least one computer program is stored in the one or more memories, and the computer program is loaded and executed by the one or more processors to implement the data collection and analysis method for the ERP system.
[0023] On the one hand, a computer-readable storage medium is provided, wherein at least one computer program is stored in the computer-readable storage medium, and the computer program is loaded and executed by a processor to implement the data collection and analysis method for an ERP system.
[0024] On the one hand, a computer program product or a computer program is provided, which includes a program code, and the program code is stored in a computer-readable storage medium. A processor of a computer device reads the program code from the computer-readable storage medium, and the processor executes the program code, so that the computer device executes the above-mentioned data collection and analysis method for an ERP system.
[0025] Through the technical solution provided in the embodiment of the present application, the system operation log of the ERP system on the target date, the transaction information of the target transaction corresponding to the ERP system, and the baseline active information of the ERP system are obtained, and the system operation log and the baseline active information are used to determine multiple active system nodes and multiple suspicious system nodes from multiple system nodes of the ERP system. Based on the usage feedback information, transaction information, multiple active system nodes, and multiple suspicious system nodes of the ERP system on the target date, it is determined whether the ERP system is abnormal on the target date, and the abnormality of the ERP system is identified. BRIEF DESCRIPTION OF THE DRAWINGS
[0026] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.
[0027] Figure 1 It is a schematic diagram of an implementation environment of a data collection and analysis method for an ERP system provided in an embodiment of the present application; Figure 2 This is a flow chart of a data collection and analysis method for an ERP system provided in an embodiment of the present application; Figure 3 is another flow chart of a data collection and analysis method for an ERP system provided in an embodiment of the present application; Figure 4 It is a structural diagram of a data collection and analysis system for an ERP system provided in an embodiment of the present application; Figure 5 It is a structural diagram of a server provided in an embodiment of the present application. DETAILED DESCRIPTION
[0028] In order to make the objectives, technical solutions and advantages of the present application clearer, the implementation methods of the present application will be further described in detail below with reference to the accompanying drawings.
[0029] In this application, the terms "first", "second", etc. are used to distinguish identical or similar items with basically the same effects and functions. It should be understood that there is no logical or temporal dependency between "first", "second", and "nth", nor is there any limitation on quantity and execution order.
[0030] ERP system: is an integrated software system designed to help companies manage their resources. It coordinates a variety of different business processes, such as supply chain management, procurement, inventory management, finance and human resources. By using an ERP system, companies can manage their resources more efficiently, thereby improving productivity and efficiency. In addition, ERP systems can also provide real-time data and analytical reports to help companies make smarter business decisions, achieve better quality control, reduce errors and waste, improve customer satisfaction and enhance corporate brand value.
[0031] System operation log: The log generated during the operation of the ERP system is used to reflect the operating status of the system.
[0032] System node: A software node in an ERP system. Different system nodes are responsible for performing different functions.
[0033] Transaction: also known as business. In the embodiment of the present application, all transactions that can be processed through ERP are called transactions. For example, transactions include supply management transactions, procurement transactions, inventory management transactions, and financial transactions.
[0034] Machine Learning (ML) is a multi-disciplinary subject involving probability theory, statistics, approximation theory, convex analysis, algorithm complexity theory and other disciplines. It specializes in studying how computers simulate or implement human learning behavior to acquire new knowledge or skills and reorganize existing knowledge sub-models to continuously improve their performance. Machine learning is the core of artificial intelligence and the fundamental way to make computers intelligent. Its applications are spread across all areas of artificial intelligence. Machine learning and deep learning usually include artificial neural networks, belief networks, reinforcement learning, transfer learning, inductive learning, and teaching learning.
[0035] Normalization: Mapping sequences with different value ranges to the interval (0, 1) makes data processing easier. In some cases, the normalized values can be directly implemented as probabilities.
[0036] Embedded Coding: Embedded coding mathematically represents a correspondence, that is, mapping data in X space to Y space through a function F, where the function F is an injective function, and the mapping result is structural preservation. The injective function indicates that the data after mapping is uniquely corresponding to the data before mapping, and structural preservation indicates that the size relationship of the data before mapping is the same as the size relationship of the data after mapping. For example, there are data X1 and X2 before mapping, and after mapping, Y1 corresponding to X1 and Y2 corresponding to X2 are obtained. If the data X1 before mapping is greater than X2, then correspondingly, the data Y1 after mapping is greater than Y2. For words, it means mapping the words to another space to facilitate subsequent machine learning and processing.
[0037] In the related art, enterprises rely on the capabilities of ERP systems, especially when facing the processing of some special affairs, the existence of ERP systems is crucial, so how to timely detect whether there are abnormalities in the ERP system is a problem that enterprises are concerned about. After adopting the technical solution provided in the embodiment of the present application, it is possible to timely detect whether there are abnormalities in the ERP system, so that when there are abnormalities, it is convenient for technical personnel to handle them in time, thereby improving the stability of the ERP system.
[0038] Figure 1 This is a schematic diagram of an implementation environment of a data collection and analysis method for an ERP system provided in an embodiment of the present application, see Figure 1 , the implementation environment may include a terminal 110 and a server 140.
[0039] The terminal 110 is connected to the server 140 via a wireless network or a wired network. Optionally, the terminal 110 is a smart phone, a tablet computer, a laptop computer, a desktop computer, etc., but is not limited thereto. The terminal 110 is installed and runs an application that supports data collection and analysis for the ERP system. In the embodiment of the present application, the terminal 110 is a terminal used by the maintenance personnel of the ERP system. The maintenance personnel can use the terminal 110 to maintain the ERP system. In addition, the terminal 110 can display the operating status of the ERP system.
[0040] The server 140 is an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, distribution networks (Content Delivery Network, CDN), and big data and artificial intelligence platforms. The ERP system runs on the server 140.
[0041] The following is a description of the data collection and analysis method for an ERP system provided in an embodiment of the present application. Figure 2 is a flow chart of a data collection and analysis method for an ERP system provided in an embodiment of the present application, see Figure 2 Taking the execution subject as a server as an example, the method includes the following steps.
[0042] 201. The server obtains a system operation log of an ERP system on a target date, transaction information of a target transaction corresponding to the ERP system, and baseline activity information of the ERP system. The ERP system includes multiple system nodes, and the baseline activity information is used to indicate baseline activity levels of the multiple system nodes.
[0043] Among them, the target date is the date on which the ERP system abnormality identification is to be performed, and the system operation log is used to record the operating status of the ERP system. The target transaction is the transaction processed using the ERP system on the target date, or refers to the transaction supported by the ERP system. Transaction information is used to describe the transaction status of the target transaction. The ERP system includes multiple system nodes, and different system nodes are responsible for different functions of the ERP system. In an embodiment of the present application, data can be exchanged between different system nodes. The baseline activity information is used to represent the baseline activity level of multiple system nodes, that is, the activity level of multiple system nodes under normal circumstances.
[0044] 202. The server determines, from the multiple system nodes, a plurality of active system nodes and a plurality of suspicious system nodes on the target date based on the system operation log and the baseline activity information, wherein the active system nodes are system nodes whose activity levels meet a first activity level condition, and the suspicious system nodes are system nodes whose activity levels meet a second activity level condition, and the baseline activity level is determined based on the baseline activity information.
[0045] Wherein, if the activity level meets the first activity level condition, it indicates that the activity level is high, and if the activity level meets the second activity level condition, it indicates that the activity level is abnormal. The first activity level condition and the second activity level condition are set by the technician according to the actual situation.
[0046] 203. The server determines whether the ERP system has an abnormality on the target date based on usage feedback information of the ERP system on the target date, the transaction information, the multiple active system nodes, and the multiple suspicious system nodes. The usage feedback information is feedback information submitted by an object that uses the ERP system on the target date.
[0047] The ERP system collects usage feedback information every day. The usage feedback information on the target date is the usage feedback information collected on the target date, which indicates the feedback of the object on the ERP system on the target date.
[0048] Through the technical solution provided in the embodiment of the present application, the system operation log of the ERP system on the target date, the transaction information of the target transaction corresponding to the ERP system, and the baseline active information of the ERP system are obtained, and the system operation log and the baseline active information are used to determine multiple active system nodes and multiple suspicious system nodes from multiple system nodes of the ERP system. Based on the usage feedback information, transaction information, multiple active system nodes, and multiple suspicious system nodes of the ERP system on the target date, it is determined whether the ERP system is abnormal on the target date, and the abnormality of the ERP system is identified.
[0049] The above steps 201-203 are a brief introduction to the data collection and analysis method for the ERP system provided in the embodiment of the present application. The data collection and analysis method for the ERP system provided in the embodiment of the present application will be more clearly explained below with reference to some examples. Figure 3 Taking the execution subject as a server as an example, the method includes the following steps.
[0050] 301. The server obtains the system operation log of the ERP system on the target date, the transaction information of the target transaction corresponding to the ERP system, and the baseline activity information of the ERP system, wherein the ERP system includes multiple system nodes, and the baseline activity information is used to indicate the baseline activity levels of the multiple system nodes.
[0051] Among them, the target date is the date on which the ERP system abnormality identification is to be performed, and the system operation log is used to record the operating status of the ERP system. The target transaction is the transaction processed using the ERP system on the target date, or refers to the transaction supported by the ERP system. Transaction information is used to describe the transaction status of the target transaction. The ERP system includes multiple system nodes, and different system nodes are responsible for different functions of the ERP system. In an embodiment of the present application, data can be exchanged between different system nodes. The baseline activity information is used to represent the baseline activity level of multiple system nodes, that is, the activity level of multiple system nodes under normal circumstances.
[0052] In a possible implementation, the server uses the target date to query in the log database to obtain the system operation log of the ERP system on the target date. In the case where the target transaction is a transaction processed by the ERP system on the target date, the server determines the target transaction from the system operation log, and uses the transaction identifier of the target transaction to query in the transaction database to obtain the transaction information of the target transaction. In the case where the target transaction is a transaction supported by the ERP system, the server queries in the transaction database based on the system identifier of the ERP system to obtain the target transaction and the transaction information of the target transaction. The server obtains the activity information of the ERP system on multiple historical dates before the target date. Based on the activity information on the multiple historical dates, the server determines the baseline activity information of the ERP system.
[0053] Among them, the number and selection method of historical dates are set by technical personnel according to actual conditions, and the embodiments of the present application do not limit this.
[0054] In order to explain the above implementation more clearly, the method of determining the benchmark active information of the ERP system based on the active information of the multiple historical dates in the above implementation is explained below.
[0055] In some embodiments, the active information includes data processing volume and data transceiver volume of multiple system nodes of the ERP system. The server determines the average of the data processing volume and data transceiver volume of multiple system nodes on multiple historical dates as the benchmark data processing volume and benchmark data transceiver volume of each system node. The server determines the combination of the benchmark data processing volume and benchmark data transceiver volume of each system node as the active information of each system node.
[0056] For any system node among the multiple system nodes, the data processing volume and data transmission and reception volume of the system node can represent the activity level of the system node.
[0057] 302. The server determines, from the multiple system nodes, a plurality of active system nodes and a plurality of suspicious system nodes on the target date based on the system operation log and the baseline activity information, wherein the active system nodes are system nodes whose activity levels meet a first activity level condition, and the suspicious system nodes are system nodes whose activity levels meet a second activity level condition, and the baseline activity level is determined based on the baseline activity information.
[0058] Wherein, if the activity level meets the first activity level condition, it indicates that the activity level is high, and if the activity level meets the second activity level condition, it indicates that the activity level is abnormal. The first activity level condition and the second activity level condition are set by the technician according to the actual situation.
[0059] In a possible implementation, the server determines the activity levels of the multiple system nodes based on the system operation log. The server determines the system nodes among the multiple system nodes whose activity levels are greater than or equal to the activity level threshold as the active system nodes. The server determines the baseline activity levels of the multiple system nodes based on the baseline activity information. The server determines the system nodes among the multiple system nodes whose deviations from the baseline activity levels are greater than or equal to the deviation threshold as the suspicious system nodes.
[0060] Among them, if the activity level is greater than or equal to the activity level threshold, it means that the activity level meets the first activity level condition; if the deviation between the activity level and the baseline activity level is greater than or equal to the deviation threshold, it means that the activity level meets the second activity level condition. The activity level threshold and the deviation threshold are set by technicians according to actual conditions, and the embodiments of the present application do not limit this.
[0061] In order to explain the above implementation more clearly, the above implementation will be explained in several parts below.
[0062] In the first part, the server determines the activity levels of the multiple system nodes based on the system operation log.
[0063] In a possible implementation, the server queries the system log to obtain the data transmission and reception volume and data processing volume of each system node. The server determines the activity level of each system node based on the data transmission and reception volume and data processing volume of each system node.
[0064] For example, the server uses the node identifier of each system node to query the system log to obtain the data transmission and reception volume and data processing volume of each system node. The server substitutes the data transmission and reception volume and data processing volume of each system node into the first relational data to obtain the activity level of each system node.
[0065] Among them, the first relationship data is a function used to express the correlation between the data transmission and reception volume, the data processing volume and the activity level. The first relationship data is set by technical personnel according to actual conditions, and the embodiment of the present application does not limit this.
[0066] In the second part, the server determines a system node whose activity level is greater than or equal to an activity level threshold among the multiple system nodes as the active system node.
[0067] In a possible implementation, the server compares the activity level of each system node with the activity level threshold, and determines a plurality of system nodes whose activity levels are greater than or equal to the activity level threshold as a plurality of active system nodes.
[0068] Part three: The server determines the baseline activity levels of the multiple system nodes based on the baseline activity information.
[0069] In a possible implementation, the server obtains the baseline data receiving / sending volume and the baseline data processing volume of each system node from the baseline activity information. The server determines the baseline activity level of each system node based on the baseline data receiving / sending volume and the baseline data processing volume of each system node.
[0070] For example, the server obtains the baseline data receiving and sending volume and the baseline data processing volume of each system node from the baseline activity information. The server substitutes the baseline data receiving and sending volume and the baseline data processing volume of each system node into the first relational data to obtain the baseline activity level of each system node.
[0071] Part 4: The server determines, as the suspicious system node, a system node among the multiple system nodes whose activity level has a deviation from a benchmark activity level that is greater than or equal to a deviation threshold.
[0072] In a possible implementation, the server subtracts the activity level of each system node from the benchmark activity level to obtain a deviation, and determines the system node whose corresponding deviation is greater than or equal to the deviation threshold as a suspicious system node, thereby obtaining multiple suspicious system nodes.
[0073] 303. The server determines a first abnormality level of the ERP system based on usage feedback information of the ERP system on the target date, where the usage feedback information is feedback information submitted by a subject who uses the ERP system on the target date.
[0074] The ERP system collects usage feedback information every day, and the usage feedback information on the target date is the usage feedback information collected on the target date, indicating the feedback of the subject on the ERP system on the target date. The first abnormality degree is determined based on the usage feedback information, so the first abnormality degree can reflect whether the ERP system has abnormalities from the perspective of subject usage.
[0075] In a possible implementation, the server clusters multiple sub-feedback information in the usage feedback information to obtain multiple sub-feedback information groups, where one sub-feedback information group includes at least one sub-feedback information, and the similarity between the sub-feedback information in the same sub-feedback information group meets the similarity condition. The server extracts features from at least one sub-feedback information in each sub-feedback information group to obtain group features corresponding to each sub-feedback information group. The server determines a first reference abnormality degree for each sub-feedback information group based on the group features corresponding to each sub-feedback information group. The server fuses the first reference abnormality degrees of each sub-feedback information group using the first weights of each sub-feedback information group to obtain the first abnormality degree, and the first weight is positively correlated with the number of sub-feedback information in the corresponding sub-feedback information group.
[0076] Among them, a sub-feedback information is feedback information submitted by an object, or refers to feedback information for a function of the ERP system. Similarity meets the similarity condition means that the similarity is high, and the similarity condition is set by the technician according to the actual situation, and the embodiment of the application does not limit this.
[0077] For example, the server performs feature extraction on each sub-feedback information to obtain the sub-feedback information features of each sub-feedback information. The server clusters the multiple sub-feedback information based on the sub-feedback information features of each sub-feedback information to obtain multiple sub-feedback information groups, and the similarity between the sub-feedback information features of the sub-feedback information in the same sub-feedback information group is greater than or equal to the similarity threshold. The server fuses at least one sub-feedback information in each sub-feedback information group to obtain the group features corresponding to each sub-feedback information group. The server fully connects and normalizes the group features corresponding to each sub-feedback information group to determine the first reference abnormality degree of each sub-feedback information group. The server substitutes the number of sub-feedback information in each sub-feedback information group into the second relational data to obtain the first weight of each sub-feedback information group. The server fuses the first reference abnormality degree of each sub-feedback information group using the first weight of each sub-feedback information group to obtain the first abnormality degree.
[0078] Among them, the second relationship data is set by technical personnel according to actual conditions, and the embodiments of the present application do not limit this.
[0079] 304. The server determines a second abnormality level of the ERP system based on the transaction information, the multiple active system nodes, and the multiple suspicious system nodes.
[0080] In a possible implementation, the server determines a plurality of reference active system nodes based on the transaction information, where the reference active system nodes are theoretically active system nodes when processing the target transaction. The server determines a second reference abnormality level of the ERP system based on the plurality of reference active system nodes and the plurality of active system nodes. The server determines a third reference abnormality level of the ERP system based on the deviation between the activity level of the plurality of suspicious system nodes and the benchmark activity level. The server fuses the second reference abnormality level and the third reference abnormality level to obtain a second abnormality level of the ERP system.
[0081] In order to explain the above implementation more clearly, the above implementation will be explained in several parts below.
[0082] In the first part, the server determines multiple reference active system nodes based on the transaction information.
[0083] In a possible implementation, the server determines multiple reference system functions of the ERP system based on the transaction information. The server determines multiple reference active system nodes based on the multiple reference system functions, where one reference active system node corresponds to one reference system function. The transaction information can describe the system functions required to process / execute the target transaction. Therefore, the transaction information can be used to determine multiple reference system functions, which are the system functions required to process / execute the target transaction. The correspondence between system functions and system nodes is set by technicians according to actual conditions, and the embodiments of the present application do not limit this.
[0084] For example, the server performs content recognition on the transaction information to obtain function description information in the transaction information. The server determines the multiple reference system functions based on the function description information. The server uses the multiple reference system functions to perform a query to obtain the multiple reference active system nodes.
[0085] In the second part, the server determines a second reference abnormality degree of the ERP system based on the multiple reference active system nodes and the multiple active system nodes.
[0086] In a possible implementation, the server determines a second reference abnormality degree of the ERP system based on the overlap degree between the multiple reference active system nodes and the multiple active system nodes.
[0087] The second reference abnormality degree is negatively correlated with the overlap degree, that is, the lower the overlap degree, the higher the second reference abnormality degree; the higher the overlap degree, the lower the second reference abnormality degree. In some embodiments, the overlap degree is determined by the ratio of the number of the same system nodes among the multiple reference active system nodes and the multiple active system nodes to the number of the multiple reference active system nodes.
[0088] Part three: The server determines a third reference abnormality level of the ERP system based on the deviation between the activity levels of the multiple suspicious system nodes and the benchmark activity level.
[0089] In a possible implementation, the server determines an average value of deviations between the activity levels of the multiple suspicious system nodes and the benchmark activity level as a reference deviation. Based on the reference deviation, the server determines a third reference abnormality level of the ERP system.
[0090] The third reference abnormality degree is positively correlated with the reference deviation, that is, the larger the reference deviation is, the larger the third reference abnormality degree is; and the smaller the reference deviation is, the smaller the third reference abnormality degree is.
[0091] Part 4: The server merges the second reference abnormality level with the third reference abnormality level to obtain a second abnormality level of the ERP system.
[0092] In a possible implementation, the server uses the second weight and the third weight to perform weighted fusion on the second reference abnormality level and the third reference abnormality level to obtain the second abnormality level of the ERP system.
[0093] Among them, the second weight is the weight corresponding to the second reference abnormality level, and the third weight is the weight corresponding to the third reference abnormality level. The second weight and the third weight are set by technical personnel according to actual conditions, and the embodiments of the present application do not limit this.
[0094] 305. The server determines a third abnormality degree of the ERP system based on the first node logs of the multiple active system nodes on the target date and the second node logs of the multiple suspicious system nodes on the target date, wherein the first node log and the second node log belong to the system operation logs.
[0095] In a possible implementation, the server determines the first interaction abnormality level and the first transceiver abnormality level of each active system node based on the first node log of the multiple active system nodes on the target date. The server determines the second interaction abnormality level and the second transceiver abnormality level of each suspicious system node based on the second node log of the multiple suspicious system nodes on the target date. The server determines the third abnormality level of the ERP system based on the first interaction abnormality level and the first transceiver abnormality level of each active system node, and the second interaction abnormality level and the second transceiver abnormality level of each suspicious system node.
[0096] In order to explain the above implementation more clearly, the above implementation will be explained in several parts below.
[0097] In the first part, the server determines a first interaction abnormality degree and a first sending and receiving abnormality degree of each active system node based on the first node logs of the multiple active system nodes on the target date.
[0098] In a possible implementation, for any active system node among the multiple active system nodes, the server obtains a data interaction log and a data transceiving log of the active system node from a first node log of the active system node. The server determines a first interaction abnormality degree of the active system node based on a node type of the active system node and the data interaction log. The server determines a first transceiving abnormality degree of the active system node based on a node type of the active system node and the data transceiving log.
[0099] Among them, the data interaction log stores the data interaction information of the system nodes, and the data interaction information is used to record the data interaction between the system nodes and other system nodes; the data transmission and reception log stores the data transmission and reception information of the system nodes, and the data transmission and reception information is used to record the data transmission and reception of the system nodes.
[0100] For example, for any active system node among the multiple active system nodes, the server obtains the data interaction log and the data transceiving log of the active system node from the first node log of the active system node. The server obtains data interaction information from the data interaction log of the active system node, and obtains data transceiving information from the data transceiving log. The server determines the baseline data interaction information and baseline data transceiving information of the active system node based on the node type of the active system node. The server determines the first interaction abnormality degree and the first transceiving abnormality degree of the active system node based on the interaction difference information between the data interaction information and the baseline data interaction information of the active system node, and the transceiving difference information between the data transceiving information and the baseline data transceiving information.
[0101] The data interaction information includes the interaction frequency with other system nodes, and the interaction difference information includes the interaction frequency difference; the data transceiving information includes the data transceiving volume, and the transceiving difference information includes the data transceiving volume difference. The correspondence between the node type and the reference data interaction information and the reference data transceiving information is set by the technician according to the actual situation, and the embodiments of the present application do not limit this.
[0102] In order to explain the above example more clearly, the manner in which the server determines the first interaction abnormality degree and the first transceiver abnormality degree of the active system node in the above example is explained below.
[0103] In some embodiments, the server substitutes the interaction difference information between the data interaction information of the active system node and the reference data interaction information into the third relationship data to obtain the first interaction abnormality degree of the active system node. The server substitutes the transceiving difference information between the data transceiving information of the active system node and the reference data transceiving information into the fourth relationship data to obtain the first transceiving abnormality degree of the active system node.
[0104] Among them, the first interaction abnormality degree is positively correlated with the interaction difference information, that is, the larger the interaction difference information is, the larger the first interaction abnormality degree is; the smaller the interaction difference information is, the smaller the first interaction abnormality degree is. The first transceiver abnormality degree is positively correlated with the transceiver difference information, that is, the larger the transceiver difference information is, the larger the first transceiver abnormality degree is; the smaller the transceiver difference information is, the smaller the first transceiver abnormality degree is. The third relationship data and the fourth relationship data are set by technical personnel according to actual conditions, and the embodiments of the present application do not limit this.
[0105] In the second part, the server determines the second interaction abnormality degree and the second sending and receiving abnormality degree of each suspicious system node based on the second node logs of the multiple suspicious system nodes on the target date.
[0106] In a possible implementation, for any suspicious system node among the multiple suspicious system nodes, the server obtains the data interaction log and the data transceiving log of the suspicious system node from the second node log of the suspicious system node. The server determines the second interaction abnormality degree of the suspicious system node based on the node type of the suspicious system node and the data interaction log. The server determines the second transceiving abnormality degree of the suspicious system node based on the node type of the suspicious system node and the data transceiving log.
[0107] For example, for any suspicious system node among the multiple suspicious system nodes, the server obtains the data interaction log and the data transceiving log of the suspicious system node from the second node log of the suspicious system node. The server obtains the data interaction information from the data interaction log of the suspicious system node, and obtains the data transceiving information from the data transceiving log. The server determines the baseline data interaction information and baseline data transceiving information of the suspicious system node based on the node type of the suspicious system node. The server determines the second interaction abnormality degree and the second transceiving abnormality degree of the suspicious system node based on the interaction difference information between the data interaction information and the baseline data interaction information of the suspicious system node, and the transceiving difference information between the data transceiving information and the baseline data transceiving information.
[0108] The data interaction information includes the interaction frequency with other system nodes, and the interaction difference information includes the interaction frequency difference; the data transceiving information includes the data transceiving volume, and the transceiving difference information includes the data transceiving volume difference. The correspondence between the node type and the reference data interaction information and the reference data transceiving information is set by the technician according to the actual situation, and the embodiments of the present application do not limit this.
[0109] In order to explain the above example more clearly, the manner in which the server determines the second interaction abnormality degree and the second sending and receiving abnormality degree of the suspicious system node in the above example is explained below.
[0110] In some embodiments, the server substitutes the interaction difference information between the data interaction information of the suspicious system node and the reference data interaction information into the third relationship data to obtain the second interaction abnormality degree of the suspicious system node. The server substitutes the transceiver difference information between the data transceiver information of the suspicious system node and the reference data transceiver information into the fourth relationship data to obtain the second transceiver abnormality degree of the suspicious system node.
[0111] The second interaction abnormality degree is positively correlated with the interaction difference information, that is, the larger the interaction difference information is, the larger the second interaction abnormality degree is; the smaller the interaction difference information is, the smaller the second interaction abnormality degree is. The second transceiver abnormality degree is positively correlated with the transceiver difference information, that is, the larger the transceiver difference information is, the larger the second transceiver abnormality degree is; the smaller the transceiver difference information is, the smaller the second transceiver abnormality degree is.
[0112] Part three: The server determines the third abnormality level of the ERP system based on the first interaction abnormality level and the first reception and transmission abnormality level of each active system node, and the second interaction abnormality level and the second reception and transmission abnormality level of each suspicious system node.
[0113] In a possible implementation, the server determines the average value of the first interaction abnormality degree of each active system node as the active interaction abnormality degree of multiple active system nodes. The server determines the average value of the first transceiver abnormality degree of each active system node as the active transceiver abnormality degree of multiple active system nodes. The server determines the average value of the second interaction abnormality degree of each suspicious system node as the suspicious interaction abnormality degree of multiple suspicious system nodes. The server determines the average value of the second transceiver abnormality degree of each suspicious system node as the suspicious transceiver abnormality degree of multiple suspicious system nodes. The server performs weighted fusion on the active interaction abnormality degree, the active transceiver abnormality degree, the suspicious interaction abnormality degree, and the suspicious transceiver abnormality degree to obtain the third abnormality degree of the ERP system.
[0114] 306. The server determines whether the ERP system has an abnormality on the target date based on the first abnormality level, the second abnormality level, and the third abnormality level.
[0115] In a possible implementation, the server merges the first abnormality level, the second abnormality level, and the third abnormality level to obtain a target abnormality level. When the target abnormality level is greater than or equal to the abnormality level threshold, the server determines that the ERP system has an abnormality on the target date. When the target abnormality level is less than the abnormality level threshold, the server determines that the ERP system has no abnormality on the target date.
[0116] Among them, the abnormality threshold is set by technical personnel according to actual conditions, and the embodiments of the present application do not limit this.
[0117] In order to explain the above-mentioned embodiment more clearly, the following describes a method for fusing the first abnormality degree, the second abnormality degree and the third abnormality degree in the above-mentioned embodiment to obtain the target abnormality degree.
[0118] In some embodiments, the server determines the average of the first abnormality degree, the second abnormality degree, and the third abnormality degree as the target abnormality degree. Alternatively, the server uses a fourth weight, a fifth weight, and a sixth weight to fuse the first abnormality degree, the second abnormality degree, and the third abnormality degree to obtain the target abnormality degree, the fourth weight being the weight corresponding to the first abnormality degree, the fifth weight being the weight corresponding to the second abnormality degree, and the sixth weight being the weight corresponding to the third abnormality degree.
[0119] Among them, the fourth weight, the fifth weight and the sixth weight are set by technical personnel according to actual conditions, and the embodiments of the present application do not limit this.
[0120] Optionally, after step 306, the server can also execute steps 307 and 308 described below.
[0121] 307. When an abnormality exists in the ERP system on the target date, the server determines at least one abnormal system node among the multiple system nodes based on the first node logs of the multiple active system nodes on the target date and the second node logs of the multiple suspicious system nodes on the target date.
[0122] Among them, the abnormal system node is the system node that causes the ERP system to be abnormal.
[0123] In a possible implementation, when the ERP system has an abnormality on the target date, the server inputs the first node logs of the multiple active system nodes on the target date and the second node logs of the multiple suspicious system nodes on the target date into an abnormal system node identification model, and extracts features of the first node log and the second node log through the abnormal system node identification model to obtain first node log features and second node log features. Based on the first node log features and the second node log features, the server determines at least one abnormal system node from the multiple system nodes.
[0124] For example, when the ERP system has an anomaly on the target date, for any active system node among the multiple active system nodes, the server will input the first sub-log corresponding to the active system node in the first node log. The server inputs the first sub-log into the abnormal system node identification model, extracts features of the first sub-log through the abnormal system node identification model, and obtains the sub-log features of the first sub-log. The server fully connects and normalizes the sub-log features of the first sub-log through the abnormal system node identification model to obtain the abnormal value of the first sub-log. When the abnormal value is greater than or equal to the abnormal value threshold, the active system node is determined as an abnormal system node.
[0125] Among them, the abnormal value threshold is set by the technician according to the actual situation, and the embodiments of the present application do not limit this.
[0126] 308. The server determines the abnormal reason of the ERP system based on the at least one abnormal system node.
[0127] In a possible implementation, the server determines the abnormal cause of the ERP system based on the node type and node function of the at least one abnormal system node.
[0128] All the above optional technical solutions can be arbitrarily combined to form optional embodiments of the present application, which will not be described one by one here.
[0129] Through the technical solution provided in the embodiment of the present application, the system operation log of the ERP system on the target date, the transaction information of the target transaction corresponding to the ERP system, and the baseline active information of the ERP system are obtained, and the system operation log and the baseline active information are used to determine multiple active system nodes and multiple suspicious system nodes from multiple system nodes of the ERP system. Based on the usage feedback information, transaction information, multiple active system nodes, and multiple suspicious system nodes of the ERP system on the target date, it is determined whether the ERP system is abnormal on the target date, and the abnormality of the ERP system is identified.
[0130] Figure 4 This is a structural diagram of a data collection and analysis system for an ERP system provided in an embodiment of the present application, see Figure 4 , the system includes: An acquisition module 401 is used to acquire a system operation log of an ERP system on a target date, transaction information of a target transaction corresponding to the ERP system, and baseline activity information of the ERP system, wherein the ERP system includes a plurality of system nodes, and the baseline activity information is used to indicate the baseline activity levels of the plurality of system nodes. The target transaction is a transaction processed by using the ERP system on the target date; a node determination module 402, configured to determine, based on the system operation log and the baseline activity information, a plurality of active system nodes and a plurality of suspicious system nodes on the target date from the plurality of system nodes, wherein the active system nodes are system nodes whose activity levels meet a first activity level condition, and the suspicious system nodes are system nodes whose activity levels meet a second activity level condition, and the baseline activity level is determined based on the baseline activity information; The anomaly identification module 403 is used to determine whether the ERP system has an anomaly on the target date based on the usage feedback information of the ERP system on the target date, the transaction information, the multiple active system nodes and the multiple suspicious system nodes. The usage feedback information is feedback information submitted by the object using the ERP system on the target date.
[0131] In one possible implementation, the node determination module 402 is used to determine the activity level of the multiple system nodes based on the system operation log; determine the system node among the multiple system nodes whose activity level is greater than or equal to an activity level threshold as the active system node; determine the baseline activity level of the multiple system nodes based on the baseline activity information; and determine the system node among the multiple system nodes whose deviation between the activity level and the baseline activity level is greater than or equal to the deviation threshold as the suspicious system node.
[0132] In a possible implementation, the anomaly identification module 403 is used to determine a first degree of anomaly of the ERP system based on the usage feedback information; determine a second degree of anomaly of the ERP system based on the transaction information, the multiple active system nodes, and the multiple suspicious system nodes; determine a third degree of anomaly of the ERP system based on a first node log of the multiple active system nodes on the target date and a second node log of the multiple suspicious system nodes on the target date, the first node log and the second node log belonging to the system operation log; determine whether the ERP system has an anomaly on the target date based on the first degree of anomaly, the second degree of anomaly, and the third degree of anomaly.
[0133] In a possible implementation, the abnormality identification module 403 is configured to cluster multiple sub-feedback information in the usage feedback information to obtain multiple sub-feedback information groups, where one sub-feedback information group includes at least one sub-feedback information, and the similarity between the sub-feedback information in the same sub-feedback information group meets the similarity condition; perform feature extraction on at least one sub-feedback information in each sub-feedback information group to obtain a group feature corresponding to each sub-feedback information group; determine a first reference abnormality degree for each sub-feedback information group based on the group feature corresponding to each sub-feedback information group; and fuse the first reference abnormality degrees for each sub-feedback information group using a first weight for each sub-feedback information group to obtain the first abnormality degree, where the first weight is positively correlated with the number of sub-feedback information in the corresponding sub-feedback information group.
[0134] In a possible implementation, the anomaly identification module 403 is used to determine a plurality of reference active system nodes based on the transaction information, where the reference active system nodes are theoretical active system nodes when processing the target transaction; determine a second reference abnormality level of the ERP system based on the plurality of reference active system nodes and the plurality of active system nodes; determine a third reference abnormality level of the ERP system based on a deviation between the activity levels of the plurality of suspicious system nodes and a baseline activity level; and fuse the second reference abnormality level and the third reference abnormality level to obtain a second abnormality level of the ERP system.
[0135] In a possible implementation, the anomaly identification module 403 is used to determine a first interaction anomaly degree and a first reception and transmission anomaly degree of each active system node based on the first node logs of the multiple active system nodes on the target date; determine a second interaction anomaly degree and a second reception and transmission anomaly degree of each suspicious system node based on the second node logs of the multiple suspicious system nodes on the target date; and determine a third anomaly degree of the ERP system based on the first interaction anomaly degree and the first reception and transmission anomaly degree of each active system node and the second interaction anomaly degree and the second reception and transmission anomaly degree of each suspicious system node.
[0136] In a possible implementation, the anomaly identification module 403 is used to obtain, for any active system node among the multiple active system nodes, a data interaction log and a data sending and receiving log of the active system node from the first node log of the active system node; determine, based on the node type of the active system node and the data interaction log, a first interaction anomaly degree of the active system node; determine, based on the node type of the active system node and the data sending and receiving log, a first sending and receiving anomaly degree of the active system node; for any suspicious system node among the multiple suspicious system nodes, obtain the data interaction log and the data sending and receiving log of the suspicious system node from the second node log of the suspicious system node; determine, based on the node type of the suspicious system node and the data interaction log, a second interaction anomaly degree of the suspicious system node; determine, based on the node type of the suspicious system node and the data sending and receiving log, a second sending and receiving anomaly degree of the suspicious system node.
[0137] In a possible implementation, the anomaly identification module 403 is used to fuse the first anomaly degree, the second anomaly degree, and the third anomaly degree to obtain a target anomaly degree; when the target anomaly degree is greater than or equal to an anomaly degree threshold, it is determined that the ERP system has an anomaly on the target date; when the target anomaly degree is less than the anomaly degree threshold, it is determined that the ERP system has no anomaly on the target date.
[0138] In a possible implementation, the anomaly identification module 403 is also used to determine, when an anomaly exists in the ERP system on the target date, at least one abnormal system node among the multiple system nodes based on the first node logs of the multiple active system nodes on the target date and the second node logs of the multiple suspicious system nodes on the target date; and determine the cause of the anomaly of the ERP system based on the at least one abnormal system node.
[0139] It should be noted that: the data collection and analysis system for the ERP system provided in the above embodiment only uses the division of the above functional modules as an example when collecting and analyzing the data of the ERP system. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the computer device is divided into different functional modules to complete all or part of the functions described above. In addition, the data collection and analysis system for the ERP system provided in the above embodiment and the data collection and analysis method embodiment for the ERP system belong to the same concept. The specific implementation process is detailed in the method embodiment and will not be repeated here.
[0140] Through the technical solution provided in the embodiment of the present application, the system operation log of the ERP system on the target date, the transaction information of the target transaction corresponding to the ERP system, and the baseline active information of the ERP system are obtained, and the system operation log and the baseline active information are used to determine multiple active system nodes and multiple suspicious system nodes from multiple system nodes of the ERP system. Based on the usage feedback information, transaction information, multiple active system nodes, and multiple suspicious system nodes of the ERP system on the target date, it is determined whether the ERP system is abnormal on the target date, and the abnormality of the ERP system is identified.
[0141] Figure 5 This is a schematic diagram of the structure of a server provided in an embodiment of the present application. The server 500 may have relatively large differences due to different configurations or performances, and may include one or more processors (Central Processing Units, CPU) 501 and one or more memories 502, wherein at least one computer program is stored in the one or more memories 502, and the at least one computer program is loaded and executed by the one or more processors 501 to implement the methods provided in the above-mentioned various method embodiments. Of course, the server 500 may also have components such as a wired or wireless network interface, a keyboard, and an input and output interface for input and output. The server 500 may also include other components for implementing device functions, which will not be described in detail here.
[0142] In an exemplary embodiment, a computer-readable storage medium is also provided, such as a memory including a computer program, and the computer program can be executed by a processor to complete the data collection and analysis method for the ERP system in the above embodiment. For example, the computer-readable storage medium can be a read-only memory (ROM), a random access memory (RAM), a compact disc read-only memory (CD-ROM), a magnetic tape, a floppy disk, an optical data storage device, etc.
[0143] In an exemplary embodiment, a computer program product or a computer program is also provided, which includes a program code, and the program code is stored in a computer-readable storage medium. A processor of a computer device reads the program code from the computer-readable storage medium, and the processor executes the program code, so that the computer device executes the above-mentioned data collection and analysis method for an ERP system.
[0144] In some embodiments, the computer program involved in the embodiments of the present application may be deployed and executed on a computer device, or on multiple computer devices located at one location, or on multiple computer devices distributed at multiple locations and interconnected by a communication network. Multiple computer devices distributed at multiple locations and interconnected by a communication network may constitute a blockchain system.
[0145] A person skilled in the art will understand that all or part of the steps to implement the above embodiments may be accomplished by hardware or by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, and the above-mentioned storage medium may be a read-only memory, a disk or an optical disk, etc.
[0146] The above are only optional embodiments of the present application and are not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application should be included in the protection scope of the present application.
Claims
1. A data collection and analysis method for an ERP system, characterized in that: The method comprises: Acquire a system operation log of an ERP system on a target date, transaction information of a target transaction corresponding to the ERP system, and baseline activity information of the ERP system, wherein the ERP system includes a plurality of system nodes, and the baseline activity information is used to indicate baseline activity levels of the plurality of system nodes; Based on the system operation log and the baseline activity information, determining a plurality of active system nodes and a plurality of suspicious system nodes on the target date from the plurality of system nodes, the active system nodes being system nodes whose activity levels meet a first activity level condition, the suspicious system nodes being system nodes whose activity levels meet a second activity level condition, and the baseline activity level being determined based on the baseline activity information; Based on usage feedback information of the ERP system on the target date, the transaction information, the multiple active system nodes and the multiple suspicious system nodes, it is determined whether the ERP system has an abnormality on the target date, wherein the usage feedback information is feedback information submitted by an object using the ERP system on the target date.
2. The method according to claim 1, characterized in that The determining, based on the system operation log and the benchmark activity information, a plurality of active system nodes and a plurality of suspicious system nodes on the target date from the plurality of system nodes comprises: Determining the activity levels of the plurality of system nodes based on the system operation log; Determine a system node whose activity level is greater than or equal to an activity level threshold among the multiple system nodes as the active system node; Determining baseline activity levels of the plurality of system nodes based on the baseline activity information; A system node among the multiple system nodes, the deviation between the activity level and the benchmark activity level of which is greater than or equal to the deviation threshold, is determined as the suspicious system node.
3. The method according to claim 1, characterized in that The determining whether the ERP system has an abnormality on the target date based on the usage feedback information of the ERP system on the target date, the transaction information, the multiple active system nodes, and the multiple suspicious system nodes includes: Determining a first abnormality level of the ERP system based on the usage feedback information; determining a second abnormality degree of the ERP system based on the transaction information, the plurality of active system nodes, and the plurality of suspicious system nodes; determining a third abnormality degree of the ERP system based on first node logs of the plurality of active system nodes on the target date and second node logs of the plurality of suspicious system nodes on the target date, the first node log and the second node log belonging to the system operation log; Based on the first abnormality level, the second abnormality level, and the third abnormality level, it is determined whether the ERP system has an abnormality on the target date.
4. The method according to claim 3, characterized in that The determining, based on the usage feedback information, a first abnormality degree of the ERP system includes: Clustering the plurality of sub-feedback information in the usage feedback information to obtain a plurality of sub-feedback information groups, wherein a sub-feedback information group includes at least one sub-feedback information, and similarities between the sub-feedback information in the same sub-feedback information group meet a similarity condition; Performing feature extraction on at least one sub-feedback information in each sub-feedback information group to obtain group features corresponding to each of the sub-feedback information groups; Determining a first reference abnormality degree of each sub-feedback information group based on a group feature corresponding to each of the sub-feedback information groups; The first reference abnormality degree of each sub-feedback information group is merged by using the first weight of each sub-feedback information group to obtain the first abnormality degree, and the first weight is positively correlated with the amount of sub-feedback information in the corresponding sub-feedback information group.
5. The method according to claim 3, characterized in that: The determining, based on the transaction information, the plurality of active system nodes, and the plurality of suspicious system nodes, a second abnormality degree of the ERP system includes: Based on the transaction information, determining a plurality of reference active system nodes, wherein the reference active system nodes are theoretically active system nodes when processing the target transaction; determining a second reference abnormality degree of the ERP system based on the plurality of reference active system nodes and the plurality of active system nodes; determining a third reference abnormality level of the ERP system based on deviations between the activity levels of the plurality of suspicious system nodes and a baseline activity level; The second reference abnormality level and the third reference abnormality level are merged to obtain a second abnormality level of the ERP system.
6. The method according to claim 3, characterized in that The determining the third abnormality degree of the ERP system based on the first node logs of the plurality of active system nodes on the target date and the second node logs of the plurality of suspicious system nodes on the target date comprises: Determine, based on the first node logs of the plurality of active system nodes on the target date, a first interaction abnormality degree and a first sending and receiving abnormality degree of each of the active system nodes; Determine, based on the second node logs of the plurality of suspicious system nodes on the target date, a second interaction abnormality degree and a second sending and receiving abnormality degree of each of the suspicious system nodes; Based on the first interaction abnormality degree and the first transceiver abnormality degree of each of the active system nodes, and the second interaction abnormality degree and the second transceiver abnormality degree of each of the suspicious system nodes, a third abnormality degree of the ERP system is determined.
7. The method according to claim 6, characterized in that The determining, based on the first node logs of the plurality of active system nodes on the target date, a first interaction abnormality degree and a first transceiver abnormality degree of each of the active system nodes comprises: For any active system node among the multiple active system nodes, obtaining a data interaction log and a data transceiving log of the active system node from a first node log of the active system node; determining a first interaction abnormality degree of the active system node based on a node type of the active system node and the data interaction log; determining a first transceiving abnormality degree of the active system node based on a node type of the active system node and the data transceiving log; The determining, based on the second node logs of the plurality of suspicious system nodes on the target date, the second interaction abnormality degree and the second sending and receiving abnormality degree of each of the suspicious system nodes comprises: For any suspicious system node among the multiple suspicious system nodes, obtain the data interaction log and the data sending and receiving log of the suspicious system node from the second node log of the suspicious system node; determine the second interaction abnormality degree of the suspicious system node based on the node type of the suspicious system node and the data sending and receiving log; determine the second sending and receiving abnormality degree of the suspicious system node based on the node type of the suspicious system node and the data sending and receiving log.
8. The method according to claim 3, characterized in that The determining whether the ERP system has an abnormality on the target date based on the first abnormality degree, the second abnormality degree, and the third abnormality degree includes: Merging the first abnormality degree, the second abnormality degree, and the third abnormality degree to obtain a target abnormality degree; When the target abnormality level is greater than or equal to the abnormality level threshold, determining that the ERP system has an abnormality on the target date; When the target abnormality level is less than the abnormality level threshold, it is determined that there is no abnormality in the ERP system on the target date.
9. The method according to claim 1, characterized in that: After determining whether the ERP system has an abnormality on the target date based on the usage feedback information of the ERP system on the target date, the transaction information, the multiple active system nodes, and the multiple suspicious system nodes, the method further includes: In the case that the ERP system has an abnormality on the target date, determining at least one abnormal system node among the multiple system nodes based on first node logs of the multiple active system nodes on the target date and second node logs of the multiple suspicious system nodes on the target date; Based on the at least one abnormal system node, a cause of the abnormality of the ERP system is determined.
10. A data collection and analysis system for an ERP system, characterized in that: an acquisition module, configured to acquire a system operation log of an ERP system on a target date, transaction information of a target transaction corresponding to the ERP system, and baseline activity information of the ERP system, wherein the ERP system includes a plurality of system nodes, the baseline activity information is used to indicate a baseline activity level of the plurality of system nodes, and the target transaction is a transaction processed using the ERP system on the target date; a node determination module, configured to determine, based on the system operation log and the baseline activity information, a plurality of active system nodes and a plurality of suspicious system nodes on the target date from the plurality of system nodes, wherein the active system nodes are system nodes whose activity levels meet a first activity level condition, and the suspicious system nodes are system nodes whose activity levels meet a second activity level condition, and the baseline activity level is determined based on the baseline activity information; an abnormality identification module, for determining whether the ERP system has an abnormality on the target date based on usage feedback information of the ERP system on the target date, the transaction information, the multiple active system nodes, and the multiple suspicious system nodes, wherein the usage feedback information is feedback information submitted by an object using the ERP system on the target date.
Citation Information
Patent Citations
Log monitoring alarm method and device, computer equipment and storage medium
CN117194191A
Systems and methods for protecting against malware attacks
US20210334374A1
Cluster anomaly detection method, apparatus, and related device
WO2022257421A1
Communication link detection method and apparatus
WO2023236874A1
KR20240129317A
